METAMORPHISM AND METAMORPHIC BELTS

Other Geology titles:

Read: RUTLEY'S ELEMENTS OF MINERALOGY

Winkler: PETROGENESIS OF METAMORPHIC ROCKS
Hatch, Wells and Wells: THE PETROLOGY OF THE IGNEOUS ROCKS
Greensmith, Hatch and Rastall: PETROLOGY OF THE SEDIMENTARY ROCKS
Allen: PHYSICAL PROCESSES OF SEDIMENTATION
Metamorphism and
Metamorphic Belts

By

AKIHO MIYASHIRO
Professor of Geology, State University of New York at Albany

London
GEORGE ALLEN & UNWIN
Boston Sydney
First published in Great Britain in 1973
Second impression 1975
Third impression 1978

This book is copyright under the Berne Convention. All rights reserved. Apart from
any fair dealing for the purpose of private study, research, criticism or review, as
permitted under the Copyright Act, 1956, no part of this publication may be
reproduced, stored in a retrieval system, or transmitted, in any form or by any
means, electronic, electrical, chemical, mechanical, optical, photocopying, recording
or otherwise, without the prior permission of the copyright owner. Enquiries should
be addressed to the publishers.
This translation © Akiho Miyashiro, 1973

Softcover reprint of the hardcover I st edition 1973

ISBN-13: 978-0-04-550026-0 e-ISBN-13: 978-94-011-6836-6

DOl: 10.1007/ 978-94-011-6836-6

Original Japanese language edition published by Iwanami Shoten, Publishers, Tokyo

Henseigan To Henseita © Akiho Miyashiro, 1965

The Gresham Press, Old Woking, Surrey

Contents

page
Preface 13
Acknowledgments for figures 15
PART I BASIS OF METAMORPHIC PETROLOGY
1 Introduction 19
1-1 Concept of metamorphism 19
1-2 Geologic classification of metamorphism 22
1-3 Compositional groups of metamorphic rocks 28
1-4 Nomenclature of metamorphic rock types 30
1-5 Distribution of metamorphic rocks 34
2 Basic Characteristics of MetamorphiC Reactions 38
2-1 Aim of this chapter and units of measurement 38
2-2 Solid-solid reactions 39
2-3 Open system and models for the behaviour of H2 0 42
2-4 Dehydration reactions 45
2-5 Mineral assemblages and element distribution 48
2-6 Decarbonation reactions 50
2-7 Oxidation and reduction in metamorphism 51
3 Basic Concepts of Metamorphic Petrology S6
3-1 Progressive regional metamorphism 56
3-2 Outline of metamorphic facies 66
3-3 Baric types and paired metamorphic belts 71
3-4 Baric types and geothermal gradients 85
3-5 Temperature and pressure corresponding to individual meta-
morphic facies 88
4 Some Important Problems in Metamorphic Geology 93
4-1 Igneous rock associations related to regional metamorphism 93
4-2 Volcanic arcs in relation to low-pressure regional meta-
morphism 100
4-3 Metamorphic facies and geologic age 106
4-4 Diversity in low-temperature metamorphism 112
4-5 Nature of contact metamorphism 115
S Diagrammatic Representation of Mineral Parageneses 120
5-1 Mineralogical phase rules and composition-paragenesis dia-
grams 120
8/METAMORPHISM AND METAMORPHIC BELTS
5-2 Eskola'sACF andA'KF diagrams 124
5-3 AFM diagrams for metapelites 130
5-4 Schreinemakers bundle 133

PART II PROGRESSIVE METAMORPHISM

6A Zeolite and Prehnite-PumpeUyite Facies Metamorphism: Its Diversity 139
6A-l Classification of burial metamorphic terranes in terms of
geothermal gradients 139
6A-2 Zeolite and prehnite-pumpellyite facies terranes of the low-
pressure type 140
6A-3 Zeolite distribution in active geothermal fields 147
6A-4 Zeolite and prehnite-pumpellyite facies terranes of the
medium-pressure type 149
6A-5 Relevant data from high-pressure metamorphic terranes 152
6A-6 Other areas 153
6B Zeolite and Prehnite-Pumpellyite Facies Metamorphism: Progressive
Mineral Changes 156
6B-l Progressive changes in zeolite assemblages 156
6B-2 Experimental studies on the stabilities of zeolites 159
60-3 Smectite, illite and mixed-layer clay minerals 163
60-4 Prehnite and pumpellyite 164
60-5 Opal, chalcedony and quartz 165
7A Metapelites: Diversity in Progressive Regional Metamorphism 167
7A-l Nature of the diversity 167
7A-2 Low-pressure metapelites in the Ryoke belt and Abukuma
Plateau, Japan 167
7A-3 Low-pressure metapelites in Australia 174
7A-4 Low-pressure metapelites in France and Spain 175
7A-5 Low-pressure metapelites in North America 180
7A-6 Medium-pressure metapelites in the Scottish Highlands 185
7A-7 Medium-pressure metapelites in the Northern Appalachians 188
7A-8 High-pressure metapelites in the Alps 189
7A-9 High-pressure metapelites in the Sanbagawa belt, Japan 191
7A-I0 High-pressure metaclastics in the California Coast Ranges 193
7B Metapelites: Progressive Mineral Changes 198
70-1 Order of discussions 198
70-2 Disappearance of clay minerals 198
70-3 Paragonite 199
70-4 Muscovite 201
70-5 Chloritoid 205
70-6 Chlorite 206
 CONTENTS/9
7B-7 Formation of biotite 208
7B-8 Plagioclase 210
7B-9 Cordierite 211
7B-1O Pyralspite garnets 214
7B-11 Staurolite 222
7B-12 Formation of andalusite and kyanite 223
7B-13 Phase relations of andalusite, kyanite and sillimanite 224
7B-14 Microcline to orthoclase 225
7B-15 Breakdown of muscovite 226
7B-16 Partial melting 226
7B-17 Breakdown of biotite 227
7B-18 AFM diagrams for progressive metamorphism of pelites 228
7B -19 Lawsonite and aragonite 228
7B-20 Formation of jadeite 230
7B-21 Oxides and sulphides of iron 232
7B-22 Organic material and graphite 232

8A Metabasites: Diversity in Progressive Regional Metamorphism 234

8A·l Mineralogical sensitivity of metabasites 234
8A-2 Low·pressure metabasites in the Ryoke belt and Abukuma
Plateau, Japan 234
8A-3 Low·pressure metabasites in Spain 238
8AA Low-pressure metabasites in Michigan 239
8A-5 Medium-pressure metabasites in the Scottish Highlands and
Norway. 239
8A-6 Medium·pressure metabasites in the Northern Appalachians 241
8A-7 Metabasites in passage from the amphibolite to the granulite
facies 242
8A-8 High-pressure metabasites in the Sanbagawa belt, Japan 244
8A-9 High-pressure metabasites of the California Coast Rallges 246

8B Metabasites: Progressive Mineral Changes 248

8B-l Chlorite 248
8B-2 Epidote and zoisite 248
8B-3 Plagioclase 249
8BA Calcic and sub calcic amphiboles 251
8B-5 Cummingtonite, anthophyllite and gedrite 258
8B-6 Muscovite and biotite 258
8B-7 Almandine garnet 259
8B-8 Calcic clinopyroxene 260
8B-9 Orthopyroxene 260
8B-1O Glaucophane 262
la/METAMORPHISM AND METAMORPHIC BELTS
8B-ll Stilpnomelane 263
8B-12 Oxides and sulphides of iron 264
9 Progressive Regional Metamorphism of Limestones 266
9-1 Compositions of limestone and fluids 266
9-2 Limestones in low-pressure regional metamorphism 267
9-3 Limestones in medium-pressure regional metamorphism 268
9-4 Limestones in high-pressure regional metamorphism 269
9-5 Progressive metamorphic reactions in siliceous dolomitic
limestones 270
9-6 Metamorphic reactions in alumina-containing siliceous lime-
stones 274
10 Progressive Contact Metamorphism and Pyrometamorphism 277
10-1 Contact metamorphism of pelites to produce andalusite as
the only A1 2 SiO s mineral 277
10-2 Contact metamorphism of pelites to produce sillimanite with
or without andalusite 279
10-3 Contact metamorphism of pelites to produce staurolite and
kyanite 282
10-4 Possible example of partial melting in a contact aureole 285
10-5 Pyrometamorphism of pelitic and psammitic rocks 286
10-6 Contact metamorphism of basic rocks 288
10-7 Contact metamorphism of limestones 290
11 Metamorphic Facies and Facies Series 293
11-1 Development of the concept of metamorphic facies 293
11-2 Metamorphic facies of the low-pressure type 299
11-3 Metamorphic facies of the medium-pressure type 304
11-4 Metamorphic facies of the high-pressure type 308
12 Eclogites and the Eclogite Facies 310
12-1 Eclogites and Eskola's concept of the eclogite facies 310
12-2 Relation between the granulite and the eclogite facies 311
12-3 Relation between the eclogite and hydrous metamorphic
facies 316
12-4 Eclogite inclusions 320
12-5 Granites and metapelites in the eclogite facies 324
PART III METAMORPHISM AND CRUSTAL EVOLUTION
13 Metamorphic Belts of North America 327
13-1 Constitution of North America 327
13-2 Metamorphism in the Canadian Shield 327
13-3· The metamorphic belt of the Appalachians 330
13-4 Metamorphic belts in the Cordilleran Mountains 332
 CONTENTS/ll
14 Metamorphic Belts of Europe 338
14-1 Constitution of Europe 338
14-2 Metamorphism in the Baltic Shield 338
14-3 Metamorphic belts in the Caledonides 340
14-4 Metamorphic belts in the west European Hercynides 342
14-5 Metamorphic belts in the west European Alpides 343
14-6 Correlation and basements of metamorphic belts in the
Atlantic regions 345
15 Metamorphic Belts in Japan and its Environs 348
15-1 Present-day arcs of the Japanese Islands 348
15-2 The Asiatic Continent and the Japanese Islands 354
15-3 Paired metamorphic belts 356
15-4 Hida metamorphic complex 360
15-5 High-pressure metamorphic terranes older than the Sangun
belt 361
15-6 Sangun metamorphic belt 362
15-7 Ryoke metamorphic belt and Abukuma Plateau 363
15-8 Sanbagawa metamorphic belt 369
15-9 Metamorphism of the Shimanto terrane and the Mizuho
orogenic belt 371
15-10 Summary of the chronology and distribution of metamorphic
belts in the Honshu Arc 373
15-11 Hidaka and Kamuikotan metamorphic belts in Hokkaido 375
15-12 Metamorphic belts of the Ryukyu Arc and Taiwan 375
16 Metamorphic Belts in the Southwest Pacific 379
16-1 Characteristics of the Southwest Pacific regions 379
16-2 The Australian Continent 379
16-3 New Zealand 382
16-4 Celebes 386
16-5 New Caledonia 388
16-6 Yap Islands 388
17 Tectonics of Regional Metamorphic Belts 390
17-1 Tectonic classification of metamorphic belts 390
17-2 Variation in the rate of plate convergence and the relation
between paired and unpaired metamorphic belts 391
17-3 Summary of geologic features of metamorphic belts 394
17-4 Tectonic significance of medium-pressure regional metamor-
phism 398
17-5 Temperature distribution in orogenic belts 398
17-6 Major faults, slab surfaces, a.ld the uplift of high-pressure
metamorphiC complexes 400
12/METAMORPHISM AND METAMORPHIC BELTS
17-7 Tectonics of island arcs and paired metamorphic belts in
the main part of Japan 404
18 Metamorphic Structure of Continental Crusts 409
18-1 Constitution of continents 409
18-2 Precambrian shields 410
18-3 Vertical differentiation of continental crusts 412
18-4 Phanerozoic orogenic belts 414
19 Ocean-floor Metamorphism and its Significance 416
19-1 Hard rocks of deep ocean floors 416
19-2 Metabasalts and metagabbros from deep ocean floors 417
19-3 Significance of ocean-floor metamorphism 419
20 Cataclastic Metamorphism along Transform Faults 421
20-1 Classification and field relations of catac1astic metamorphic
rocks 421
20-2 The San Andreas fault zone in California 423
20-3 The Alpine fault of N~w Zealand 424
20-4 Catac1astic rocks along transform faults in ocean floors 425
Appendix History of the Study of Metamorphism 427
References 441
Index 481
Preface

My book Metamorphic Rocks and Metamorphic Belts (in Japanese) was

published by Iwanami Shoten, Publishers, in Tokyo in 1965. A few years later,
Mr D. Lynch-Blosse of George Allen & Unwin Ltd contacted me to explore the
possibility of translating it into English. Thus, translation accompanied by
rewriting of substantial parts of the book was made in subsequent years,
resulting in the present book Metamorphism and Metamorphic Belts. This title
was chosen to emphasize the tectonic Significance of metamorphic belts.
Metamorphic geology has a long history. The microscopic description and
classification of metamorphic rocks began in the late nineteenth century. The
theory of equilibrium mineral assemblages began in the first half of the
twentieth century. Detailed mineralogical studies and the experimental
determination of the pressure-temperature conditions of metamorphism began in
the 1950s. The importance of metamorphic petrology in our understanding of
the tectonic processes has been realized only in the past decade. This book is
intended to synthesize the mineralogic, petrologic" and tectonic aspects of
metamorphism. Advanced treatment of the thermodynamic and structural
aspects is not intended.
Part I of this book treats the basic concepts of metamorphic petrology and
geology. Part II deals mainly with the progressive mineral changes and their
diversity in regional metamorphism. Part III deals with the tectonic aspects of
regional, ocean-floor and transform-fault metamorphism in relation to the
evolution of the crust and lithosphere. A short history of metamorphic geology
is given in the Appendix to help readers understand the historical background of
concepts and terms and to emphasize the great contribution of the
Finno-Scandinavian school of metamorphic petrology in the first half of the
twentieth century.
From the tectonic viewpoint, regional and ocean-floor metamorphism are the
most important of all categories of metamorphism. Abundant data are available
for regional metamorphism. Hence, the mineralogy, metamorphic facies, facies
series, and pressure-temperature and tectonic conditions of regional meta-
morphism are discussed in detail. Available data, though they are not abundant, of
ocean-floor, transform-fault and contact metamorphism are reviewed.
In 1960, the hardworking geologist could read virtually all the papers in
metamorphic geology published in the world. At present, probably no one would
be able to do so, because the amount of publications and the diversification of
their contents have so drastically increased in the past decade. About 800 papers
14/METAMORPHISM AND METAMORPHIC BELTS

and books are listed toward the end of this book. Approximately 1 per cent of
them were published in the period 1795-1899, 51 per cent in the period
1900-65, and 48 per cent in 1966-72 (i.e. in the 7 years since the publication of
the Japanese edition). These figures will give an idea about the recent increase of
publications and the extent of rewriting made for this edition. No doubt, many
important publications escaped the attention the}l deserve.
The researches made by my former co-workers in Japan have played an
essential role in this as well as in the Japanese edition. Among them Drs Yotaro
Seki, Fumiko Shido, Shohei Banno, Masao Iwasaki, Mitsuo Hashimoto, Tatsujiro
Uno, Shunro Ueta and Akira Ono made particularly important contributions,
which helped me formulate some of the basic ideas presented in this book.
Some differences between the Japanese and this edition resulted because
recent ocean-floor studIes have given a much wider perspective to geology. My
acquaintance with ocean-floor studies was acquired during my stay at
Lamont-Doherty Geological Observatory of Columbia University, during which
time half of this manuscript was written. I am most grateful to Dr Maurice
Ewing, former Director of the Observatory, for his interest and help in my
ocean-floor research and for his encouragement to write this book. The rest was
written after I had moved to the State University of New York at Albany.
I am greatly indebted to Drs Fumiko Shido and Roger Mason, who read
through the manuscript and gave a lot of helpful advice for its improvement. Drs
Hugh Greenwood, Mineo Kumazawa and Peter Benedict read parts of the
manuscript with friendly criticism.
I am most grateful to Mr Yujiro Iwanami, President of Iwanami Shoten, for
his permission and encouragement to make this rewritten English edition from
the Japanese version published by him, and also for his lasting friendship to me
since our high school days. Messrs Yutaka Ogawa and Nobuyoshi Urabe of
Iwanami Shoten were helpful in the publication of the Japanese edition. The late
Mr David Lynch-Blosse and Mr David Grimshaw of George Allen & Unwin Ltd
have been very helpful in the preparation of the present edition.

AKlHO MIYASHIRO
Albany, New York
October 1972
Acknowledgments for Figures

I am indebted to the following friends and organizations for permission for the
use of figures they published:
Iwanami-Shoten for Figs. 1-3,5-1 and 5-2.
Dr P. C. Hess for Fig. 3-9.
Dr R. L. Oliver and the Geological Society of Australia for Fig. 4-3.
Dr J. B. Thompson and the Mineralogical Society of America for Figs. 5-7
and -S.
Dr T. Matsuda for Fig. 6A-1.
Dr Y. Seki for Fig. 6A-3.
Universitetsforlaget (Norway) for Figs. 6B-1, -2, -3 and -4 and for Fig. 7B-11.
Dr A. Ono for Fig. 7 A-I.
Dr J. P. Bard for Fig. 7A-3.
Drs M. R. W. Johnson and G. A. Chinner for Fig. 7A-1O.
Dr E. Niggli and Fortschritte der Mineralogie for Fig. 7A-II.
Dr S. Banno for Fig. 7A-12.
Dr W. G. Ernst and the Geological Society of America for Fig. 7A-13.
Dr B. McKee and American Journal of Science for Fig. 7 A-14.
Drs B. W. Evans and C. V. Guidotti and Springer-Verlag for Fig. 7B-3.
Dr E-an Zen and the Mineralogical Society of America for Fig. 7B-4.
Dr G. A. Chinner for Fig. 7B-8.
Dr L. C. Hsu and Clarendon Press for Figs. 7B-9 and -10.
Dr B. E. Leake for Fig. SB-2.
Dr H. G. F. Winkler for Fig. 9-2.
Dr D. H. Green for Figs. 12-1 and -6.
Drs J. F. Lovering and A. J. R. White and Springer-Verlag for Figs. 12-4 and
-5.
Dr J. G. Moore for Fig. 13-2.
Dr E. H. Bailey for Fig. 13-3.
Dr H. J. Zwart and Geological Society of Denmark for Figs. 14-2 and -3.
Dr A. Sugimura for Fig. 15-2.
Dr S. Uyeda and American Geophysical Union for Fig. 15-4.
Dr N. Kawai for Fig. 15-7.
Dr M. Hashimoto for Fig. 15-9.
Dr Y. Ueda for Fig. 15-11.
Dr D. S. Coombs and Elsevier Publishing Company for Fig. 16-2.
Note on the Transliteration
of Japanese Names

A large number of Japanese geographical names appear in this book. There are
two commonly used systems for the transliteration of Japanese names in western
languages. The main differences between them are as follows:
Hepburn system: shi shu sho tsu chi fu ji mb
Governmental system: si syu syo tu ti hu zi nb
For example, the same names may be written in the two folowing ways:
Hepburn system: Honshu Shikoku Kyushu Sambagawa
Governmental system: Honsyu Sikoku Kyusyu Sanbagawa
Practically all the atlases published in English-speaking countries adopt the
Hepburn system. In this book, however, both systems are used intentionally
mixed in order to follow the commoner usages in the relevant papers.
I Basis of Metamorphic
Petrology
Chapter 1

Introduction

1-1 CONCEPT OF METAMORPHISM

Threefold Classification of the Rocks

Since the last quarter of the nineteenth century, it has been a widespread custom
among geologists to classify rocks into three categories, igneous, sedimentary
and metamorphic. There are some difficulties in this classification, however. For
example, pyroclastic rocks may be either igneous or sedimentary, while granitic
rocks may be either igneous or metamorphic. Some authors proposed other
systems of classification, which did not however come into common use.
Therefore, I will accept the threefold classification as the starting point of our
discussions.
Igneous rocks are aggregates of minerals that crystallized out of magmas,
which were high-temperature melts mainly composed of silicates. The usual
temperature of crystallization appears to be in the range of 600-1300 °C. The
crystallization takes place either on the surface of the earth or at some depth, but
magmas, if rapidly cooled, may produce glassy rocks.
Rocks may be weathered, broken down and chemically decomposed on or
near the surface of the earth. The resultant products may be then transported by
water or air, and their deposition on the surface of the earth will lead to the
formation of sedimentary rocks. After deposition, the rocks usually become
gradually more coherent and harder owing to the process of diagenesis, which
includes compaction, cementation and recrystallization. Sedimentary rocks
commonly contain clay minerals which are stable at low temperatures on or near
the surface of the earth.
Some igneous and sedimentary rocks are subjected to conditions that dIffer
from those under which the rocks were originally formed. As a result, changes in
mineral composition and structure may take place in such rocks. Van Hise
(1904) used the term 'metamorphism' to represent all such changes. In this
meaning, metamorphism naturally includes weathering and diagenesis.
However, the use of the term metamorphism in such a broad sense is both
unconventional and undesirable, and in normal usage, metamorphism does not
include weathering and diagenesis. Rather, it refers to mineralogical and
20/ INTRODUCTION

structural changes of rocks which take place in deeper parts of the earth, usually
at higher temperatures than those encountered on its surface. The rocks that
were subjected to such changes are called metamorphic rocks (Daly, 1917).
In metamorphism, the mineralogical reconstitution takes place in essentially
solid rocks. This process is called metamorphic recrystallization, or simply
recrystallization. (The term recrystallization is used in different meanings in
metallurgy and chemistry.)

Aspects o/the Concept 0/ Metamorphism

Insofar as metamorphism is defined as a certain group of mineralogical and
structural changes of rocks, it may be naturally regarded as a process comparable
to, for example, weathering and cementation. From this viewpoint, the
characteristics of metamorphism are displayed in such small-scale features as are
observable in thin sections, hand-specimens or a few outcrops. In the study of
small-scale metamorphism (such as pyrometamorphism to be discussed below),
this aspect usually plays the only or the most important role.
However, certain kinds of metamorphism (such as regional metamorphism to
be discussed below) are a direct result of large-scale tectonic processes. Though
in this case also, the mineralogical changes observable in thin sections are of
interest .for themselves, many geologists are interested in such large-scale
metamorphism for the reason that the metamorphosed geologic bodies preserve
records of the characte.istics, distribution and variations of P- T (pressure and
temperature) and other conditions during the tectonic processes, and hence
may serve as a clue to these processes themselves. From this viewpoint, the
mineralogy of metamorphic rocks is of interest because it can give information
of the conditions during the tectonic processes. When we emphasize this aspect
of large-scale metamorphism, such metamorphism may be defined as the
mineralogical and structural recording of the physico-chemical conditions in
geologic bodies involved in tectonic processes.
Of course, small-scale metamorphism is also a recording of the
physico-chemical conditions involved. In this case, however, the pertinent
process has little or no tectonic significance, and petrologic studies of such
metamorphism usually have no direct connection with tectonic problems.
The simple definition of metamorphism as mineralogical and structural
changes of rocks, therefore, does not reveal the full meaning of the term. A
metamorphic terrane has a definite distribution of P-T conditions and chemical
composition as a result of its tec!onic development. The study of such a terrane
(or metamorphic complex) has its own raison d'etre apart from that of
individual minerals and mineral assemblages in it.
The problems related to minerals and mineral assemblages will be treated
mainly in Part II of this book, while the problems related to the tectonic aspects
of large-scale metamorphic complexes will be treated mainly in Part III.
 1-1 CONCEPT OF METAMORPHISM/21

Low-Temperature Limit of Metamorphism

In the 1950s, Coombs (1954) and others began to study the intermediate field
between diagenesis and metamorphism, and the study of metamorphism came to
merge with that of diagenesis.
We may define diagenesis to include those changes taking place at essentially
the same temperature as that of the original deposition, and metamorphism to
include those changes taking place at essentially higher temperatures (Coombs,
1961). The boundary should be defined by definite mineralogical changes. Since
the stability of a mineral or a mineral assemblage is partly related to the bulk
chemicai composition of the rocks, the temperature for such a boundary should
vary with the composition. Our present knowledge on the mineralogy of
low-temperature metamorphic rocks is not ample enough for effective
application of this definition.
As will be discussed in §§ 3-5 and 4-4, the low-temperature limit of
metamorphism is probably around 150°C. Ordinary crystalline schists would
form at temperatures above 300 DC.

High-Temperature Limit of metamorphism

If a rock is heated until it is completely or largely melted within the earth, the
melt should be· regarded as a magma, and rocks formed from consolidation of
this melt should be classed as igneous. It follows that metamorphism should be
defined as processes taking place in essentially solid rocks which are roughly
below the temperature of the beginning of melting. The possible existence of a
small proportion of silicate melt or aqueous fluid in the interstices between
mineral grains, however, is not prohibited.
The experimental determination of the melting temperature of rocks yields
an estimate for the high-temperature limit of metamorphism. Within the pressure
range realized in the typical continental crust, i.e. 0-10 kbar, granitic rocks
begin to melt at a lower temperature than other common rocks. In the presence
of an aqueous fluid, the temperature for the beginning of melting of granite is
around 950°C at 1 bar, and decreases with increasing pressure to about 620 °c
at 10 kbars (Yoder and Tilley, 1962; Boettcher imd Wyllie, 1968). Pelitic rocks
begin to melt usually at slightly higher temperatures than granite as shown in fig.
1-1 (Wyllie and Tuttle, 1961). The temperature of the beginning of melting of
basalts is about 100Q-I100°C at 1 bar, but decreases at a greater rate with
increasing pressure than that of granite, approaching that of granite at 10 kbars
(Yoder and Tilley, 1962). Therefore, the high-temperature limit of
metamorphism may be estimated at about 700-900°C for most rock
compositions in the presence of an aqueous fluid.
It is an unsettled problem, however, whether there is an aqueous fluid in a
large part of the crust. The high-temperature limit of metamorphism should be
considerably raised under dry conditions.
22/INTRODUCTION

10
G

.....
.c
6

:
~
:;,
III
~
c...
4

o~--~---L--~~~----~~~--~~~
600 800 1000 1200
Temperature (OC)

Fig. 1-1 Decrease of melting temperature of rocks with pressure in the

presence.of an aqueous fluid. Curves G, P. and Bl represent the beginning
of melting of granite, pelites and basalt (olivine tholeiite) respectively.
Curve B2 represents the complete melting of basalt (olivine tholeiite).
(Wyllie and Tuttle, 1961; Yoder and Tilley, 1962.)

1-2 GEOLOGIC CLASSIFICATION OF METAMORPHISM

Metamorphism takes place under a great variety of geologic conditions. It is

desirable to divide it into classes on this basis.
Since the classification of metamorphism began in the middle of the
nineteenth century, metamorphic geology has been burdened with confused
terminology related to it. Petrologists might well have some knowledge on the
historical background of classification and terminology (as will be given in the
 1-2 GEOLOGIC CLASSIFICATION OF METAMORPHISM/23

appendix of this book). Here, however, we will confine ourselves to a

present-day view of this problem.

Metamorphism and Basic Geologic Processes

The importance of the study of metamorphism depends largely on the
information it can give toward the clarification of basic geologic processes. It
appears that ocean-floor spreading and movements of the lithospheric plates are
the most basic processes in the reconstruction of the crust and upper mantle of
the earth (Vine, 1966; Morgan, 1968; Le Pichon, 1968; Isacks, Oliver and Sykes,
1968; Wilson, 1968). These processes are accompanied by metamorphism on a
grand scale in the belts where the plates are created and disappear, that is, in
mid-oceanic ridges and in orogenic belts. Grand-scale metamorphism in orogenic
belts has been called regional metamorphism. whereas that in mid-oceanic ridges
will be called ocean-floor metamorphism (Miyashiro, Shido and Ewing, 1971, p.
602; Miyashiro, 1972a). These two classes of metamorphism are of the greatest
geologic significance. Metamorphism takes place along transform faults also and
hence along almost all the plate boundaries (cf. Miyashiro, 1972b).
We have detailed data on the geology, petrology and mineralogy of regional
metamorphic rocks, which form the sound basis of metamorphic petrology. On
the other hand, the data now available on ocean-floor metamorphism and
metamorphism along transform faults are scarce. In view of their geologic
importance, however, special attention will be directed to them in chapters 19
and 20 respectively.
The stable continental crust (especially of Precambrian shields) is usually
more than 30 km thick and the temperature in its lower part is probably a few
or several hundred degrees Centigrade. It is natural to presume that
recrystallization takes place Widely in the lower crust even outside the regions
which are undergoing orogeny. Such metamorphism may be an extremely slow
process. No adequate name has been proposed for it, and available data are
scarce. The metamorphic nature of the lower crust in general will be discussed in
chapter 18. Probably some of the metamorphic changes which will be discussed
in that chapter took place within the stabilized crust, and others had taken place
during orogenies prior to the stabilization.
Metamorphism should not be considered as being confined to the crust. Some
rocks in the upper mantle should have been subjected to metamorphism, though
our knowledge in this matter is very meagre. Since the oceanic crust is thinner
than the continental one, mantle materials appear on the surface of the earth more
easily in oceanic regions (especially in mid-oceanic ridges) than in continents.
The consideration of possible metamorphism in the upper mantle is, therefore,
more pressing in the study of oceanic rocks.
Brief preliminary comments will be given below on regional and ocean-floor
metamorphism together with a few other classes of minor geologic importance.
24/ INTRODUCTION

Regional or Orogenic Metamorphism

Since regional metamorphism is a part of orogeny, the name orogenic
metamorphism could be more appropriate for it. However, we will use the
former name simply because its use is well established in geology. The term
orogeny has been used in a number of different meanings. In this book, it
denotes an association in time and space of large-scale tectonic, metamorphic
and magmatic events along convergent junctures of lithospheric plates, and does
not necessarily include the formation of topographical mountains. We will not
mind the etymological inadequacy of this usage.
In common cases, a cycle of orogeny includes the formation and development
of a geosyncline. However, the formation of a geosyncline could not be an
inevitable part of orogeny if the ultimate cause of orogeny lies in processes in
the deeper part of a lithosphere and asthenosphere. Regional metamorphism
may take place at a distinctly later time than the formation of the pertinent
sediment piles, as in the case of the Ryoke metamorphism in Japan during which
a contemporaneous geosyncline was present but was situated to the south of the
belt of the regional metamorphism.
Rocks subjected to regional metamorphism usually occur in a great belt,
hundreds or thousands of kilometres long and tens or hundreds of kilometres
wide, within an ancient orogenic belt on a continent or an island arc. They
include phyllites, schists and gneisses, being strongly deformed and more or less
schistose. For example, a regional metamorphic belt extends from the Scottish
Highlands to Norway along the Caledonian orogenic belt. Another belt runs
almost along the full length of the Appalachian Mountains in North America.
In a geosynclinal pile subjected to orogeny, the temperature would generally
rise with depth, though it should be influenced greatly by plutonic intrusions
and the migration of any aqueous fluid. With increasing denudation, deeper parts
of the orogenic belt which recrystallized at higher temperatures will be exposed
on the surface. Such a belt of regional metamorphic rocks has one or more axes,
representing the maximum temperature of recrystallization, generally running
parallel with the trend of the orogenic belt. The temperature of metamorphism
increases toward such a thermal axis from both sides, if later tectonic
movements did not destroy the primary structure. The first successful
demonstration of such a relation was made by Goldschmidt (1915) in the
Trondheim (Trondhjem) area of the Caledonian belt of Norway. A simplified
form of his geologic map is shown in fig. 1-2.
In most examples of regional metamorphism, granitic and/or ultramafic
masses are intruded into metamorphic and adjacent terranes. The relationship
between regional metamorphism and granitic masses is in dispute. The general
distribution of the temperature of metamorphic recrystallization in such a
terrane is on a grand scale, and is more or less independent of the shapes and
distribution of individual intrusive masses (Harker, 1932, p. 177). Isothermal
 1-2 GEOLOGIC CLASSIFICATION OF METAMORPHISMf25

Unrecrystallized area ~f~;] Almandine zone

Chlorite zone Plutons

Biotite zone

Fig. 1-2 Classical metamorphic terrane of the Trondheim area, Norway.

(Modified from Goldschmidt, 1915.) The terrane is mainly composed of
Cambro-Silurian metapelites, and the zone boundaries are approximate
isothermal curves. The thermal axis lies within the almandine zone. The
blank area seen in the lower-right and left parts of the figure is composed
of Precambrian gneisses. Tht Atlantic Ocean (A) is present in the
upper-left part, with Trondheim (indicated by T) on the coast of a fjord.

curves plotted on a map of regional metamorphic terranes commonly tend to be

parallel to the thermal axis of the metamorphic belt. The formation boundaries
also tend to be parallel to the axis and hence to the isothermal curves.
In some cases of regional metamorphism, syn- or late-tectonic granite is
abundant in higher temperature parts of terranes. In the Caledonides of the
Scottish Highlands, for example, small granitic masses that formed
simultaneously with the regional metamorphism are characteristically abundant
in high-temperature parts of the terrane. Even if the temperature of
26/ INTRODUCTION

metamorphism does not show a noticeable increase toward the individual small
masses, the general rise of temperature in the axial zone of the metamorphic
terrane may have been largely controlled by the swarm of small intrusives
(Barrow, 1893). Conversely, some of the granitic rocks now under consideration
may have been generated by incipient melting or chemical migration during the
relevant regional metamorphism.
Regional metamorphism appears to be a long process including a number of
phases of recrystallization and of deformational movements. Successive phases
of deformation were usually separated from one another by a static period.
Active recrystallization does not necessarily take place during an active phase of
deformation. The deformational movements in individual phases appear to have
definite characteristics and directions, possibly being distinguished from each
other and put into a time sequence by field work. Schistosity and other
structural features may be assigned to individual phases. Microscopic observation
of relationships between the structural features and mineral growth contributes
to the understanding of such time relations (Sturt and Harris, 1961; Johnson,
1962,1963; Zwart, 1962, 1963).
It has been observed in some metamorphic belts that folds in an early phase
are of a recumbent and isoclinal type, whereas those in a later phase are of a
steeply inclined and more open type (Johnson, 1963; Fyson, 1971). It was also
found in some metamorphic belts that deformational movements producing
large-scale structures such as nappes took place at an early stage of regional
metamorphism, when the prevailing temperature was still low. They were
followed by deformations producing smaller-scale structures and by temperature
increases (e.g. Zwart, 1962; Johnson, 1963; Chatterjee, 1961).

Ocean-Floor Metamorphism
Until quite recently our knowledge of metamorphic rocks was confined to the
realm of continents and islands. In the last few years, however, it became clear
that metamorphic rocks are probably widespread in the deep ocean floors.
Dredging on mid-ocean ridges has yielded a great variety of metamorphic rocks.
The surface layer of the hard crust in the deep ocean floor is largely
composed of basaltic rocks. Judging from the observed patterns of strong
magnetic anomalies, however, the basaltic layer is probably only 0'5-2'0 km
thick. The underlying layer is virtually demagnetized and is probably mainly
composed of metamorphosed mafic rocks (Miyashiro, Shido and Ewing, 1970a,
1971). The ocean-floor metamorphic rocks are mostly of basic and ultra basic
compositions, being formed by the recrystallization of deeper parts of volcanic
piles and associated basic and ultrabasic intrusions. The oceanic crust beneath
normal ocean basins is too low in temperature to cause intensive metamorphic
recrystallization. Ocean-floor metamorphism would take place mainly beneath
the crest of mid-ocean ridges as the geothermal gradient would be relatively high
 1-2 GEOLOGIC CLASSIFICATION OF METAMORPHISM /27
there. The metamorphic rocks thus produced are probably moved laterally by
ocean-floor spreading, leading to the formation of the oceanic crust.
Coombs (1961) coined the term burial metamorphism to represent large-scale
recrystallization of deeply buried rocks without marked penetrative
movements. The resultant rocks are lacking in schistosity. This name was
originally intended to apply to zeolite-facies rocks of New Zealand, which
represent the lowest temperature part of the regional metamorphic belt in the
New Zealand geosyncline. Large-scale recrystallization without marked
penetrative movements, however, is common in ocean-floor metamorphism also.
Therefore the term burial metamorphism is considered in this book to represent
a category to include parts of both regional and ocean-floor metamorphism.

Other Classes of Metamorphism

Contact metamorphism is recrystallization of rocks in an aureole around an
intrusive igneous body due to rise in temperature. The width of
contact-metamorphic aureoles varies, but is in most cases in the range of several
metres to a few kilometres. The most typical example of contact-metamorphic
rock is the non-schistose rock called hornfels. However, schistose rocks are
occaSionally present. The temperature rise in the aureole is due partly to heat
conduction and partly to permeation of an aqueous fluid derived from the
igneous body.
Comparison of contact metamorphism with regional metamorphism
contributes to a better understanding of the latter. Contact metamorphism will
be discussed mainly in § 4-5 and chapter 10.
Pyrometamorphism is recrystallization that shows effects of particularly
high temperatures and takes place commonly in xenolithic fragments included in
volcanic rocks or in the wall rocks of some minor intrusions. Partial melting is
common, and in this respect pyrometamorphism may be regarded as being
intermediate between metamorphism and igneous processes. Such a partially
melted rock derived from shale or sandstone is called buchite. Pyrometa-
morphism may be regarded as an extraordinary kind of contact metamorphism
(chapter 10).
Hydrothermal metamorphism is the process of recrystallization which takes
place under the influence of a hydrothermal solution introduced from the
outside. When the recrystallization is confined to small areas such as walls of
veins, hydrothermal alteration appears to be a more familiar designation.
However, in some geothermal fields, recrystallization takes place over such wide
areas as to justify the name of hydrothermal metamorphism (§ 1-5).
Cataclastic metamorphism (or cataclasis) means crushing and grinding of
rocks which usually take place as a result of fault movement. Though this
process has Widely been called dislocation or dynamic metamorphism, these
terms were initially coined to represent what is now called regional
28/ INTRODUCTION

metamorphism, and are still used in this meaning by some authors (see
appendix). In order to avoid misunderstanding, the name 'cataclastic
metamorphism' will be used in this book. This category of metamorphism takes
place at low temperatures where recrystallization is not active. If the process
takes place near the surface, the resultant rocks are incoherent, whereas if it
takes place at a considerable depth, the resultant rocks remain coherent in spite
of crushing.
Major faults are commonly accompanied by a belt of cataclastic rocks, up to
several kilometres wide. Cataclastic metamorphism along major transform faults
will be treated in chapter 20, because it is of some importance from the tectonic
viewpoint. Our knowledge of cataclastic metamorphism is, however, very
limited.

1-3 COMPOSITIONAL GROUPS OF METAMORPHIC ROCKS

When a metamorphic rock has been derived from a particular type of original
rock, we may call the metamorphic rock by the name of the original rock with
the prefix meta, as is exemplified by the terms metasediment, metaclastics,
metapelite, metagraywacke and metavolcanics.
If the original rock had an igneous porphyritic texture, which was modified
by later metamorphic recrystallization, but was not entirely obliterated, the
resultant metamorphic texture is called blastoporphyritic. In general, the prefix
blasto denotes relict textures, while in contrast the suffix blast refers to textural
features which are newly formed by metamorphic recrystallization. For
example, the porphyroblastic texture means a texture that resembles an igneous
porphyritic one but is produced by metamorphic recrystallization. The
pseudo-porphyritic crystals in such rocks are called porphyroblasts.
In the course of metamorphism, some degree of material migration takes
place, resulting in metasomatism, which may be very intense on a small scale. On
a large scale, however, metasomatism is usually not so marked as to efface the
chemical features of the original rocks beyond recognition. Hence, we may
classify most metamorphic rocks into the following five broad groups based on
the composition of the original rocks.

Pelitic Metamorphic Rocks

Metamorphic rocks derived from pelitic sediments are characterized by relatively
high contents of Al z 0 3 and Kz 0, which during metamorphism usually give rise
to the generation of abundant micas (tables 1-1 and 3-1). Muscovite is common
in low-temperature metapelites, but tends to decompose at high temperatures.
Biotite begins to form usually at a higher temperature than muscovite. Owing to
the abundance of micas and other flaky minerals, metapelites commonly show
strong schistosity, resulting in mica schists. Highly aluminous minerals such as
 1-3 COMPOSITIONAL GROUPS OF METAMORPHIC ROCKS /29
andalusite, kyanite, sillimanite and cordierite are common. Almandine, garnet
and staurolite may also occur.
Rocks of this group are widespread in geosynclinal piles and are susceptible to
changes in temperature and pressure during metamorphism. In other words,
small changes in temperature and pressure commonly cause mineral reactions.

Quartzo-Feldspathic Metamorphic Rocks

Rocks of this group are usually derived either from quartzose sedimentary rocks
or from felsic igneous rocks. They are characterized by a high content of Si0 2
and low contents of FeO and MgO. Schistosity, if present, is usually weak. They
are abundant in geosynclinal piles, but are not susceptible to changes in
temperature and pressure.

Metamorphic Rocks Derived from Calcareous Sediments

If the original rock is composed of nearly pure CaC0 3 , metamorphic
recrystallization changes it to calcite marble. If the original rock contains in
addition some MgO and Si0 2 , metamorphism produces tremolite, diopside,
wollastonite and other calc silicates. Were the original rock to contain abundant
Al 2 0 3 also, plagioclase, epidote and hornblende would be produced, leading to
generation of a rock type mineralogically similar to metamorphic derivatives of
mafic igneous ones. They are susceptible to changes in temperature and pressure.

Basic Metamorphic Rocks

It is a well-established custom in petrology to call igneous rocks with about
4<r45 per cent of Si0 2 ultrabasic, those with about 50 per cent Si0 2 basic,
those with about 60 per cent Si0 2 intermediate, and those with about 70 per
cent Si0 2 acidic. Basaltic rocks, for example, are basic. This custom may be
extended to metamorphic derivatives of igneous rocks, as is exemplified by such
terms as basic metamorphic rocks. The term metabasites was originally proposed
by V. Hackman in Finland in 1907 for basic metamorphic rocks, and will be
adopted in this book.
Many authors like the term mafic better than basic, but we will not be as
scrupulous about this trifle.
Basaltic volcanics are commonly included in sedimentary piles of orogenic
belts. We can correlate the mineral assemblages of metabasites with those of
associated metasediments which were recrystallized under practically the same
temperature and pressure.
Since the basic rocks are rich in MgO, FeO, CaO and A1 2 0 3 , they commonly
produce chlorite, actinolite and epidote in metamorphism at low temperatures
(greenschists), and hornblende at higher temperatures (amphibolites). At still
higher temperatures, clino- and orthopyroxene-plagioclase rocks are formed.
301 INTRODUCTION
Metabasites are very susceptible to changes in temperature and pressure, and
hence are of great petrologic interest.

Ultrabasic Metamorphic Rocks

Ultrabasic rocks occur in many metamorphic terranes. Serpentinites, for
example, are common in glaucophane-bearing metamorphic terranes, such as the
Franciscan Formation of California. In many cases, they have been
metamorphosed together with the surrounding rocks.
In contrast to other classes of metamorphic rocks, however, no detailed
petrographic data are available on the progressive changes to take place in this
class.

1-4 NOMENCLATURE OF METAMORPHIC ROCK TYPES

The textural feature is emphasized in the definitions of such names as schist and
gneiss, whereas the mineral composition is delineated in the definitions of such
names as amphibolite. Texture is, however, more or less related to mineral
composition. For example, abundant flaky minerals could produce strong
schistosity. If this relationship is emphasized, most rock names are related to
both texture and mineral composition.
The same names have been used in different meanings by different authors.
The meanings of most names in common usage have changed to some extent
with time. In recent years, the concepts of metamorphic facies have become so
popular as to modify the meaning of some rock names. For example, there is a
strong tendency to limit the use of the names granulite and eclogite to rocks of
strictly defined granulite and eclogite facies. In this respect, the existing
textbooks of descriptive petrography do not give information necessary for the
proper application of terms. In view of this situation, brief comments on main
rock names and related textural terms will be given below.
Metamorphic rocks which contain abundant flaky minerals shOwing parallel
disposition, tend to split into thin irregular plates or lenticles. This property is
called schistosity.
Well-recrystallized metamorphic rocks shOWing strong schistosity are called
schists. Many old petrographers liked to limit the use of the name schists to
rocks poor or practically lacking in feldspars. They called more feldspathic
metamorphic rocks gneisses. This usage is, however, obsolete. Mica schists are a
variety of schists mainly composed of quartz, feldspar and micas, with or
without garnet, kyanite and so on, usually being derived from pelitic sedi-
mentary rocks. Greenschists (to be spelled as a single word) are mainly
composed of feldspar, chlorite and epidote with or without quartz and
actinolite, being formed by the low-temperature metamorphism of basic rocks.
 1-4 NOMENCLATURE OF METAMORPHIC ROCK TYPES/31

Foliation is an unfortunate term which is widely used in America to denote

planar stuctures of metamorphic and plutonic rocks; usually due to parallel
arrangement of minerals. In this meaning, schistosity and gneissosity are a
typical foliation. Harker (1932) and many British authors used the term to
denote compositional layering (banding) in metamorphic rocks. Many but not all
schists show compositional layering. Schistosity, rock cleavage, jOints and
bedding produce planar structures in rocks. Any kind of such planar structures
may be called s-surfaces as proposed by Bruno Sander (1930) regardless of its
origin. An elaborate compilation of such structural terms with critical comments
is given by Dennis (1967).
Table 1-1 Average chemical compositions of metamorphic rocks
(anhydrous basis)

4
2 3 Quartzo-
1 Mica Two-mica feldspathic 5
Phyllites schists gneisses gneisses Amphibolites
Si0 2 60·0 64·3 67·7 70·7 50·3
Ti0 2 1·1 1·0 0·5 1·6
Ah 0 3 20·7 17·5 16·6 14·5 15·7
Fe203 3·0 2·1 1·9 1·6 3·6
FeO 4·8 4·6 3·4 2·0 7·8
MnO 0·1 0·1 0·1 0·2
MgO 2·9 2·7 1·8 1·2 7·0
CaO 1·2 1·9 2·0 2·2 9·5
Na20 2·0 1·9 3·1 3·2 2·9
K20 4·0 3·7 3·5 3·8 1·1
P20 S 0·2 0·2 0·2 0·3
100·0 100·0 100·0 100·0 100·0

Note: 1,2,4,5 after Poldervaart (1955); 3 after Lapadu-Hargues (1945).

If the parallel disposition of flaky or prismatic minerals is more or less

disturbed, or if the proportion of flaky or prismatic minerals decreases with a
complementary increase of quartz and feldspar, the tendency of the rocks to
split up into plates decreases. Such rocks commonly show compositional
layering. These textural properties are termed gneissosity. Medium- or coarse-
grained metamorphic rocks with gneissosity are called gneisses. The term gneiss
has been most commonly applied to metapelites and quartzo-feldspathic rocks,
but may be used for other compositional groups a,s well.
Schists and gneisses are widespread in regional metamorphic terranes. Typical
pelitic rocks tend to become schists owing to the abundance of micas, whereas
quartzo-feldspathic rocks tend to become gneisses. With increasing temperature,
the parallel disposition of flaky minerals tends to be disturbed, and some flaky
32/ INTRODUCTION
minerals such as chlorite and muscovite may be broken down. Hence, even
pelitic rocks tend to change from schists to gneisses at high temperatures.
Pelitic rocks that were recrystallized during regional metamorphism at lower
temperatures than schists are usually very fine-grained, showing a lustre on
cleavage surfaces due to parallel arrangement of muscovite and chlorite. Such
rocks are called phyllites.
As the degree of recrystallization decreases, phyllites would grade into slates
(clay slates). The constituent minerals of slates are too fine-grained to be
distinguished by the unaided eye. Slates are indurated rocks with a tendency to
split into thin plates, the direction of which is independent of the original
bedding. Usually slates are a link between unmetamorphosed and meta-
morphosed terranes.
Thus, we have the following series of common pelitic metamorphic rocks
with rising temperature: slates ~ phyllites ~ mica schists ~ gneisses. Average
chemical compositions of these rocks are shown in table 1-1. Since quartzo-
feldspathic rocks tend to become gneisses rather than schists, the average
°
chemical compositions of gneisses show higher Si0 2 and Na2 contents than that
of schists.
Some quartzo-feldspathic rocks recrystallized at low temperatures show a
gneissose texture. According to the traditional petrographic nomenclature, there
can be no objection to calling such rocks gneisses. In recent years, however, most
authors prefer to confine the use of the term gneiss to medi um- and high-
temperature metamorphic rocks.
Some gneissic rocks are megascopically heterogeneous and look like mixed
rocks composed of seemingly metasedimentary and granitic materials. Such
rocks are called migrrultites. Though this name was coined by Sederholm (1907)
to denote truly mixed rocks, it is now widely used in the above-mentioned
non-genetic meaning. The apparently granitic material in migmatite may have
been introduced from the outside, formed by partial melting in situ, or
segregated by a metamorphic process. Migmatites usually occur in association
with granitic rocks in the higher temperature part of regional metamorphic
terranes.
Amphibolite is mainly composed of hornblende and plagioclase, but in some
cases plagioclase may be scarce. A subordinate amount of quartz, almandine,
biotite, clinopyroxene or cummingtonite may also be present. Amphibolite may
or may not be schistose, and mayor may not show compositional banding.
When the presence of schistosity or gneissosity is emphasized, the term
hornblende-plagioclase schist or gneiss is more appropriate. Most amphibolites
are derived from basic igneous rocks, but some apparently form from impure
calcareous sediments.
Amphibolite is stable in medium-temperature metamorphism. If the tempera-
ture is lower, epidote amphibolite which is mainly composed of hornblende,
 1-4 NOMENCLATURE OF METAMORPHIC ROCK TYPES /33
sodic plagioclase and epidote forms instead. Formerly epidote amphibolite was
treated as a subgroup of the amphibolites. In and prior to the 1930s, it was
common to call all the basic metamorphic rocks rich in actinolite or hornblende
by the name of amphibolite and to include actinolite in hornblendes. Since the
concept of metamorphic facies became popular, however, it has become
common to use the name amphibolite for amphibolite-facies metabasites and the
name epidote amphibolite for epidote-amphibolite facies metabasites, as defmed
above. Greenschist-facies metabasites, even if they may contain abundant
actinolite, are not called amphibolite in the current usage.
Granulites originally meant metamorphic rocks with granulitic texture. The
granulitic texture (as distinct from the granular) is characterized by alternation
of megascopic or microscopic streaks and bands of different mineral composi-
tions, where the individual streaks and bands are made up of more or less
equigranular and equidimensional grains and are nearly or entirely lacking in
flaky and prismatic minerals. Naturally this texture occurs typically in rocks rich
in quartz, feldspars, garnets and/or pyroxenes. This old usage of the term
granulites is ~ti1l accepted by many geologists.
However, Eskola (1939) proposed the name of granulite facies to include a
group of metamorphic rocks recrystallized under a range of high-temperature
conditions. Since then, it has become gradually common to call some or all of
the rocks of this facies by the name of granulite. Many but not all of the rocks
of this facies have a granulitic texture (White, 1971).
There is some confusion in the usage of the three textural terms granulitic,
granulose and granular. The term granulitic is recommended here for the meta-
morphic texture characteristic of granulites as described above. The terms
granulose and granular (or granoblastic) should be avoided, because their
meanings are obscure. Granular, for example, was used sometimes as a synonym
of equigranular (Le. non-porphyritic), sometimes in the meaning of equidi-
mensional, and sometimes for rocks composed of rounded grains.
There are wide exposures of Precambrian crystalline rocks characterized by
the presence of orthopyroxene in Peninsular India. This group of rocks with
diverse compositions was named the Charnockite Series. Acidic members of the
Series were called charnockite by Holland (1900). Some later authors, however,
used this rock name to designate all the rocks, acidic and basic, of the Series.
Similar rocks throughout the world came to be called by this name. Char-
nockites were considered to be igneous by many authors, but are now widely
believed to be metamorphic rocks of the granulite facies. They are usually
equigranular and xenomorphic with or without parallel structure. The most
characteristic feature is a dark bluish or brownish colour of quartz and feldspars,
owing to which not only the basic but also the acidic charnockites are very dark
coloured.
Eclogite, in a recent common usage, is a basic rock composed mainly of
34/ INTRODUCTION
omphacite (clinopyroxene) and rather magnesian garnet. Eskola (1939) pro-
posed the name of eclogite facies for eclogites and related rocks. However, some
modification of his defmition is necessary as will be discussed in chapter 12.
Certain recent workers prefer to confine the use of the name eclogite to
clinopyroxene-garnet rocks belonging to a more strictly defined eclogite facies
(§ 12-2).
Hornfels is typically a fine-grained, non-schistose, contact-metamorphic rock
of pelitiC composition, sometimes containing porphyroblasts of andalusite or
cordierite. It may be entirely massive, but may show parallel structure inherited
from the original sedimentary rock. Such a parallel structure, however, does not
give the rock a tendency to split into thin plates. This is the diagnostic property
of hornfels as distinct from schist. It is believed that this property is a result of
the absence of penetrative movements. The term hornfelses may be used in a
wider sense to include non-schistose contact metamorphic rocks of any other
compositions.
Some geologists consider that the names of rock types should be defined so as
to be independent of genetic hypotheses. Since descriptions are a mere
preliminary to genetic discussions in many cases, however, it is natural that the
definitions of rock names change with genetic views so that genetic discussions
are facilitated and confusions are avoided.
The nomenclature of cataclastic metamorphic rocks will be treated in chapter
20, since it has no relation to other chapters.

1-5 DISTRIBUTION OF METAMORPHIC ROCKS

Precambrian Shields
Shield areas were subjected to phases of orogeny and metamorphism in
Precambrian, and became stabilized in Palaeozoic and later time. They usually
occur as cores of continents. The proportions of areal extents of various rock
types have been estimated for the Baltic Shield in Finland (Sederholm, 1925)
and in Norway (Barth, 1961) and for the Canadian Shield (Engel, 1963; Reilly
and Shaw, 1967). A rough estimate of the average relative areal extents of the
rock types exposed on shields is as follows:
Granitic rocks and quartzo-feldspathic
gneisses 70-80%
Other metasediments 10% or a little more
Metabasites 10% or less

For average chemical compositions, see table 18-2.

Large· parts of the metamorphic rocks exposed on the shields are in the
amphibolite and the epidote-amphibolite facies. It is to be noted that the
intensity or temperature of metamorphism does not tend to increase with
 O~D
~~

VI·
til
...."
>-1
~
....
til
c:
....>-1
0
:z:
0
':t!
is:
ttl
>-1
>
is:
d~1 0
ShO('lds '" ~
~

D PhanerozOIC orOQenoc belts

....('l:I:
~
.'. 0
n
~
Fig. 1-3 Precambrian shields and Phanerozoic orogenic belts. til

W
VI
-
36/ INTRODUCTION
geologic age within shields. The Superior Province is the oldest part of the
Canadian Shield, but generally speaking the intensity of metamorphism observed
there is not high. The Grenville Province is the youngest part of the same shield,
but in general has been metamorphosed at higher temperatures, resulting in the
abundant formation of rocks of the granulite facies. Most of the granulite facies
rocks occur in Precambrian shields.
Precambrian granitic and metamorphic complexes were found to exist also as
basements of younger orogenic belts.

Phanerozoic Orogenic Belts

Phanerozoic (Le. Palaeozoic and later) orogenic belts tend to be disposed
surrounding Precambrian shields. Such disposition becomes especially clear, if
we reconstruct the configuration of old continents prior to the beginning of
drifting in Mesozoic time (chapters 13 and 14). Some of the younger orogenic
belts are on continental margins and in well-developed island arcs. Granitic and
regional metamorphic rocks are abundant in most orogenic belts on continental
margins and in well-developed island arcs, but not so abundant as in shields.
In Japan, for example, granite rocks and well-recrystallized metamorphic rocks
underlie about 13 and 4 per cent respectively of the whole area (chapter 15).
The Izu-Bonin and Mariana Arcs jut out into the Pacific Ocean to the south
of Japan. These arcs are made up of young volcanics and sediments (unmeta-
morphosed). The Yap Arc to the west of the Mariana, however, has developed a
regional metamorphic terrane in spite of its great distance from continents
(chapter 16).
It is interesting that the kinds of metamorphic rocks exposed in Phanerozoic
orogenic belts differ to some extent from those observed in Precambrian shields.
In the former, metamorphic rocks of the glaucophane-schist and zeolite facies are
relatively common, and those of the granulite facies are very rare (§ 4-3).
It appears that continental crusts have grown by addition of successively
newer orogenic belts around the pre-existing sialic crustal masses. Regional
metamorphism should have played an essential part in transforming soft
geosynclinal piles into hard crystalline crust.

Deep Ocean Floors

The deep ocean floor is mostly covered by sediments, which are usually
underlain by a layer of basaltic rocks, which ih turn probably grades downward
into metamorphosed basalts and gabbros. Some of the metamorphic rocks are
brought up to the surface of the ocean floor by tectonic processes and as
inclusions in peridotites or serpentinites. Virtually all the oceanic metamorphic
rocks appear to have been derived from basic and ultrabasic igneous rocks. As
stated before, the metamorphism appears to take place in mid-oceanic ridges and
 1-5 DISTRIBUTION OF METAMORPHIC ROCKS /37

the resultant metamorphic rocks would be transported laterally by ocean-floor

spreading (chapter 19).

Present-Day Metamorphism, Particularly of Geothermal Fields

Metamorphism is probably taking place now on a grand scale beneath some
active orogenic regions in island arcs and beneath mid-oceanic ridges (chapters
15-17 and 19). On a smaller scale, present-day metamorphism is taking place in
many active geothermal fields.
In the Salton Sea area to the north of the Gulf of California, a Pliocene and
Quaternary sedimentary deposit, which reaches a thickness of 6 km in some
places, was formed as a part of the delta of the Colorado River. Under the
influence of an ascending hydrothermal fluid rich in carbon dioxide as well as in
sodium, calcium and potassium chlorides, metamorphic recrystallization is
taking place within the deposit. The sediments on and near the surface are
unmetamorphosed clays, silts and sands, containing montmorillonite, illite,
kaolinite, calcite, dolomite and quartz. With an increase of depth, montmoril-
lonite and kaolinite disappear, whereas ankerite, chlorite, and furthermore
epidote appear. Rocks recrystallized at temperatures around 3000 C have the
quartz+epidote+chlorite+potassium feldspar+albite assemblage characteristic of
the greenschist facies, though schistOSity is lacking (Muffler and White, 1969;
Keith, Muffler and Cromer, 1968).
In some active geothermal areas of New Zealand and Japan, metamorphic
recrystallization is taking place, producing zeolite-bearing assemblages at shallow
depths, and albite-bearing assemblages at greater depths (Steiner, 1953; Coombs
et al. 1959; Seki et al. 1969b). A detailed review of such areas is given in § 6A-3.
The presence of zeolite-bearing assemblages in this case and their absence in the
Salton Sea area are considered to be due in part to a lower concentration of CO 2
here than in the Salton Sea (§ 4-4).
Chapter 2

Basic Characteristics of Metamorphic

Reactions

2-1 AIM OF THIS CHAPTER AND UNITS OF MEASUREMENT

The mineral changes in metamorphism are chemical reactions from the view-
point of thermodynamics. Thermodynamics and crystal chemistry are important
for a better understanding of such reactions. There is an extensive literature in
this field. The present book, however, is not intended to cover them in any way,
but is directed toward the mineralogic, petrographic and tectonic aspects of
metamorphism. However some elementary knowledge of the thermodynamic
characteristics of metamorphic reactions is needed even for understanding
mineralOgic and petrographic problems. In view of this need, some of the basic
characteristics are summarized in a simple way in this chapter, largely as a
preliminary to the discussion of progressive metamorphism to be treated in the
next chapter. In this respect, this chapter is essentially a part of the introduction
of this book. Readers particularly interested in the thermodynamic problems in
metamorphism are referred to other books and papers (e.g. Kern and Weisbrod,
1967; Thompson, 1955; Fyfe, Turner and Verhoogen, 1958; Korzhinskii, 1959).
Since our direct concern here is with semi-quantitative relations, fluids will
always be regarded as ideal gases. Four main classes of metamorphic reactions
will be treated, in this order:

1. Solid-solid reactions, i.e. reactions among solid phases, involving no libera-

tion of volatiles.
2. Dehydration reactions, i.e. reactions involving liberation of H2 0 with rise in
temperature. The majority of metamorphic reactions belong to this class.
3. Decarbonation reactions, i.e. reactions involving liberation of CO 2 with rise
in temperature.
4. Oxidation-reduction reactions. Since many rock-forming minerals contain
iron, this class of reactions is also important.

The units of pressure and heat most commonly used in the past were the bar,
kilobar, calorie and kilocalorie. In future, however, SI units, pascal (Pa) and
joule (J), will become common for pressure and heat respectively (Pa = N m -2; J
 2-2 SOLID-SOLID REACTIONS/39

= N m). The conversion factors between them are given in table 2-1. The bar and
J will be mainly used in this chapter.
The internationally agreed thermochemical standard state is 25°C (298·15 K)
at 1 atm (1·013 bars). Most thermochemical data have been given for these
conditions (e.g. Robie and Waldbaum, 1968). The equilibrium pressure directly
calculated from such data is in atmospheres. However, the difference between
the atmosphere and the bar is small (table 2-1), and may be neglected for the
purposes of this chapter. Therefore, such calculated pressures will be regarded
below as being in bars.

Table 2-1 Some important symbols and units

Symbol Physical quantity Unit

T Temperature
P Pressure }
Ps Rock-pressure
Pfluid Pressure of a fluid phase bar, kilobar
PH,O, Pco" Po, Pressure of H2 0, CO 2 and O 2
within an imaginary semi-
permeable cell in a rock
Standard free energy of
a reaction J (joules)

1 bar = 10 5 Pa = 0·987 atm = 14.5 p.s.i.

1 kbar= 0·1 GPa (gigapascal)
1 cal = 4·184 J = 41·8 cm3 bar
1 J = 10 cm3 bar = 10 7 ergs
Note: The pressure at a depth of h km in the crust with density of d g cm- 3 is
about 98dh bar. Approximate pressures within a continental crust are as
follows:
Depth (km) 10 35 SO
Pressure (kbar) 2·6 10 15

2-2 SOLID-SOLID REACTIONS

As a simple and important case, we consider phase transformations between the

three polymorphs of Al 2 SiO s (andalusite, kyanite and sillimanite):
kyanite ~ sillimanite (2-1)
andalusite ~ sillimanite. (2-2)
The phase relations corresponding to these transformations are observed in many
metamorphic terranes, and are of prime importance in determining the tempera-
ture and pressure of metamorphism. The right-hand sides of these equations
represent a higher temperature. Such reactions may take place with or without
40/ BASIC CHARACTERISTICS OF METAMORPHIC REACTIONS

the catalytic action of a fluid. Since these equations involve solid phases only,
however, the equilibrium P- T relations are independent of the presence and
absence of a fluid.
A number of experimental studies have been carried out, revealing that there
exists a triple point (fig. 2-1). The P-T values of the point, however, have not

o
30~O~--~40~O~---=50~O~--~6~OO~--~7~O~O----~8~OO~
Temperature (DC)
Fig. 2-1 Equilibrium curves for polymorphic transformations of A12 SiO s
(Richardson et al. 1969) and for solid-solid reactions involving jadeite.
(Birch and Le Comte, 1960; Newton and Kennedy, 1968.)
 2-2 SOLID-SOLID REACTIONS /41
been well established yet. A synopsis of recently-obtained values is given in table
2-2. Althaus (1967) and Richardson, Gilbert and Bell (1969) gave values near
600 °c and 6 kbar, whereas Newton (1966b) and Holdaway (1971) gave values
near 510 °c and 4 kbar. The natural minerals are not of pure Al2SiOs, and may
show strain and structural disorder. The stability relations may be influenced by
these factors as well as the surface free energy when the relevant minerals are
fine-grained.

Table 2-2 Temperature and pressure of the triple point ofAl2 SiO s
Author and year Pressure (kbar)

Newton, 1966b 520 4·0

Weill, 1966 410 2·4
Althaus, 1967 595 6·5
Pugin and Kbitarov, 1968 540 7·6
Richardson et al. 1969 622 5·5
Holdaway, 1971 501 3·76

As examples of a more common kind of solid-solid reactions, we consider

reactions to form jadeite from albite.
NaAlSi 2 0 6 + Si0 2 ~NaAlSi308 (2-3)
jadeite quartz albite
(24)
jadeite albite nepheline
The right-hand side of these equations represents higher temperatures and lower
pressures (Birch and I.e Comte, 1960; Newton and Kennedy, 1968), as shown in
fig. 2-1.
Natural jadeite and albite do not have the idealized compositions given above,
but are solid solutions containing some other components, which should have
effects on their stability relations (e.g. Ramberg, 1952, pp. 18-37). For
example, natural jadeite may contain diopside and acmite components, which
should decrease the pressure necessary for the stabilization of jadeite.
In all reactions, the high-temperature side has a greater entropy than the
low-temperature side, and the high-pressure side has a smaller volume than the
low-pressure side.
The slope of such equilibrium curves could be calculated by the Clausius-
Clapeyron equation:
dP 10·0 t:.H 10·0 t:.s
-= (2-5)
dT T.::lV .::lV
Here, P and T represent the pressure (bars) and the absolute temperature, and
42/ BASIC CHARACTERISTICS OF METAMORPHIC REACTIONS

Ml, A V, and AS represent the heat of reaction (J) and the changes in volume
(cm 3 ) and in entropy (J K- 1 mol-I) respectively. (When Ali is expressed in cal
and AS in cal K- 1 mol-I, a coefficient of 41'8 should be used instead of 10.0.)
In solid-solid reactions, the change of heat capacity is usually very small. By
neglecting it, we have the follOwing approximate relation:
AG = Ml298 - TAS;98 + (P - I)AV/I0·0, (2-6)
where AG = free energy of the reaction (J) as a function of absolute temperature
T and pressure p(bars), Ali298 and AS298 = heat (J) and entropy (J K- 1
mol-I) of the reaction in the standard state, and AV = volume change (cm 3 ) of
the reaction. The values of Ali298 and AS 298 can be calculated from data given
in tables of thermodynamic constants (e.g. Robie and Waldbaum, 1968).
The equation of the equilibrium curve for the reaction is obtained by the
condition AG = 0 in equation (2-6).
Solid-solid reactions are important for the estimation of temperature and
pressure, since their equilibrium curves are independent of the chemical
potentials of H2 0 and CO 2 ,

2-3 OPEN SYSTEM AND MODELS FOR THE BEHAVIOUR OF H20

Ps (flock-Pressure) and PH 2 0
Pelitic sediments contain a large amount of H2 0, part as interstitial water
between mineral grains, and part as a constituent of minerals. Metamorphism
causes marked dehydration of them. On the other hand, basaltic rocks are
originally poor in H2 0. Their metamorphic transformation to greenschists
involves marked hydration. Since dehydration and hydration are widespread
occurrences, H2 0 must be mobile to some degree within a rock complex
undergOing metamorphism. Similarly, since the metamorphism of calcareous
sediments involves decarbonation, CO2 also should be mobile to some degree. In
other words, a rock complex undergoing metamorphism is probably open to
H2 0 and CO 2 ,
The degree of mobility, however, is not clear. It could differ in different
cases. As an ideal case, we may imagine that a component can move perfectly
freely through a rock complex. In such a case, the chemical potential of the
perfectly mobile component is uniform throughout, and is controlled by
external conditions just like the temperature and pressure acting on the solid
phases. In reality, the degree of mobility is probably much less than in such a
case (e.g. Zen, 1961a; Carmichael, 1970; Guidotti, 1970). The behaviour of H2 0
and other mobile components is of vital importance for understanding meta-
morphic reactions. This problem has been discussed in general forms by Ramberg
(1952), Thompson (1955), Korzhinskii (1959) and Greenwood (1961).
 2-3 OPEN SYSTEM AND MODELS FOR THE BEHAVIOUR OF H20/43

The pressure acting on the solid phases will be denoted as Ps and called
rock-pressure in this book.
Chemical potential may be regarded as the free energy that one mole of a
component has in a solution. Chemical potential is an intensive property, being
of the same value for the same component in all phases in a system in
equilibrium. The chemical potential of a pure substance is equal to its molar free
energy. The molar free energy or chemical potential of an ideal gas is related to
pressure P by the following equation:
/l =/le +RTlnP. (2-7)
Here, /l-&, R and T represent the chemical potential at unit pressure, the gas
constant, and absolute temperature respectively. The same relationship holds for
a component gas of an ideal gas mixture when P is regarded as its partial
pressure. In other words, the chemical potential of a component gas increases
with its partial pressure in a gas mixture.
Let us imagine a small cell made up of rigid walls permeable only to H2 0.
This cell is contained in a rock. H2 0 molecules should diffuse through the walls
until osmotic eqUilibrium is reached between the inside and outside of the cell.
The chemical potential of H2 0 should be equal between them. The pressure
within the cell will be denoted as PH, o. It is related to the chemical potential of
H2 0 by equation (2-7).

Four Simple Models

Four models for the behaviour of H2 0 in a rock complex undergoing meta-
morphic recrystallization will be discussed below. Each of these models could be
approximately applied to certain 'cases of metamorphism.

Model (A). A rise in temperature should cause progressive dehydration of

minerals. The liberated H2 0 will move gradually out of the rock. If the mobility
is not great in comparison to the rate of the dehydration reactions, an aqueous
fluid phase would continue to exist between mineral grains. It may be assumed
that the temperature and pressure of the fluid are equal to those of the adjacent
solid phases. The fluid is not of pure H2 0 composition, but contains a small
amount of dissolved mineral substance. If we neglect this for a first approxi-
mation, we have the relation: Ps =Pfluid =PH, o.
Before the middle of the 1930s, the existence of such an inter granular fluid
was advocated by practically all the authorities in metamorphic petrology,
including Becke, Niggli, Harker and Eskola. The fluid was assumed to behave as a
solvent to help metamorphic reactions and as a carrier of dissolved substances.
This view is supported by many contemporary authors also (e.g. Korzhinskii,
1959; Winkler, 1967). The metamorphic reactions in such a case are identical in
character to those observed in simple hydrothermal experiments.
44/BASIC CHARACTERISTICS OF METAMORPHIC REACTIONS

Model (B). If pelitic sediments are mixed with calcareous ones, metamorphic
reactions should produce an inter granular fluid containing both H2 0 and CO 2 ,
The temperature and pressure of the fluid are assumed to be equal to those of
the adjacent solids.
The chemical potential of H2 0 and 'PH 2 0 should decrease with a decreasing
proportion of the component in the mixture. In this model, Ps = Pfluid > PH 2 O'

Model (C) The intergranular spaces containing a fluid may be mutually con-
nected for a vertical distance. In an extreme case, they may be continuous up to
the surface of the earth. If equilibrium is approached in such a channel, the rate
of downward increase of pressure therein should depend on the density of the
fluid column. It should be much less than the rate of pressure increase in the
surrounding solid phases. In this model, if the dissolved mineral substances are
neglected and the fluid is composed of pure H2 0, Ps > Pfluid = PH 2 0 .
This model appears to be applicable at least to incipient recrystallization of
sediments and hydrothermal metamorphism in active geothermal fields. The
situation represented by this model is mechanically and thermodynamically
unstable. Metamorphic recrystallization would tend to disconnect the inter-
granular spaces. However, penetrative deformations in regional metamorphism
may continually produce new channels.

Model CD). Since the 1940s, a number of petrologists have concluded that an
inter granular fluid phase is not necessary for metamorphic reactions to take
place, and that probably it does not exist in many or most cases of meta-
morphism. This view was particularly emphasized by radical advocates of
granitization around 1950 (Holmes and Reynolds, 1947;· Ramberg, 1952).
It must be noted that the absence of a fluid phase in a rock does not
necessarily mean the absence of H2 0 and CO 2 molecules. Such molecules can be
present in the grain boundaries between minerals, and move by diffusion down
the gradients of their chemical potentials, as was clearly discussed by Ramberg
(1952).
The total mass of H2 0 and CO 2 molecules in grain boundaries should be
negligible as compared with that of the rest of the rock. If a rock is closed to
such molecules, their effects on mineral changes should be negligible. If the rock
is open, however, such components may be successively liberated within the
rock and diffuse away, or may be successively introduced from the outside and
react with the existing minerals within the rock. Continued reactions with
continuous migration of such mobile components would result in a great extent
of change. Therefore, this model has significance only when the pertinent rock
is open for H2 0 and/or CO 2 , This situation is quite different from that of
ordinary experimental studies.
In this model, Pg > PH", 0, and the chemical potential of H2 0 is smaller than
 2-4 DEHYDRATION REACTIONS /45
the molar free energy of pure H2 0 under the pertinent temperature and
pressure. If the chemical potential is increased to approach the molar free energy
of pure H2 0, a fluid phase of virtually pure H2 0 composition may form.
In the early phase of regional metamorphism, the temperature should increase
progressively. H2 0 and CO 2 should be liberated from hydrous and carbonate
minerals through a sequence of reactions which take place with rising tempera-
ture. The models for this case are particularly important, and will be treated in
§ 3-1.

2-4 DEHYDRATION REACTIONS

Most metamorphic reactions occurring with rising temperature cause liberation

of H2 0 from solid phases as exemplified below.
(2-8)
pyrophyllite kyanite quartz

muscovite quartz sillimanite K-feldspar

(2-10)
muscovite corundum K-feldspar

Equation (2-8) represents a possible reaction to produce kyanite in meta-

pelites. Equations (2-9) and (2-10) represent the breakdown of muscovite in
quartz-bearing and quartz-free rocks. The side having liberated H2 0 in such
equations represents higher temperatures in almost all cases, because the entropy
of the liberated H2 0 is usually very great and the side with a greater entropy
represents higher temperatures.

Equilibrium Curves in Model (A)

We assume that the liberated H2 0 forms an aqueous fluid phase at the same
temperature and pressure as the solid phase in contact. This corresponds to
model (A) discussed in the preceding section. The total volume and entropy of
the products in such reactions are usually greater than those of the reactants in
the pressure range up to several kilobars, because the volume and entropy of the
liberated H2 0 are great. The Clausius-Clapeyron equation (2-5) holds for
equilibrium curves in this case. Here, ~ V means an increase in the total volume.
Since both ~ V and AS are usually positive, dP/dT is positive. In other words,
the equilibrium curves usually have a positive slope (fig. 2-2). At pressures below
1 kbar, ~ V is very great and hence dP/dT is very small. An increase of pressure
to about 3 kbar markedly reduces the volume of the liberated H2 0, and hence
dP/dT increases. Thus, the eqUilibrium curves for dehydration reactions usually
show a sharp bend in the pressure range of 0'1-3'0 kbar. At pressures above 3
kbar, they are nearly straight and are steeper than the eqUilibrium curves of most
46/ BASIC CHARACTERISTICS OF METAMORPHIC REACTIONS

15 t:\
\
I
\
I
I
I
I
I
10 16
I~
I
I

-;:
'"
.0
.>t.
o.N
rt

It
~ I
~ I
I
I

o50~0~----~~------~~------~~------~-

Temperature (OC)

Fig. 2-2 Equilibrium curves for dehydration reactions in the presence of

an aqueous fluid. Full lines represent equilibria for the reaction of
muscovite and quartz (2-9), and for the breakdown of muscovite (2-10)
(Evans, 1965; Ve1de, 1966). The dashed line FT represents equilibria for
the reaction: forsterite+talc = enstatite +H2 O. This curve shows a negative
slope at pressure above 10 kbar. (Kitahara and Kennedy, 1967.)
 2-4 DEHYDRATION REACTIONS /47
solid-solid reactions. In other words, the effect of pressure on eqUilibrium
temperature is small under such pressures.
It has been found empirically that the total volume of the solids commonly
decreases in such reactions (table 5 in Thompson, 1955). At pressures higher

°
than 10 kbar, an aqueous fluid is highly compressed and the volume of the
liberated Hz is small. If the volume decrease of the solids is relatively large, the
total volume of the system including both solid and fluid may decrease. In such
a case, an equilibrium curve should take on a negative slope. An example of the
change of a curve from positive to negative slopes with increasing pressure is
shown in fig. 2-2 (curve FT).
In the pressure range of more than 30 kbar (corresponding to the upper
mantle), most dehydration curves may have a negative slope (e.g. Fry and Fyfe,
1969). In such a case, increasing fluid pressure should cause dehydration.
Some reactions representing the breakdown of zeolites have a large decrease
in volume of the solid phases. Hence, their equilibrium curves come to have a
negative slope at relatively low pressures (Coombs et al. 1959), as is illustrated
by the breakdown of analcime in the presence of quartz (fig. 6B-5).
Some breakdown reactions of zeolite involve not only a decrease of volume
but also a decrease of entropy. The follOWing are examples: analcime =jadeite +
HzO, and laumontite = lawsonite + 2 quartz + 2 HzO (Fyfe and.Valpy, 1959;
Thompson, 1970b). The equilibrium curves of these reactions have a positive
slope, as shown in fig. 6B-S.

Equilibrium Curves in More General Cases

T4e equilibrium curves for any dehydration reaction can be calculated by the
following approximate relation:

(2-11)

Here, t:.cr and t:. Vs represent the free energy of reaction (J) under 1 bar and the
volume change of solid phases (cm 3 ) respectively, both per mole of liberated
Hz 0, and hence t:.cr is a function of absolute temperature T. P Ha 0 and Ps are
in bars and log indicates the common logarithm. If we accept model (A), Ps =
P Ha 0 in this equation.
In models (B), (C) and (0), Ps > P Ha o. The dehydration temperature in this
case is lower than the corresponding one in model (A) under the same rock-
pressure, and decreases with decreasing P Ha o. Equation (2-ll) is applicable to
this case also.
The general shapes of eqUilibrium curves are shown in fig. 2-3. The curves for
PH 2 0 = constant shown there meet the curve for P s = PH a 0 at their lower end
where Ps is equal to the pertinent constant value of PHa o. The difference
between PHz 0 and Ps becomes greater upwards along the curves of constant
48/BASIC CHARACTERISTICS OF METAMORPHIC REACTIONS

Dehydrated
Minerals
+H 2 0

Hydrated
Minerals

Temperature T
Fig. 2-3 Common forms of equilibrium curves for dehydration reactions
under the conditions of Ps = PH, 0 and of constant PH, o. (Thompson,
1955; Greenwood, 1961.)

PH, 0, which usually have a negative slope. In other words, increasing Ps when
PH,o = constant tends to dehydrate minerals.
The relation between the slope of a curve under a constant PH, 0 and that of
the corresponding curve under Ps = PH, 0 is given by the following relation
(Greenwood,1961):

('Op)
a; P =
10·0 AS
.lVs
(2-12)
H,O

When (dP/dT)p g = PH, 0 is known from a synthetic study, the slope of the curve
for a constant PH,o can be directly calculated by this equation. In this
equation, P and AS are in bars and J K -1 mol- 1 respectively.

2-5 MINERAL ASSEMBLAGES AND ELEMENT DISTRIBUTION

It is instructive here to call attention to the effect of the presence of quartz on

the P-T relations of metamorphic reactions. Equilibria for equations (2-4) and
 2-5 MINERAL ASSEMBLAGES AND ELEMENT DISTRIBUTION /49

(2-10) represent the high temperature stability limit of jadeite and muscovite
respectiveiy. If jadeite or muscovite is associated with quartz, reactions would
take place as shown by equations (2-3) and (2-9) respectively. Thus, the stability
fields of these minerals are reduced, as shown in figs. 2-1 and 2-2.
Generally speaking, if quartz can participate in a reaction, its presence will
change the equilibrium temperature. This effect is very important in meta-
morphism, since quartz is widespread and is able to participate in a large number
of reactions (Fyfe, Turner and Verhoogen, 1958; Miyashiro, 1960).
The stability field of a mineral is widest when the mineral is present alone. A
lot of synthetic studies have been made to clarify the breakdown temperature
and pressure of single minerals. Such studies yield only the maximum stability
fields of the pertinent minerals. The stability fields of mineral assemblages
containing these minerals are the same as or smaller than the maximum stability
fields.
The recent progress of metamorphic petrology has enhanced the importance
of mineral assemblages. Description of metamorphic minerals without due regard
to their paragenetic relations is now becoming less important. If two coexisting
minerals in equilibrium have a component in common, Nernst's partition law
holds for the distribution of the component as follows:
x
-=K (2-13)
Y
where x and y represent the mole fractions of the component in the two
minerals, which are regarded as ideal solutions. The partition coefficient K is a
function of temperature and rock-pressure.
The chemical compositions of solid solution minerals should vary with the
bulk chemical composition of their host rock. If two ferromagnesian minerals
coexist, an increase in the Mn/(Mg+Fe2++Mn) ratio of the rock, for example,
should induce an increase of the same ratio in both minerals. The Mg-Fe
distribution between associated ortho- and clinopyroxenes, for example, can be
treated from the viewpoint of equilibrium for the reaction:

(2-14)
The values of atomic ratio Fe/(Mg + Fe) in ortho- and clinopyroxenes are
designated as x and y respectively. If the two minerals are regarded as ideal
solutions, we have the following relation:

In(_X
I-x
)j(---L-)
1
-y
=_
RT·
llG (2-15)

Here, llG represents the free energy change for equation (2-14) under the
condition that each term indicates an end-member mineral. This relation may
SO/BASIC CHARACTERISTICS OF METAMORPHIC REACTIONS

be rewritten as follows:
(Fe/Mg) in orthopyroxene =ex (-!J.G/.'RT) =K T ~ (2-16)
(Fe/Mg) in clinopyroxene p ( , s)

where (Fe/Mg) represents an atomic ratio, and the equilibrium constant K is a

function of temperature and rock-pressure (Mueller, 1960; Kretz, 1961).
The graphical analysis of the relationship between mineral paragenesis and
bulk chemical composition is important for a better understanding of thermo-
dynamic relations. This problem will be treated in chapter 5.

2-6 DECARBONA TION REACTIONS

The metamorphism of calcareous rocks liberates CO 2 , as is exemplified by the

follOWing equation:
CaC0 3 + Si0 2 = CaSi0 3 + CO 2 (2-17)
calcite quartz wollastonite

The equilibrium curve for this reaction under the condition ofPco 2 =Ps is shown
in fig. 2-4.

~O~O~-----6~O~O~----~70bO~----~8~OO~
Temperature (DC)

Fig. 2-4 Equilibrium curve for the reaction to form wollastonite (2-17)
in the presence of a fluid phase with the composition of pure CO 2 •
(Harker and Tuttle, 195(i.)

The equilibrium curves for decarbonation reactions show a marked bend in

the pressure range of 0,1-2,0 kbar just as those for dehydration reactions do.
Calcite has a low density, and hence, the !J.Vs of decarbonation reactions usually
has a large negative value. When Pco 2 =Ps , the equilibrium curves may tum to
have a negative slope at pressures above several kilobars.
The thermodynamiCS of decarbonation reactions has been discussed by
Danielsson (1950) and Weeks (1956). The entropy increase of decarbonation is
 2·7 OXIDATION AND REDUCTION IN METAMORPHISM/51
roughly similar to that of dehydration. At 298 K and 1 bar, the former is about
170 J K- 1 mol- 1 of liberated CO2 , whereas the latter is about 150 J K- 1 mol- 1
of liberated H2 0. The equilibrium curves can be calculated by equation (2·11) in
whichPcoa is substituted for P Ha O.
In calcareous rocks undergoing metamorphism, a fluid phase if any would
usually be a mixture of CO2 and H2 O. Equilibria for mixtures of CO 2 and H2 0
will be treated in detail in chapter 9.

2·7 OXIDATION AND REDUCTION IN METAMORPHISM

Iron Oxides and O2
In the atmosphere, oxygen is mostly in the free molecular state with a partial
pressure of about 0·2 bar near the surface of the earth, whereas within the crust
and uppermost mantle it is mostly in combined forms as silicates and oxides.
Just as in the case of H2 0, we can imagine here a small cell with rigid walls,
permeable only to O2 , which is contained in a rock. The oxygen in the cell is at
a pressure Po 2 ' being in osmotic equilibrium with that outside.
The most useful indicators for Po 2 in the crust are iron oxides, i.e. haematite,
magnetite and wiistite. These minerals are related to one another by the
follOWing equations:
6 Fe203 = 4 Fe304 + O2 (2.18)
haematite magnetite

2 Fe304 =6 FeO + O2 (2·19)

magnetite wiistite

2FeO = 2 Fe +0 2 (2.20)
wiistite native iron

! Fe304 =!Fe +0 2, (2.21)

magnetite native iron

P02 for these equilibria can be calculated by the following equation:

lld'"= -19·14 TlogPo a - PsllVs/lO·O (2.22)

Here, llG~, T, and llVs represent the free energy of reaction (J) under 1 bar,
absolute temperature, and the volume change of solid phases (cm 3) respectively.
POa andPs are in bars. The last term is small compared with the others, and may
usually be neglected (Ernst, 196(;); Miyashiro, 1964). Note that equation (2-22)
has the same form as equation (2-11). The calculated equilibrium curves are
shown in fig. 2·5.
As is clear from the figure, haematite is stable at high Po 2 . AsPo 2 decreases,
magnetite, wllstite, and finally native iron become stable in turn. Haematite
52/ BASIC CHARACTERISTICS OF METAMORPHIC REACTIONS
occurs in some and magnetite in many metamorphic rocks, but wiistite and
native iron do not. nus suggests thatP02 in rocks undergoing metamorphism is
usually of the order of 10-10 to 10-40 bar, that is, much lower than the partial
pressure of O2 in the atmosphere.
Fig. 2-5 indicates that the magnetite + haematite assemblage show a deftnite

'2 \<-naf _ - - - - -----

-5
t>\lfe '.Nate
f
_ - -, - - - , naf _
.",."'- \J\Iatet • - ..... -

-10 .".
.. ............
....-
.. .",..""
t>~~--- (2.18)

........ Haematite
.....

-15

-.::
a'" -20
rE'"
CD
.2

Native Iron

-30

-35
600 700 800 900 1000 1100
Temperature (OC)

Fig. 2-5 Stability relations of oxides of iron and native iron. Curves
(2-18), (2-19), (2-20) and (2-21) represent equilibria for the equations
which are denoted in the text with these numbers. The effect of pressure
on solid phases is neglected. (Miyashiro, 1964.) 'Wus' means the wiistite
field.

P02 at a deftnite temperature (and rock-pressure), so far as the two minerals

belong to the Fe-O system. Natural magnetite and haematite contain some Ti0 2
(see e.g. ftg. 8B-6). Under such a situation, the three-mineral assemblage
magnetite + haematite + ilmenite has a definite P02 at a specifIc temperature
and rock-pressure. The effect of Ti0 2 , however, will be neglected here for
simplicity of discussion.
 2-7 OXIDATION AND REDUCTION IN METAMORPHISM/53

Dissociation of HzO
The Hz 0 present in rocks undergoing metamorphism would be approximately in
dissociation equilibrium with Hz and Oz as follows:
HzO ~Hz + ,Oz (2-23)

A~ =-19·14 Tlog[(PH , . .yPo, )/PH, 0]. (2-24)

Here, PH" Po, and PH, 0 are in bars. If we assume that an aqueous fluid has
the pure water composition, i.e. an atomic ratio of H : 0 equal to 2, Po,
becomes a function of PH, 0 and T. The calculated.curves of Po, for PH, 0 = 1
bar and 2 kbar are shown in fig. 2-5. These curves indicate that if an aqueous
fluid has the pure water composition, haematite is to be stable for almost the
entire temperature range of metamorphism so long as PH, 0 is higher than 1 bar.
This is not the case, since magnetite is very widespread (§ § 7B-21 and 8B-12).
Therefore, Po, in such rocks should be lower than that produced by
dissociation of pure water. In other words, Hz must be present in excess of the
pure water composition. Such a reducing condition may have been fostered
partly by the conceivable ascent of Hz and CO from the mantel of the earth.
However, it is more likely that the reducing condition is produced within the
crust by organic materials and graphite. Such substances are almos.t omnipresent
in metapelites.

Role of Graphite
Black carbonaceous matter in metamorphic rocks has usually been called
graphite in the geologic literature. Recent investigations have shown, however,
that the carbonaceous matter in low-temperature metamorphic rocks are organic
materials comparable to semi-anthracite and anthracite. Their composition and
atomic structure change gradually with increasing metamorphism, until graphite
is formed in the amphibolite facies (Izawa, 1968). Since the individual com-
pounds in them have not been identified, assignment of numerical values to their
reducing power is not possible. As an approximation, we will take up graphite.
The following equilibrium would hold between graphite and Oz:
C + O2 = CO 2 . (2-25)
By neglecting the effect of Pg , we have
Peo
A~= -19·14 Tlog::....::::::1.. (2-26)
Po,
Here, Peo , and Po , are in bars, and AG" is in J. Thus, Po 2 is a function of
Pco , and T. The value of Peo , within the crust could not be greater than 10
kbar. The calculated curves of Po, for the conditions Peo , = 10 bar and 10
kbar are shown in fig. 2-5. These curves indicate that the values of Po, in the
54/BASIC CHARACTERISTICS OF METAMORPHIC REACTIONS

presence of graphite is within the magnetite field over almost the entire
temperature range of metamorphism (Miyashiro, 1964).
In rocks containing graphite (or organic matter) and H20, not only CO2 , H2
and O2 but also CO and CH4 (methane) are formed by such reactions as CO2 =
CO + !02 and C + 2H2 = C~. Under a relatively reducing condition, the
amount of CH4 can be considerable (French and Eugster, 1965). The presence
of CH 4 has been detected in bore-holes made through metamorphic rocks in the
Bessi and other mines in the Sanbagawa belt (Izawa, 1968).

Effect and Mobility of O2

An FeO-bearing compound having a defmite composition is stable over a range
of Po: at a defmite temperature and rock-pressure. For example, fayalite
(Fe2Si04) is stable in the range of Po: = about 10-16 - 10-26 bar at 700°C
(with a slight variation with rock-pressure). If POa is higher or lower than this
range, fayalite decomposes to produce magnetite or native iron, as follows:
2 Fea04 + 3 Si0 2 =3 Fe2Si04 + O2 (2-27)
magnetite quartz fayalite

(2-28)
fayalite iron quartz
Similarly, a silicate solid solution containing FeO is stable in a range of Po:
at a definite temperature and rock-pressure. As Po: increases, the FeO-bearing
component gradually decomposes to produce magnetite and other products.
Therefore, the FeO/MgO ratio of a silicate solid solution associated with
magnetite and other decomposition products is an indicator of Po: . The higher
the Po : ,the poorer in FeO becomes the silicate solid solution (Mueller, 1960).
In contrast to H20 and CO 2 , which are relatively highly mobile during
metamorphism, O2 and H2 are not as mobile (Thompson, 1957; Eugster, 1959;
Mueller, 1960; Chinner, 1960). In metamorphosed iron formations in the
Canadian Shield, a haematite-bearing layer is found to be in contact with a
magnetite-bearing layer. The actinolite in the former layer is lower in FeO/MgO
ratio than that in the latter. This suggests a higher P02 in the former. The
boundaries between such layers are sharp in some examples and gradational in
others. At any rate, such relations indicate that the mobility of O2 in rocks
undergoing metamorphism is extremely small.
Equation (2-24) indicates the dependence of PHa on POa and PH: O. If Po:
varies greatly across a short distance, PH2 also must do so. Accordingly, the
mobility of H2 was also very small in the above case. Such small mobilities of O2
and H2 are probably a result of a very low value of Po: and of PH:. The
gradients of concentration for O2 and H2 were probably too small to cause
effective diffusion.
It has been shown in some terranes that the Fe 3+ /Fe2+ ratio of metamorphic
 2-7 OXIDATION AND REDUCTION IN METAMORPHISM Iss
rocks tends to decrease with increase in temperature of metamorphism (e.g.
Shaw, 1956; Miyashiro, 1958). Probably this is not due to long-distance flow or
diffusion of O2 , In pelitic rocks undergoing metamorphism, Fe20rbearing
minerals lose some oxygen by reaction with organic matter or graphite to form
CO 2 , The CO 2 would flow or diffuse over a long distance, leading to a general
decrease of the Fe 3+ IFe2+ ratio of the rocks. Flow of an intergranular fluid or
diffusion would have spread the reducing condition of metapelites to the
surrounding rocks. Such a flow or diffusion should not be uniform, and the
degree of reduction should be variable in metamorphic rocks. It would cause
diversity in mineral composition. Some metapelites were found to. contain no
graphite and hence to show highly variable degrees of reduction with resultant
generation of variable mineral compositions (Chinner, 1960).
In experimental work, Po 2 can be controlled by using a set of oxygen buffers
(Eugster and Wones, 1962; Ernst, 1960). For example, a buffer composed of a
mixture of magnetite and iron keeps P02 at a value corresponding to the
equilibria of reaction (2-21). The curves of Po 2 for common buffers are shown
in fig. 7 B-1 O. The fayalite-magnetite-quartz buffer, which corresponds to the
equilibria of (2-27), gives P02 values within the magnetite field, and hence
relatively close to those in ordinary pelites undergoing metamorphism.
Chapter 3

Basic Concepts of Metamorphic

Petrology

3-1 PROGRESSIVE REGIONAL METAMORPHISM

Concept of Progressive Metamorphism

Metamorphism usually takes place in response to a rise of temperature.
Progressive changes brought about in rocks by rising temperature are called
progressive metamorphism. Successive rises of temperature bring about a series
of reactions, which might be understood as changes of mineral assemblages
approximately at equilibrium. In many cases, however, some minerals which
formed at a lower temperature remain as unstable relics in the core of minerals
that formed later.
The progressive nature of metamorphic changes was strongly emphasized by
Alfred Harker (1932) in order to characterize his line of approach or attitude to
the study of metamorphism. He regarded the old German school of petrography,
particularly Grubenmann (1904-6), as being too static, classificatory and
descriptive. Grubenmann classified metamorphic rocks according to their
chemical composition and the assumed depths of formation. G,enetic relations
were mostly ignored. In contrast, Harker characterized his line of approach as
genetic, and attempted to treat metamorphism as a sequence of progressive
changes in response to changing conditions. The Harker line of approach is
generally adopted in this book.
The effectiveness of this approach results mainly from the empirical fact that
metamorphic rocks, recrystallized at progressively increasing temperature, are
exposed side by side so that the sequence of changes can be confidently
followed in the field. It has been found in many regional metamorphic terranes
that the temperature of metamorphism is distributed in a relatively simple,
regular pattern, as exemplified by fig. 1-2. In a metamorphic belt as a whole, the
temperature increases toward the thermal axis, and a series of zones can usually
be delineated, each one representing a fairly definite range of temperature.
Since metamorphism is a long process involving a number of phases, it is
important to recognize which phase is recorded in the mineral composition of
metamorphic rocks. It is widely believed that reactions take place easily with
 3-1 PROGRESSIVE REGIONAL METAMORPHISM/57

rising temperature, but do not occur as easily with declining temperature (e.g.
Harker, 1932; Fyfe, Turner and Verhoogen, 1958). Insofar as this holds true, the
mineral compositions of metamorphic rocks are usually representative of the
state of the highest temperature reached by the individual rocks.
However, it is also common for some mineral changes to take place in the
declining phases of metamorphism. We can recognize such changes usually
without great difficulty through observation of the textural relations of the
minerals.

Ba"ovian Zones as a Classical Example

George Barrow, a British geologist, was the first to succeed in showing the
existence of progressive mineral zoning on a regional scale in a metamorphic
terrane. His work was carried out toward the end of the nineteenth century in a
part of the Dalradian Series in the Scottish Highlands, about 50 km southwest of
Aberdeen. As shown in fig. 3-1, granitic rocks are exposed in the north of the
area, and pelitic schists and gneisses in the south. He found that sillimanite-
bearing metapelites were exposed in a particular zone in the northernmost part

River South Esk

..
t[;f~ Newer Granites Kyani e zone

v:·:. :] Permea t ion area intrUSIve mass s

o
S aurolite zone
: ...:.,. (Older Granites)

Sillimani te zone o 5 10 km
....J'
'1 0 . _ _. . . .' . _ _

Fig. 3-1 Classical metamorphic terrane mapped by Barrow (1893) in the

Scottish Highlands. (Simplified from the original map.)
S8/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

of the metamorphic terrane, kyanite-bearing metapelites occurred in an adjacent

zone to the south, and staurolite-bearing ones still further south (Barrow, 1893).
He considered that the prevailing temperature had been successively higher
toward the north.
About twenty years later, Barrow extended his survey of progressive meta-
morphic zones to the southern limit of the terrane, the Highland Boundary fault
(Barrow, 1912). The general validity of his work was later confirmed by Tilley
(I92S) and Harker (1932). The following series of metamorphic zones (fig. 3-2)
have been delineated on the basis of mineral changes in metapelites in response
to increasing temperature:
1. Chlorite zone, characterized by chlorite-muscovite phyllite or schist.
2. Biotite zone, characterized by the appearance of biotite, which persists to all
the higher zones. The boundary between the chlorite and the biotite zone,
i.e. the line that marks the first appearance of biotite is called the biotite
isograd.
3. Almandine zone (garnet zone), characterized by the appearance of almandine
garnet, which persists to the higher zones. The boundary between the biotite
and the almandine zone is called the almandine isograd.
4. Staurolite zone, characterized by the appearance of staurolite.
5. Kyanite zone, characterized by the appearance of kyanite:
6. Sillimanite zone, which is on the high-temperature side of the sillimanite
isograd marking the first appearance of sillimanite and the disappearance of
kyanite. This change corresponds to reaction (2-1) of § 2-2.
The zones in this and similar series, recognized for pelitic rocks, are called
Barrovian zones. Pelitic rocks tend to show textural changes with rising
temperature as follows: phyllites -+ mica schists -+ gneisses. Barrovian zones have
been found in several other areas in the world.
In general, an isograd may be defined as the line of outcrops on which a
mineral assemblage begins to appear or disappear. It represents the intersection
of the surface of a mineralogical change within the geosynclinal pile with the
present topographic surface. The metamorphic rocks present within a zone that
is limited on both sides by two isograds are considered to have recrystallized
within a definite range of temperature and rock-pressure.
In the Dalradian metamorphics, the original geosynclinal pile contained
pyroclastic beds and dikes of basic composition. All these rocks were subjected
to metamorphism (e.g. Wiseman, 1934). The mineral compositions of the
metabasites have been correlated with those of the surrounding metapelites.
Zonal mapping of metamorphic terranes may well be based on rocks of any
chemical composition. Basic and calcareous rocks are sensitive to changing
temperature, and no less appropriate for this purpose than pelitic ones, if they
are widespread in the terrane under consideration.
 3-1 PROGRESSIVE REGIONAL METAMORPHISM /59
Barrovian Region of Scottish Highlands
Greenschist Epidote-
Metamorphic facies amphibolite Amphibolite facies
facies
facies

Mineral zoning Chlorite and Almandine Staur. and Sillimanite

biotite zones zone kyan. zones zone

Sodic plagioclase
Interm. and calcic
.....,.,'"
'in
plagioclase
Epidote -- ---
.,
...
.c Actinolite
-Blue-green
Green
and brown
'"
;2;
Hornblende
Chlorite --
Almandine ~----- ------ -----
Chlorite --
Muscovite
Biotite
'"'"
.'!:! Almandine
a;
....,c-
Staurolite
Kyanite
'"
;2;
Sillimanite
Sodic plagioclase
Quartz

Fig. 3-2 Progressive mineral changes in the region of the Barrovian zones
in the Scottish Highlands. (Harker, 1932; Wiseman, 1934.)

Significance of Isograds
In the progressive metamorphism of pelitic sediments, a series of reactions
proceeds with successive rises of temperature. The medium- and high-
temperature parts of regional metamorphic terranes usually approach a state of
chemical equilibrium. Isograds are defined in terms of mineral variations in the
near-equilibrium state.
Toward the lower temperature, a metamorphic terrane grades imperceptibly
into an unrecrystallized area. The low-temperature limit of a metamorphic
terrane depends on the rate of, and the time available for, recrystallization.
Such a boundary is not an isograd in the proper sense.
Isograds give a general picture of the temperature distribution in a meta-
morphic terrane. Strictly speaking, however, a definite isograd (e.g. the
sillimanite isograd) does not necessarily represent a definite temperature. The
sillimanite isograd in the Barrovian zones is due to a solid-solid reaction and
represents a definite temperature only under a definite rock-pressure (fig. 2-1).
The isograds defined in terms of dehydration reactions depend on Ps and PH, 0,
both of which may vary within a terrane.
60/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

Isograds may be classified into three groups as regards the character of the
relevant reactions. One group is characterized by the true appearance or
disappearance of a mineral due to the stability limits of the mineral by itself.
The sillimanite isograd, for example, indicates the disappearance of kyanite and
the appearance of sillimanite. The breakdown of muscovite by equation (2-10) is
another example that might be used as such an isograd.
The second group of isograds is due to a change in tielines in a paragenesis
diagram. All the individual relevant minerals are stable by themselves on both
sides of an isograd, but a tieline between two minerals becomes unstable with
respect to another tieline.
The third group of isograds is due to a change of the composition field of
solid solution minerals. Examples of the second and third groups will be
discussed in detail in relation to the formation of biotite in § 7B-7 (fig. 7B-6).
In all the three groups, variations in Fe 2+/Mg2+ and Mn 2+/Fe 2+ ratios should
cause a change in reaction temperature. Apparent isograds observed in some
progressive metamorphic terranes are actually due to variations in chemical
composition of predominant rocks.
The term grade of metamorphism has been used widely in the geologic
literature. In a current common usage, a low and a high grade designate
petrologic effects of low and high temperatures respectively. So far as the term is
used for comparison of rocks within a single progressive metamorphic terrane, it
has a clear and empirically determinable meaning. When applied to rocks in
different regions, its meaning is not always clear, since entirely reliable values of
recrystallization temperature are not known for most metamorphic rocks. Many
geologists called metamorphic rocks recrystallized at higher rock-pressures,
high-grade rocks. If this usage were to be accepted, we could not tell which of
the following is higher in grade: rocks recrystallized at high pressure and low
temperature, or rocks recrystallized at low pressure and high temperature.

Compositional Changes in Solid-Solution Minerals

In the chlorite and biotite zones of the Dalradian and other similar metamorphic
areas, the stable plagioclase is albite regardless of the composition of the
host-rocks. Some of the original rocks (e.g. basaltic ones) must have contained
intermediate or calcic plagioclase. When they were subjected to low-temperature
metamorphism, the anorthite molecule of the plagioclase was converted into
epidote with the resultant formation of the albite+epidote assemblage. With a
rise in temperature of metamorphism, the albite reacts with epidote to form
intermediate or calcic plagioclase, as is the case with metabasites of the kyanite
and sillimanite zones (§ 8B-3). If epidote is very scarce or absent, albite will
persist into the high-temperature zones.
As epidote contains H2 0 and Fe203, the reaction of albite and epidote to
form more calcic plagioclase is affected by PH • O and po •. As epidote is not
 3-1 PROGRESSIVE REGIONAL METAMORPHISM /61
strictly isochemical with anorthite, the compositional difference must be
adjusted by a change in composition of associated minerals, and thus the
reaction is affected by the associated minerals. Therefore, the anorthite content
of plagioclase accompanied or unaccompanied by epidote does not increase
regularly with rise in temperature in a progressive metamorphic terrane.
Moreover, a discontinuous change in composition appears to take place in sodic
plagioclase (§ 78-8).
The monoclinic amphibole, stable in ordinary metabasites of the chlorite and
biotite zones, is actinolite, whereas that of higher temperature zones is common
hornblende with higher Ah 0 3 and Na20 contents. The colour (Z) of common
hornblende usually changes from blue-green to green or green-brown with rise
in temperature (§ 884).
In an isochemical series of metamorphic rocks, the amounts of minerals and
the distribution relation of elements in a particular pair of minerals should
change with temperature. If such rocks are in equilibrium, the chemical
composition of a mineral is a function of the bulk chemical composition of the
host rock, the amounts of the minerals present, and the distribution relations of
elements. A progressive compositional change of a solid solution mineral takes
place owing to a change in the amounts of minerals and in the distribution
relations of elements. The progressive compositional change of SOn:le minerals is
mainly controlled by such eqUilibrium relations (Miyashiro and Shido, 1973).
On the other hand, the composition of garnet, particularly formed at lower
temperatures, appears to be mainly controlled by disequilibrium growth. Garnets
in low-temperature metapelites are usually rich in MnO. The MnO content tends
to decrease with rising temperature (§ 7B-10). Such garnets usually show strong
compositional zoning with the MnO content decreasing to the rim. Hollister
(1966) and Atherton (1968) have shown that the MnO decrease in some garnets
can be semi-quantitatively explained by making the assumption that chemical
diffusion in garnet crystals is negligible and only the outermost layer of garnet is
in equilibrium with the surroundings. In such a case, garnet, once formed, is
immediately removed from the reacting system resulting in a progressive change
in the effective bulk chemical composition of the reacting system, and hence a
progressive change in the composition of newly formed garnet. This will be the
case even if there is no noticeable change in the distribution relation of MnO
between garnet and associated minerals.

Migration of Materials
The kind, extent and mechanism of the migration of materials in rock complexes
undergoing regional metamorphism have been subjects in dispute for many
years. Some geologists envisaged large-scale migration of various materials, while
others denied it. The migration, if it occurs, may cause a change in bulk chemical
62/BASIC CONCEPTS OF METAMORPHIC PETROLOGY
composition of rocks. The compositional change of rocks caused by the
introduction of materials is called metasomatism (Goldschmidt, 1922).
Decreases in H2 0 and CO2 contents and in Fe203/FeO ratio are the most
commonly observed compositional changes in progressive regional meta-
morphism (Shaw, 1956; Miyashiro, 1958; Engel and Engel, 1958, 1962a;
Vallance, 1960). H2 0 molecules liberated by dehydration of minerals should
either flow or diffuse out of the rock complex. Generally, most of the H2 0
would migrate upwards to the surface of the earth.
Not only liberated H2 0 but also H20 of magmatic or mantle origin may
move through a rock complex undergoing regional metamorphism. In the
progressive metamorphic area of Dutchess County in southeastern New York it
has been demonstrated that the 18 0/ 16 0 ratios of whole-rock samples of
metapelites tend to become lower with an increase in temperature of meta-
morphism, ultimately approaching that of pegmatitic granite and plutonic rocks
in the highest temperature zone. The 18 0/ 16 0 ratios of the original pelites were
probably high, and simple dehydration does not change the ratio so markedly.
Probably H2 0 ofmagrnatic, or some other deep-seated origin with low 18 0/ 16 0,
has been mixed and moved through the metamorphic terrane, resulting in a
significant change of the 18 0/ 16 0 ratio of metapelites on a regional scale. It is
interesting that a quartz-rich pod enclosed in limestone, and an amphibolite,
showed a much higher and lower 18 0/ 16 0 ratio respectively than the sur-
rounding metapelites. Probably the moving H2 0 did not permeate into the
limestone and amphibolite sufficiently to homogenize their 18 0/ 16 0 ratios.
Similar relationships have been noticed in some other metamorphic terranes
also. In other areas, however, metapelites show a more or less irregular
distribution of 18 0/ 16 0 ratios probably owing to an incomplete exchange
reaction of oxygen isotopes (Garlick and Epstein, 1967; Epstein and Taylor,
1967).
CO 2 which is mainly produced by decarbonation would also move through
rock complexes during metamorphism. A decrease in Fe203/FeO would be due
to reduction of ferric iron by organic matter or graphite to result in the
production of CO 2 and H2 o. There is evidence suggesting that the mobility of
O2 is very low (§ 2-7).
It has been claimed in a number of metamorphic terranes that introduction of
Na20 and/or K20 on a regional scale has taken place in the process of
transformation of slates and phyllites to gneisses (e.g. Goldschmidt, 1921;
MacGregor and Wilson, 1939; Read, 1957; Miyashiro, 1958). Such introduction
may be valid in some cases, and a possible mechanism was proposed by Orville
(1962). The hitherto available evidence for the introduction is, however, mostly
equivocal and the observed variation in composition can be well or better
accounted for by the original diversity of the relevant rocks (e.g. Milller, 1970;
Muller and Schneider, 1971; Miyashiro, 1967a). A marked constancy in the
 3-1 PROGRESSIVE REGIONAL METAMORPHISM/63
chemical composition of metamorphic rocks has been demonstrated in some
terranes ranging from a slate to a gneiss zone (e.g. Shaw, 1956; Uno, 1961;
Vallance, 1960, 1967; R. W. White, 1966).
Table 3-1 shows average chemical compositions of unmetamorphosed and
metamorphosed pelitic rocks. Between two groups of pelitic rocks which have
undergone different intensities of metamorphic recrystallization, significant
differences are noticed only in HzO, COz and CaO contents, and in FeZ03/FeO
ratio. The difference in CaO is probably due to collecting bias of samples (Shaw,
1956).

Table 3-1 Average compositions of pelitic rocks of the world. After

Shaw (1956)

Clays, shales, and slates Phyllites, schists, and gneisses

(85 analyses) (70 analyses)
Average Standard Standard
deviation Average deviation

SiOz 59·93 6·33 63·51 8·94

TiO z 0·85 0·57 0·79 0·67
Alz 0 3 16·62 3·33 17·35 5·08
Fez0 3 3·03 2·08 2·00 1·66
FeO 3·18 1·84 4·71 2·44
MgO 2·63 1·98 2·31 1·82
CaO 2·18 2·54 1·24 0·92
Na20 1·73 1·27 1·96 1·06
K 20 3·54 1·33 3·35 1·31
H2O 4·34 2·38 2·42 1·53
CO 2 2·31 a 2·60 0·22 b 0·22

a Determined in only 43 analyses.

b Determined in only 19 analyses.

NazO, KzO, CaO, Alz03 and SiOz can move in solution in an aqueous fluid
at relatively low temperatures. If a fluid is not available, migration may take
place by diffusion in solids. However, the effect of such diffusion would be very
limited. In the Abukuma plateau, Japan, grains of corundum and quartz, only
0·3 mm apart, were found to remain unchanged in an amphibolite facies zone,
other incompatible assemblages commonly occur at a distance of a few
millimetres in a low amphibolite facies zone and of a few centimetres in a higher
amphibolite facies zone in the same region (Miyashiro, 1958, p. 262). The
common occurrence of zoned plagioclase also suggests a narrow range of
diffusion in solids.
At temperatures higher than the middle of the amphibolite facies, however,
partial melting would take place. The resultant melts are generally of more or
64/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

less granitic composition and may move to permeate or inject at some places,
leaving solid residues behind. This may cause drastic changes in the bulk
chemical composition of the relevant rocks.
In regional metamorphic terranes, quartz veins occur commonly in low
temperature zones (especially in those of the greenschist facies), whereas
pegmatite usually occurs in high-temperature zones (more particularly in those
of the amphibolite facies). Such relationships suggest that their formation is due
to migration associated with the metamorphism of the surrounding rocks
(Ramberg, 1948).
In Dutchess County, New York, quartz veins and pegmatites tend to occur in
lower and higher temperature zones respectively of regional metamorphism. The
18 0/ 16 0 ratio of quartz in the veins and pegmatites decreases with temperature
rise in accordance with that of the surrounding metamorphic rocks. This suggests
that quartz veins and pegmatites are genetically related to their surrounding
rocks (Garlick and Epstein, 1967). Quartz veins could form through the activity
of an aqueous fluid, whereas many pegmatites could form by the consolidation
of magma produced by the partial melting of pelitic and psammitic metamorphic
rocks.

Possible Behaviour of H 2 0 in Progressive Metamorphism

Newly deposited pelitic sediments have a high proportion of open spaces fllied
with an aqueous fluid. With increasing depth of burial, the porosity decreases,
reaching a value of several per cent usually at a depth of few kilometres. The
decrease is mainly due to mechanical squeezing-out of fluid combined with
cementation. Rise of temperature in burial and incipient metamorphism induces
recrystallization to cause generally upward expulsion not only of intergranular
fluid but also of potential water that is contained in minerals as OH-, H20 and
H3 0+. The intergranular fluid would be continuous through long narrow
channels for a great vertical distance, possibly even to the surface of the earth.
This situation corresponds to model (C) of § 2-3.
Some rocks in a geosynclinal pile such as unaltered basalt do not have such a
high initial content of H2 O. However, pelitic and other hydrous rocks are the
main constituents of ordinary geosynclinal piles. Hence, a geosynclinal pile as a
whole should usually have a high original H20 content. In the process of
recrystallization at low temperatures, H20 should move from hydrous rocks to
some less hydrous ones in the vicinity to produce new hydrous mineral
assemblages in the latter.
Progressive rise in temperature during metamorphism should cause gradual
dehydration of minerals and reduce intergranular spaces, thus tending to cut off
channels of fluid. Individual intergranular spaces would be isolated from one
another. The fluid filling them would be nearly at the same P and T as the
surrounding solids. This situation corresponds to models (A) and (B) of § 2-3.
 3-1 PROGRESSIVE REGIONAL METAMORPHISM /65

Deformational movement and volume increase by dehydration should tend to

produce new paths of flow for fluid during metamorphism. Mechanical flow
combined with chemical diffusion should lead to a considerable extent of
regional migration of H20, as suggested by oxygen isotope studies as well as to a
progressive decrease of H20 content. The migrating H20 should tend to move
generally upwards, partly reaching the surface of the earth.
Prolonged metamorphism at relatively high temperatures may cause a virtu-
ally complete disappearance of the intergranular fluid phase, though discrete
H20 molecules should still be present in an adsorbed state along the inter-
granular surfaces. Under this condition, PH 2 0 within the rock may become
lower than the value that would be obtained in the presence of an aqueous fluid
phase. This situation corresponds to model (D) of § 2-3.
In some cases the temperature of metamorphism may become high enough to
cause partial melting of rocks. Such a melt phase should absorb a large part of
H20 present in the rock if the system is nearly closed with respect to H2 O. This
should result in an abrupt decrease of PH2 0, which may lead to the breakdown
of hydrous minerals. The breakdown should to some extent give a buffering
effect to PH. o. If the system is ideally open for H20, such a change in PH. 0
does not take place. It is conceivable that the formation of strongly dehydrated
granulite facies mineral assemblages observed in high-temperature regional
metamorphic terranes is usually promoted by a decrease of PH. 0 due to the
beginning of partial melting.
If a rock mass formerly metamorphosed at medium or high temperatures is
subjected repeatedly to metamorphic recrystallization at a later geologic time,
the initial H2 0 content of the mass should be small and this would cause a
condition different from that in the case discussed above.

Scrutiny of Complex History

The high-temperature zones of progressive metamorphism must have passed
through low temperatures in the course of temperature rise. However, recrystall-
ization may not have taken place at the low temperatures if the rise was rapid.
Even if recrystallization took place, the equilibrium state in such cases may not
have been identical to the state of the low-temperature zones now observed of
the same region. Different location in a geosynclinal pile could have resulted in
differentPs and PH, o.
In certain metamorphic terranes, original sedimentary structures such as false
bedding are known to be completely destroyed in low-temperature zones, but to
be preserved in some high-temperature zones. A greater degree of deformational
movements may have taken place in the lower-temperature zones (Read, 1957,
pp. 24-5, 31-2).
66/ BASIC CONCEPTS OF METAMORPHIC PETROLOGY

It appears common that metasomatism is more intense in low-temperature

zones than in the higher temperature ones of the same area. The existence of a
flowing fluid phase in the low-temperature part may cause intense chemical
migration. The higher temperature zones could have passed low-temperature
conditions too rapidly to undergo any appreciable metasomatism.
The history of a progressive metamorphic terrane may be much more
complicated than might be considered on a cursory investigation. Detailed
investigations of apparently progressive metamorphic terranes in the Akaishi and
Tanzawa Mountains in Japan have revealed their complicated and auto-
cannibalistic histories (Matsuda and Kuriyagawa, 1965; Seki, Oki, Matsuda,
Mikami and Okumura, 1969a). Higher temperature metamorphic rocks had
formed at an early stage of an orogeny. They were later exposed and eroded, and
their fragments were transported, to be deposited in an adjacent area. Then the
new deposits were metamorphosed at a later stage of the same orogeny at lower
temperatures, with a resultant generation of a zeolite facies zone which is
apparently continuous with the older higher temperature ones (§ 6A-2).
Recent analysis of deformation phases and accumulation of radiometric age
data have also revealed that some apparently simple metamorphic terranes have
complex histories. Different parts of a terrane may have different histories and
an apparently simple progressive metamorphic terrane may be the product of
overprinting of two or more episodes of metamorphism. Even the Dalradian
Series may be polymetamorphic (e.g. Brown, Miller, Soper and York, 1965). If
the temperature of recrystallization is highest in the last phase, however,
minerals formed in earlier phases may be entirely reconstructed. In such a case,
it may be justified to ignore the earlier phases as far as mineral reactions are
concerned.

3-2 OUTLINE OF METAMORPHIC FACIES

Metamorphic Facies
The concept of metamorphic facies was first proposed by Eskola (1920, 1939)
and was later modified by other writers, especially by Korzhinskii (1959). A
brief outline of the concept will be given here. Detailed discussions on it and
related problems will be made in chapter 11.
lt is assumed that the temperature, rock-pressure and the chemical potential
of H2 0 (or PH 2 0) in a metamorphic complex are controlled by external
conditions. Under a given set of values for these variables, a rock with a chemical
composition defmite except for H2 0 content shows a defmite mineral assem-
blage. With a slight change of any externally controlled variable, there would be a
slight variation in the compositional relations of coexisting minerals. At some
value of the variable, a new phase would appear. Thus, the paragenetic relations
of minerals serve as indicators of the externally controlled variables.
 3-2 OUTLINE OF METAMORPHIC FACIES/67

The rocks recrystallized within definite ranges of temperature, rock-pressure

and chemical potential of H2 0 may be designated as belonging to a meta-
morphic facies. It is conventional to define a metamorphic facies using
boundaries on which major mineral changes occur.
Each of the zones, based on progressive mineral changes, corresponds to
definite ranges of temperature, rock-pressure and chemical potential of H2 O. In
practice, therefore, we may treat the rocks of such a zone as belonging to a
metamorphic facies. Individual zones are usually limited by major mineral
changes. Thus, a progressive series of metamorphic facies can be defined on the
basis of mineral zoning in a terrane.
Metamorphic rocks in other areas which show similar mineral assemblages for
similar chemical compositions as rocks in a zone of a standard terrane, are
regarded as belonging to the same metamorphic facies as the rocks of the zone.
Eskola (1939) adopted eight metamorphic facies, which are accepted in this
book with some modifications. Many later authors proposed a number of new
facies and facies names, which however are not used in this book (for the reason
given in § 11-1), except for the two new important facies established by
Coombs (1960,1961; Coombs etal. 1959).

Four Common Metamorphic Facies

Three of Eskola's metamorphic facies will be introduced here in terms of the
Barrovian zones of the Scottish Highlands, and one more facies of Eskola will be
added at the high-temperature end. The occurrence of these four facies is very
common in the world, suggesting that each of them represents a wide range of
rock-pressure.
The progressive mineral changes in the Scottish Highlands are summarized in
fig. 3-2.

Greenschist Facies. The metabasites occurring in the chlorite and biotite zones
commonly show the mineral assemblage actinolite + chlorite + epidote + albite,
which is characteristic of greenschists. Actinolite may be absent and quartz may
be present. This assemblage is diagnostic of the greenschist facies. Not only such
metabasites but also all the other rocks including pelitic and quartzo-feldspathic
ones within the two zones belong to the greenschist facies. Pelitic rocks contain
no biotite in the chlorite zone, but do contain it in the biotite zone. When
desirable, the greenschist facies may be divided into the chlorite-zone and
biotite-zone sub facies.
If the metabasites in another area show the greenschist assemblage, those
rocks and the associated ones recrystallized within the same ranges of externally
controlled variables belong to the greenschist facies.

Epidote-Amphibolite Facies. The metabasites in the almandine zone commonly

show the mineral assemblage albite + epidote + hornblende, which is diagnostic
68/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

of the epidote- amphibolite facies. All the rocks in this zone belong to the
epidote- amphibolite facies.
Oligoclase may substitute for albite, and quartz may join the assemblage.
Almandine may occur in metapelites and metabasites, but is not regarded as
essential to the definition of this facies.
Epidote amphibolites are distinguished from greenschists by the presence of
hornblende instead of actinolite, and from amphibolites by the presence of the
albite + epidote assemblage instead of intermediate or calcic plagioclase.

Amphibolite Facies. The metabasites in the kyanite and sillimanite zones

commonly show the mineral assemblage intermediate or calcic plagioclase +
hornblende, which is diagnostic of the amphibolite facies. All the rocks in these
zones as well as amphibolites and associated metamorphic rocks in other areas
belong to this facies.

Granulite Facies. In the Dalradian metamorphics, amphibolite is stable to the

highest observed temperature of metamorphism. In certain other metamorphic
regions, however, a zone of the amphibolite facies grades with increasing
temperature into another zone belonging to Eskola's granulite facies. In this
facies, ortho- and clinopyroxenes take the place of hornblende in metabasites,
and garnets of the almandine-pyrope series may be common. Since granulite-
facies rocks are highly dehydrated, their formation should be promoted by lower
PH20 •
Thus, generally in order of increasing temperature, we have the following
series of metamorphic facies: greenschist -+- epidote-amphibolite -+- amphibolite
-+- granulite.
Such a series corresponding to progressive changes is called metamorphic
facies series (Miyashiro, 1961a). Each progressive metamorphic terrane has its
own facies series. The facies series observed throughout the world are very
diverse in character. Nonetheless they can be classified according to aspects of
similarity and difference, as will be discussed in § 3-3.
All these metamorphic facies have been defined primarily on the basis of
mineralogical variations in metabasites, which are sensitive to changes in
temperature, rock-pressure and chemical potential of H2 0. If, for Simplicity, we
neglect the effect of the chemical potential of H2 0, each metamorphic facies
represents a definite field in a Ps - T diagram.

Metamorphic Facies Characterized by High Rock-Pressure

Eskola (1939) proposed two facies as representing relatively high pressures.

Glaucophane-Schist Facies. Some metamorphic rocks of ordinary basaltic

compositions contain glaucophane. The P-T conditions for their formation
 3-2 OUTLINE OF METAMORPHIC FACIES/69

should differ from those of any of the facies so far discussed. Hence, such rocks
are said to belong to the glaucophane-schist facies. Some authors prefer the
name of blueschist facies instead.
It has been demonstrated that glaucophane by itself is stable under a P- T
range much wider than that here considered for this facies (Ernst, 1961, 1963a).
Moreover, the stability of glaucophane should vary with its composition.
Therefore, a proper definition of the facies must be based on a rigorous
paragenetic analysis, which, however, is not available at present.
In this book, we proceed under the provisional definition that glaucophane-
bearing metabasites and associated isophysical metamorphic rocks belong to the
glaucophane-schist facies, so long as there are no positive indications against it.
Under this definition, a rock complex only rarely containing glaucophane is also
classed in this facies. It may really be intermediate between the more typical
glaucophane-schist facies and some other facies (e .g. greenschist facies).
The occurrence of lawsonite and of the jadeite + quartz assemblage is
confined to the typical part of the glaucophane-schist facies. Thus, we may
subdivide this facies into -a typical and atypical sub facies. If necessary, we may
well use such names as lawsonite-glaucophane-schist sub facies for the typical
part.
Jadeite not associated with quartz is stable over a wider P- T range than the
jadeite + quartz assemblage. So far as I am aware, however, such jadeite occurs
only in glaucophane-bearing metamorphic areas and associated serpentinites.
With rising temperature, a glaucophane-schist facies zone usually grades into a
greenschist-facies or epidote-amphibolite facies zone. Glaucophane schist-facies
rocks are widespread in Mesozoic and Cenozoic metamorphic belts of the world.

Eclogite Facies. This facies is characterized by eclogites which show the

assemblage clinopyroxene (omphacite) + magnesian garnet + quartz (or kyanite),
devoid of feldspar. (The original definition of Eskola has been modified here.
Detailed discussions will be given in chapter 12.) The rocks of this facies do not
show a regular zonal distribution in progressive metamorphic terranes, but occur
in irregular distribution in some gneiss and glaucophane-schist areas, or as
inclusions in peridotite and kimberlite.

Metamorphic Facies Characteristic of Contact and Pyrometamorphism

These facies were proposed by Eskola for contact and pyrometamorphism under
relatively low rock-pressures.

Pyroxene-Hornfels Facies. This facies is characterized by the presence of ortho-

and clinopyroxenes and the absence of almandine and pyrope in rocks of
common chemical compositions. This facies represents high temperature com-
bined with relatively low rock-pressure, as is common in high-temperature
70/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

contact metamorphism around shallow intrusions. The type areas are the inner
contact aureoles of alkali plutonic masses near Oslo, Norway (Goldschmidt,
1911). Contact metamorphic rocks recrystallized at relatively low temperatures
usually belong to the amphibolite facies and not to the pyroxene-hornfels facies.

Sanidinite Facies. This facies represents the highest temperature combined with
low rock-pressure as is common in pyrometamorphism of inclusions within
volcanic rocks and minor basic intrusions.
It is important to realize that the definitions of metamorphic facies are based
only on T, Ps , and PH 20, and on the paragenetic relations of minerals, and not
on the geologic settings. If the same paragenetic relations are observed in groups
of regional, ocean-floor and contact metamorphic rocks, all these rocks are
regarded as belonging to the same facies. The amphibolite facies, for example, is
common to all the three kinds of metamorphism. In the past, some authors used
different facies names between regional-metamorphic and contact-metamorphic
rocks, even when all the rocks were considered to show practically the same
paragenetic relations. This is not consistent with the concept of metamorphic
facies, as will be discussed again in § 4-5 and chapter 11.

Two Metamorphic Facies Common in Burial Metamorphism.

In many progressive metamorphic terranes (e.g. the Ryoke belt and Dalradian
Series), the lowest temperature zone is in the greenschist facies, whereas in
others the lowest temperature zone is in the glaucophane-schist facies. In either
case, the greenschist or glaucophane-schist facies zone grades into an unre-
crystallized area with decreasing temperature.
On the other hand, it has been found in some other regional as well as more
localized metamorphic areas that the zone of the greenschist facies grades with
decreasing temperature into a zone with prehnite and/or pumpellyite, which in
tum grades into another zone containing various zeolites, which then grades into
an unrecrystallized area. Examples have been documented from the regional
metamorphic terrane on the South Island of New Zealand (Coombs et aZ. 1959)
and in the central Kii Peninsula of Japan (Seki, Onuki, Oba and Mori, 1971)
and from a smaller scale metamorphic area of the Tanzawa Mountains in Japan
(Seki et al. 1969a).
The metamorphic facies corresponding to these prehnite- and/or pumpellyite-
bearing zones is called the prehnite-pumpellyite facies, whereas that corres-
ponding to the ze01ite-bearing zone is called the zeolite facies after Coombs
(1960,1961).
The rocks in these two facies are usually non-schistose and preserve textural
features of their original sedimentary or igneous character owing to lack of
penetrative movements. In other words, the metamorphism may be said to be of
the burial type (§ 1-2). Recrystallization of the rocks is usually incomplete.
 3-3 BARIC TYPES AND PAIRED METAMORPHIC BELTS 1'71
Within the zeolite facies, the stable zeolite assemblages vary with the chemical
composition of the rocks as well as T, Ps , and PH 2,o. With rising temperature,
progressive dehydration takes place. Thus, in the presence of quartz and an
aqueous fluid, the stable calcium zeolites appear to vary with rising temperature
in the order: stilbite ~ heulandite ~ laumontite ~ wairakite (Miyashiro and
Shido, 1970). The stages of stilbite and of wairakite were observed only in a few
zeolite areas, probably of high geothermal gradients. The analcime-quartz
assemblage is stable in the heulandite stage, whereas albite becomes stable in the
laumontite and wairakite stage.

3-3 BARtC TYPES AND PAIRED METAMORPHIC BELTS

Concept of Baric Types

Before World War II, British petrologists took the lead in the study of
progressive metamorphism through elaborate descriptions of the Barrovian zones
in the Scottish Highlands (e.g. Tilley, 1925; Harker, 1932; Wiseman, 1934).
Under their influence, petrologists throughout the world at that time considered
that regional metamorphism as observed in the Barrovian zones was normal and
widespread in the world, and that other kinds of regional metamorp~sm, if any,
might well be regarded as anomalous and unimportant. Moreover, it was also
widely believed that glaucophane schists were formed by local metasomatism
due to materials coming from associated ultramafic and mafic intrusions (e.g.
Taliaferro, 1943). '
After the war, however, progress in metamorphic petrology in the circum-
Pacific region and especially in Japan has indicated that the above view is not
valid. Glaucophane-producing metamorphism was shown to have taken place on
a regional scale. This type as well as the andalusite-producing type of regional
metamorphism was found to be very widespread in the circum-Pacific region.
These findings have led to a marked expansion of the concept of regional
metamorphism, resulting in a threefold classification of regional metamorphism
in terms of rock-pressure (Miyashiro, 1961a). Each of the three categories
contains a number of characteristic series of metamorphic facies covering a wide
range of temperature.
The geothermal gradient varies from place to place on the earth. The depth at
which a given temperature is reached differs from region to region. Hence, the
rock-pressure therein also differs. In a Ps-T diagram, a series of Barrovian zones
would represent a certain definite curve (or band) showing increasing rock-
pressure with increasing temperature. Since kyanite is stable in the lower
temperature part and sillimanite in the higher, the curve should be located as
shown in fig. 3-3 in relation to the stability fields of Al 2 SiO s polymorphs. The
slope of the curve represents the pertinent geothermal gradient.
72/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

40

E
30 ::
...
.s::
0.
'"
Cl
~
20.§x
2
0.
0.
o:!
10

Temperature (OC)
Fig. 3-3 Classification of metamorphic facies series in relation to the
stability fields of Al2 SiOs minerals and jadeite.

Now we know that there are many metamorphic terranes that are re-
crystallized at lower rock-pressures and others that are recrystallized at higher
rock-pressures than the Barrovian zones.
Therefore, we term the kind of metamorphism as observed in the Barrovian
zones as the medium-pressure type, and the other two categories as the
low-pressure type and the high-pressure type respectively (table 3-2). These three
types should correspond to curves with different slopes in a Ps- T diagram. These
three will be called the baric types. They represent a classification of meta-
morphic facies series and hence a classification of metamorphism and of
metamorphic terranes.
It is to be noted that such names as 'high-pressure type' and 'high-pressure
metamorphism' are somewhat misleading, since the baric types as defined here
refer to the slopes of the geothermal curves corresponding to facies series, and
not the values of Ps themselves. As is clear from fig. 3-3, for example,
low-temperature zones in high-pressure metamorphism may represent lower
pressures than high-temperature zones in low-pressure metamorphism. It is
logically more appropriate to use such names as high PIT type (Miyashiro,
1967b) and high-pressure series metamorphism, instead. In this book, however,
high-pressure type, high-pressure metamorphism and analogous names will be
used only for simplicity of expression.
In this section, the threefold classification will be discussed mainly in terms
of regional metamorphism. Since the classification is based only on the Ps- T
Table 3-2 Threefold classification offacies series of regional metamorphism

Characterized Common Common metamorphic w

Baric type by minerals facies series Associated magmatism W
1:0
Low-pressure Andalusite Biotite, Greenschist -+ amphibolite Geosynclinal volcanics, >
cordierite, -+ granulite ranging from basic to ::c
staurolite, acidic, are present but n
o-l
sillimanite usually scarce. Granites -<"C
are very abundant, trl
being accompanied in some til

cases by andesite and >

Z
rhyolite o
"C
Medium-pressure Presence of Biotite, Greenschist -+ epidote- Both ophiolites and >
kyanite and almandine, amphibolite -+ amphibolite granites are present ::c
trl
absence of staurolite, -+ granulite o
glaucophane sillimanite ~
trl
o-l
High-pressure Glaucophane, Almandine, Glaucophane schist Ophiolites, ranging >
jadeite, barroisite, -+ epidote amphibolite; from ultrabasic to ~
lawsonite stilpnomelane Glaucophane schist basic, are abundant. o
::c
-+ greenschist; Granites are usually "C
:I:
Prehnite-pumpellyite absent
("J
-
-+ glaucophane schist
1:0
trl
t"
Note: For a greater detail, refer to tables 4-1, 7A-1 and 8A-1. o-l
til
-..J
W
-
74jBASIC CONCEPTS OF METAMORPHIC PETROLOGY

relations, it also may be applied to other geologic classes of metamorphism such

as ocean-floor and contact ones. Some cases of contact metamorphism appear to
have a metamorphic facies series identical to that of the low-pressure regional
one, but others appear to have a different series. If the rock-pressure in such
contact metamorphism is the same as, or lower than that of low-pressure
regional metamorphism, we will include it in the category oflow-pressure type.

Low-Pressure Type
This type is characterized by andalusite, which is stable at lower rock-pressures
than kyanite at the same temperature. Andalusite occurs in the lower and
sillimanite in the higher temperature part of metamorphic terranes of this series,
if the chemical compositions of the rocks permit it. Hence, Miyashiro (1961a)
designated this series as the andalusite-sillimanite type. A Ps- T curve for it is
shown in fig. 3-3. The occurrence of cordierite and the rarity of pyralspite garnet
are among the common features of this type.
The progressive mineral changes in low-pressure metamorphism in the Shiojiri
area of the Ryoke belt, Japan, are shown in fig. 3-4. Since metabasites are
relatively rare in this area, it is not clear to what extent a zone of the
epidote-amphibolite facies is developed between those of the greenschist and
amphibolite facies.
Miyashiro (1961a) originally used the central Abukuma Plateau in Japan as
the type terrane for the low-pressure type. It also shows the facies series:
greenschist """* amphibolite (fig. 3-5). Recently, however, kyanite has been found
to occur occasionally besides andalusite and sillimanite in this area (Hara et al.
1969). The Ps-T curve corresponding to this area would pass on or near the
triple point of A1 2 SiO s (Tagiri, 1971). In other words, the metamorphism of
this area appears to be transitional between the low- and medium-pressure types.
The orthopyroxene-bearing metamorphic areas in Lapland (Eskola, 1952) and
Uusimaa (Parras, 1958), both in Finland, appear to represent the highest
temperature part of low-pressure regional metamorphism. They fall into an
intermediate state between the granulite facies of medium-pressure meta-
morphism and the pyroxene- hornfels facies characteristic of low-pressure
contact metamorphism. In this book, such a Ps-T condition will be included
within the range of the granulite facies. Thus, we have a common low-pressure
facies series of regional metamorphism: greenschist """* amphibolite """* granulite.
There is a considerable diversity in the metamorphic facies series belonging to
this type, conceivably owing to the wide ranges of geothermal gradients and
PH o. Other observed facies series belonging to this type will be described in
2
§ § 4-4 and 4-5 and chapters 10 and 11.

Medium-Pressure Type
This type corresponds to the Barrovian zones as stated above. Since kyanite
 3-3 BARIC TYPES AND PAIRED METAMORPHIC BELTS /75
Shiojiri Area

Metamorphic facies Greenschist Amphibolite facies

facies

Mineral zoning la Ib Ic lIa lib III

Albite
01 igoclase- Labradorite ?
- ----
.
....
OJ
Orthoclase
Chlorite
Epidote
'in
'"
.0 Muscovite
....'"OJ Biotite, green
2
Biotite, brown - --- -- --
Actinolite
Hornblende, blue-green
Hornblende, green
Clinopyroxene

Plagioclase ?
Microcline ?
Orthoclase

..
OJ
:!::
Chlorite
Muscovite
a; Biotite
0.
....'"OJ Andalusite ? - ---
2
Cordierite - -- -- --
Sillimanite
Pyralspite ---- ---- ---- ----
Tourmaline, green
Tourmaline, brown
Graphite

Dolomite

0
OJ
<:
Tremolite
Diopside
? -
~
OJ Grandite
E
:J Wollastonite r----
Scapolite

Fig. 3-4 Progressive mineral changes in the Ryoke terrane of the Shiojiri
area (northern Kiso), Japan. (Katada, 1965.) See fig. 7A-1.
occurs in the lower, and sillimanite in the higher temperature part, Miyashiro
(1961a) designated the series as the kyanite-sillimanite type. Cordierite is usually
absent, and almandine garnet is common.
It usually shows the metamorphic facies series: greenschist -+ epidote-
amphibolite -+ amphibolite -+ granulite. The greenschist, amphibolite and
granulite facies are shared in common between the low- and medium-pressure
76/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

Central Abukuma Plateau

Metamorphic facies Greenschist

Amphibolite facies
facies
Mineral zoning A B C

Sodic plagioclase
Interm. and calcic
plagioclase
Epidote --
Actinolite
'"
~
.;;; Hornblende - --
Blue:green Green and brown
co
.0 Cummingtonite --- --------
~
OJ Chlorite --
::2 Calcite - --
Clinopyroxene - - - - - _.. -
Magnetite ? - ---- ?
Ilmenite
Pyrite
Pyrrhotite -------
Chlorite -- -
Muscovite --
Biotite
Pyralspite MnO)18% MnO = 18-10% MnO (10%
'"
-.~
OJ
Andalusite -------
c. Sillimanite
~
OJ
::2
Cordierite - --- --------
Plagioclase
K-feldspar ------
Quartz
Magnetite ? - ---- ?
Ilmenite
Pyrrhotite - -- - - -- --
Calcite
Epidote - --
'"cOJ Actinolite ~------- ----- ---
0 Hornblende -- ------- -------
t>OJ
E
Clinopyroxene - - --
:J Grandite --
Wollastonite -- - --
K-feldspar -------- -------- ---------
Plagioclase
Quartz

Fig. 3-5 Progressive mineral changes in the central Abukuma Plateau,

Japan. (Miyashiro, 1958; Banno and Kanehira, 1961.)
 3-3 BARIC TYPES AND PAIRED METAMORPHIC BELTS/77

types. When necessary, we can designate the metamorphic facies .of a particular
case by the combination of a baric type and a metamorphic facies; e.g.
low-pressure amphibolite facies.
It is to be noted that the occurrence of kyanite does not warrant assignment
to the medium-pressure type because the mineral occurs in some high-pressure
type terranes also (Niggli and Niggli, 1965; Banno, 1964). Hence, the lack of
glaucophane, jadeite and lawsonite is a necessary condition for assignment to the
medium-pressure type.

High-Pressure Type
This type is characterized by jadeite, lawsonite and glaucophane. Jadeite,
especially in association with quartz, is a well-demonstrated high-pressure
mineral as shown in fig. 3-3. However, the equilibrium curve shown refers to
pure jadeite NaAlSiz 0 6 . The jadeites that occur in association with quartz in
metamorphic rocks usually contain significant amounts of diopside and acmite
components. It follows that the pressure necessary for their formation should be
slightly lower than that shown by the equilibrium curve for pure jadeite. Such
impure jadeites may usually form on the low-pressure side of, but in the vicinity
of the curve (Ernst and Seki, 1967; Ernst, Seki, Onuki and Gilbert 1970;
Coleman and Clarke, 1968). The Ps- T curve for the high-pressure type in fig. 3-3
is based on this view.
In some high-pressure regional metamorphic terranes, jadeite, lawsonite and
glaucophane all occur. However, there are many other areas where glaucophane
occurs but jadeite and lawsonite do not. As was discussed in § 3-2, glaucophane
is not necessarily a reliable indicator of high pressures. When it occurs in rocks of
ordinary basic compositions, however, it may be regarded as representing
relatively high pressures (i.e. the glaucophane-schist facies). Since the precise
definition of 'ordinary basic compositions' is very difficult, this criterion may
not give an unequivocal assignment.
All the facies series that contain the glaucophane-schist facies in a broad sense
as defmed in § 3-2, may be regarded as belonging to the high-pressure type.
Since glaucophane is a familiar mineral, high-pressure metamorphism was
sometimes called glaucophanitic metamorphism.
The facies series of the high-pressure type are diversified. When the jadeite +
quartz assemblage is stable in a part or the whole of a facies series, this will be
designated as a typical high-pressure series. When it is not, the name of an
atypical high-pressure series will be used.
Figs. 3-6, 3-7, and 3-8 indicate three examples of high-pressure terranes. They
appear to show three successive steps in the passage from the medium-pressure
to the typical high-pressure type. In the first two examples, an epidote-
amphibolite or a greenschist facies zone occurs on the high-temperature side of a
78/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

Bessi-Ino Area

Glaucophane- Epidote-
Metamorphic Transi- amphibolite
facies schist facies tiona I facies
Mineral zoning A B C 0 E

Albite
Oligoclase
Actinolite
Barroisite
Hornblende
Glaucophane --
...'"
VI

'in
Diopside
II>
-c Pumpellyite
...
II>
Epidote
~'" Chlorite -- -
Pyralspite - - --
Muscovite
Biotite - - - - - - ---
Haematite
Ilmenite
Rutile

Chlorite --
Muscovite
VI

'"
Biotite - ---
.~
a; Pyralspite -
...
Q,
II> Albite
'"
~
Oligoclase ----
Epidote -- - - ---- --- -- - - - - - - - - - -
Calcite

_.. - - - - -
Quartz
Ilmenite ---

Fig. 3-6 Progressive mineral changes in the Sanbagawa terrane of the

Bessi-Ino area, Japan. (Banno, 1964.) See fig. 7A-12.

glaucophane-schist facies zone. Other facies series of this type will be discussed
in § 4-4 and especially in chapter 11.

Petrogenetic Grids
The three baric types have been defined here on the basis of Al 2 SiO s
polymorphs and glaucophane-schist facies minerals. Table 3-2 indicates, how-
ever, that cordierite is common in low-pressure metamorphic terranes, whereas
barroisite and stilpnomelane are common in high-pressure ones. Chloritoid,
staurolite and garnet also show some extent of preferential occurrence in certain
baric types. Such problems will be discussed in detail in chapters 7B and BB.
 3-3 BARIC TYPES AND PAIRED METAMORPHIC BELTS/79
Kanto Mountains

Metamorphic Green-
Glaucophane-schist facies schist
facies
facies
Mineral zoning I II III IV V VI

Sodic plagioclase
Jadeite
Lawsonite
.,'"
+-'
Pumpellyite --
.;:;;
Epidote
~'"
.,'"
+-'
Glaucophane
Actinolite
::2:
Chlorite
Stilpnomelane
Quartz

Chlorite
Muscovite
-0'"
c::l!l
",.-
Pyralspite - - - - --
'" E Stilpnomelane
l!l E
.,c.c.'"
:'::«1 Piemontite - - - -- - - --
.,'" '".,
+-,+-'
Lawsonite
Jadeite
::2: E
Albite
Quartz

Fig. 3-7 Progressive mineral changes in the Sanbagawa terrane of the

Kanto mountains, Japan (Seki, 1958, 1960, 1961a). Recrystallization is
very poor in Zone 1. Some of the 'actinolite' in Zone VI may actually be
barroisite.
It is important to emphasize that the occurrence of these minerals depends
not only on T, Ps and PH 2 0 but also on the bulk chemical compositions of the
metamorphic rocks. The formation and composition of cordierite and garnet, for
example, are strongly controlled by the Al 2 0 3 /(MgO + FeO) and FeO/MgO
ratios and the oxidation condition of the rocks. The presence or absence of such
minerals is not a reliable indicator of baric types and Ps - T conditions, until a
rigorous analysis of their paragenetic relations becomes available. Theoretical
efforts in this direction have been made by Albee (l965a), Hoschek (1969), Hess
(1969), and Hensen (1971).
Hess (1969) has given a semi-quantitative analysis of the P- T relations of
cordierite-bearing parageneses observed in metapelites with quartz and muscovite
(or quartz and K-feldspar). The topology of the univariant curves was derived
from Schreinemakers rule (§ 5-4) and observed paragenetic relations. Approx-
imate slopes of the curves were calculated by the Clausius-Clapeyron equation
from the estimated volume and entropy changes_ This system of curves was
80/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

Panoche Pass Area

Metamorphic facies Glaucophane-schist facies

Albite f-----+-- - ----

Aegirine
Actinolite
Crossite
Lawsonite
Pumpellyite t - - - - - - t - - - ---
Quartz
White mica
Chlorite
Stilpnomelane 1----+------ - - - - - - - -
Sphene
Calcite
Aragonite
Apatite
Haematite
Magnetite

Albite
Jadeitic pyroxene
Glaucophane
Lawsonite - - - - - - --------------1
Quartz
White mica
Chlorite
Stilpnomelane
Sphene
Calcite
Aragonite
Apatite
Haematite-I imon ite
Magnetite
Pyrite

Fig. 3-8 Progressive mineral changes in the Franciscan terrane of the

Panoche Pass area, California. (Ernst, 1965.) See Fig. 7A-13.

placed on the P- T diagram with reference to the experimentally determined

curves for the stability relations of Al 2 SiO s polymorphs and for the reaction of
muscovite with quartz. The Fe 2 + /Mg ratios of ferromagnesian minerals were
somewhat arbitrarily fIxed in order to simplify the calculation. The results (for
metapelites containing an aqueous fluid phase) is shown in fig. 3-9. It is
noticeable in this diagram that increasing pressure causes the transformation of
assemblages containing cordierite into those containing andalusite, sillimanite,
kyanite, biotite, staurolite and chlorite.
 Abbreviations:

8 A = Andalusite
B = Biotite
Cd = Cordierite Vol
Cd' = Mg-cordierite W
7 Ch = Chlorite
Ch' = Mg-chlorite t:tI
G = Garnet >
Ky = Kyanite ;tI
Mu = Muscovite (=j
.... 6 Ph = Phlogopite >oj

..0
'" Q = Quartz -<
.."
~ S = Staurolite t%1
~ Sa = Sanidine en
:l
en 5 Si = Sillimanite >
en
~ H = Hypersthene Z
a.. 1:1
.."
>
....
4 ;tI
t%1
1:1
il::
t%1
3 >oj
>
il::
o
;tI
2 .."
500 600 700 :I:
Temperature (DC) (=j
t:tI
Fig. 3-9 Semi-quantitative stability relations of cordierite and related t%1
t"'
minerals in association with muscovite (or K-feldspar) and quartz in >oj
metapelites. (After Hess, 1969; with modification based on his personal en
00
communication, 1970.) -
82/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

Facies Series of Metamorphic Belts

The observed facies series varies to some extent from area to area within a single
regional metamorphic belt. For example, the Caledonian metamorphic terrane in
the Scottish Highlands, including the Barrovian zones of the Dalradian, shows
predominantly the medium-pressure facies series, whereas its northeastern end in
Aberdeensmre and Banffshire belongs to the low-pressure series (fig. 7A-I0). A
part of the northern Appalachians belongs to the medium-pressure series, and
the rest to the low-pressure ones (fig. 7 A-6).
A low-pressure metamorphic area appears to grade into a medium-pressure
one without any structural discontinuity in the Scottish Highlands and northern
Appalachians. It suggests the possibility that metamorphism of the two types in
such areas occurs by the same or similar geologic causes, and that the difference
between the two simply reflects a difference in some subsidiary conditions (see
chapter 17 for a more detailed discussion).
On the other hand, some high-pressure metamorphic areas appear to grade
into medium-pressure areas within the same belt, though available data for this
are not complete.
It is not inconceivable that there exist only two distinct categories of regional
metamorphism as regards geologic settings: low-pressure and high-pressure, and
that medium-pressure metamorphic areas are variants belonging to one or the
other of them. This possibility is important especially in the problem of paired
metamorphic belts and of the magmatism genetically connected with meta-
morphism, both of which will be discussed later.

Paired Metamorphic Belts and their Significance

In Japan and many other parts of the circum-Pacific regions, two regional
metamorphic belts of similar geologic ages but of contrasting characters run side
by side, forming a pair (Miyashiro, 1961a). One belt is of the low-, and the other
is of the high-pressure type, though parts of each may be of the medium-pressure
type, as shown in fig. 3-10. Outside Japan, the paired belts in California
(Hamilton, 1969a), in Chile (GonzaIez- Bonorino, 1971) and in New Zealand
(Landis and Coombs, 1967) are well documented. The formation of paired
metamorphic belts appears to be a result of underthrusting of an oceanic plate
Qeneath an island arc or a continental margin (Miyashiro, 1961a, b, 1967b;
Takeuchi and Uyeda, 1965).
The high-pressure metamorphic belt of a pair usually lies on the oceanic side
of the aSl!ociated low-pressure belt. The underthrusting of an oceanic plate would
bring a sediment pile in the trench zone to great depth (fig. 3-11). The tectonic
descent should cause an unusually low geothermal gradient, leading to high-
pressure metamorphism.
 3-3 BARIC TYPES AND PAIRED METAMORPHIC BELTS/83

o 400km
'' - _ - ' - _ ' - - - - ' - _ . . . J'

Hodaka metamor belt (Iow- P )

{
Kamulkotan metamor belt

Permian - JurasSIC pair

Hlda melamor. belt (!ow- P )

{
Sangun melamor. bell Ihlgh- P I

I
Ryoke melamor bell (low - P} Cretaceous
JuraSSIC'
Sanbagawa melamor. belt (high· P I pair

Fig.3-1O Three pairs of regional metamorphic belts in Japan (Miyashiro,

1961a, 1972b). A pair is made up of the Hida and Sangun belts, ano'ther
pair of the Ryoke and Sanbagawa belts, and a third pair of the Hidaka and
Kamuikotan belts. The eastward extension of the Ryoke belt is regarded as
being exposed in the Abukuma Plateau. F- F = Itogawa-Shizuoka Line (a
Tertiary fault). For a detailed description see chapter 15.
 00
.:t.
til

Marginal Island arc Trench Ocean

>
00
Sea Inner arc (volcanIC I '"'
n
VolcanIC pile
n
o
Z
Crust (} ----- Ocean water n
melamorphlsm ~-0cean1C CruSt t!l
'"1:1
~ i-l
OceanIC plate 00
f:!..as.! £!" .£'ate ~ fv;) Q 0 () o
--_2 V ...,
<::) ~ - - - - ___ 9 ~ \)
,~ 0 0 --;-- _---Base of plate
s:
"-. ,'f;) t!l
i-l
>
~ ~ .. <;O'L 0\ ~ , s:
o
" Q ::tI
'"1:1
~\ ID \ ~
\ ~ \~ c- t ....::c
n
'"1:1
t!l
i-l
::tI
o
t"'
\ ~\ ~ ,
o
o
1 -<
t

Fig. 3-11 Schematic figure representing the origin of paired metamorphic

belts with special reference to the Northeast Japan Arc from the Miocene
to the present (Miyashiro, 1972b). For a detailed account see chapter 17.
 3-4 BARIC TYPES AND GEOTHERMAL GRADIENTS / 85
Paired metamorphic belts are not confined to the circum-Pacific regions. The
Scottish Caledonides, for example, appear to have paired belts in which one belt
is only poorly developed (§ 14-3).
When the velocity of underthrusting is not rapid enough to cause high-
pressure metamorphism, medium-pressure metamorphism would take place
instead, and the contrasting characters of two parallel belts, if present, may
become obscure, or one belt of the two may not develop well (Miyashiro,
1972b). At any rate, most metamorphic belts in the Atlantic region are
apparently unpaired. For example, the Hercynian belts in Europe are mostly of
the low-pressure type, whereas the Alpine belt is of the high-pressure type. It is
likely, however, that some of them are actually paired and one belt of the pair is
very poorly developed or has not been exposed yet (§ 17-1).
Metamorphic belts of the low-pressure type are characteristically accom-
panied by abundant granitic plutons (§ 4-1). If erosion has been less advanced,
well-recrystallized regional metamorphic terranes may not be exposed, but
high-level granitic plutons and associated contact aureoles may appear on the
surface. It is likely that chains of andesitic volcanoes in island arcs and
continental margins are a surface manifestation of low-pressure metamorphic
complexes and associated granitic rocks as will be discussed in § 4-1. In this
meaning, the low-pressure metamorphic belt may be regarded as a belt of
granitic plutonism and andesitic volcanism.
Most of the typical paired metamorphic belts in the circum-Pacific regions are
Mesozoic and Cenozoic in age. Most of typical glaucophane schists in the world
were formed in Mesozoic and Cenozoic time. Conceivably, lithospheric plates
may have become thicker and/or have come to move more rapidly since the
beginning of Mesozoic time, particularly in the Pacific regions (Miyashiro,
1972b). More rapid underthrusting of plates should create lower geothermal
gradients and should tend to form more typical high-pressure metamorphic belts
and more typical paired belts (§ § 3-4 and 17-1).

3-4 BARIC TYPES AND GEOTHERMAL GRADIENTS

Average Geothermal Gradients for Regional Metamorphism

If the triple point of A1 2 SiO s lies at about 600°C and 6·0 kbar (Althaus, 1967;
Richardson, Gilbert and Bell, 1969), and if we assume that the temperature rises
at a constant rate with depth, the gradient of the geothermal curve which passes
through the triple point is about 25°C km -1.
If the existing data on the phase relations of jadeite (Newton and Kennedy,
1968) is accepted and a zone having the jadeite + quartz assemblage in typical
high-pressure metamorphism is assumed to have been recrystallized at 400°C or
a lower temperature (§ 3-5), the average geothermal gradient should be about
10°C km- 1 or less.
86/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

Thus, we obtain a very rough estimation of average geothermal gradients as

follows:
Low-pressure regional metamorphism: greater than 2S °c km -1,
Medium-pressure regional metamorphism: about 20°C km -1,
High-pre~sure regional metamorphism: about 10°C km -1 or less.

The P-T values adopted above for phase relations and the temperature stated
above for the jadeite + quartz zone may be in error. Ifthey are, the estimates of
average geothermal gradients must be modified. The triple point of Al 2 SiO s
minerals, for example, was located at SIO °c and 4·0 kbar by Newton (1966b)
and Holdaway (1971). In case we accept these values, the necessary modification
in the estimate of the average geothermal gradient that divides the low- and
medium-pressure types is very slight, though the estimates of the temperature of
the sillimanite isograd must be decreased by about 100°C.

Possible General Shapes of Geothermal Curves

In the high-pressure area of the Kanto Mountains (fig. 3-7), albite occurs
throughout the metamorphic terrane, whereas. the jadeite + quartz assemblage
occurs only in a zone of medium temperature. The configuration of the P- T
diagram of fig. 3-3 suggests that this relation can be explained by assuming a
geothermal curve convex upward. If this is the case, the geothermal gradient
would have been very low (say, S °c km -1) to a depth of about 20 km, and then
increased at greater depths.
This curvature is contrary to that of generally accepted ordinary geothermal
curves, as is given, for example, by Ringwood (1969, fig. 3). The ordinary
geothermal curve is controlled mainly by the production and conduction of heat
within the crust, whereas the curve of the Kanto Mountains must be a result of
some other factors. Possible factors include the rapid accumulation of overlying
sediments, rapid tectonic sinking of the metamorphic complex to greater depths
and endothermic reactions in metamorphism. Other high-pressure metamorphic
terranes would have geothermal curves similar to that of the Kanto Mountains, if
high-pressure metamorphism generally occurs by a similar mechanism.
No evidence is available for the general shapes of geothermal curves for
medium- and low-pressure regional metamorphism. However, abundant granitic
rocks are usually associated with such metamorphic regions and especially with
low-pressure terranes. The temperature rise in such regional metamorphism
would not simply be due to the heat conduction, but would be largely related to
the upward movement of magma, H2 0 and other materials through the crust
(§ § 4-1 and 4-2). High-temperature zones are commonly accompanied by syn-
or late-tectonic granitic masses. As is clear from oxygen isotope studies, some
oxygen-containing material (probably mainly composed of H2 0) migrates
through such regional metamorphic terranes, but the migration is not uniform.
 3-4 BARIC TYPES AND GEOTHERMAL GRADIENTS /87

Thus, the geothennal gradient would vary more or less irregularly in the lateral
and vertical directions.
If the temperature distribution at depth in the medium- and low-pressure
metamorphic belts is largely controlled by ascending magma and H2 0, the
geothermal curves in fig. 3-3 should be concave upward in contrast to those of
the high-pressure belts.
In contact metamorphic aureoles, the temperature rise would take place
toward the pluton, and hence downwards, horizontally, or even upwards. If so,
the average geothermal gradient calculated down to a point lying within a rock
complex undergoing contact metamorphism should have no relation to the
actual geothennal gradient at that point.

Estimation of Depths and Geothermal Gradients

There are many metamorphic terranes where none of the diagnostic minerals for
baric types occur. In some such cases, we may be able to get indirect evidence
suggesting a presumptive type.
Such evidence could be feasible in particular in some zeolite-facies terranes,
since the deformation there is relatively weak so that the depths of recrystalliz-
ation can be estimated with some degree of reliability. The temperature reached
at a depth may be inferred from the occurrence of metamorphic re.actions such
as analcime + quartz = albite + H2 0. In this way, we may obtain an estimate of
the average geothennal gradient. (Examples will be given in chapter 6A.)
However, the- reliability of such an estimate is very low.
The reliable estimation of the depths of metamorphism is very difficult in
most regional metamorphic terranes where deformation is intense. The depth at
which a rock lies is quite different from the thickness of an overlying
stratigraphic column in tectonically disturbed terranes. Zwart (1963) has
estimated the depths of two mineral zones in a low-pressure terrane in the
Pyrenees on a structural basis (§ 7 A-4), giving a geothermal gradient of about
150°C km- I .

Width of Metamorphic Belts

In areas of higher geothermal gradients, the widths of individual mineral. zones
tend to be narrower, though they depend not only on the gradient but also on
the angle between the surface of the earth and the isothermal surfaces. However,
the reliable determination of this angle is usually difficult, since isograds are not
sharp lines but broad zones depending upon the variation in bulk chemical
composition of the rocks.
The width of a metamorphic belt, as measured from the line of beginning of
marked recrystallization to the thermal axis (line of maximum temperature) is
commonly of the order of 1(}-20 km in the low-pressure facies series, and
50-150 km in the medium- and high-pressure series. Wide metamorphic
88/ BASIC CONCEPTS OF METAMORPHIC PETROLOGY

terranes of the low-pressure type such as the Hercynides in western Europe and
the northern Appalachians contain two or more metamorphic belts or thermal
axes running parallel. Fig. 7A -6 shows that the wide metamorphic belt of the
northern Appalachians has three thermal axes running approximately in a
NE-SW direction, that is, along the elongation of the pertinent orogenic belt.
Moreover, along one of the thermal axes, there is an unusually wide zone
representing nearly uniformly high temperatures (about 80 km in width). In this
zone, the present erosion surface is probably roughly parallel to an original
isothermal surface.
On the other hand, the high-pressure Alpine metamorphic belt in Switzerland
contains only one thermal axis.

3-5 TEMPERATURE AND PRESSURE CORRESPONDING TO INDIVIDUAL

METAMORPHIC FACIES

Difficulties in P- T Estimation
The low- and high-temperature limits of metamorphism were defined in § 1-1 as
the lowest temperature forming minerals unstable on the earth's surface and as
the incipient melting temperature of rocks respectively. Since a lot of experi-
mental data on the phase relations of minerals have acc1.l;mulated in the last
twenty years, we can now estimate the temperature and pressure of individual
metamorphic facies. This estimation, however, involves various difficulties, as
will be outlined below.
Many published synthetic studies give metastable relations, and hence are not
directly applicable to metamorphic reactions. This difficulty is especially serious
at low temperatures. For example, most studies on zeolites have given meta·
stable phase relations (e.g. Koizumi and Roy, 1960).
Even if true stability relations may be revealed, almost all experiments have
been carried out on samples having greatly simplified compositions. For example,
breakdown P- T curves have been determined for many minerals under the
condition that they are not accompanied by any other minerals. In reality,
mineralogical changes in a rock take place commonly by reaction between
coexisting minerals. The lower and upper breakdown temperatures of a mineral
that is present by itself represent only its maximum possible stability limits
(§ 2-5).
Experiments are usually carried out first on end members of solid solution
series. The mixing of other components may sometimes exert a profound
influence not only on the reaction temperature but also on the type of stable
assemblages. Many such difficulties will be removed in the near future with
increasing experimental data.
Most hydrothermal syntheses have been carried out in the presence of an
aqueous fluid of nearly pure H2 0 composition. This condition corresponds to
 3-5 TEMPERATURE AND PRESSURE OF METAMORPHIC FACIES/89

model (A) of § 2-3. Decarbonation reactions have been studied with a fluid of
nearly pure CO 2 composition, or with a mixture of CO 2 and H20. The latter
corresponds to model (B) of the same section. It is not clear to what extent
these models represent real metamorphic processes. It is likely that models (C)
and (0) of the same section represent some or many cases of metamorphism. If
so, the experimental data cannot be applied quantitatively. Solid-solid reactions,
which do not suffer from this difficulty, accordingly, play an important role in
the estimation.
Another difficulty results from the lack of relevant field data. For example,
we have some experimental data on the reaction of kaolinite and quartz to
form pyrophyllite and of the breakdown of pyrophyllite to form an Al2 SiO s
mineral and quartz (fig. 7B-l). However, we have little petrographic data on
the distribution of kaolinite and pyrophyllite in metamorphic rocks. Hence,
such experimentally determined eqUilibrium curves, reliable or not, cannot be
effectively used in coordinating metamorphic facies in the P-T diagram.

P- T Estimation by Solid-Solid Reactions

Relatively reliable, pertinent equilibrium curves have been compiled in fig. 3-12.
Let us consider solid-solid reactions first.
The phase relations of andalusite, kyanite and sillimanite that have been
determined by Althaus (1967) and Richardson, Gilbert and Bell (1969), give an
important framework for the Ps-T relations of metamorphism (figs. 2-1 and
7a-1). The triple point for the polymorphs has been located approximately at
5·5 kbar and 620 °c by the latter authors. Petrographic experience shows that
this triple point lies in the low amphibolite facies as shown in fig. 3-12. By
definition, the triple point separates the geothermal curves of low-pressure
metamorphism from those of medium-pressure.
The equilibrium curve for the calCite-aragonite transformation is shown as
curve (3) in fig. 3-12 (Crawford and Fyfe, 1964; Johannes and Puhan, 1971).
Aragonite appears to be stable in a high-pressure part of the glaucophane-schist
facies and an adjacent range of conditions.
The eqUilibrium curves for the reactions: jadeite + quartz = albite, and jadeite
= albite + nepheline, have been determined by Birch and Le Comte (1960) and
Newton and Kennedy (1968). Nearly pure jadeite occurs in serpentinite in some
high-pressure metamorphic terranes. Such jadeite is not associated with quartz
(e.g. Coleman and Clark, 1968), and hence is considered to have formed under
conditions which lie on the high-pressure side of the curve for the reaction:
jadeite = albite + nepheline, which in turn lies on the low-pressure side of the
curve for jadeite + quartz = albite.
On the other hand, jadeites in ordinary metamorphic rocks are commonly
associated with quartz, but usually contain a considerable amount of other
pyroxene components. These components should decrease the pressure necessary
90/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

for the formation of jadeites. Hence, it is widely belived that such jadeites
formed under the condition on the low-pressure side of the equilibrium curve for
the reaction: pure jadeite + quartz =albite (Ernst and Seld, 1967; Coleman and
Clark, 1968). Thus, even the conditions for the most typical high-pressure
metamorphism as in the Franciscan Formation may not have reached the
high-pressure side of the equilibrium curve of the reaction: pure jadeite + quartz
=albite, shown as curve (4) in fig 3-12.

10

8
-.:
OJ
.0
~
7
:!!
~

rt:
II
6
~
E
:> 5
~
E
a...
4

o
o 100 200 300 400 500 600 700 800 900 1000
Temperature (OC)
Fig. 3-12 Temperature and pressure of metamorphic facies. The equili-
brium curves are for the following reactions. (The left-hand side of each
equation is stable on the low-temperature side of the corresponding curve.)
(1) analcime + quartz = albite + H2 0;(2) lawsonite + 2 quartz + 2H2 0)=
laumontite; (3) aragonite = calcite; (4) jadeite + quartz = albite; (5)
kyanite = andalusite; (6) kyanite =sillimanite; (7) andalusite =sillimanite;
(8) muscovite + quartz = K-feldspar + Ah SiO s + H2 0; (9) beginning of
melting of granite; (10) beginning of melting of olivine tholeiite.
 3-5 TEMPERATURE AND PRESSURE OF METAMORPHIC FACIES/91

P- T Estimation by Dehydration Reactions

Here we will provisionally accept model (A) of § 2-3. In other words P g =P fluid
=PH,O •
The curves for the formation and decomposition of pyrophyllite are of little
use, since relevant petrographic data are too meagre.
The reaction: analcime + quartz = albite + H2 0 takes place in the middle of
the zeolite facies, as shown by curve (1) in fig. 3-12. The low-pressure limit of
the lawsonite + quartz assemblage is shown as curve (2) on the basis of recent
work by A. B. Thompson (1970b).
The equilibrium curve of the reaction: muscovite + quartz = Al 2 SiO s +
K-feldspar + H2 0 (fig. 7B-l) is relatively reliable (Evans, 1965). This reaction
takes place within the amphibolite facies in some areas of regional meta-
morphism (§ 7B-15), but is said to take place approximately at the boundary
between the amphibolite and granulite facies in the northern Appalachians
(Thompson and Norton, 1968).
The experimental data outlined above, combined with our knowledge of the
facies series of metamorphic rocks enable us to locate metamorphic facies
roughly as shown in fig. 3-12. The curves for the beginning of melting of granite
and olivine tholeiite determined by Yoder and Tilley (1962) are also shown.
It is impressive that all these dehydration curves, which were determined in
the presence of an aqueous fluid, when plotted together with curves for
solid-solid reactions on a P- T diagram, give relations generally consistent with
our knowledge of the sequence of reactions in progressive metamorphism. This
means that we may provisionally accept model (A) for dehydration reactions,
though it does not prove the universal validity of the model.
Decarbonation reactions will not be used in our P- T estimation, because even
if we assume the presence of a fluid phase during metamorphism, we cannot
specify the concentration of CO 2 in the fluid.

Oxygen Isotope Geothermometry

The oxygen isotopes 18 0 and 1 60 show a measurable extent of systematic
fractionation between coexisting minerals in metamorphic rocks. In most cases,
there is a close approach to isotopic equilibrium in a rock specimen. Since the
equilibrium distribution of isotopes depends mainly on the temperature of
formation of mineral assemblages, the 18 0/ 16 0 ratios measured on coexisting
minerals combined with experimentally determined fractionation curves can be
used as a geothermometer.
Experimentally determined fractionation curves are available for the pairs
quartz-muscovite, quartz-calcite, quartz-magnetite and so on. The fraction-
ation becomes less marked with risIng temperature, and hence the precision of
temperature determination usually becomes very poor at temperatures above
500°C.
92/BASIC CONCEPTS OF METAMORPHIC PETROLOGY

A considerable number of temperature determinations have been made by

this method on the regional metamorphic rocks of the medium-pressure type in
the northern Appalachians and the high-pressure rocks of the Franciscan group
in California (Epstein and Taylor, 1967; Garlick and Epstein, 1967; Taylor and
Coleman; 1968). The results are summarized as follows:

Glaucoph ane-schist laCles

&: • { Franciscan, type III 270-315°C
F· IV
ranClscan, type 410-535°C
Biotite zone 350-400°C
Appalachians { Almandine zone 410-490°C
Staurolite-kyanite zone 490-600°C
Sillimanite zone 600-750°C

(For the meaning of types III and IV, refer to § 7 A-10.)

These figures are mutually consistent and are fairly close to the temperature
values inferred from phase equilibrium data. This fact suggests that the isotope
distribution obtained at the maximum temperature has been approximately
preserved to the present.
The wide range of temperature of the glaucophane-schist facies is noted. This
facies covers at least the same temperature range as the greenschist and
epidote-amphibolite facies combined, and probably extends toward a lower
temperature.
Chapter 4

Some Important Problems in

Metamorphic Geology

4-1 IGNEOUS ROCK ASSOCIATIONS RELATED TO REGIONAL META-

MORPHISM

Contrasting Igneous Rocks between Low- and High-Pressure Terranes

In a classical view, one cycle of orogeny includes a long series of events
beginning with geosynclinal sedimentation, followed by phases of deformation,
regional metamorphism, and plutonic intrusion, and ending with post-meta-
morphic disturbances. The duration of a cycle is usually of the order of tens or
hundreds of million years. In some regions, the- geosynclinal sedimentaton is
followed immediately by tectonic disturbances and metamorphism. In other
regions, there is so long a lapse of time between the end of sedimentation and
metamorphism that the existence of genetic connection may well be doubted.
Many authors believed that three distinct phases of magmatism exist in a
cycle of orogeny, namely: basaltic magmatism in the geosynclinal stage, granitic
intrusion in the orogenic stage, and post-orogenic magmatism (Tyrrell, 1955; de
Sitter, 1956; Aubouin, 1965). However, they paid no attention to the relation-
ship of igneous rocks to the associated metamorphism. The characteristics of
igneous rocks vary with the baric type of associated regional metamorphism as
emphasized by Miyashiro (1961a, 1967b, c) and by Zwart (1967a,b, 1969). The
associated igneous rocks are mainly of acidic and intermediate compositions in
low-pressure metamorphic terranes, but of basic and ultrabasic compositions in
high-pressure terranes. Medium-pressure terranes tend to be intermediate in
character in this respect also (table 3-2).
Table 4-1 summarizes the contrasting petrologic and tectonic characteristics
of orogenic belts with low-pressure and high-pressure metamorphic terranes,
particularly on the basis of the available data from the Hercynian and Alpine
belts.
We may add that a relationship also exists between the baric types of regional
metamorphism and the sedimentation in relevant metamorphic terranes. Typical
aluminous pelites and limestones are usually widespread in low-pressure meta-
morphic terranes, but are scarce in high-pressure ones. In the latter, the most
94/S0ME IMPORTANT PROBLEMS IN METAMORPHIC GEOLOGY

Table 4-1 Contrasting characters of orogenies. After Zwart (J967a)

Hercynotype orogeny Alpinotype orogeny

Metamorphism 1. Shallow, low-pressure 1. Deep, high pressure

metamorphism metamorphism
2. Narrow progressive mineral 2. Wide progressive mineral
zones zones
3. Mineral changes due to 3. Mineral changes due to
increasing temperature decrease of rock-pressure
commonly observed tend to take place possibly
owing to rapid erosion
resulting from rapid
isostatic uplift
Igneous rocks 1. Abundant granites and 1. Few granites and migmatites.
migmatites 2. Abundant ophiolites with a
2. Few ophiolites. Ultrabasic considerable amount of
rocks practically lacking ultrabasic rocks
Width of orogen Very wide (thousands of Relatively narrow (hundreds of
kilometres) kilometres)
Isostatic uplift after Small and slow uplift. Large and rapid uplift.
disappearance of (Average rate of uplift is (Average rate of uplift is
compressive force about 1 mm per 10-25 about 1 mm per year)
years)
Structure Nappe structures missing Nappe structures predominant
or rare

typical sedimentary rocks. are graywackes. The K2 O/Na2 0 ratio and the
maturity tend to be higher in clastic sediments of the former than in those of the
latter (e.g. Miyashiro and Haramura, 1966).

Ophiolites. Abundant metabasalts and metagabbros occur in high-pressure

regional metamorphic terranes such as the Franciscan and Sanbagawa. The
structural features suggest that originally some of the metabasalts were pillow
lavas and others were pyroclastics. They are usually accompanied by serpenti-
nite. The basic and ultrabasic rocks of such an association have traditionally
been called ophiolites (e.g. Bailey, Irwin and Jones, 1964).
Some high-pressure metamorphic terranes are so chaotically disturbed that
reliable interpretation of the field relations of ophiolites are not possible. In the
Sanbagawa terrane of central Shikoku, however, disturbance was not so strong
and the modes of occurrence of the ophiolite have been clarified to a
considerable extent. There are many layers of basic schists which are apparently
conformably interbedded with pelitic schists. Some layers can be traced for a
distance of many kilometres within the deformed geosynclinal pile. This is
clearly shown in the geologic maps, for example, by Kawachi (l968, fig. 3) and
by Ernst et al. (1970, plates 4 and 5). Thus these layers appear to have been
derived from pyroclastics (and possibly also from associated lavas) which were
 4-1 IGNEOUS ROCK ASSOCIATIONS /95
erupted in the geosyncline concurrently with sedimentation. Basic rock layers in
only slightly recrystallized parts of the terrane commonly preserve the original
pyroclastic structure. Serpentinite usually occurs as independent masses in the
neighbourhood. There is no definite stratigraphic sequence in the rocks. Such
basic and ultrabasic rocks may well be referred to as non-sequence type
ophiolites.
There is another group of ophiolites, here called the sequence-type ophiolites
for convenience, which usually form layered masses showing a more or less
regular stratigraphic sequence of rocks: ultrabasic rocks ~ gabbros ~ basalts
from the bottom upwards. The Troodos massif, Cyprus (Gass, 1968; Moores
and Vine, 1971) and the Vourinos complex, Greece· (Moores, 1969) are
well-documented examples of this type. These two examples occur in
unmetamorphosed parts of the Alpine terranes and probably represent basaltic
volcanoes in ancient island arcs (A. Miyashiro, in preparation).
Most of the well-developed high-pressure metamorphic belts are accompanied
by abundant ophiolites. This suggests the existence of some genetic connection
between ophiolites and high-pressure metamorphism. Such ophiolites may have
been derived from rocks in oceanic regions.
In the last several years, a large number of papers have been published on the
origin of ophiolites. Most of them have supported the hypothesis that ophiolites
represent fragments of oceanic crust and upper mantle pushed into a pre-existing
geosynclinal sediment pile during strong disturbances caused by plate under-
thrusting (e.g. Dietz, 1963; Coleman, 1971; Moores and Vine, 1971). This view
may be valid for some sequence-type ophiolites which represent conceivable
structures of an oceanic crust and the underlying mantle, but may not be valid
for many of the non-sequence-type ophiolites, as discussed above.
Some ophiolites, such as those in Puerto Rico, show a marked Similarity in
chemical composition to common ocean-floor rocks (Donnelly, Rogers, Pushkar
and Armstrong, 1971). However, most metabasaltic rocks in the ophiolitic
associations in high-pressure terranes differ in chemical composition from
common ocean-floor igneous rocks. No one has examined whether or not the
difference can be ascribed to the metamorphism in mid-oceanic ridges and
orogenic belts as well as to ocean-floor weathering. Thus, we are not prepared to
make any reliable comprehensive evaluation of the relevant hypotheses.

Spilite-Keratophyre group and Granitic Rocks. Metamorphosed basaltic rocks

are present usually only in a small quantity in low-pressure metamorphic
terranes (e.g. the Ryoke and Hercynian belts). There, intermediate and acidic
volcanics may be relatively common. An association of slightly metamorphosed
basic, intermediate and acidic volcanic rocks forms a spilite-keratophyre group
(Rocci and Juteau, 1968). In rare cases, ultrabasic rocks also occur in the
association.
96/s0ME IMPORTANT PROBLEMS IN METAMORPHIC GEOLOGY
A large amount of granitic rocks, ranging from quartz diorite to granite in the
strictly-defined petrographic sense, occurs in all low-pressure regional meta-
morphic terranes. It appears that granitic magmas and associated fluids are
important carriers of heat in low- (and medium) pressure regional meta-
morphism. Low-temperature zones of regional metamorphism would be formed
by a general rise of temperature in thick geosynclinal piles, but high-temperature
zones might be mostly formed through heat transfer by upward movement of
granitic masses and associated fluids. In deeper levels, such regional meta-
morphism would cause partial melting of rocks, resulting in the formation of
granitic magma, whose rise would carry up a large amount of heat to higher
levels.
The cooperative heat effect of granitic masses and regional metamorphism
can take place only where they are synchronous. In most low-pressure regional
metamorphic terranes, granitic masses appear to have been emplaced at a late
phase of and after the end of regional metamorphism.

Examples of Granitic Rocks

Finnish Precambrian Terrane. Granitic rocks are widely exposed in the Svecofen-
nian low-pressure metamorphic terrane of the Baltic Shield (§ 14-2). Some
granites are gneissose, but others are not. Such differences were ascribed to the
different relationships of the granites to the tectonic movements in a cycle of
orogeny. Thus, the idea of the tectonic classification of granitic rocks was
formulated by Eskola (1932), Simonen (1960a, b), Marmo (1962) and others in
Finland, as follows:
(a) Synkinematic granites. These are gneissose and concordant with the
structure of the surrounding metamorphic rocks, and contain abundant basic
inclusions. They are presumed to have been emplaced at an early stage of an
orogenic movement, and the gneissosity has been ascribed to orogenic deform-
ation. The so-called Older Granite of Finland belongs to this group. Rocks of
this group are mostly granodioritic and quartz-dioritic in composition, though
some ultrabasic and gabbroic rocks also occur.
(b) Late-kinematic granites. These are seldom gneissose, and clearly cross-cut
the country rocks or show a diapir-like mode of intrusion within the axial
culminations of metamorphosed terranes. They are presumed to have been
emplaced at a later stage of orogenic movement. Rocks of this group in Finland
are commonly called Younger Granite. They are micro cline-rich granites or
aplites, and are not accompanied by ultrabasic and basic rocks. Sederholm
(1907, 1923, 1926, 1967) investigated the process of migmatization related to
the Younger Granite in southwestern Finland.
(c) Post-kinematic granites. These occur as discordant bodies entirely devoid
of gneissosity, and are presumed to have been emplaced after all the deform-
 4-1 IGNEOUS ROCK ASSOCIA TIONS /97
ationai movements had ceased. The Rapakivi Granite of southern Finland
belongs to this group.
Recent radiometric dating has shown that both Older and Younger Granites
were formed about 1800 m.y. ago, and the Rapakivi Granite about 1650 m.y.
ago (Eskola, 1963). No detailed analysis has been made of the tectonic
movements. The meaning of syn- and late-kinematic, therefore, is not definitive
in this case.

Scottish Caledonides. The granitic rocks of the Scottish Highlands are usually
classified into two groups: the Older and the Newer Granites (§ 14-3). The
Older Granites are gneissose and occur in many small concordant bodies in
high-temperature parts of regional metamorphic areas. No contact meta-
morphism is noticed. Barrow (1893) ascribed the regional metamorphism to the
temperature rise resultant from intrusion of the Older Granites. Recent investiga-
tions, however, suggest that the Older Granites are syntectonic masses resulting
from regional metamorphism, metasomatism and migmatization (Harry, 1958;
Dalziel, 1966).
In the Dalradian metamorphics, four (or five) phases of deformation have
been distinguished. The temperature of metamorphism increased progressively,
reaching the maximum value between the third and fourth phases. The regional
rnigrna tization and the formation of the Older Granites and sillimanite zone
probably took place in this period (Chinner, 1966a, b; Johnson, 1963; Harte and
Johnson, 1969), probably about 500 m.y. ago or earlier (Brown, Miller, Soper
and York, 1965).
The Newer Granites occur commonly in large discordant masses, including
not only granitic but also dioritic and gabbroic rocks. They show clear contact
metamorphism producing a hornfels zone. The time of intrusion is about
415-390 m.y. ago, that is, Late Silurian to Early Devonian (Brown, Miller and
Grasty, 1968). At the same time, many masses of granitic rocks were intruded
into unmetamorphosed terranes in the Midland Valley and the Southern Uplands
to the south of the Highlands (Read, 1961). This period of intrusion roughly
agrees with that of the last phase of deformation.

Northern Appalachians. Three plutonic series, named the Olive rian , the New
Hampshire and the Late Devonian, have been distinguished among the Devonian
Plutons of the northern Appalachians. These series show systematic differences
in relation to the wall rocks, metamorphic effects, modes and place of
emplacement, internal structure, mineralogy and so on. Page (1968) regards the
three series as representing pretectonic, syn- and early post-tectonic, and late
post-tectonic intrusions respectively in relation to the Acadian (Devonian)
orogeny. The Oliverian and a part of the New Hampshire plutonic series are
gneissose. The masses of the New Hampshire series occur characteristically in
98/s0ME IMPORTANT PROBLEMS IN METAMORPHIC GEOLOGY

high-temperature zones of the regional metamorphic terrane, suggesting the

existence of a close genetic connection with the metamorphism. The masses of
the Late Devonian series produced contact-metamorphic aureoles.
However, many other authors consider that the rocks of the Oliverian series
are actually metamorphic derivatives of volcanic and associated plutonic rocks,
and form the cores of mantled gneiss domes. The cores are encircled concordantly
by a mantle of well-stratified metamorphic rocks (especially of the Ammonoosuc
Volcanics). The volcanic rocks within the cores are regarded as stratigraphically
underlying the envelope rocks (Ordovician). During the metamorphism, the core
rocks would have risen to form domes owing to their lower densities (Thompson,
Robinson, Clifford and Trask, 1968; Naylor, 1968).

Cooma Granite, New South Wales. The Cooma granite is exposed over an area of
about 31 km 2 roughly along the thermal axis of a Palaeozoic low-pressure
regional metamorphic belt in southeastern Australia (§ § 7A-3 and 16-2). The
granite is gneissose and contains quartz, feldspars, biotite and muscovite,
together with subordinate sillimanite, andalusite and cordierite. The outer limit
of the metamorphic belt is at a distance of about 10 km or less from the granite
body (Joplin, 1942).
Pidgeon and Compston (1965) have determined the Rb- Sr ages and the initial
87 Sr/ 86Sr ratios (for total-rock samples) of the granite and associated meta-
sediments with the following results.

Rb-Sr age (m.y.) Initial 87Sr/86Sr

Cooma granite 415 ± 12 0·7179 ± 0·0005
Migmatites (surrounding 406 o·no
the granite)
Metasediments of high- 399 ± 29 0·719 ± 0·002
temperature zone
Metasediments of low- 460 ± 11 0·710 ± 0·002
temperature zones

The Rb-Sr age of the Cooma granite is the same as that of the associated
high-temperature metasediments within experimental error, whereas it is signifi-
cantly smaller than that of the associated low-temperature metasediments. It is
conceivable that the low-temperature metamorphism occurred about 460 m.y.
ago, whereas the high-temperature event and granite emplacement occurred
about 410 m.y. ago. Since the original sediments seem to be Ordovician, the 460
m.y. metamorphism would have immediately followed the sedimentation.
The 87 Sr/ 86 Sr ratio of the upper mantle is generally considered to be about
0·702-0·704. The initial 87 Sr/ 8 6Sr ratios of most Phanerozoic granitic rocks are
about 0·705-0·710, whereas the same ratio of the Cooma granite is much higher
 4-1 IGNEOUS ROCK ASSOCIATIONS /99
(about 0·718), and is virtually the same as that of the associated high-
temperature metasediments. The 87Sr/86Sr ratio of the metasediments in the
low-temperature zones should have gradually increased so as to become similar
to the initial ratio in high-temperature zones and in the granite at the time of
granite emplacement (410 m.y. ago). Therefore, a likely explanation for the high
initial 87Sr/86Sr ratio of the granite is that the granite magma was formed by
the melting of associated sediments, as far as the Sr isotope evidence is
concerned.

Holum Granite, Norway. It is conceivable that some granites in orogenic belts

are of magmatic origin, but others are of metamorphic origin, though the
available evidence is mostly equivocal. As an example of granitic plutons that
have been interpreted as being of metamorphic origin, the Holum granite in a
Precambrian migmatite region at the southern tip of Norway will be described
below (Smithson and Barth, 1967). Here, the term migmatite is used in the
non-genetic meaning (§ 1-4).
Migmatites occur widely in the area. There is a zone of augen gneisses,
0-2 Ian wide, between the migmatite area and the Holum granite. The augen
gneisses contain well-developed porphyroblasts of K-feldspar perthite together
with smaller grains of K-feldspar, plagioclase, quartz, hornblende and biotite.
The Holum granite is a pluton elongated in the N-S direction with a
maximum width of about 7 km. The rocks usually show a weak parallel texture,
and are mainly composed of K-feldspar perthite, plagioclase, quartz, biotite and
hornblende, just like the adjacent augen gneisses; petrographically, they are
adamellite. The K-feldspars of the granite and the adjacent augen gneisses are
usually microcline, but are nearly monoclinic in some specimens.
Mineralogically and probably chemically, the granite closely resembles the
augen gneisses. The main difference between them lies in texture. The augen
gneisses show a stronger parallel texture and a stronger tendency to xenomorphic
crystal outlines. The trend of the parallelism in the granite, augen gneisses and
migmatites generally follows the direction of elongation of the pluton. The
augen gneisses grade into the granite with a transitional zone up to 500 m wide.
The main chemical variation within the granite is produced by different
proportions of K-feldspar perthite (OrsoAb 2o ) and plagioclase (An 26 - 30 ),
where the composition of each feldspar remains nearly constant. The com-
positions of various parts of the granite fall mostly outside the low-temperature
valley of the pertinent experimental systems. These data suggest that the Holum
granite was formed not by magmatic crystallization, but by metamorphic
recrystallization of the associated augen gneisses. The recrystallization appears to
have taken place in a synform, possibly being accompanied by incipient partial
melting.
tOO/SOME IMPORTANT PROBLEMS IN METAMORPHIC GEOLOGY

4-2 VOLCANIC ARCS IN RELATION TO LOW-PRESSURE REGIONAL

METAMORPHISM

Low-Pressure Metamorphic Belts and Volcanic Arcs

The Quaternary volcanism of continental margins and island arcs is characterized
by the great abundance of intermediate and acidic rocks, though basaltic rocks
also occur. (It will be shown later that this is due to the abundance of rocks of
the calc-alkali series in a narrow sense.) The volcanism of the low-pressure
terranes and granitic belts such as the Ryoke belt (§ 15-7) and the Sierra
Nevada- Klamath zone (Dickinson, 1962) is also characterized by the abundance
of intermediate and acidic rocks, though the ratios of intermediate rocks to
acidic ones differ in different cases. It may be presumed that andesitic-dacitic
volcanic chains on island arcs and active continental margins are surface
manifestations of low-pressure metamorphic terranes and granitic belts.
The main differences in the ratio of volcanic to granitic rocks and in the grade
of metamorphism in such areas could be ascribed to differences in the degree of
erosion. Table 4-2 shows an attempt to explain the different characteristics of
Table 4-2 Hypothetical series a/increasing depth

Andesitic-
rhyolitic Regional
General depths Examples volcanics Granites metamorphism

Surface Quaternary Abundant Absent Absent

volcanic arcs
Late Tertiary Abundant Scarce Zeolite and
terrane on prehnite-
the Japan Sea pumpellyite
side of north- facies
east Japan
Sierra Nevada- Abundant Abundant Mainly green-
Klamath zone schist facies(?)
of N. America
Relatively Ryoke belt of Present Abundant Mainly amphibo-
deep Japan lite facies

Note: Contact metamorphism by granitic plutons is ignored. For the Late Tertiary terrane
of northeast Japan, refer to § 15-9.
four regions in this way (Miyashiro, 1972b). With increasing depth, the grade of
metamorphism should increase and the ratio of volcanic rocks should decrease.
It is not clear whether the ratio of granitic rocks increases with depth, or
whether it first increases and then becomes constant or even decreases.
Thus, a low-pressure regional .metamorphic complex would be present
beneath a belt of volcanic piles in island arcs and continental margins. Therefore,
volcanic rocks in island arcs will be discussed in some detail below.
4-2 VOLCANIC ARCS IN RELATION TO REGIONAL METAMORPHISM/lOl

Igneous Petrology of Volcanic Arcs

Igneous Rock Series. In the first half of the twentieth century it was realized
that the chains of volcanoes in island arcs and active continental margins contain
large amounts of basalt, andesite, dacite and rhyolite, in which the Si02 and
alkali contents tend to increase and the MgO and Fe203 + FeO contents tend to
decrease from basalt to rhyolite in the above-named order. Bowen (1928)
regarded this order as representing progressive fractional crystallization which is
controlled by two main reaction series, one for colourless and the other for
coloured minerals. The coloured minerals were considered to crystallize in the
order: olivine ~ orthopyroxene ~ clinopyroxene ~ amphibole ~ biotite. Such
volcanics were usually classed as sub alkali or calc-alkali rocks.
Fenner (1929), and more spectacularly Wager and Deer (1939) demonstrated
the existence of the tholeiitic series of igneous rocks which is not alkaline but is
distinct from the above calc-alkali rocks. The tholeiitic series shows little or no
increase in Si0 2 content and instead a considerable increase in Fe203 + FeO
content in the early (or main) part of its course of crystallization. This series has
olivine and pyroxenes as the main coloured minerals but has little or no
amphibole and biotite. The tholeiitic series is usually composed of abundant
mafic, less abundant intermediate and few acidic rocks, whereas the calc-alkali
series usually includes abundant intermediate and acidic rocks.
Thus, Nockolds and Allen (1953, 1954, 1956) have discussed the composi-
tional variation of igneous rocks in terms of a threefold classification into the
tholeiitic, calc-alkali and alkali series. Kuno (1959), Taylor and White (1965),
Jakes and White (1972) have accepted this classification with certain modifica-
tions in their studies of island arc volcanism in Japan and southwest Pacific
regions. Alkali basalts in some areas have K2 0/Na2 0 ratios much lower than
unity, whereas those in others have K2 0/Na2 0 ratios near unity. Joplin (1964)
and Jakes and White (1972) have accepted the name shoshonite for the latter
group of alkali basalts.
Textbooks of petrography may define the name andesite mainly in terms of
Si0 2 content, colour index, or the ratios of plagioclase-alkali feldspar-silica
minerals. In such definitions, all the three series of volcanic rocks include
andesites as members. However, the 'andesites' so defined show different
chemical and mineralogical characteristcs in the three different series. This
situation is undesirable as the rock series classification has great genetic
implications. Thus, Macdonald (1960) has proposed to use the name hawaiite
and mugearite for the so-called 'andesites' of the alkali series. Carmichael (1964,
p. 442) coined the name icelandite for 'andesites' of Iceland, and this name may
well be accepted for the 'andesites' of the tholeiitic series. In this way, the use of
the name andesite can be limited to the 'andesites' of the calc-alkali series.
Thus, we have the following three series of volcanic rocks in island arcs and
active continental margins.
102/ SOME IMPORTANT PROBLEMS IN METAMORPHIC GEOLOGY

1. Tholeiitic series, including tholeiitic basalt, icelandite and some dacite. The
Si02 contents are mostly in the range 48-63 per cent by weight.
2. Calc-alkali series, including abundant andesite and dacite, and some
rhyolite. The Si0 2 contents are mostly in the range 52-70.
Series (1) and (2) are 'calc-alkalic' in the broader meaning that they both have
Peacock's (1931) alkali-lime indices as high as 56-57.
3. Alkali series, which may well be subdivided into: (a) the sodic alkali group,
including alkali olivine basalt, hawaiite, mugearite, trachyte and alkali rhyolite,
and (b) the shoshonite group, including shoshonite, latite and leucite-bearing
rocks.
Compositional Variation Across Volcanic Arcs. In island arcs which have a
Benioff zone inclined towards the continent as in northeast Japan, the Kurile
Islands and Indonesia, the volcanic rocks tend to be increasingly alkaline toward
° ° °
the continent. In other words, the K2 and Na2 + K2 contents and the K/Na
ratio of volcanic rocks tend to increase towards the continent, if we compare
rocks with the same Si02 content (Sugimura, 1960, 1968; Kuno, 1959, 1960,
1966; Katsui, 1961; Dickinson, 1968; Dickinson and Hatherton, 1967; Kawano,
Yagi and Aoki, 1961; Yagi, Kawano and Aoki, 1963).
In active island arcs in mature stages of development such as northeast Japan
and Kamchatka, petrographic provinces of the tholeiitic series, calc-alkali series
and alkali series are present succeSSively in this order from the oceanic to the
continental side of the volcanic belt, though there may be marked overlapping of
provinces (fig. 15-3).
Nature of ihe Calc-Alkali Series. The status and character of basalt in the
calc-alkali series are not clear. Many volcanoes made up of rocks of the calc-
alkali series have no basalt. The series may begin with andesite. Nockolds
and Allen (1953) found in variation diagrams that acidic and intermediate rocks
of the calc-alkali series fall on a smooth curve for each area, whereas more basic
rocks show more or less random scattering. They interpreted this as suggesting
that the parental magmas were intermediate (Le. andesite or dioritic) in
composition (with Si02 = 52-56 per cent), and that more basic rocks of the
series were formed by the accumulation of early crystals. Green and Ringwood
(1966, 1968) demonstrated experimentally the possibility of the formation of
primary andesitic magma.
Some basalt occurs in many andesitic volcanoes, however. Such andesitic
rocks may have been derived from basaltic magmas (e.g. Jakes and White, 1972;
Aoki and Oji, 1966; Kuno, 1968). Osborn (1962) has emphasized the possible
importance of oxygen fugacity in the derivation of the tholeiitic and the
calc-alkali series from more or less similar basaltic magmas.
Volcanic rocks of the calc-alkali series are confined to orogenic regions (table
4-3). The occurrence of calc-alkali rocks, or of a great amount of andesitic and
4-2 VOLCANIC ARCS IN RELATION TO REGIONAL METAMORPHISM /103
Table 4-3 Younger volcanic rocks and tectonic environment
Stable Oceanic Mid-oceanic
continents Orogenic belts islands ridges
Mature, very
active island
arcs and
Immature, active con-
very active tinental Less active
island arcs margins island arcs

Tholeiite ++ ++ ++ + + ++
series
Calc-alkali ++ ++
series
Alkali ++ + + ++ +
series

+ +, abundant; +, subordinate

dacitic rocks may be regarded as suggesting the former existence' of an island arc
or an active continental margin.
The volcanic rocks of the calc-alkali series resemble granitic rocks. in chemical
composition and they both occur in orogenic belts. They could have been
derived from the same andesitic (dioritic) magma (e.g. Nockolds and Allen,
1953; Dickinson, 1970). It is of interest to discover whether granitic rocks show
a regular compositional variation across an island arc and a continental margin
like the volcanic rocks. The existence of such a variation in granitic rocks has
been demonstrated by Moore (1959) and Moore, Grantz and Blake (1961) on
the west coast of North America, and by Taneda (1965) in Japan.

Factors Controlling the Diversity of Magmas. Magmas which cause volcanism in

island arcs and active continental margins should be created in some genetic
relationship to the descending lithospheric slab. Since volcanic rocks, in particular
those of the calc-alkali series, are confined to volcanic arcs (table 4-3), they
could be directly related to the descending slab. For example, the primary
andesitic magma which gives rise to this series could be created by partial melting
of the oceanic crust which forms the uppermost layer of the descending slab, as
shown in fig. 3-11 (cf. Ringwood, 1969).
Basaltic magmas of the tholeiitic and alkali series could be produced by
partial melting either in the descending slab or in the upper mantel overlying it.
The oceanic crust with abyssal tholeiitic composition and the underlying
presumably peridotitic layer of the descending lithospheric slab should undergo
a series of phase changes with increasing depth, that is, with increasing pressure.
The equilibrium relations between melt and solid residue should differ at
104/ SOME IMPORTANT PROBLEMS IN METAMORPHIC GEOLOGY

different pressures. Recent experimental results suggest that the melt formed in
peridotite under greater pressure tends to be more undersaturated with silica
(e.g. Green and Ringwood, 1967; Kushiro, 1968). Even under the same P-T
conditions, the composition of the magmas tends to be less alkalic as the
proportion of the melt increases. The pressure and the proportion of melt could
vary regularly with depth in the descending slab, resulting in a regular
compositional variation in volcanic rocks across the volcanic arc.
Even if we may assume with Ringwood (1969) that basaltic magmas are
generated by partial melting not in the descending slab but in the overlying
upper mantle where convective upwelling or diapiric rise occurs, the upwelling
and the generation and separation of basaltic magmas should be controlled
largely by the descending slab, and hence the resulting magmas could vary
regularly across the arc.

Rate of Plate Motion and the Evolution of Island Arcs

There exists a clear correlation between the rate of plate motion and the
activeness and petrography of island arcs (Miyashiro, 1972b), as summarized in
table 4-4. Here, the activity of arcs may be represented by the depths of mantle
earthquakes and the associated trenches. The depth of a trench should be
influenced by the rate of burial by sediments, but nevertheless it shows a good
correlation with the rate of plate motion (cf. Sugimura, 1968).
In table 4-4, group I of island arcs is characterized by a rate of relative motion
(convergence) of plates as high as 8-9 cm/year. Here, the maximum depth of
earthquakes is 600-700 km, that of the trench is 10-11 km, and volcanic rocks
of the typical tholeiitic series are well developed. Rocks of the calc-alkali and
alkali series are abundant only in arcs in the mature stage of development in this
group. Fig. 15-3 illustrates petrographic provinces in such a mature stage.
In group II of island arcs, the rate of convergence of plates is slower and the
maximum depths of earthquakes and of the trench are smaller, than those in
group I. Here, typical tholeiite is absent, though some rocks present show an
affinity with the tholeiitic series. Calc-alkali volcanic rocks are abundant. Alkali
rocks, though usually present, are not abundant.
Group III is characterized by a still lower rate of plate convergence (about
0- 2 cm/year). Alkalic rocks are well developed in such arcs, though a few
atypical tholeiites and calc-alkali rocks may also occur. Thus, in the order of
decreasing rate of plate convergence, the main volcanic rocks of island arcs vary
from tholeiites to calc-alkali rocks, and then to alkali rocks.
Since the K20 and usually also the Na20 content decrease across an arc
towards the ocean, the volcanic rocks in the oceanic-side zone within the
volcanic belt usually have smaller K 20 and Na20 contents than the volcanic
rocks in the continental-side zone. Fig. 4-1 shows a comparison of the K2 0
contents of volcanic rocks in the oceanic-side zone within the volcanic belts of
 ~
<
o
t"
C':l
>
Z
Table 4-4 Correlation between the activeness and volcanic rock series in island arcs. After Miyashiro (1972b) ....
C':l
>
~
Rate of plate Maximum depth Maximum depth
C':l
convergence of earthquakes of the trench Volcanic rock til
Group Arc and trench (em/year) (km) (km) series Z
~
ttl
I Tonga 9 700 11 Th t"
Izu-Bonin 9 600 11 Th + (C) + (A) >
N.E. Japan 9 600 11 Th+ C + (A) ....>..,]
Kurile-Kamchatka 8 600 10 Th+C +A
o
z
II Aleutian 6 300 8 (Th) + C + A >..,]

Indonesia 5-6 600 7 (Th) + C + A o

~
Ryukyu ? 300 7 (Th) + C ttl
North Island (N.Z.) Slow (3?) 300 4 C+A Q
Hellenic (Aegean) Slow (3?) 200 4 (Th) +C + A o
Z
III Calabrian (Sicily) Very slow (2?) 300 Buried A >
Macquarie Very slow 100 Shallow (Th)+C+ A t"
s::
ttl
Note: The activeness of arcs decreases in the order of group I ~ II ~ III. Th = tholeiitic series, C = calc-alkali series, A = alkali series. The >..,]
rocks of the series shown in parentheses are not typical of the series, and very small in quantity. The rates of plate convergence are mainly >
after Le Pichon (1968). s::
o
~
'tl
:I:
....
til
s::
.....
-
~
106/ SOME IMPORTANT PROBLEMS IN METAMORPHIC GEOLOGY

island arcs. The very active arcs (group I in table 4-4) always show lower KzO
contents than less active arcs (groups II and III).
In group I, the alkali basaltic rocks are sodic, whereas arcs of groups II and III
appear to have both sodic and potassic (shoshonitic) types of alkali basaltic
rocks. The Calabrian Arc (group III) has especially well-developed potassic
basaltic rocks.

Aleutian (II)
Ryukyu (II)
N.E. Japan (I)

~;;j~~~~::;;;:;;~~::==~==:~~~~:~Kurile (I)
Izu·Bonin (\)

Fig. 4-1 K2 0 versus Si0 2 relation in volcanic rocks of the oceanic-side

(i.e. trench-side) zone within the volcanic belts of island arcs. The Roman
numerals in parentheses represent the group numbers shown in table 4-4.
(Miyashiro, 1972b.)
The activity of the northeast Japan Arc began in the early Miocene, and
magmatism of the tholeiitic, calc-alkali and sodic alkali series took place from
the beginning (Ozawa, 1968; Miyagi, 1964). Jakes and White (1972) claimed
that the occurrence of shoshonite begins at a later stage in the development of
island arcs.

4-3 METAMORPHIC FACIES AND GEOLOGIC AGE

Distribution Rules
Some metamorphic facies show clear relationships to geologic age. Most of the
glaucophane-schist facies rocks now observed in the world were formed by
metamorphism in Cenozoic and Mesozoic time. Fig. 4-2 shows the distribution
of glaucophane in the world. Its relationship with younger orogenic belts is clear
(Eskola, 1939, p. 368; de Roever, 1956, 1964; van der Plas, 1959; Miyashiro,
1961a; Zwart, 1967a; Dobretsov, 1968; Ernst 1972).
 4-3 METAMORPHIC FACIES AND GEOLOGIC AGE /107
There are some glaucophane schists in Palaeozoic metamorphic areas. How-
ever, such rocks occur extremely rarely in Precambrian metamorphic terranes
(Sobolev: et aZ. 1967).
Zeolite-facies rocks appear to be much more common in Cenozoic and
Mesozoic formations than in the older ones (e.g. Hay, 1966). On the other hand,
most of the granulite facies rocks appear to have formed in Precambrian time.
They usually occur in the shields, but are sometimes present in Phanerozoic
orogenic belts as uplifted blocks of Precambrian basements (Oliver, 1969). A
small proportion of granulite facies rocks, however, appears to be truly young.
Fig. 4-3 shows the distribution of granulite facies terranes in the world .

..
)

Fig. 4-2 Distribution of glaucophane in the world. (Largely based on van

der Plas, 1959 and Sobolev et al. 1967.)

Many authors have pointed out the existence of a close relationship between
anorthosites and geologic age. It has been claimed that there are at least three
types of anorthosites. One type occurs as a member of differentiated gabbroic
complexes such as the Bushveld of South Africa, and can be of any geologic age.
Another type occurs in great dome-like or batholithic masses mainly composed
of andesine or labradorite, and of middle Precambrian age (11 00-1700 m.y.
old). The third type occurs as a member of layered masses, being mainly
composed of bytownite or anorthite, and of early Precambrian age (more than
3000 m.y. ago). The latter two types show close association with granulite or
amphibolite facies metamorphic rocks (Isachsen, 1969; Windley, 1970).
On the other hand, low-pressure regional metamorphic rocks of the green-
schist and amphibolite facies show no relationships to geologic age. Such rocks
 o
-
~
~ Dtr · . ,p {Il
o
s:
t>::I

~
..... ~
~
.'
~ ."
o
:=
I-j
>
Z
I-j
."
:=
o
lh t:I:I
t"'
t>::I
:.~(}/J"'~,~ d>a..... s:
{Il

~, Z
s:t>::I
I-j
~ Ii> d
~ >
s:
o
:=
."
::£:
.~(;..

". . . . . t( , ..V' ·d:~~ ... . ~ n

( '
r ~:.'1i~"{:~':'
.' C'l
t>::I
o
t"'
o
C'l
-<
Fig. 4-3 Distribution of granulite-facies metamorphic areas in the world.
(Oliver, 1969.)
 4-3 METAMORPHIC FACIES AND GEOLOGIC AGE/I09

occur, for example, in the Precambrian Svecofennian terranes, Palaeozoic

terranes of eastern Australia, the Hercynian terranes of western Europe, and
some Mesozoic terranes in Japan (table 4-5).
Apparent or Real?
It is an unsettled question whether these distribution rules have any real
significance.
Many geologists have doubted the validity of the idea that glaucophane-schist
facies rocks are confmed to younger metamorphic terranes. Some have con-
sidered that since Precambrian terranes are very extensive, large areas of
glaucophane-schist facies rocks may remain unnoticed. Others have suggested
that glaucophane schists formed in Precambrian time may have been lost in later
geologic time.
Zwart (1967a) has attempted to explain the rarity of glaucophane in old
geologic ages as follows: Glaucophane-schist facies metamorphism takes place in
the Alpinotype orogenic belts as already shown in table 4-1. Rapid and large
uplift occurs characteristically in such belts, resulting in removal of all the
low-temperature part (containing glaucophane) by erosion. Another possibility is
that glaucophane may have been replaced by minerals of the greenschist facies,
since su.ch changes are common in high-pressure metamorphics.
The first explanation leads to the possibility that only glaucophane-schist
facies rocks were lost from the Precambrian terranes, and rocks of high-
temperature facies of the high-pressure type may be widely exposed unnoticed
there. At present, we have no factual data to support or negate any of these
possibilities.
Phanerozoic high-pressure metamorphic belts containing glaucophane schists
appear to form along trenches in island arcs and continental margins, usually as
the ocean-side members of paired belts. If such belts were formed in Pre-
cambrian time, and if a later Precambrian or Phanerozoic orogeny produced new
paired belts roughly along the same zone but slightly on the ocean side of the
older belts, the low-pressure metamorphism and granite emplacement of the new
pair might take place on the same zone as the older high-pressure belt, which
thus could be lost by the new recrystallization. If continental growth produces
successively younger paired metamorphic belts slightly on the ocean side of the
older ones, the high-pressure belts of the older pairs might be successively
destroyed, and only the youngest pair might be observed undestroyed. This
would give the appearance that all the older phases of metamorphism were of
the low-pressure type.
Progressive Dehydration as a Possible Cause of the Distribution Rules
If the observed rules concerning the distribution of metamorphic facies through
geologic time are tentatively assumed to be real, what could be the possible
causes for them?
 -
Q
til
o
a::
t>l

sa
"CI
Table 4-5 Baric types of Regional metamorphism and geologic ages o
:;:tI
>.,]
Precambrian Palaeozoic Mesozoic-Cenozoic >
Z
>.,]

Low- Svecofennides Hercynides Ryoke-Abukuma belt (Japan) "CI

:;:tI
pressure Karelides Appalachians (partly) Hidaka belt (Japan) o
metamorphism Canada (partly) Eastern and South Australia =
t'"
Australia Hida belt (Japan) t>l
N.E. China Pichilemu series (Chile) a::til
Medium- Canada (partly) Caledonides North American Z
pressure Appalachians (partly) Cordillera (partly) a::
t>l
metamorphism >.,]
>
a::
High- Anglesey (Wales) Kiyama (Japan) Alps
Sangun belt (Japan) Franciscan group (California) o
pressure :;:tI
Curepto series (Chile) Sanbagawa belt (Japan) "CI
metamorphism :c
Penjna Range (NW Kamchatka) Kamuikotan belt (Japan) n
New Caledonia C)
Central Kamchatka t>l
o
t'"
o
C)
-<
 4-3 METAMORPHIC FACIES AND GEOLOGIC AGE /111
First, a large part of the rules appears to result from increasing dehydration
with geologic age. Rocks of the zeolite- and glaucophane-schist facies are
generally higher in H2 0 content than those of other facies, whereas rocks of the
granulite facies are generally the lowest in H2 0 content among regional
metamorphic rocks. Therefore, the abundance of the former in younger ages and
of the latter in Precambrian time may mean a tendency for increasing
dehydration with age. This trend can be noticed even when we focus our
attention on zeolites and feldspars. Hay (1966) has emphasized the correlation
between zeolites in sedimentary and metasedimentary rocks and their geologic
age. The sum of phillipsite, clinoptilolite, erionite and chabazite appears to
decrease, whereas analcime appears to increase with age within Cenozoic time.
Analcime appears to decrease with age through Palaeozoic time, whereas albite
and K-feldspar increase from Cenozoic to Precambrian. Such variations clearly
indicate advancing dehydration (Miyashiro and Shido, 1970).
A deeper level of the orogenic belt would tend to be exposed by advancing
denudation with age, leading to a generally higher temperature of metamorphism
and a higher degree of dehydration on the exposed level. This is a possible, but
not very convincing explanation.
Another possible process is the following: when a geosynclinal pile is
subjected to regional metamorphism for the fIrst time, pelitic and psammitic
rocks contain abundant H2 0 between and within mineral grains. Progressive rise
of temperature would cause progressive dehydration of minerals. In the initial
stages, metamorphic recrystallization would proceed in the presence of aJ}
intergranular aqueous fluid. A zeolite or glaucophane-schist facies zone would
form uf!,der such a condition. If the duration of metamorphic recrystallization is
prolonged, the H2 0 in the rock complexes would be gradually expelled upward,
and the com~ition would become drier. The zone of the zeolite and glaucophane-
schist facies would tend gradually to disappear. Bryhni, Green, Heier and Fyfe
(1970) pointed out that the onset of partial melting would also greatly
contribute to dehydration.
Recent accumulation of radiometric age data has shown that many regional
metamorphic terranes were subjected to two or more pl).ases of metamorphism.
With increasing age, the possibility of repeated metamorphism would also
increase. In particular, most Precambrian rocks may be polymetamorphic.
Repeated metamorphism should be similar in effect to prolonged recrystalliza-
tion, and would result in greatly advanced dehydration (Winkler, 1967. p. 139).
Progressive metamorphic reactions in hydrous sediments are strongly endo-
thermic. A large part of the heat transferred should be consumed in such
reactions, resulting in a relatively small rise of temperature, which may favour
the formation of rocks of high-pressure series. On the other hand, in repeated
metamorphism, endothermic reactions would not be as intense, and the
l12/s0ME IMPORTANT PROBLEMS IN METAMORPHIC GEOLOGY

temperature rise may be more rapid, promoting the formation of lower pressure
series rocks, especially in the Precambrian.
If this is true, the Precambrian shields must be generally drier than younger
orogenic belts. Hyndman and Hyndman (1968) have pointed out that there is a
layer of high electrical conductivity in the lower crust of young orogenic belts,
and the high conductivity is probably due to the saturation of the rocks with
H2 0, whereas such a layer is absent in shield regions.

Changes in the Thickness and the Motion of Plates as Possible Causes of the
Distribution Rules
If the oceanic plates have become thicker, the descending slabs have become
steeper, and the movements of the slabs have become more rapid in younger
geologic time, the occurrence of high-pressure metamorphism should have
become more common (Ernst, 1972a; Miyashiro, 1972b). Thus, we may assume
that plates were thin and/or moved slowly in Precambrian and early Palaeozoic
time. Under such conditions, metamorphism of the low- and medium-pressure
types should be common, and occasionally atypical glaucophane schists could
form.
In late Palaeozoic and early Mesozoic time, the plates may have become
thicker and/or may have come to move more rapidly, with resultant common
formation of high-pressure metamorphic rocks and paired metamorphic belts.
The beginning of the present cycle of continental drift in early Mesozoic time
may be another result of such a change in the plates.
Magnetic anomalies suggest that plate motion was more rapid in the Pacific
than in other oceans at least in late Mesozoic and later time (Heirtzler et al.
1968). It is conceivable that similar rapid motion took place in early Mesozoic
and late Palaeozoic time as well, leading to the common formation of
high-pressure metamorphic rocks in the circum-Pacific regions.
Runcorn (1965) has suggested that a great change in the pattern of currents
in mantle convection took place in early Mesozoic time, and as a consequence
the present cycle of continental drift began. This change in the pattern of
currents may have resulted also in an increase in the velocity of plate motion,
particularly in the Pacific region and may have caused the common formation of
glaucophane schists and paired metamorphic belts.

4-4 DIVERSITY IN LOW-TEMPERATURE METAMORPHISM

Observed Diversity. The metamorphic rocks formed at low temperatures,

including the zeolite, prehnite-pumpellyite, glaucophane-schist and greenschist
facies, show a greater mineralogical diversity than those formed at higher
temperatures. Though Ps and PCO 2 are factors influencing mineral paragenesis at
any temperature, their effects appear to be more evident at low temperatures.
Table 4-6 shows a synopsis of the observed diversity. In some regions, an
unrecrystallized area grades directly into a greenschist-facies zone. In others, a
 4-4 DIVERSITY IN LOW-TEMPERATURE METAMORPHISM /113
Table 4·6 Classification of the low-temperature parts of metamorphic facies
series

Baric type Facies series Examples

Low-pressure (a) Unrecrystallized zone Ryoke metamorphic belt, Japan

~ greenschist facies zone (Katada, 1965)
(b) Unrecrystallized zone Probably Mogami area, Japan
~ zeolite facies zone (Utada, 1965)
~ greenschist facies zone
(c) Unrecrystallized zone Tanzawa Mountains, Japan
~ zeolite facies zone (Seki, et aZ. 1969a)
~ prehnite-pumpellyite facies Probably Akaishi Mountains,
zone ~ greenschist facies zone Japan (Matsuda and Kuriyagawa,
1965)

Medium- (a) U nrecrystallized zone Part of the Scottish Highlands

pressure ~ greenschist facies zone (Wiseman, 1934)
Part of the northern
Appalachians, U.S.A. (Zen,
1960)
(b) Unrecrystallized zone Central Kii Peninsula, Japan
~ zeolite facies zone (Seki et aZ. 1971)
~ prehnite-pumpellyite facies Probably part of the South Island
zone ~ greenschist facies zone New Zealand (Coombs et aZ.
1959)

High-pressure (a) Unrecrystallized zone Bessi area, Japan (Banno, 1964)

~ glaucophane-schist facies Kanto Mountains, Japan (Seki,
zone 1958, 1960)
(b) Unrecrystallized zone Katsuyama area, Japan
~ prehnite-pumpellyite facies (Hashimoto, 1968)
zone ~ glaucophane-schist
facies zone

zeolite facies zone occurs in between. In still others, an unrecrystallized area

grades into a prehnite-pumpellyite facies zone, which in tum grades into a
glaucophane-schist facies zone.
A part of, the diversity could be due to a difference in the threshold
temperature of recrystallization, which would decrease with the increased
duration of metamorphic recrystallization. Other possible factors will be dis-
cussed below.

Effect of p co • . Higher PC02 tends to decompose Ca-bearing silicates with the

resultant formation of carbonates. This greatly contributes to the mineralogical
diversity of low-temperature metamorphic rocks.
The role of Pco 2 and PH 2 0 in the formation of zeolites and a zeolite facies
zone has been discussed particularly by Zen (1961b). The following composi·
tional relations hold for Ca-zeolites:
114/ SOME IMPORTANT PROBLEMS IN METAMORPHIC GEOLOGY

heulandite + CO 2 =
calcite + pyrophyllite (or kaolinite) + quartz + H2 0,
laumontite + CO 2 =
calcite + pyrophyllite (or kaolinite) + H2 0.

A high PH, 0 should promote the formation of Ca-zeolites, whereas a high Peo,
should tend to decompose them with the resultant formation of calcite and an
aluminous mineral. It follows that under a high Pco , ' an unrecrystallized area
may directly grade into a greenschist facies zone, and a decrease in Peo may
produce a zone of the zeolite facies between these two areas. •
Lawsonite, prehnite and pumpellyite, which are characteristic of the glauco-
phane-schist and prehnite-pumpellyite facies, are rich in Ca. They should be
unstable under high Peo, .
In the greenschist facies, actinolite and epidote are common Ca-bearing
silicates. These minerals are abundant in particular in greenschists recrystallized
at low Peo ,. With increasing Peo " actinolite fIrst and then epidote decom-
pose to produce calcite and dolomite (or ankerite) together with Ca-free
ferromagnesian minerals such as chlorite and stilpnomelane (§ 8A-6).
In the Sanbagawa and other low·temperature metamorphic terranes of Japan,
Seki (1965) has observed that the frequency of occurrence of calcite in
metabasites differs from area to area within the same metamorphic belt, that
Ca-bearing silicates (pumpellyite, epidote and actinolite) are not common in
areas where calcite is widespread in metabasites, and that the frequency of
occurrence of calcite tends to decrease with increasing temperature of meta-
morphism. This suggests that Peo differs from area to area within the same
•
metamorphic belt, resulting in a difference in the degree of carbonate formation.

Effect of PH. o. Even if an aqueous fluid may exist during metamorphism at low
temperatures, models (A), (B) and (C) described in § 2-3 give widely different
PH 0 under the same Ps • Highe"r PH 0 should tend to form a greater proportion
•
of hydrated minerals such as zeolites•and chlorites.
In some regions, metamorphic recrystallization immediately follows the
sedimentation of the geosynclinal pile, whereas in others recrystallization occurs
long after the completion of sedimentation. Conceivably, the abundance of H2 0
in a newly formed sediment pile may cause a higher PH 2 0, and hence may tend
to form zeolite facies rocks.

Relation to Geothermal Gradient. Zeolite facies zones form under a wide ran&e
of geothermal gradients, that is, in metamorphic terranes of the low-, medium-
and high-pressure types. However, the paragenetic relations of zeolites appear to
differ in different baric types (§ 6A-1).
 4-5 NATURE OF CONTACT METAMORPHISM /115
Zeolite facies areas, representing a relatively high geothermal gradient, show a
great variety of zeolites, including mordenite, stilbite, yugawaralite, wairakite,
heulandite, laumontite and analcime; e.g. the Tanzawa Mountains (Seki et al.
1969a). The present topographic surface in such areas represents a shallow depth
of low-pressure regional metamorphic belts, or of contact- or hydrothermally
metamorphosed areas. In contrast, zeolite facies areas, representing a low
geothermal gradient, have only a very limited variety of zeolites such as
heulandite, laumontite and analcime; e.g. the Taringatura area (Coombs et al.
1959). Such areas are parts of regional metamorphic terranes of the high-
pressure type or close to it.
The physical factors directly controlling the difference in zeolite paragenesis
between different baric types are not clearly known. Under the condition Ps =
PHz 0, the stability fields of some zeolites could be limited to a low-pressure
range so that their occurrence is confined to metamorphism of the low-pressure
type. It is conceivable also that the behaviour of H2 0 may differ in different
baric types. The threshold temperature of recrystallization would differ in
different types.

4-5 NATURE OF CONTACT METAMORPHISM

Baric Types
It has been widely believed that contact metamorphism takes place under lower
rock-pressures than does regional metamorphism. This idea is partly incorrect,
however, since there is a great diversity in the P-T conditions of contact
metamorphism.
Among the Al 2 SiO s polymorphs, andalusite is common in contact aureoles,
but sillimanite also occurs in many cases. Kyanite has been found occurring
together with andalusite and sillimanite in a few aureoles. Thus, we may classify
contact aureoles in terms of Al 2 SiO s minerals into three categories as follows:
(a) Aureoles containing andalusite as the only Al 2 SiO s mineral; e.g. the Oslo
area, Norway (Goldschmidt, 1911).
(b) Aureoles containing andalusite in a low-temperature part and sillimanite
in a high; e.g. the Arisu area, Japan (Seki, 1957, 1961b).
(c) Aureoles containing all three polymorphs of Al 2 SiO s ; e.g. the Kwoiek
area, British Columbia (Hollister, 1969a, b).
Polymorphs of Al 2 SiO s may form by metastable crystallization. This possi-
bility will be discussed in §§ 7B-13 and 10-3. Whether the Al 2 SiO s minerals
crystallized as stable or as metastable forms, it is probable that class (c)
represents a relatively high rock-pressure in a transitional state between low- and
medium-pressure metamorphism. Class (b) includes contact-metamorphic
aureoles with a facies series more or less similar to that of low-pressure regional
116/s0ME IMPORTANT PROBLEMS IN METAMORPHIC GEOLOGY

metamorphism. As was shown in fig. 2-1, andalusite is stable up to a higher

temperature under lower pressures. The inner aureole of some examples of
class (a) appears to have reached a very high temperature, and yet not to have
produced sillimanite, suggesting that the pressure was very low. There would be
no regional metamorphism corresponding to such aureoles. However, in cases
where the maximum temperature attained is low, andalusite will form through-
out the aureole even under a pressure comparable to that of low-pressure
regional metamorphism. Therefore, class (a) probably represents a considerable
range of pressure.
In chapter 10, well documented examples of contact aureoles will be
described on the basis of this classification.
Syn- and late-tectonic plutons associated with medium-pressure regional
metamorphism may give contact effects of the medium-pressure series. In most
of such cases, however, it is not possible to distinguish contact from regional
effects.

Metamorphic Facies
Amphibolite, Pyroxene-Hornfels, and Granulite Facies. The defmitions of indi-
vidual metamorphic facies are based on the externally controlled conditions and
paragenetic relations of minerals and n.ot on geologic relations of rocks. If the
paragenetic relations in a contact aureole are virtually the same as those in some
areas of regional metamorphism, the same name of metamorphic facies may well
be applied to both.
The amphibolite facies is probably the most prevalent in contact aureoles.
Many contact aureoles around large granitic masses in orogenic belts show
amphibolite-facies mineral assemblages in all of the recrystallized zones. The
paragenetic relations in such amphibolite facies are similar to those of the
amphibolite facies in the low-pressure regional metamorphism. Though such
aureoles are common, they have aroused little interest among petrologists in
recent years.
In some aureoles, however, the outer zone belongs to the amphibolite facies,
whereas the inner zone belongs to the pyroxene-hornfels facies. Such an aureole
has been found around some masses of gabbroic and dioritic rocks, as
exemplified by the diorite of the Comrie area in Scotland (Tilley, 1924) as well
as around alkali plutonic masses (e.g. syenites and alkali granites) of the Oslo
area (Goldschmidt, 1911).
In some andalusite-sillimanite-bearing contact aureoles, the innermost
aureole, containing orthopyroxene, would belong to the granulite facies. This is
the case in particular with contact metamorphism caused by syn- or late-tectonic
plutons in low-pressure regional metamorphic terranes. Since schists and not
hornfelses usually form in such cases, the outer limit of the contact aureoles is
 4-5 NATURE OF CONTACT METAMORPHISM /117
not clear. It appears to be common, however, that such contact metamorphism
produces a metamorphic facies series very similar to, but reaching a higher
temperature than, that of the associated regional metamorphism, as exemplified
by the Nakoso aureole in the central Abukuma Plateau, where an ortho-
pyroxene-bearing zone is formed in aureoles surrounding gabbros but not
granites (Shido, 1958).

Low-Temperature Part of Contact Aureoles. The area on the outer side of an

amphibolite facies zone is commonly too poorly recrystallized to be assigned to
any metamorphic facies. However, some aureoles around late-tectonic plutons
show the same facies as the surrounding regional metamorphism of the
low-pressure type. In such a case, a zone of the greenschist facies occurs on the
outer side of the amphibolite facies zone (e.g. Shido, 1958, for Nakoso), though
distinction of the zone from the regional metamorphic terrane is difficult.
In typical contact aureoles composed of hornfelses, however, a zone of the
greenschist facies is very rare. Instead, a zone characterized by metabasites with
the actinolite + calcic plagioclase assemblage occurs commonly just on the outer
side of an amphibolite facies zone. The actinolite in this case is colourless or very
pale green with low refractive indices. The plagioclase is labradorite or andesine.
Such an assemblage was observed in the Iritono aureole (Shido, 1958) and Arisu
area (Seki, 1961b), both in Japan, and in the Sierra Nevada (Loomis, 1966), but
has not been observed in any case of regional metamorphism.
The amphibolite is mainly composed of hornblende and a relatively calcic
plagioclase. A decrease in temperature in this case is accompanied by replace-
ment of the hornblende by actinolite, whereas the calcic plagioclase remains

°
stable.
If H2 and CO 2 are mobile, the actinolite + calcic plagioclase assemblage is
virtually isochemical with common greenschist assemblages such as chlorite +
epidote + albite + quartz and chlorite + calcite + albite + quartz, as is clear from
the equations:
9 actinolite + 114 anorthite + 64 H2 °
= 10 chlorite + 66 epidote + 77 quartz,
9 actinolite + 15 anorthite + 31 H2 ° + 33 CO 2
= 10 chlorite + 33 calcite + 77 quartz.
Here, the formula of chlorite is taken as Mg4.SAl3Si2.S010(OH)s. Therefore,
the occurrence of the actinolite + calcic plagioclase assemblage would depend
mainly on external conditions. If so, this assemblage is entitled to characterize a
metamorphic facies. Thus, the metamorphic facies characterized by the actino-
lite + calcic plagioclase assemblage was provisionally called the actinolite-calcic
plagioclase hornfels facies (Miyashiro, 1961a, p. 307). Since available data are
118/ SOME IMPORTANT PROBLEMS IN METAMORPHIC GEOLOGY

not ample, however, we will avoid the term facies and use the name actinolite-
calcic plagioclase zone in this book.
In the first of the two preceding equations, the total volume of the solid
phases is much smaller on the right-hand side. In the second, the relation is
reversed. Therefore, the actinolite + calcic plagioclase assemblage could form
under a low rock-pressure if Pco , is relatively low, but could not form under a
rela tively high Pc 0, even if the rock-pressure may be low. Since epidote usually
contains much Fe203, the oxidation condition would also have some effect.
No Al 2SiO s minerals were found in the Iritono and Sierra Nevada aureoles.
In the Arisu aureole, there is a wide medium-temperature zone containing
andalusite and a narrow highest temperature zone containing sillimanite. The
narrowness of the sillimanite zone suggests the possibility that the rock-pressure
in this aureole may have been lower than that in ordinary low-pressure regional
metamorphism. It is conceivable that all these aureoles represent rock-pressures
lower than that of low-pressure regional metamorphism in combination with a
low Pco , .
In the Iritono Aureole, the actinolite-calcic plagioclase zone represents the
lowest temperature of recrystallization, whereas in the Arisu aureole, the
actinolite-calcic plagioclase zone is followed on the outer side by a zone of the
greenschist facies. Thus, this area could represent a state intermediate between
those of the Iritono-like aureole and the low-pressure regional metamorphic area
where a zone of the greenschist facies is well developed.
As will be summarized in table 8B-l, the composition of plagioclase at the
temperature at which hornblende substitutes for actinolite in metabasites, tends
to be more calcic with decreasing rock-pressure. The generation of the actinolite
+ calcic plagioclase assemblage in some contact aureoles may be regarded as a
result of this trend.

Migration of Materials
Many contact aureoles contain a few unusual rocks that suffered intense
metasomatism. It is known, however, that the majority of the hornfelses in such
aureoles approximately preserve their pre-metamorphic compositions. This was
first shown in the second half of the nineteenth century by Th. Kjerulf in the
contact aureoles of the Oslo area (Barth, 1962, p. 262) and by H. Rosenbusch
(1877) in the aureole around the Barr-Andlau granite in Alsace. The only
marked change is dehydration and decarbonation.
However, the original pelites and psammites, for example, usually show a
considerable range of composition, and hence it is difficult to prove the strict
constancy of composition in metamorphism.
Oxygen isotope studies give a clue to this problem. Shieh and Taylor (1969)
measured the 18 0/ 16 0 ratios of metapelites in contact aureoles of the Santa
Rosa Range and other places. The ratio in metapelites is similar to that in
 4-5 NATURE OF CONTACT METAMORPHISM /119
associated unmetamorphosed pelites and is much higher than that in associated
plutons. In other words, there is virtually no change in 18 0/ 16 0 of. the
metapelites due to the influence of the granitic plutons, except for a narrow
zone, one metre thick or so, in direct contact with the plutons. This could mean
that the outward migration of H2 0 from a pluton into the aureole was
negligible. This is in marked contrast with regional metamorphism, where the
18 0/ 16 0 ratio of metamorphic rocks changes over a very wide region (§ 3-1). In
general, the introduction of materials into contact aureoles would not be
intense.
In high temperature zones, particularly in the pyroxene-hornfels facies,
partial melting of pelitic and psammitic homfelses could occur. Melts of broadly
granitic compositions would form, and the removal of such melts would produce
diverse solid residues of unusual chemical compositions. Highly desilicated
homfelses rich in corundum and/or spinel have been found in high·temperature
contact aureoles of the Comrie area (Tilley, 1924), the Cashel-Lough Wheelaun
area (Leake and Skirrow, 1960), the Belhelvie area (Stewart, 1946) and others.
Such rocks may have been derived from very unusual sediments, but may
conceivably represent solid residues (§ 104).
Processes of partial melting of metapelites and resultant formation of
desilicated rocks have been traced in some cases of pyrometamorphism (§ 10-5).
Chapter 5

Diagrammatic Representation of
Mineral Parageneses

5-1 MINERALOGICAL PHASE RULES AND COMPOSITION-PARA-

GENESIS DIAGRAMS

When a group of rocks is subjected to a defmite set of externally controlled

conditions, the rocks should show variations in mineral composition corres-
ponding to their chemical variations. In equilibrium, there should be a defmite
relationship between the chemical and the mineral compositions of rocks. In this
case, the term chemical composition refers only to components that are not
mobile. If a rock is open to H2 0, for example, its H2 0 content varies with the
external conditions.
The relationship between the chemical and the mineral compositions should
vary with external conditions. A definite relationship should hold under definite
values or small ranges of external conditions. Since such ranges of external
conditions represent a metamorphic facies, each metamorphic facies is charac-
terized by a definite relationship between the chemical and the mineral
compositions. Thus the study of such a relationship is an essential part of the
theory of metamorphic facies.
A considerable number of methods for the diagrammatic representation of
such a relationship have been proposed. Here, we will begin with a simplified
accoun t of the connection of such diagrams with the phase rule of Willard Gibbs.
According to the phase rule,
F=c+2-p~O

where F = number of degrees of freedom, c = number of independent

components necessary to defme the compositions of all phases in the system, and
p = number of phases. The number of rlegrees of freedom means the maximum
number of intensive thermodynamic variables which we may independently vary
without changing the number of phases in the system in eqUilibrium.

Oosed System
We now aSsume that a rock undergoing metamorphism is a closed system with
no exchange of substance with the environment. The temperature T and
 5-1 MINERALOGICAL PHASE RULES /121
rock-pressure Ps of the system are controlled by the external conditions. In
other words, these two intensive variables depend on the external conditions. We
assume that a mineral assemblage commonly found over a metamorphic area is
stable in certain ranges of temperature and rock-pressure. Mineral assemblages
stable only at a definite temperature and/or a definite rock-pressure would not
be widespread. For a system to be in equilibrium in certain ranges of
temperature and rock-pressure, the number of the degrees of freedom must be
two or more. Thus,

F=c + 2 -p~2,

therefore, c ~ p.

If there is no fluid phase in the system, p represents the number of minerals. If

there is a fluid phase, p - 1 is equal to the number of minerals. Thus, the
number of minerals that could coexist in stable equilibrium in a closed system
for certain ranges of temperature and rock-pressure is equal to or less than the
number of independent components. This is called Goldschmidt's mineralogical
phase rule (Goldschmidt, 1911, 1912b).
If such rocks are composed of three components, the number of minerals in
the commonly observed assemblages should be 3, 2, or 1. This means that when
the mineral compositions of such rocks are represented in a triangular diagram
with components [A], [B] and [C] at the "apexes as shown in fig. 5-1, the
paragenetic relations can be shown by dividing the diagram into a series of
smaller triangles. In this figure A, B, C, D, E, and F represent minerals with
definite compositions, and the diagram is divided into smaller triangles ADF,
DEF, DBE, ... A point within or on a side of the smaller triangles represents the
coexistence of three or two phases respectively. A corner of the smaller triangles
represents the presence of only one phase. Such a diagram is generally called a
composition-paragenesis diagram.
Most metamorphic minerals, however, are solid solutions with considerable
ranges of composition. Their compositions are represented by areas in the
triangular diagram as shown by A, B, C, D, E, and F in fig. 5-2. Within these
areas, c = 3 and p = 1, and hence F =c + 2 - p =4. Two out of the four degrees
of freedom refer to temperature and rock-pressure. The remaining two corres-
pond to the divariant variation in composition. When two phases coexist, c = 3
and p = 2, and hence F = 3. Thus the compositional variation is univariant. In
other words, the composition of each of two coexisting minerals is on a curve in
the diagram. Such a curve represents the outline of the composition field of a
mineral. In fig. 5-2, a point on the curve for a mineral is connected by a straight
line to the point for another mineral in eqUilibrium. When three phases coexist,
the composition of each of them is fixed, and the paragenesis is represented by a
defmite smaller triangle in the diagram.
122/DIAGRAMMATIC REPRESENTATION OF MINERAL PARAGENESES

Fig. 5-1 Composition-paragenesis diagram of the three-component closed

system [A]-[B]-[C] with minerals A, B, C, D, E and F of definite
compositions. The same relation holds for a system which has fixed
components [A], [B], [C] and is open for other components.

Open System
We assume that a rock undergoing metamorphism is open as regards mobile
components (e.g. H2 0 and CO2 ). These components are here assumed to have
extremely great mobility and hence are called perfectly mobile components after
Korzhinskii (1959). The components that are not mobile are called fixed
components. It is assumed for theoretical simplification that all the components
of a rock undergoing metamorphism can be assigned as either perfectly mobile
or fixed components. The temperature, the rock-pressure and chemical potentials
of perfectly mobile components of the system are controlled by external
conditions. The state of the system varies with the external conditions.
If Cm = number of the perfectly mobile components and Cx =number of the
fixed components, the number of the degrees of freedom of the system must be
equal to or greater than 2 + C m for eqUilibrium to hold over certain ranges of
externally cop,trolled variables.
Hence,
C=C m +c x

F= (c m +cJ + 2 -p~2 +cm

cx~p.
 5-1 MINERALOGICAL PHASE RULES /123

[AJ

Fig. 5-2 Composition-paragenesis diagram of the three-component closed

system [AJ-[BJ-[CJ with solid-solution minerals A, B, C, D, E and F. The
same relation holds for a system which has fixed components [AJ, [BJ,
and [CJ and is open for other components.

Thus, the number of minerals that could coexist in stable equilibrium in an open
system for certain ranges of temperature, rock-pressure and chemical potentials
of perfectly mobile components is equal to or less than the number of the fixed
components. This is called Korzhinskii's mineralogical phase rule (Korzhinskii,
1936, 1959, 1965).
The state of a system is determined by the external conditions and the
proportions of the fixed components in the system. Hence, the paragenetic
relations for an open system with three fixed components can be represented by
a triangular diagram with the fixed components at the apexes. We assume that an
open system contains three fixed components [A], [B] and [C], as shown in
fig. 5-1. According to Korzhinskii's phase rule, the number of minerals
in commonly observed mineral assemblages should be 3,2 or 1. If such minerals
have fixed compositions as shown by points A, B, C, D, E and F in the figure,
the triangular diagram is to be divided into smaller triangles to represent
paragenetic relations. If the minerals are solid solutions, relations will be as
shown in fig. 5-2. Thus, the composition-paragenesis diagram for an open system
is apparently the same as that for a closed system, except that only fixed
components are taken at the apexes. This is an important consequence of the
mineralogical phase rules.
l24/DIAGRAMMATIC REPRESENTATION OF MINERAL PARA GENESES

It is to be noted that the mineralogical phase rules of Goldschmidt and

Korzhinskii are based on the somewhat disputable assumption that a mineral
assemblage commonly found over a metamorphic area is stable over ranges of
temperature and rock-pressure if the stystem is closed, and over ranges of
temperature, rock-pressure and chemical potentials of the perfectly mobile
components if the system is open. Moreover, the classification of the com-
ponents into perfectly mobile and fixed ones is an idealization. Many com-
ponents should have intermediate mobility. In most cases, we have no sound
criteria for assigning a component to one or the other group. Accordingly, there
are many difficulties in the proper application of these principles to natural
rocks (Weill and Fyfe, 1964, 1967; Mueller, 1967).
The mineral assemblages in closed systems containing only three components
or in open systems containing only three fixed components can be represented
in triangular diagrams. Siliceous dolomitic limestones undergoing meta-
morphism, for example, are sometimes successfully dealt with as an open system
with three fixed components CaO, MgO and Si0 2 and two perfectly mobile
components CO 2 and H20. Most natural rocks, however, appear to contain
more fixed components. This imposes a great difficulty on the diagrammatic
analysis of paragenetic relations.
There are two contrasting trends in such analysis: (a) a large number of
components and a wide range of composition in natural rocks may be
represented in a diagram, though rigorous paragenetic analysis is not possible; (b)
a more rigorous paragenetic analysis may be attempted, though the range of
composition of the rocks to be treated cannot be wide. Eskola's ACF and A'KF
diagrams are examples of the former method whereas the AFM diagram
proposed by Thompson is an example of the latter.
In the case of fig. 5-2, if a rock contains three minerals, the compositions of
all the three are definite under definite external conditions, and should vary
with external conditions but not with the variable bulk chemical composition of
the rock. The effect of the variable bulk chemical composition on solid solution
minerals could be eliminated by using such a rock. For such a reason, rocks
containing a greater number of phases are generally more useful in the
investigation of the relationship between the compositions of minerals and P- T
conditions.

5-2 ESKOLA'S ACF AND A'KF DIAGRAMS

The major components of ordinary metamorphic rocks are Si0 2 , Ti0 2, Al2 3 , °
Fe203, FeO, MgO, MnO, CaO, Na20, K2 0, H20 and P20S' Some of these,
however, are not important. It is the Al 2 0 3 : CaO : (FeO + MgO) : K2
ratios that have the dominant effects on the mineral composition of metamorphic
°
rocks. The A CF and A'KF diagrams proposed by Eskola (1915) are based on this
 5-2 ESKOLA'S ACF AND A 'KF DIAGRAMS /125
empirical fact. These triangular diagrams are intended to focus our attention on
the ratios of Al2 0 3 : CaO : (PeO + MgO) and of Al 20 3 : K 20 : (PeO +
MgO) respectively. We will examine the construction of these diagrams with
reference to the mineralogical phase ru1e.

ACFDiagram
Metamorphic rocks usually contain small amounts of magnetite, hematite,
ilmenite, sphene, rutile, apatite and sulphides. The ACF and A'KF diagrams,
however, are intended for representing paragenetic relations among more
abundant minerals. Accordingly, these accessory minerals are disregarded. Ti02 ,
P20S and S, being contained in them, are also disregarded.
ACF and A'KF diagrams are used only for rocks containing quartz (or very
rarely another stable form of Si02 ). Such rocks are called rocks with excess
Si0 2 • Quartz is virtually pure Si02 . An increase or a decrease of Si02 content of
such rocks causes only an increase or a decrease in quartz without any effects on
the paragenetic relations of minerals. Even if the component Si0 2 and the phase
quartz are disregarded simu1taneously, the number of the degrees of freedom F =
c + 2 - p remains unchanged. Hence, this disregard of Si02 is justified from the
viewpoint of the phase rule.
When a group of rocks have been metamorphosed as closed systems, a
component which is always present as a particu1ar phase in the rocks under
consideration as Si02 is in the present case, is called an excess component
(Korzhinskii, 1959, p. 67). If the rocks have been metamorphosed as open
systems, an excess component means a fIXed component that is always present as
a particular phase, either on its own or in combination with any amounts of
perfectly mobile components.
Eskola disregarded the component H 20 on the basis of his statement that H 20
is always present in excess during metamorphism. Probably he meant that an
intergranular fluid existed during metamorphism. Indeed, if an intergranular
fluid phase, with virtually pure H2 0 composition, at the same pressure as the
surrounding solid phases, exists as in model (A) of § 2-3, we may disregard the
component H 20 and the fluid phase in our diagram, just as we did Si02 and
quartz. When we assume a rock undergoing metamorphism to be open for H 20,
H 20 will not be taken as a component at the apexes in the composition-para-
genesis diagram, and hence we come to a diagram which is the same in effect as
Eskola's.
The Na20 present in metamorphic rocks is usually mainly contained in the
albite molecule except in the glaucophane-schist and eclogite facies. In order to
reduce the number of components, Eskola disregarded the component albite,
though this is not justified from the phase rule.
Eskola added Fe203 to Al203 on the assumption that they play similar roles
in solid-solution minerals. He added MgO and MnO to FeO for an analogous
126/ DIAGRAMMATIC REPRESENTATION OF MINERAL PARAGENESES

reason. Nowadays we know that the replacement of Al 3 + by Fe 3+ and of Fe2+

by Mg2+ and Mn2+ does not justify the grouping of these components from the
viewpoint of the phase rule. These atoms have distinctive properties, and can
substitute for one another in different ways in minerals. They should be treated
as different components in more rigorous paragenetic analysis.
This procedure, however, is a necessary evil in order to reduce the number of
compositional variables. Thus, we have the following four components: Al 20 3 +
Fe203, CaO, FeO + MgO + MnO, and K2 0. For the purpose of making
triangular diagrams, we have to choose three components out of the four.
Here, we choose Al 2 0 3 + Fe2 0 3 , CaO and FeO + MgO + MnO. Since K2 0 is
disregarded in this case, K-feldspar cannot be represented in the diagram. As was
explained above, albite has also been disregarded. Therefore, correction of the
Al 2 0 3 content is necessary. The molecular amount of Al 20 3 in the alkali
feldspars is the same as that of (Na2 0 + K 20). Thus, we have the following three
groups of components:

A = Al 20 3 + Fe203 - (Na20 + K20)

C=CaO
F= FeO + MgO + MnO.

The amounts of these three groups of components are calculated on a molecular

basis. A triangular diagram having the three on the apexes is called an ACF
diagram, as exemplified by figs. 5-3 and 5-4.
The micas and hornblendes contain considerable amounts ofK 20 and Na20.
Strictly speaking, they cannot be represented on the ACF diagram. However,
they are customarily shown.
Calcite may be shown or disregarded in the diagram. If we regard a rock as
being open for CO 2 , the formation of calcite depends on the chemical potential
of CO 2 •
The merit of the ACF diagram is that all the common metamorphic rocks
including pelitic, psammitic, calcareous, and basic ones can be plotted to show
their mutual relations and broad paragenetic features. As is clear from the above,
however, the ACF diagram is based on a number of procedures that are not
justifiable from the phase rule. Accordingly, it is not a rigorous composition-
paragenesis diagram. The ACF diagram cannot be divided into smaller triangles
rigorously showing parageneses. In many cases, four or five minerals on the
diagram occur in the same rock. ACF diagrams in the literature, however, are
usually divided into smaller triangles indicating parageneses. This might be for
aesthetic reasons rather than for the truth. We follow this custom in this book.
Solid-solution minerals should appear properly as lines or areas in the ACF
diagram. However, as the diagram is not rigorous, very accurate representation of
 5-2 ESKOLA'S ACF AND A'KF DIAGRAMS /127
A

C_--:-------_......______
Wollastonite Diopside
- - ~. . . .;_(Bi:tite)
Anthophyllite
Cummingtonite
Fig.5-3 ACF diagram for rocks with excess Si0 2 and deficient in K20 in
the low-pressure amphibolite facies of the Orijiirvi area, Finland. Usually
K-feldspar is virtually the only K20-bearing mineral in rocks with excess
Si0 2 near the C apex. Near the C apex, therefore, only the rocks virtually
devoid of K20 fall into Eskola's category of 'rocks deficient in K2 0'.
the solid-solution range on the diagram is of little value. In this book, most
minerals are schematically shown by dots.
Moreover, the ACF diagram has little significance for the glaucophane-schist
facies where transformations in Na-bearing minerals play an important role.
The composition fields of some rock types in the ACF diagram are shown in
fig. 5-5.

Excess and Deficiency in K 2 0

In the epidote-amphibolite facies and the low-temperature part of the amphi-
bolite facies, muscovite and biotite are stable minerals. Since K-feldspar has a
higher content of K2 0 than micas, it reacts with K-free minerals on the AF side
of the ACF diagram such as andalusite, kyanite, cordierite, almandine and
anthophyllite to form micas. Accordingly, only rocks very poor in K 2 0 could
contain such K-free minerals. With increasing K2 0, K-free minerals are gradually
transformed into micas. After complete transformation, a further increase in
K2 0 causes the appearance ofK-feldspar. This relation was found by Eskola in a
paragenetic study of the amphibolite-facies rocks in the Orijiirvi area.
128/DIAGRAMMATIC REPRESENTATION OF MINERAL PARAGENESES

________________ ______ __________

Wollastonite Diopside F
Fig.5-4 ACF diagram for rocks with excess Si0 2 and K2 0 in the
low-pressure amphibolite facies of the Orijlirvi area, Finland.

C F
Fig. 5-5 Composition fields of some rock types in the ACF diagram.
 5-2 ESKOLA'S ACF AND A'KF DIAGRAMS /129
Since this variation is related to the K20 content, it cannot be represented in
the ACF diagram. Eskola attempted to show it qualitatively by using two ACF
diagrams: one for rocks deficient in K 2 0 (fig. 5-3) and the other for rocks with
excess K 2 0 (fig. 5-4) for the same metamorphic facies.
Here, 'rocks dificient in K2 0' means rocks containing a mineral or minerals
that cannot coexist with K-feldspar. 'Rocks with excess K2 0' means rocks not
containing such a mineral or minerals. This classification has significance only
for the rocks plotting on and near the AF side of the diagram. In the rocks
plotting near the C apex, practically all the K2 0 is contained in K-feldspar, so
long as Si0 2 is in excess.
At higher temperatures (Le. the highest amphibolite facies and the granulite
facies), mu~covite and then biotite become unstable, and hence K-feldspar can
coexist with minerals on the AF side.

A'KF Diagram for Metapelites

Pelitic rocks are usually low in CaO, containing Al 20 3 > Na20 + K 20 + CaO
(molecular basis). Such rocks are said to have excess A12 0 3 . Their Al 20 3
contents are higher than the amount necessary to combine with all the Na20,
K20 and CaO present to form feldspars. They are plotted within the triangle
A-anorthite-F of the ACF diagram (fig. 5-4).
In pelitic rocks of the amphibolite and higher facies, the CaO is contained
mainly in the anorthite molecule. The effects of K2 0 on the minerals on the AF
side of the ACF diagram can be clarified by examining the Al2 0 3 remaining
after the formation of Na-, K- and Ca-feldspars. (In the epidote-amphibolite and
greenschist facies, epidote would be formed instead of the anorthite molecule.
Since epidote and anorthite have somewhat different compositions, this must
have some effect to other minerals. However, it might well be neglected.)
For rocks with excess Al 2 0 3 , we can use the A'KF diagram with the
follOWing groups of components (molecular basis) on the apexes:

A' =Al 20 3 + Fe203 - (Na20 + K 20 + CaO)

K=K 20
F= FeO + MgO + MnO.

Eskola named such a diagram an AKF diagram. However, the meaning of A in

this case differs from that of A in the ACF diagram. To avoid possible confusion,
the name of the A'KF diagram is used in this book after Winkler (1967).
In the A'KF diagram, Si02 is an excess component and H2 0 may be regarded
as a perfectly mobile component (fig. 5-6). The construction of the A'KF
diagram also involves procedures unjustifiable from the viewpoint of the phase
rule. Hence, four phases on the diagram may stably coexist.
130/DIAGRAMMATIC REPRESENTATION OF MINERAL PARAGENESES

K~~ ______________________________ ~ ______ F

Microcline, Anthophyllite
Cummingtonite
Fig. 5-6 A 'KF diagram for pelitic rocks with excess silica in the low-
pressure amphibolite facies of the Orijiirvi area, Finland.

5-3 AFM DIAGRAMS FOR METAPELITES

In the ACF and A'KF diagrams, FeO and MgO are jointly .treated as F. Among
the minerals of metapelites, however, garnet, chloritoid, and staurolite are high
in FeO/MgO, whereas cordierite is low. The FeO/MgO ratio of rocks has a great
influence on the mineral assemblage. Hence, a triangular diagram with FeO and
MgO at different apexes is helpful in paragenetic analysis.
Most minerals in metapelites fall on and near the A'F side of the A'KF
diagram. Accordingly, paragenetic relations could be represented to a con-
siderable extent by a triangular diagram with A': FeO and MgO at the apexes, or
the like, as was used, for example, by Korzhinskii (1959, fig. 49) and Chinner
(1962,1967).

Thompson AFM Diagram.

Thompson (1957) has proposed a more rigorous method of representation.

°
Metapelites are regarded as being composed of six components Si02, AI20 3 ,
MgO, FeO, K20 and H2 by neglecting minor ones such as Ti02 , Fe2 0 3 , MnO,
CaO, and Na20. He regards the rocks as being open for H20, and hence H20
does not appear in the composition-paragenesis diagram. Most metapelites
 5-3 AFM DIAGRAMS FOR METAPELITES/131

contain quartz, and his discussion is confined to quartz-bearing rocks. Accord-

ingly, Si02 is an excess component as dermed in the preceding section, and may
be neglected in the diagram.
Thus, only four main components A1 2 0 3 , MgO, FeO and K2 0 remain in our
graphical representation, and the composition of a metapelites is approximately
represented by a point in a tetrahedron as shown in fig. 5-7.
Muscovite is very widespread in metapelites. Thompson confined his dis-
cussion to muscovite-bearing rocks, and hence this mineral need not be
expressed in the composition-paragenesis diagram. The muscovite composition
has been assumed to be KAl 3 Si 3 0 10 (OHh = KAl 3 0 S . 3Si02 • H2 0, and all the
other minerals within the tetrahedron are projected from the muscovite point
KAl 3 0 s onto the A1 2 0 3 -Feo-MgO plane, as shown in fig. 5-7. The maximum

,
,,
""
,"
, ,, "
,,
a-a' = Biotite (approx.) ,, ""
bob' =Projection of a-a'
(mole percent) b '"

Fig.5-7 Thompson's projection through the idealized muscovite compo-

sition onto the plane determined by Al2 0 3 , FeO and MgO.
132/DIAGRAMMATIC REPRESENTATION OF MINERAL PARAGENESES

number of phases is four according to the mineralogical phase rule. Excepting

muscovite, the remaining three phases can be strictly represented on a triangular
diagram on the Al20 3-FeQ-MgO plane.
Compositions in the sub-tetrahedron Al 20 3-FeQ-MgQ-KA130 s will project
onto the triangle Al2 0 3-FeQ-MgO, whereas those in the volume KAL30 s -
FeQ-MgQ-K2Mg304-K2Fe304 will project onto the extension of the plane
Al 20 3-FeO-MgO beyond the line FeO-MgO. Compositions in the sub-tetra-
hedron K2Q-K2Fe304-K2Mg304-KA130s will project onto the extension of
the plane Al20 3-FeQ-MgO beyond Al 20 3 . However, the last-mentioned pro-
jection beyond Al20 3 is usually unimportant, since each possible assemblage is
usually represented in the other parts of projection. Hence, this part of
projection may be neglected in ordinary cases. The tie lines from almandine,
cordierite or biotite toward K-feldspar extend downwards away from the Al 20 3
apex. In this sense, K-feldspar may be regarded as being situated downwards.
The actual plotting of mineral compositions on the diagram can be made by
calculating the following two parameters in molecular proportions:

(Al 20 3 - 3 K20)/(Al20 3 - 3K2 0 + FeO + MgO)

MgO/(FeO + MgO).

These parameters determine the height and horizontal position of the projected
point on the diagram, as shown in fig. 5-8.
In the coexisting minerals, the Fe2 + /Mg ratio increases in the following order:
cordierite -+ chlorite -+ biotite -+ staurolite (and chloritoid) -+ almandine
(Thompson, 1957; Albee, 1968). This relation is of use in the construction of
the diagram.
In this diagram, the ideal muscovite composition is used as the projection
point. Muscovites in high-temperature metamorphic rocks have such a composi-
tion, but those in low·temperature metamorphic rocks are now known to be
considerably deviated from it.
Among the neglected minor components, Na20 could enter muscovite and
K-feldspar, and CaO and MnO could enter garnet. This should cause some
modification of paragenetic relations. Generation of gamet, for example, would
be promoted by a higher content of CaO and MnO.
Muscovite in metapelites usually breaks down in the high-temperature part of
the amphibolite facies, and instead K-feldspar becomes nearly ubiquitous.
Another type of AFM diagrams to be used for metapelites with quartz and
K-feldspar was constructed by Barker (1961). In this case, points representing
minerals and rocks situated within the Al 20 3-FeQ-MgO-K20 tetrahedron are
projected onto the Al2 0 3-FeQ-MgO plane from the projection point of the
K-feldspar composition.
 5-4 SCHREINEMAKERS BUNDLE /133

'·0 oOl
:2
~u..
o'"' 0·6
~ Cordierite 0.5
0·5 M
I
o 0·4
'"' 'A=::=---;:;;:;::;-:::-"""I 0·3

o--
~ Chlorite

0·0

Biotite

-0·5

t
K·Feldspar

0·0 0·' 0·2 0·3 0-4 0·5 0·6 0·7 0·8 0·9 '·0
MgO
(Mole)
FeO+MgO
Fig. 5-8 Thompson AFM diagram showing possible phases in equilibrium
with quartz and muscovite.

5-4 SCHREINEMAKERS BUNDLE

SO far in this chapter, we have discussed diagrammatic representation of

variations of mineral parageneses with chemical composition. Now we intend to
give a brief general discussion on the variation of mineral parageneses in relation
to temperature and pressure.
If a system composed of c components contains c + 2 phases of fixed
composition, the phase rule tells us that the number of the degrees of freedom
(F) is zero. In a rectangular diagram with temperature, and pressure as the
coordinate axes such as fig. 5-9, the assemblage with c + 2 phases is represented
by a definite point (invariant point). If anyone of the phases is removed, the
system comes to have c + 1 phases and hence F = 1, and should be stable on a
curve (univariant curve) in the diagram. There are c + 2 diffe~ent ways in the
removal of a phase. Thus, in general, c + 2 curves exist, which should radiate
from the invariant point representing the coexistence of c + 2 phases. Such a
134/DIAGRAMMATIC REPRESENTATION OF MINERAL PARAGENESES

[AJ

(2)

(3)

(4)

Temperature

Fig. 5-9 Schreinemakers bundle for a three-component system [AI -[B l-

[el with five possible phases 1,2,3,4 and 5 in the compositional relation
illustrated in the large triangular diagram at the top. Five univariant curves
radiate from the invariant point representing the coexistence of all five
phases. Full lines represent stable equilibria, whereas broken lines repre-
sent metastable ones. Univariant curve (1), for example, represents the
four-phase assemblage that is obtained by removing phase I from the five
phases. The paragenetic relations for each divariant field are shown by a
small triangular diagram in which a shaded area gives the phase assemblage
characteristic of the field.
 5-4 SCHREINEMAKERS BUNDLE /135
group of curves around an invariant point has been called by the name
'Schreinemakers bundle'. Each of the univariant curves is divided into two parts
by the invariant point. One part represents stable and the other metastable
equilibrium. Each curve may be regarded as corresponding to a chemical
equation among the phases coexisting on it.
These curves divide the temperature-pressure diagram into c + 2 fields
(divariant fields), in each of which c or less phases can coexist in equilibrium.
Within each field, mineral assemblages change with their chemical composition
according to the rule characteristic to the field. Such a rule can be represented
by a diagram as illustrated in fig. 5-9.
As a simple example, we take the one-component system Al 2 SiO s , in which a
three-phase point (triple point) is fixed in a temperature-pressure diagram as was
shown in fig. 2-1. Three curves representing two-phase equilibria radiate from
the point. Only one phase is stable in each of the fields bounded by these curves.
The shapes and positions of individual univariant curves may be known by
synthetic experiments, or can be calculated if enough thermodynamic data are
available. The slope of the curves can be calculated by the Clausius-Clapeyron
equation, if the entropy and volume changes are known. Even when the shapes,
positions and slopes of individual curves are not known, there is a general rule
governing the sequence of univariant curves around the invariant point. It was
found by F. A. H. Schreinemakers in the 1910s. The sequence depends only on
the chemical formulae of the relevant phases. Fig. 5-9 was drawn in accordance
with it. For a detailed account, refer to Zen (1966).
An example of the use of the Schreinemakers rule was shown in § 3-3.
II Progressive Metamorphism

Outline
The recent advance in petrographic study throughout the world has clarified the
great diversity of metamorphic terranes. A cursory survey of the diversity was
given in § § 3-3 and 3-4. Part II treats detailed classifications and descriptions of
well-documented examples of progressive metamorphic terranes.
The nature of the diversity is revealed by examination of progressive
metamorphic reactions, which vary not only with T, Ps , PH. 0 and Peo • but
also with chemical composition of the dominant metamorphic rocks.
Thus, the backbone of Part II is composed of chapters 7, 8, and 9, which are
allotted respectively to the progressive regional metamorphism of pelitic, basic
and calcareous rocks. Chapters 7 and 8 are divided into two sub-chapters: A and
B. In each case, sub-chapter A treats a classification and description of pertinent
metamorphic terranes in order to show the extent of their diversity, whereas
sub-chapter B treats the progressive metamorphic reactions, except for those of
the zeolite and prehnite-pumpellyite facies, in order to clarify the nature and
cause of the diversity.
The diversity is greatly enhanced by the presence or absence of zeolite and
prehnite-pumpellyite facies zones in the low-temperature part. Moreover, zeolite
facies zones themselves show diversity. These problems are treated in chapters
6A and 6B prior to the above-mentioned backbone of Part II.
Chapter 10 treats the diversity of progressive contact metamorphism.
As a summary of all these descriptions, three generalized series of meta-
morphic facies will be formulated and described in chapter 11. A rigorous
definition of metamorphic facies will also be discussed.
Eclogite does not occur as a usual member of progressive metamorphic zones.
Therefore, the problem of eclogites and the eclogite facies will be discussed in
chapter 12 at the end of Part II.
Chapter 6A

Zeolite and Prehnite - Pumpellyite

Facies Metamorphism: Its Diversity

6A-l CLASSIFICATION OF BURIAL METAMORPHIC TERRANES IN

TERMS OF GEOTHERMAL GRADIENTS

Burial metamorphism, usually in the zeolite and prehnite-pumpellyite facies,

takes place in the lowest temperature part of regional and ocean-floor meta-
morphic terranes, though occasionally it results in apparently independent small
metamorphic areas. Rocks of the zeolite and prehnite-pumpellyite facies in all
geologie settings will be treated together in chapters 6A and 6B for the following
three reasons: (a) Comparison of burial metamorphism in different geologic
settings is instructive; (b) The limited variation in the chemical composition of
rocks which are extensively recrystallized under burial metamorphism means
that they can hardly be treated in terms of the compOSitional classification
which will be used in our treatment of regional metamorphic rocks in chapters 7,
8 and 9; (c) Many regional metamorphic terranes have no burial metamorphic
part, and hence burial metamorphism may well be treated separately for
convenience.
Recrystallization is usually incomplete in the Zt:olite and prehnite-
pumpellyite facies. Pyroclastic rocks and lavas as well as graywackes containing
abundant pyroclastic materials are susceptible to recrystallization at low tem-
peratures, whereas other rocks associated with them may show little recrystalliz-
ation. The temperature of recrystallization is too low to produce anhydrous
A1 2 SiO s minerals. However, the geothermal gradients and baric types may be
inferred from the estimated depth and temperature of recrystallization.
Recent petrographic studies, particularly in Japan, have revealed that the
paragenetic relations in the zeolite facies vary considerably from area to area
(§ 4-4). It appears that the paragenetic relations are partly related to the
geothermal gradient. The terranes with higher geothermal gradients tend to
produce a greater number of zeolites including mordenite, stilbite, yugawaralite,
wairakite, heulandite, laumontite and analcime. On the other hand, the terranes
with lower geothermal gradients produce only the last three of the above-
mentioned zeolites, or even only laumontite.
l40/DIVERSITY IN ZEOLITE FACIES METAMORPHISM

Thus, metamorphic terranes with zeolite and prehnite-pumpellyite facies

zones will be reviewed below on the basis of a classification in terms of
geothermal gradients, or baric types.

6A-2 ZEOLITE AND PREHNITE-PUMPELLYITE FACIES TERRANES

OF THE LOW-PRESSURE TYPE

Many of the low-pressure regional metamorphic terranes, such as the Ryoke belt
of Japan, have no rocks belonging to the zeolite and the prehnite-pumpellyite
facies, so far as we are aware. Their lowest temperature zone is in the greenschist
facies. On the other hand, Coombs, Horodyski and Naylor (1970) have found a
wide zone belonging to the prehnite-pumpellyite facies in northeastern Maine.
This zone probably represents· the lowest temperature part of the low-pressure
facies series in the northern Appalachian Mountains.
Three well-documented areas of zeolite facies rocks in Japan will be reviewed
below. The geothermal gradient appears to have been very great in all of them.
Hence, probably they belong to the low-pressure type, though Al 2 SiOs minerals
do not occur.

Akaishi Mountains, Japan

Matsuda and Kuriyagawa (1965) have described a metamorphosed Mesozoic and
Tertiary terrane, about 40 km wide, with abundant volcanic and pyroclastic
rocks ranging from the zeolite through the prehnite-pumpellyite to the green-
schist facies in the eastern Akaishi Mountains to the west of Mt Fuji in central
Japan (figs. 15-8 and 15-9). The metamorphism is of the burial type on a
regional scale showing no relationship to plutonic masses. The temperature of
metamorphism generally increases westward (fig. 6A-l).
The zeolites identified in metamorphosed basalts, dolerites and andesites were
mordenite, heulandite (c1inoptilolite), laumontite, analcime and thomsonite.
Iron saponite and celadonite occur in the zeolite facies zone, whereas prehnite
begins to occur in a higher temperature part of the same zone and persists to the
prehnite-pumpellyite facies zone. Relict minerals are abundant not only in the
zeolite but also in the prehnite-pumpellyite facies zone.
The isograds cross formational boundaries obliquely. The boundary between
the zeolite and the prehnite-pumpellyite facies zone was probably at a depth of
2-4 km during the metamorphism. The temperature that prevailed on the
boundary would be around 300-350°C. Accordingly, the average geothermal
gradient should have been 75-175 ° C km -1 . Even if we make a liberal allowance
for error in our estimation, the metamorphism is probably of the low-pressure
type.
The rocks of the zeolite facies have no noticeably preferred orientation,
whereas those in the prehnite-pumpellyite facies occasionally show preferred
 6A-2 LOW-PRESSURE TYPE/141

ML
, KS

I
I
P I II
I
I
I
I
/
I
I
I
I
I III

I
I

,,
I
I
I

,,
I \
I \

,
I
I
I
I II
/
I (w
/ 1\
I \
I \
I
I
I
I
I
I
I
/
/
I
I
Suruga-wan
(Bay of Suruga)

o, ,
10 20km
,

Fig. 6A-1 Zones of burial metamorphism in the eastern Akaishi

Mountains. (Matsuda and Kuriyagawa, 1965.)
I: Zeolite-facies zone.
II: Prehnite-pumpellyite facies zone.
II': A zone similar to II but more advanced in recrystallization.
III: Greenschist facies zone.
P: Metamorphosed Palaeozoic formations; ML: Median Tectonic Line; and
KS: Kobuchizawa-Shizuoka thrust.
142/DIVERSITY IN ZEOLITE FACIES METAMORPHISM

orientation. Many rocks in the greenschist facies show weak schistosity.

The metamorphic facies of the different stratigraphic formations may be
summarized as follows.

Formation (age) Metamorphic facies

Fujikawa Series (middle and zeolite facies
late Miocene)
zeolite facies
Misaka Series (early Miocene) {
prehnite-pumpellyite facies
zeolite facies
Shimanto Group (early Tertiary (
prehnite-pumpellyite facies
to MesOZOic)
greenschist facies

Thus, the younger formations were generally metamorphosed at lower tempera-

tures than the older.
Though the three mineral zones are arranged regularly, they are not in
progressive relations. Rock fragments of the greenschist and prehnite-
pumpellyite facies are included in rocks of the Fujikawa Series in the zeolite
facies zone. Therefore, a part of the Shimanto Group had been metamorphosed
and exposed on the surface before the deposition of the Fujikawa Series, that is,
before the middle Miocene. The site of the depositional basin and the burial
metamorphism made a gradual movement eastward from the area of the
Shimanto Group to that of the Fujikawa Series during Tertiary time. Meta-
morphism in the Fujikawa Series should be middle Miocene or later.

Mogami~rea,Japan
The metamorphism of a small Tertiary terrane of the Mogami area, in northeast
Honshu has been described by Utada (1965). The area is about 6 km across,
containing abundant lavas and pyroclastics. A zeolite facies area grades directly
into a greenschist facies area; there is no zone of the prehnite-pumpellyite
facies. The sediments were deposited in Miocene and Pliocene times. The
metamorphism took place presumably in the Pliocene. The metamorphism as a
whole appears to be of the burial type, though the highest temperature zone has
small intrusions of granite, rhyolite, diorite and dolerite.
The terrane was divided into the following five zones in order of increasing
depth (fig. 6A-2).

Zone I (unrecrystallized zone). Recrystallization is negligible. Even volcanic

glass remains fresh. Interstitial spaces remain mostly open, though sometimes
filled with opal.
 6A-2 LOW-PRESSURE TYPE/143
Mogami

Metamorphic facies Unrecrys· Zeolite facies Greenschist

tallized facies

Mineral zoning I II III IV V

Mordenite
Cli nopti 101 ite
Analcime
Heulandite
-
;---
Laumontite I--
Smectite
Mixed layer mineral
Chlorite
Celadonite f-
Sericite
I Epidote
I Albite
Adularia
Opal
Chalcedony
Quartz
Calcite

Fig. 6A-2 Progressive mineral changes in pyroclastic rocks in the Mogami

area, Japan. (Utada, 1965.)

Zone II (mordenite-clinoptilolite zone). Mordenite and clinoptilolite together

with smectite replace glass. Interstitial spaces of pumice and vitric tuffs and of
clastic rocks are commonly filled with opal and zeolites. Augite and hornblende
are frequently replaced by smectite (saponite), but most plagioclase remains
unchanged.

Zone III (analcime-heulandite zone). The opal, mordenite and clinoptilolite

disappear, whereas analcime, heulandite, albite, adularia and chlorite begin to
appear. Volcanic glass is mostly replaced by analcime and/or heulandite
associated with clay minerals and sometimes by adularia. Vesicles in pumice are
filled with analcime or heulandite associated with quartz. Augite and hornblende
are replaced by chlorite, while plagioclase by albite and heulandite. Epidote
begins to occur.

Zone IV (laumontite layer). This is actually a thin layer, about 2 m thick. It

contains laumontite together with albite, adularia and quartz.

Zone V (chlorite-albite-epidote zone). Greenschist-facies mineral assemblages

with chlorite, albite, epidote and calcite occur.
144/DIVERSITY IN ZEOLITE FACIES METAMORPHISM

The mineral changes are synoptically shown in fig. 6A-2. It is said that quartz
or opal is present in most assemblages, but no attention was paid to the possible
effect of the presence of these minerals. The figure indicates that analcime and
albite apparently begin to occur at the same zone. Since analcime + quartz =
albite + H2 0, the occurrence of analcime and albite in the presence of quartz
cannot be understood. It is not clear where there is always associated quartz.
The upper boundary of the greenschist facies zone is estimated to lie at about
3 km from the top of the Tertiary. If the temperature at the boundary may be
assumed to be about 350-400 °c, the geothermal gradient must have been over
100°Ckm-1 •
Prehnite, pumpellyite and actinolite are absent in this area, conceivably owing
to a relatively highPcoz . Calcite is present in zones III, IY and Y.

Tanzawa Mountains, Japan

A Tertiary metamorphic complex, about 10 km in width, is exposed on the
southern slope of the Tanzawa Mountains, about 70 Ian southwest of Tokyo and

... ...... ... ...

... ... ... ...
\ \
\ \ I
III I I
I \
I \
I \
/ 11\
----- --!~--- //// /
--- III _.... .../1
-K~---- II
----1-------.";.,,,.
... ........

o ,
'5 I
10 km
,

Fig.6A-3 Zones of metamorphism in the Tanzawa Mountains, Japan.

I-Y represent Zones I-Y respectively in fig. 6A-4. Line K is the boundary
between the Tanzawa Group on the north and the Ashigara Group on the
south. (Seki et al. 1969a.)

halfway between Tokyo and the Akaishi Mountains (fig 6A-3). The meta-
morphism ranges from the zeolite to the amphibolite facies. The high-tempera-
ture parts of the complex are strongly deformed and schistose, and even partly
overturned, and have also been influenced by thermal effects of the associated
mass of quartz diorite. On the other hand, the low-temperature rocks are
non-schistose and are a product of burial metamorphism. The original rocks were
mostly submarine pyroclastics and lavas of basaltic and andesitic compositions
that formed in early Miocene to early Pliocene time. Seki et al. (1969a) have
 6A-2 LOW-PRESSURE TYPE/145

divided the terrane into the following five zones in order of increasing
temperature of metamorphism (fig. 6A-4).

Zone I (low zeolite facies). Both sedimentary and mafic igneous rocks preserve
their original texture and show no schistosity. Only the fine-grained matrix and
groundmass are recrystallized. The recrystallized part is characterized by clino-
ptilolite, heulandite, stilbite and mordenite as well as by mixed-layer smectite-
vermiculite and vermiculite-chlorite. Chlorite proper does not occur.
Tanzawa

Preh.·pump. G reensch ist Amph.

Metamorphic facies Zeolite facies
facies facies facies
Mineral zoning I II III IV V

Mordenite
Heulandite
Stilbite
Clinoptilolite
Laumontite r--?
Thomsonite
Analcime
Yugawaralite -
Wairakite -
Opal r-
Mixed layer mineral
Chlorite
Celadonite
Muscovite
Biotite
Prehnite
Pumpellyite
Epidote
Actinolite
Hornblende
Albite %An 10 20 30
Plagioclase ?

Fig. 6A-4 Mineral changes in basic and intermediate metamorphic rocks

in the Tanzawa Mountains, Japan. (Seki et at. 1969a.) Probably this
diagram is based on both quartz-bearing and quartz-free rocks, though it is
not explicitly mentioned.

Zone II (high zeolite facies). The rocks are still usually non-schistose and
incompletely recrystallized. This zone is characterized by the occurrence of the
laumontite-quartz assemblage. A mixed-layer clay mineral is common in the
low-temperature half of this zone, whereas chlorite is common instead in the
146/DIVERSITY IN ZEOLITE FACIES METAMORPHISM

high-temperature half. Wairakite and yugawaralite were rarely found to occur in

association with quartz in the higher half.

Zone III (prehnite-pumpellyite facies). The original textures are still preserved,
though some rocks show weak schistosity. This zone is characterized by the
occurrence of prehnite and pumpellyite. Chlorite, epidote, and albite also
occur.

Zone IV (greenschist facies). The entrance to this zone is marked by the rlIst
appearance of actinolite in metabasites, whereas prehnite and pumpellyite
disappear. The metabasites are greenschists or greenstones mainly composed of
actinolite, chlorite, epidote, and albite with minor amounts of quartz and
sphene. Calcite, white mica or biotite may occur. In the higher temperature part
of this zone, however, the stable calciferous amphibole appears to be horn-
blende, and the plagioclase is sodic oligoclase.

Zone V (amphibolite facies). The flIst appearance of plagioclase with 20 per

cent An has been regarded as the lower boundary of this zone. The maximum
An content observed is 73 per cent. Common metabasites are schistose or
non-schistose amphibolites. Recrystallization is complete. The hornblende is
blue-green in the lowest temperature part, and becomes green or brownish green
with increasing temperature of metamorphism. The colour generally deepens
with advancing metamorphism. The amphibolites are associated with small
amounts of metabasites having different compositions: a cummingtonite-
plagioclase rock, an anthophyllite-plagioclase rock, and a grandite-salite-plagio-
clase rock.
The distance on the surface from the line where albite is stabilized to the
onset of the amphibolite facies appears to be roughly 5-8 km. The corres-
ponding average geothermal gradient is of the order of 40-60 DC km -1.
Stratigraphically, the metamorphic terrane is divided into two units: Tanzawa
and Ashigara Groups. Quartz diorite is exposed in the northernmost part of the
area. The Tanzawa Group (early and middle Miocene) is exposed on the south
side of it, and successively forms metamorphic zones V, IV and III and a part of
zone II (fig. 6A-3). The southern limit of this Group is at the Kannawa thrust, to
the south of which is exposed the Ashigara Group (late Miocene and Pliocene)
that forms the rest of zone II and zone I. Autocannibalism is clear. Metamorphic
rock fragments derived from the Tanzawa Group occur as pebbles in conglom-
erates of metamorphosed parts of the Ashigara Group. In the upper horizons of
the Ashigara, quartz diorite also occurs as pebbles (Seki etal. 1969a, pp. 6-11).
It follows that the submarine volcanic deposition of the Tanzawa Group in
early and middle Miocene time was probably simultaneous with an early major
phase of metamorphism, which formed a progressive series from a low-tempera-
 6A-3 ACTIVE GEOTHERMAL FIELDS/147

ture part Up to zone V. Then, the quartz diorite mass was intruded, which
resulted in some modifications in the highest temperature part. Later uplift and
erosion had exposed the metamorphic and plutonic complex by late Miocene
time. A new depositional basin was formed to the south of the uplifted
mountains, and many fragments of the exposed rocks were transported southward
and deposited, leading to the formation of the Ashigara Group. Metamorphism
continued and was spatially extended into the newly formed sediments of the
Ashigara. Thus, a part of zone II and zone I formed there in late Miocene and
Pliocene time.

6A-3 ZEOLITE DISTRIBUTION IN ACTIVE GEOTHERMAL FIELDS

In many areas that have hot springs and fumaroles, wells have been drilled to
provide steam and hot water, or rarely carbon dioxide. Mineralogical investi-
gations of drill cores and cuttings gave unusual opportunities for correlating
mineral assemblages with temperatures actually measured and other conditions.
Comparison of such mineral assemblages with those in burial or regional
metamorphism helps us understand the nature of metamorphic assemblages. The
geothermal gradients in such areas are still higher than those in the low-pressure
metamorphic areas described in the preceding section.
In active geothermal fields, zeolites occur in the follOWing generalized order
with increasing depth (in the presence of excess silica): mordenite (and
heulandite) -+ laumontite (and yugawaralite) -+ wairakite. This is very similar to
the zeolite sequence in low-pressure metamorphiC terranes, though a zone
characterized by prehnite and pumpellyite has not been noticed.

Wairakei, New Zealand

A well-documented example of active geothermal fields is the Wairakei area in
North Island of New Zealand (Steiner, 1953, 1955; Coombs et al. 1959). The
following zones were observed in hydrothermally altered rhyolite tuffs and
breccias, from the surface downwards:
(a) Acid leached zone. The characteristic minerals are kaolinite, alunite and
opal.
(b) Montmorillonite zone. Rhyolite glass is altered to montmorillonitic
clays.
(c) Upper zeolite zone (roughly ISO-200°C). Mordenite (ptilolite) is
characteristic, filling vesicles in pumice. It is sometimes accompanied by
minor heulandite.
(d) Lower zeolite zone (roughly 200-230 0C). Wairakite replaces andesine,
and also fills veins and caVities by direct crystallization from solution.
 148/DIVERSITY IN ZEOLITE FACIES METAMORPHISM

Laumontite was found roughly in a zone transitional between the upper

and lower zeolite zones (roughly 195-220°C).
(e) Feldspathization zone (roughly 230-250°C). An upper part is charac-
terized by albitization and a lower one by adularia replacing plagioclase.
Rarely prehnite occurs, and calcite is locally precipitated.
if) Hydromica zone. Primary plagioclase is replaced by a hydromica as a
result of leaching of alkalis by rising carbon dioxide solutions.
The hydrothermal solutions are rich in silica which precipitates at all levels.
The temperature and depth of each zone differ in different wells. The depth of
the upper and lower zeolite zones ranges roughly from 90 to 400 m. The
geothermal gradient is roughly 200-1000 °c km -1.
The hot solutions in the deepest zone tapped have a high concentration of K,
which is progressively lowered by the formation of adularia. In contrast, the
concentrations of Na and Ca are increased until some of these ions are
precipitated in the zone of zeolite formation. In the deeper zones, the water
contains sufficient CO 2 to make it slightly acid with respect to pure water at the
same temperature. Towards the surface, the CO 2 is lost from the water, which
hence becomes slightly alkaline.
In Wairakei, the pressure of the aqueous fluid measured in fissures is often
close to that of the water column, that is, much lower than the rock-pressure. It
corresponds to model (c) discussed in § 2-3.

Onikobe, Japan
Seki (1966, 1969) and Seki, et al. (1969b) studied drill cores from the Onikobe
active geothermal area in northeast Japan. The following zones were noticed in
descending order.
(a) Native sulphur zone.
(b) Smectite zone.
(c) Mordenite zone. Analcime and albite also occur.
(d) Laumontite zone. Yugawaralite, analcime, and albite occur in some
rocks.
(e) Wairakite zone. Albite also occurs.
if) Albite zone.
Quartz is present in all rocks. The depth and temperature of each zone differ in
different drill holes. Laumontite occurs in the range of 80-240 m and
80-170 °c. The geothermal gr<)dient is roughly 300-1000 °c km -1.

Geothermal Fields Unfavourable to Zeolite Formation

The composition of the hot water ascending from depths differs in different
fields, and should have important effects on the formation of zeolites and other
 6A-4 MEDIUM-PRESSURE TYPE/149

minerals. In some geothermal fields, there is scarcely any formation of zeolites,

probably owing to a higher concentration of CO 2 and other chemical charac-
teristics of the water (§4-4).
For example in the North Island of New Zealand, there is another well-
documented geothermal field in the Waiotapu area. Though the area is near
Wairakei, the observed mineral sequence differs greatly. In terms of the
alteration of plagioclase in acidic volcanics, three zones were noticed in
descending order as follows: (a) zone of sulphuric acid alteration, characterized
by alunite, (b) zone of potassium silicate alteration characterized by the
formation of hydromica and K-feldspar, and (c) zone of albitization associated
with formation of calcite. Wairakite occurs occasionally in the zone of albitiz-
ation. No other zeolites occur (Steiner, 1963).
No zeolites were found in the geothermal field of the Salton Sea area,
California, which is characterized by solutions containing abundant CO 2 (§ 1-5).

6A-4 ZEOLITE AND PREHNITE-PUMPELLYITE FACIES TERRANES OF

THE MEDIUM-PRESSURE TYPE

Coomb's zeolite facies terrane of the Taringatura area, New Zealand, had a much
lower geothermal gradient than the areas discussed above. The difference in the
varieties of zeolites observed is partly related to this difference in geothermal
gradient. Mordenite, stilbite, yugawaralite and wairakite were not found,
whereas heulandite, laumontite and analcime do occur. This terrane is a part of
the New Zealand geosyncline, other parts of which contain lawsonite and
crossite. Hence, the zeolite facies metamorphism of this area is probably of the
medium-pressure type, close to the high-pressure type.
The Sanbagawa belt of Japan is largely of the high-pressure type. However, no
glaucophane occurs in the central Kii Peninsula, where instead zeolite facies
rocks occur. Therefore, the central Kii terrane also is probably of the medium-
pressure type, relatively close to the high-pressure type. Laumontite was the
only zeolite found there.
Judging from the incompatibility of glaucophane and zeolites in these cases,
perhaps a zeolite facies zone hardly form in high-pressure metamorphic areas,
and at least in typical high-pressure ones. (Relevant data concerning zeolites in
the Franciscan Formation and a prehnite-pumpellyite facies zone in high-
pressure terranes will be reviewed in § 6A-5.)

Taringatura Area, New Zealand

This area is situated near the southern end of South Island (Coombs, 1954;
Coombs et al. 1959). The rocks form a part of the sediment pile of the New
Zealand geosyncline, being Triassic in age and about 10 km thick. They are
1S0/DIVERSITY IN ZEOLITE FACIES METAMORPHISM

mainly composed of tuffs and volcanic graywackes of the andesite-dacite-

rhyolite group.
In the upper part of the stratigraphic section, recrystallization is poor.
However, glass in tuff beds has been completely replaced by heulandite or less
commonly by analcime (fig. 6A-S). Both zeolites can coexist with quartz and
fine-grained phyllosilicates. Igneous plagioclase remains unchanged. In succes-
sively lower horizons, however, plagioclase is increasingly replaced by pseudo-
morphs of dusty albite with sericite inclusions. Simultaneously with the

Taringatura

Zeolite facies
Prehnite-
Metamorphic facies Heulandite- Laumontite pumpellyite
analcime stage facies
stage

Heulandite
Analcime
Laumontite
Celadonite
Smectite
Chlorite
Prehnite
Pumpellyite
Albite
Quartz

Fig. 6A-5 Progressive mineral changes of volcanic graywackes and tuffs

in the Taringatura area, New Zealand. (Coombs, 1954; Coombs et al.
1959.)

replacement, analcime and heulandite disappear, and laumontite associated with

quartz begins to occur. Adularia also occurs. Toward the base of the section,
pumpellyite and prehnite appear.
Thus, with regard to recrystallized mineral assemblages, the following three
stages in order of increasing temperature were noticed.
(a) Zeolite facies; heulandite-analcime stage (upper 5 km):
Quartz + heulandite + smectite
Quartz + heulandite + celadonite
Quartz + analcime + celadonite
(b) Zeolite facies; laumontite stage:
Quartz + laumontite + albite + celadonite + adularia
Quartz + albite + smectite + adularia
 6A-4 MEDIUM-PRESSURE TYPE/lSI

(c) Prehnite-pumpellyite facies:

Quartz + albite + pumpellyite + adularia
Albite + chlorite + pumpellyite
Quartz + albite + chlorite + prehnite
If the post-Triassic cover that has been removed may be assumed to have been
a few km thick, the reaction analcime + quartz = albite + H2 0 at the boundary
between the heulandite-analcime and the laumonite stage appears to have taken
place at a depth of about 8 km. If we may regard this reaction as taking place at
about 180°C, the goethermal gradient is estimated at about 22°C km -1.

Canterbury Region, New Zealand

Another wide terrane of graywackes which occurs in Canterbury, South Island,
has been recrystallized at a grade to produce prehnite and pumpellyite (Coombs
et al. 1959). This is probably in the prehnite-pumpellyite facies, and is a part of
the accumulations in the New Zealand geosyncline. Toward the west, it appears
to grade into a greenschist facies zone, which grades into an epidote-amphibolite
facies zone, which in turn grades into an amphibolite facies zone. This group of
schists is called the Alpine schists.
In the transitional part between the prehnite-pumpellyite facies and the
greenschist facies, prehnite disappears first, and then pumpellyite does. The first
appearance of actinolite takes place roughly at the same temperature as the
disappearance of prehnite.
No intrusive igneous rocks are present. Penetrative deformation was intense in
the zone of the greenschist and higher facies, but was weak in the prehnite-
pumpellyite facies zone.

Central Kii Peninsula, Japan

Though the Sanbagawa belt is generally of the high-pressure type, an area in the
central part of the Kii Peninsula (fig. 15-8) within the belt has no glaucophane,
nor lawsonite, nor jadeite. There has developed, instead, the follOWing series of
metamorphic facies: zeolite (laumontite) facies -+ prehnite-pumpellyite facies -+
greenschist facies (Seki et al. 1971).
This area is made up of late Palaeozoic and Jurassic sediments with
interbedded mafic volcanics. The degree and temperature of recrystallization
increase northward over a distance of about 40 km. There are no plutonic rocks
influencing the temperature distribution. Metaclastic rocks show only slight
mineralogical changes, being graphite-white mica-chlorite-albite-quartz schist
with porphyroblastic development of albite in the highest temperature part. On
the other hand, metavolcanic rocks show marked progressive changes (Fig.
6A-6). On the basis of their mineral assemblages, the following zones have been
delineated from south to north:
152/DIVERSITY IN ZEOLITE FACIES METAMORPHISM

Central Kii Peninsula

Metamorphic Laumontite Prehnite-pumpellyite Greenschist

facies facies facies facies
Mineral zoning I II III IV

Laumontite
Prehnite
Pumpellyite
Epidote
Chlorite
Actinolite
Albite
Quartz
White Mica
Stilpnomelane

Fig_ 6A-6 Progressive mineral changes in the central Kii Peninsula in the
Sanbagawa metamorphic belt. (Seki et al. 1971.)

(a) Zone of laumontite. The characteristic metamorphic assemblage in

metavolcanics is: quartz + albite + laumontite + chlorite. No zeolites other
than laumontite have been found.
(b) Zone of prehnite-pumpellyite facies. Neither laumontite nor actinolite
occurs. The characteristic assemblage is: quartz + albite + prehnite +
pumpellyite + epidote + chlorite.
(c) Transitional zone with both pumpellyite and actinolite.Prehnite dis-
appears and actinolite begins to appear at the threshold to this zone.
Since pumpellyite is still present, this zone is regarded as belonging to the
prehnite-pumpellyite facies, and not to the greenschist facies. The
characteristic assemblage is: quartz + albite + pumpellyite + epidote +
actinolite + chlorite.
(d) Zone of greenschist facies. Pumpellyite disappears and the follOWing
greenschist-facies assemblage appears: quartz + albite + epidote + actino-
lite + chlorite + stilpnomelane.
The whole rock Rb-Sr age of metamorphic rocks of this area is about 110
m.y. (Yamaguchi and Yanagi, 1970).

6A-S RELEVANT DATA FROM HIGH-PRESSURE METAMORPHIC

TERRANES

A zone of the prehnite-pumpellyite facies occurs in the lowest temperature part

of the high-pressure metamorphic terrane in the Katsuyama area, Japan, but no
 6A-6 OTHER AREAS/IS3

zone of the zeolite facies has been observed. The significance of laumontite in
the Franciscan metamorphics is not clear.

Katsuyama Area, Japan

The Sangun metamorphic belt of Japan contains sporadically glaucophane-
bearing areas but no zeolite facies areas, though a thorough search has not been
made.
In the Katsuyama area (fig.lS-8) within the belt in western Honshu, Hashimoto
(1968) found a zone of the prehnite-pumpellyite facies which grades with
increasing temperature into a zone of the glaucophane-schist facies. Prehnite is
confined to the former zone, whereas pumpellyite occurs in both. The glauco-
phane-schist facies zone in tum grades with increasing temperature into a
greenschist facies zone with actinolite but without pumpellyite.

Franciscan Formation, California

In the Panoche Pass and Pacheco Pass areas of the Franciscan Formation,
California (Ernst, 1965; Ernst et al. 1970), a well-recrystallized zone having
glaucophane, lawsonite, jadeite and aragonite, grades into an area where
recrystallization is less advanced, and pumpellyite occurs besides albite, calcite,
actinolite, chlorite and stilpnomelane. This might be regarded as a transition
from the typical glaucophane-schist facies toward the prehnite-pumpellyite
facies (fig. 7 A-IS).
In the western part of the Franciscan terrane in California, laumontite is
widespread. The relation of this part to the glaucophane-schist facies area to the
east has not been clarified (Bailey et al. 1964). Apparently, the laumontite-
bearing rocks in the west, the pumpellyite-bearing rocks in the central part, and
the typical glaucophane-schist facies rocks in the east form a progressive
metamorphic facies series. However, it is possible that the laumontite-bearing
rocks in the west were deposited and metamorphosed at a later time than the
typical glaucophane-schist facies rocks in the east (fig. 13-3).

6A-6 OTHER AREAS

Aragonite-bearing Prehnite-Pumpellyite Facies Area in Washington State

The occurrence of aragonite in a burial metamorphic terrane is of some interest,
since aragonite is stable only under high rock-pressures. Vance (1968) described
it in the San Juan Islands, west of Bellingham in the State of Washington. The
terrane is mainly composed of Upper Palaeozoic and Lower Mesozoic pyroclastic
and sedimentary rocks, which were subjected to recrystallization in the prehnite
-pumpellyite facies in Mesozoic time. Metabasalts and meta-andesites show the
mineral assemblage chlorite + albite + prehnite + pumpellyite. Prehnite is more
abundant than pumpellyite. Epidote was rarely found.
IS4/DIVERSITY IN ZEOLITE FACIES METAMORPHISM

The limestone lenses in this terrane are composed of aragonite and calcite.
The largest lens measures about 200 m in length. Vance considers from
structural data that the rock-pressure of the prehnite-pumpellyite facies terrane
was probably lower than the experimentally determined stability field of
aragonite. He doubted the Validity of the use of aragonite as a criterion of high
rock-pressure. Newton, Goldsmith and Smith (1969) have supported this
argument by demonstrating that deformed calcite can be transformed into
aragonite at a lower pressure than undeformed one (see § 7B-19).

Quartz-free Zeolitic Rocks in Eastern Iceland

Quartz occurs in most rocks of the petrographically well-documented areas of
the zeolite facies. Our knowledge on quartz-free zeolite assemblages is very
meager.
A variety of quartz-free zeolitic rocks was described by Walker (1960) in the
Tertiary basaltic lava plateau of eastern Iceland. The lava piles are roughly 5 km
thick, and were subjected to rise of temperature sufficient to cause weak
recrystallization in the zeolite facies. In the recrystallized olivine basalt lavas,
silica minerals are absent, even in amygdales. These rocks contain such zeolites as
gismondine, thomsonite, chabazite, scolecite and phillipsite, which all are stable
probably only in the absence of free silica (§ 6B-I).
Zones based not only on the kinds, but also on the relative abundance, of
zeolites have been distinguished.
(a) Unzeolitized basalt zone (about 200 m thick).
(b) Slightly zeolitized zone. Chabazite and thomsonite are common, and
levyne, phillipsite, and gismondine also occur in amygdales.
(c) Analcime-heulandite zone (about 150 m thick). Analcime, mesolite,
chabazite, thomsonite, levyne, phillipSite, stilbite and heulandite occur in
amygdales.
(d) Scolecite-laumontite zone. This extends from a maximum altitude of
760 m down to sea level. Zeolitization is most intense here. Zeolites
occur not only in amygdales but also in interspaces between or replacing
igneous minerals in ordinary parts of rocks. The zeolites include mesolite,
scolecite, chabazite, thomsonite, analcime, stilbite, heulandite and levyne,
and in the lower part laumontite.
A strongly hydrated zeolite, gismondine, occurs only in the uppermost zone,
whereas relatively weakly hydrated zeolites, scolecite and laumontite, occur in
the lowest zone. The zeolite zones are probably approximately parallel to the
original top of the lava pile and clearly cut across the stratification of the lavas,
and zeolitization should have taken place at a time long subsequent to the
effusion and cooling of individual lavas.
 6A-6 OTHER AREAS/1SS

Quartz-free Zeolite Assemblages in the Mid-Atlantic Ridge near 24°N

A large amount of quartz-free metabasalts of the zeolite facies was dredged from
a transverse fracture zone near 24°N on the Mid-Atlantic Ridge (Miyashiro et al.
1971). These metabasalts were probably recrystallized at some depth and were
caught and brought up by serpentinite masses rising along the fracture zone.
Natrolite, thomsonite, analcime, chabazite, laumontite and stilbite were
identified, among which analcime is the most widespread. Natrolite and
thomsonite occur due to the absence of quartz in these rocks. Natrolite, being
deficient in silica, does not coexist with stilbite rich in silica. The zeolite
assemblages are considered to be close to chemical equilibrium. Laumontite and
chabazite, differing only in H20 content, occur. This suggests that the dredge
hauls contain metabasalts recrystallized over a considerable range of tempera-
ture.
The original basalts were probably abyssal tholeiites. Recrystallization under
zeolite facies conditions was accompanied by introduction of Na20 and H2 O.
The Na20 content varies sympathetically with the H20 content.
The mafic minerals stable in zeolite facies metabasalts are smectite and
mixed-layer minerals made up of chlorite, smectite and vermiculite components.
Chlorite proper does not occur.
Chapter 6B

Zeolite and Prehnite - Pumpellyite

Facies Metamorphism:
Progressive Mineral Changes

6B-l PROGRESSIVE CHANGES IN ZEOLITE ASSEMBLAGES

Though considerable amounts of petrographic and synthetic data are available

on zeolites, they have not given a clear picture of the progressive changes in
zeolite assemblages. It is regrettable that most existing deSCriptions are of
individual zeolites and not of zeolite-bearing assemblages.
Miyashiro and Shido (1970) have therefore attempted to formulate a
theoretical scheme for the possible stability relations of zeolites on the basis of a
few basic assumptions and the available petrographic data, as discussed below.
Since necessary petrographic data are not always available, the scheme is not
complete, and will probably have to be modified in the light of future progress
in the study of zeolite assemblages.
It is assumed in this scheme that zeolites in metamorphic rocks form in
equilibrium with a fluid phase composed of virtually pure H 20, and that the
pressure of the fluid phase is equal to that acting on the solid phases. In some
cases, however, a fluid phase may contain a large proportion of CO 2 (Zen,
1961b), may be highly saline (Hay, 1966), or may not exist. The present scheme
is not applicable to such cases. We further assume that a rise in temperature
always tends to form more strongly dehydrated zeolite assemblages.
The zeolites in general are a large group of hydrated alumino-silicates of
alkalies and alkaline earths. Some of them contain a large amount of Ba, Sr, or
K. However, most are poor in these elements, and may be approximately
regarded as being composed of the following four components: An (= CaA1 2Si2-
Os), Ne (= Na2A12Si20s), Si02 and H2 0. In zeolite structures, Ca 2 + can
usually be replaced by 2Na +. Only the zeolites composed of these four

°
components will be treated here. Such zeolites will be divided into two groups:
Ca-zeolites composed of An, Si02 and H2 and Na-zeolites composed of Ne,
Si0 2 and H2 0. This division is artificial, since natural zeolites always contain
both Ca and Na. We assume, however, that the main features of the stability
relations of zeolites are preserved in this simplification.
 6B-I PROGRESSIVE CHANGES IN ZEOLITE ASSEMBLAGES/IS?

According to the phase rule, the maximum number of stable phases

coexisting over a range of temperature and of pressure in a three-component
system is three. Since we have assumed the existence of an aqueous fluid phase,
the maximum number of the coexisting mineral phases is only two. Thus, the
stable assemblages should be represented by smaller triangles having an apex at
the H2 0 comer of the triangular diagram An-Si0 2 -H 2 0, or Ne-Si0 2 -H 2 0.

°
The position of the two minerai phases in stable assemblages should move away
from the H2 comer with progressive dehydration.

Ca-Zeolite Assemblages
The composition relations of Ca-zeolites are shown in fig. 6B-l. Possible steps of
dehydration are shown in fig. 6B-2. Brief comments will be made on the
individual steps.

Step (a). The Ca-zeolite assemblages with the highest possible H2 contents°
are quartz-stilbite, stilbite-chabazite and chabazite-gismondine, as shown in fig.
6B-2a. These assemblages are thought to represent the lowest temperature of
recrystallization. Stilbite occurs in the lowest temperature part of the Tanzawa
area (fig. 6A-4). Chabazite and gismondine occur in the low-temperature part of
basaltic piles of eastern Iceland (Walker, 1960).

\\
\\
\\
\\
\ \
,Chabazite. Gmelinite, Levyne
Scolecite. \ '\ \ \
Laumontite'. "4tStilbite
\ '<;::~Heulandite
'."'
\~ -'Yugawaralite
\

.
Watrakite " ",\

'" " "' \

" "' \":\
, ""
,,~
,,~
,,~

~ ~~

Fig. 6B-I Composition relations of Ca-zeolites. An = CaAl2 Si 2 Os·

ISS/MINERAL CHANGES IN ZEOLITE FACIES

Quartz
~------------------~
SiO:z
(b)

~ __________________~~Q.uartz

(d) SiO:z

Fig.6B-2 Progressive dehydration of Ca-zeolite assemblages. (Miyashiro

and Shido, 1970.)

Levyne, gmelinite and chabazite have the same or similar chemical composi-
tion. Levyne and gmelinite occur commonly in association with, but less
frequently than, chabazite (Walker, 1960). Only chabazite is shown in fig. 6B-2a
on the assumption that levyne and gmelinite are metastable.
Heulandite occurs in the same zone as stilbite in Tanzawa, and thomsonite
occurs in the same zone as gismondine in eastern Iceland. Heulandite and
thomsonite could form by dehydration of stilbite and gismondine respectively,
as is clear from fig. 6B-1. If our assumptions are satisfied, heulandite and
thomsonite cannot be stable in the same zone as stilbite and gismondine.
Unfortunately the paragenetic relations of these zeolites are not clearly known.

Step (b). With a rise in temperature, stilbite and gismondine disappear, and
heulandite and thomsonite appear instead, as shown in fig. 6B-2b. If the
threshold temperature for the beginning of recrystallization in a metamorphic
terrane lies at this step, stilbite and gismondine would not form at all, as in
 6B-2 EXPERIMENTAL STUDIES/159

Taringatura (fig. 6A-5) and Mogami (fig. 6A-2). In the higher temperature part,
sc01ecite may begin to form, as shown by broken lines in the figure.

Step (c). With a further rise of temperature, chabazite and heulandite

become unstable, and laumontite appears instead, as shown in fig. 6B-2c. The
laumontite + quartz assemblage is stable in the higher temperature part of the
zeolite facies zone in Taringatura (fig. 6A-5) and Tanzawa (fig. 6A-4). Scolecite
and thomsonite should occur in quartz-free rocks, though pertinent petrographic
data are not available.

Step (d). Wairakite occurs in the highest temperature part of the zeolite
facies zone of Tanzawa (fig. 6A-4). If our basic assumptions are satisfied,
laumontite cannot be stable in this part. Wairakite is the only zeolite in drill
cores from the deepest part of the active geothermal fields of Wairakei in New
Zealand and Onikobe in Japan.
The zones observed in individual areas may be correlated roughly with the
above steps as summarized below.
Taringatura (Coombs et al. 1959) (b) (c)
Tanzawa Mountains (Seki et al. 1969a) (a) (b) (c) (d)
Eastern Iceland (Walker, 1960) (a) (b) (c)
Wairakei geothermal field (Coombs et al. 1959) (b) (c) (d)
Onikobe geothermal field (Seki et al. 1969b) (c) (d)

Na-Zeolite Assemblages
A similar line of discussion has been put forward for Na-zeolites by Miyashiro
and Shido (1970). The composition relations and four possible steps are shown
in figs. 6B-3 and 6B-4 respectively.
The analcime + quartz assemblage in the heulandite-analcime stage of
Taringatura (fig. 6A-5) corresponds to step (c'), and the occurrence of albite in
the laumontite stage to step (d') in fig. 6B-4.

6B-2 EXPERIMENTAL STUDIES ON THE STABILITIES OF ZEOLITES

Many authors have carried out synthetic experiments on zeolites. However, most
experiments show metastable phase relations. Charges in starting materials and in
time duration give a great variation in the results (e.g. Coombs et al. 1959;
Koizumi and Roy, 1960). Only a few experimental studies giving important
results will be reviewed here.
Synthetic analcime shows a wide range of solid solution from the albite-like
composition NaAlSi 3 0 s .1·5H2 0, through the idealized analcime composition
160/MINERAL CHANGES IN ZEOLITE FACIES

I
I
I
I
I
I
: ~.PhilliPsite\
Mesolite + '\.
... \.\.",pinoptilolite
Natrolite
"
. \
\. \~Mordenite
\.~.
Analcime \ ~~
"
, ,~~

, ,'~~
\~~

"
\~
,\
,~
,\I
NeL---------------------------------.-------~Si02
Albite
" ~

Fig.6B-3 Composition relations of Na-zeolites. Ne = Na2 Ah Si2 Os.

NaAlSi z 0 6 .Hz O, to the natrolite-like composition NaAlSi1'SOS.0·75HzO (Saha,

1961).
Natural analcime appears to range from Na: AI :'Si = 1: 1: 2·7 to 1: 1: 2
composition. Analcimes formed by reaction of siliceous volcanic glass with water
commonly have a 1: 1: 2·5 or more siliceous composition. They may be
metastable, having formed in environments with an unusually high chemical
potential of silica. Analcimes formed in environments free from siliceous
volcanic glass and quartz usually have a composition around 1: 1: 2 (Coombs and

°
Whetten, 1967).
The eqUilibrium curve for the reaction analcime + quartz = albite + Hz
corresponds to the boundary between step (c') and (d') in fig. 6B4. According
to field evidence in Taringatura, this agrees apprOximately with the boundary
between steps (b) and (c) in fig. 6B-2. The equilibrium curve for the reaction has
a negative slope above 0-4 kbar, as shown in fig. 6B-5 (Campbell and Fyfe, 1965;
Thompson, 1971).
The equilibrium curve for the breakdown reactions of laumontite and
wairakite, determined by Liou (1970), are also shown in the same figure. These
correspond to the boundary between steps ( c) and (d) and the upper limit of
step (d) respectively in fig. 6B-2. Liou (1971b) gives a breakdown curve of stilbite
into laumontite + quartz.
 6B-2 EXPERIMENTAL STUDIES/161

(a') (b')

(c') (d')

Fig. 6B-4 Progressive dehydration of Na-zeolite assemblages. (Miyashiro

and Shido, 1970.)

In § 2-4, when the equilibrium curves of dehydration reactions were

discussed, it was pointed out that they usually have a very small positive slope at
pressures below 0'1 kbar, show a marked bend in the pressure range 0'1-3,0
kbar, and become very steep at pressures above 3 kbar. At pressures above 10
kbar, an aqueous fluid is so highly compressed that equilibrium curves may come
to have a negative slope. The volume changes of the solid phases in dehydration
reactions of certain zeolites are greater than those in other dehydration
reactions. This tends to cause some zeolite dehydration curves to turn to a
negative slope at a relatively low pressure. The curve for the reaction analcime +
quartz = albite + H2 0 is an example. On the other hand, the breakdown curve of
laumontite has a positive slope.
A possible configuration of equilibrium curves for Ca-zeolites based on their
volume relations is shown schematically in fig. 6B-6. The high-pressure parts of
the stability fields of Ca-zeolites are cut off in natural rocks by the fields of
lawsonite, pumpellyite, prehnite and epidote, as shown for example in figs. 6B-5
and 7B-14.
162/MINERAL CHANGES IN ZEOLITE FACIES

4
...
.:'"
.0
3
0
N
J:
Q..
2 ~'l>~'

--,.,.,,.,.,, "
., _ _
,'-:----- .. ----
~o9_

o ... -
o 100 200 300 400
Temperature (OC)

Fig. 6B-5 Stability fields of the analcime + quartz assemblage and

laumontite (Campbell and Fyfe, 1965; Liou, 1970). Broken lines 'Tar' and
'Mog' indicate possible geothermal curves of the Taringatura and Mogami
areas. The high-pressure stability limit of laumontite is shown after
Thompson (1970b).

oN
r?

Temperature
Fig. 6B-6 Possible form of phase boundaries between Ca-zeolites in the
presence of an aqueous fluid. (Modified from Coombs et ai. 1959.)
'Heuland' and 'Wair' mean heulandite and wairakite. The high-temperature
and low-pressure part of this figure is more precisely shown in fig. 6B-5.
 6B-3 SMECTITE, ILLITE AND MIXED-LAYER CLAY MINERALS/163

6B-3 SMECTITE, ILLITE AND MIXED-LAYER CLAY MINERALS

In rocks of the low greenschist facies, chlorite and muscovite are the most
common phyllosilicates. These two minerals continue to be stable down to the
prehnite-pumpellyite facies. In the zeolite facies, chlorite disappears, whereas
smectite, illite and various members of mixed-layer clay minerals appear.
The name smectite (montmorillonoids) here means minerals of the mont-
morillonite group, including montmorillonite, nontronite and saponite. The name
illite means mica-type clay minerals, including white mica (sericite) and celadon-
ite (glauconite).
Identification of these minerals requires special techniques. A conventional
procedure of sample preparation for the X-ray method is as follows: a rock is
crushed into a fine powder, which is dispersed in water and left to settle for
several hours. Fine-grained clay minerals remain suspended in the upper part of
the water column. This suspension is then settled out by the centrifuge. To make
the identification of smectite and vermiculite certain, the following pre-
treatment before X-ray study is strongly recommended. A part of the sample is
treated with MgCI solution and another treated with KCI or NH 4N0 3 solution
to obtain Mg2 +- and K+- or NH4+ -saturated clays. When treated with NH 4N0 3 ,
several hours of boiling in about 1N solution is recommended to assure the
saturation of a certain type of vermiculite with NH 4N0 3 . After these treat-
ments, the samples are washed repeatedly with distilled water and concentrated
with the centrifuge. They are then transferred onto slide glasses and dried in air.
The mounted samples are glyceroled (i.e. immersed in glycerol for more than
several hours) or heated at 550-600 °c for about 1 hour. The changes in angles
and relative intensities of their basal reflections are clues for identification (table
6s-1 ).
The refractive indices of smectite and vermiculite vary according to the
immersion medium. When measured values are reported, the liquid in which they
were determined should be recorded. Illites and chlorites do not show such a
variation.
Present-day marine sediments contain smectite, kaolinite, illite, and chlorite.
Smectite is abundant where volcanic ash has been an important source material.
Kaolinite tends to be abundant near shore. At least some of illite in sediments
seems to be detrital. Recent K- Ar dating of deep-sea illitic clays has given
various geologic ages which suggest an airborne derivation from continental
regions. Discrete chlorite does not occur in zeolite facies rocks, but does occur in
rocks recrystallized at higher temperatures. Therefore, chlorite in present-day
sediments is probably of detrital origin. Chlorite is abundant in sediments on
ocean floors at high latitudes, because the mineral contained in exposed
continental metamorphic rocks is transported and deposited without chemical
weathering owing to the cold climate (Griffin, Windom and Goldberg, 1968).
 164/MINERAL CHANGES IN ZEOLITE FACIES

Table 6B-I Some characteristics used in identification of low-temperature phyllosilicates

Bas~

Untreated Mg 2 +
Treatment (
Heated at Heated at
Dried in air 300°C 550°C Dried in air ,
Smectite 12·4-15-4 9·5-9·8 9·5-9·8 15-4
1(001) > 1(002) etc
Vermiculite 12·6-15 9 9 14·2-14·4
(quickly rehyd. 1(001) ~ 1(002) etc
in air)
Chlorite 14·0-14·3 X 13·8-14·1 X
1(001), (003) < 1(001) ~ 1(002) etc.
1(002), (004)
Swelling 14·0-14·3 X 13·8-14·1 X !

chlorite 1(00l), (003) < 1(00 1) ~ 1(002) etc.

1(002), (004)
Illite 10 X X X
Kaolinite 7·15 X X X
Halloysite 7·2 X Collapse X

x = No change as compared with sample untreated and dried in air.

'etc.' = (002) and other higher order reflections.

Thus, smectite, kaolinite and illite appear to be stable in unmetamorphosed

pelites. Slight recrystallization under zeolite facies conditions would begin to
form mixed-layer minerals with a chlorite component. Then, with a further
increase in temperature, chlorite as a discrete mineral forms in the prehnite-
pumpellyite and the glaucophane-schist facies.
In a few sedimentary piles it has been observed that smectite is abundant in
the uppermost layer, 1-2 km thick, and it is joined by mixed-layer smectite-illite
at a deeper level, until smectite ultimately disappears at a depth of a few
kilometres (Burst, 1959; Powers, 1959). However, the chemical environment
would also have an important effect on their stability. The relationship between
physical and chemical controls has not been analysed.

6B-4 PREHNITE AND PUMPELLYITE

These two minerals are hydrosilicates of Ca and AI, though pumpellyite

contains considerable amounts of Mg, Fe 2 + and Fe 3 + also.
Prehnite Ca2AI3Si301O(OH)z
Pumpellyite Ca4(Mg, Fe2~(AI, Fe3~5Si6023(OHh· 2H 20
They are chemically similar to wairakite, laumontite, chabazite, scolecite,
 6B-5 OPAL, CHALCEDONY AND QUARTZ/165

cing (A) Optic properties

lrated K+- or NH;-saturated
immersed in Dried in Sign of
water Glyceroled air Glyceroled "(-cx elongation

> 18 17·8 12-4 17·8 0·03-0·04 +

< 14·8 14·3-14·5 10·4-10·8 10-4-10·8 0·02-0·03 +

X X X X 0·000-0·007 ±

? 16-18 ? ? ±

X X X X 0·03-0·04 +
X X X X 0·006-0·007 +
X X X X 0 ?

lawsonite, epidote, zoisite and anorthite. In progressive metamorphism, prehnite

and pumpellyite would form by breakdown of some zeolites, and with further
increase in temperature would be decomposed to produce epidote, and in some
cases of glaucophanitic metamorphism lawsonite. The high-temperature stability
limit ofprehnite is at about 400°C for 2-5 kbar PH 2 0 (Liou, 1971a).
Prehnite occurs in low-pressure as well as in high-pressure series of regional
metamorphic rocks. However, it hardly occurs in typical lawsonite-bearing
high-pressure metamorphic rocks.
Pumpellyite also occurs in low-pressure as well as in high-pressure series of
regional metamorphic rocks. It is associated with lawsonite in some areas.
Hashimoto (1968, p. 132) has shown that the refractive indices of pum-
pellyite tend to decrease with increasing metamorphic temperature, probably
owing to a decrease of the Fe 3 + content as a result of progressive reduction.

6B-5 OPAL, CHALCEDONY AND QUAR TZ

Opal is a common mineral in the low-temperature zones of active geothermal

fields. In high-temperature zones, quartz occurs instead, probably owing to a
higher rate of reaction leading to the formation of stable minerals.
166/MINERAL CHANGES IN ZEOLITE FACIES

Opal was reported as occurring in a low-temperature part of the zeolite facies

zones in Mogarni (Utada, 1965) and Tanzawa (Seki et al. 1969a), which both
have a high geothermal gradient. Chalcedony occurs in a high-temperature part
of the zeolite facies zones of Mogami. Quartz is the only silica mineral in the
greenschist facies zones of these areas.
Zeolite facies regions with smaller geothermal gradients such as Taringatura
and central Kii have only quartz, and not opal or chalcedony. Deposition of opal
would indicate an unusually high chemical potential of Si02 • This may induce
the metastable formation of highly siliceous minerals, such as mordenite.
Chapter 7A

Metapelites: Diversity in Progressive

Regional Metamorphism

7A-l NATURE OF THE DIVERSITY

In this chapter an attempt will be made to systematize the diverse features of

regional metamorphism of pelitic rocks all over the world, largely on the basis of
baric types. Many features can be explained in this way (table 7 A-I). Among the
metamorphic terranes of each type, however, there are further variations, the
causes of which have not always been clearly understood.
For example, certain high-pressure metamorphic terranes have aragonite and
the jadeite + quartz assemblage, whereas others do not. This difference may be
ascribed to a difference in rock-pressure within the category of the high-pressure
type (§ 3-3). On the other hand, certain low-pressure terranes have abundant
chloritoid, staurolite and/or almandine, whereas others do not. Such differences
should be due to a combination of T, Ps , PH. 0, Peo " and the chemical
composition of the predominant rocks. In some cases, one of these factors (e.g.
Ps) may have played a dominant role. This could be determined only by a
rigorous paragenetic analysis.
Albee (1965a), Hoschek (1969), Hess (1969) and Hensen (1971) have made
theoretical attempts to coordinate observed mineral assemblages of met ape lites
in terms of temperature and pressure (fig. 3-9). Future progress in this direction
will enable us to speak of minor variations of temperature and pressure and of
relationships between controlling factors.

7A-2 LOW-PRESSURE METAPELITES IN THE RYOKE BELT AND

ABUKUMA PLATEAU, JAPAN

Regional metamorphic rocks of the low-pressure type are widely exposed in the
Ryoke belt running along the Southwest Japan Arc. Metapelites ranging from
slates to gneisses are abundant. Andalusite and sillimanite occur in a lower and a
higher temperature part respectively.
The metamorphic terrane of the Abukuma Plateau could represent a north-
eastern extension of the Ryoke belt. The rock-pressure of metamorphism there,
however, appears to have been slightly higher than that in the Ryoke belt, since
 ......
0\
00
~
i:"l
>o,j
>
"1:1
Table 7A-I Baric types o/regional metamorphism and characteristic minerals o/metapelites i:"l
t'"
....
>o,j
i:"l
(Jl
Glat1.cophane-schist Epidote-amphibolite
Baric type facies Greenschist facies and amphibolite facies Granulite facies o
:;:
i:"l
Low-pressure Andalusite, Andalusite, Orthopyroxene, ::!l
(Jl
spessartite-almandine, sillimanite, sillimanite, ....
>o,j
chloritoid, cordierite, cordierite, -<
chlorite staurolite, almandine
Mn-rich almandine
Ii
"1:1
::!l
Med ium -pressure Mn-rich almandine, Kyanite, Orthopyroxene, o
G)
chloritoid, sillimanite, sillimanite, ::!l
chlorite, staurolite, pyrope-almandine i:"l
(Jl
(Jl
(stilpnomelane) almandine,
chlorite :;:
i:"l
;s:
High-pressure Lawsonite, Chloritoid, Kyanite, i:"l
>o,j
jadeite, chlorite staurolite,
aragonite, almandine
>
;s:
chloritoid, o
::!l
chlorite, "1:1
(stilpnomelane) ::z:
....
(Jl
;s:
 7A-2 LOW-PRESSURE METAPELITES IN THE RYOKE BELT /169

in addition to andalusite and sillimanite, the occasional occurrence of kyanite

has been noted.
For the geologic relations of the Ryoke belt and Abukuma Plateau, refer to
chapter 15, and especially § 15-7.
Three areas within these terranes will be reviewed below.

Shiojiri-Takato Area in the Ryoke Belt

This area is at the eastern end of the Ryoke belt (fig. 15-8). The progressive
mineral changes in the western part of this area have already been summarized in
fig. 3-4 (Oki, 1961a, b; Katada, 1965), whereas those in metapelites and
metapsammites in the eastern part of the area will be reviewed here on the basis
of recent studies by Ono (1969a, b; 1970).
In this region, the sporadic occurrence of recrystallized biotite begins roughly
at the same temperature as the beginning of recrystallization, and the chlorite
zone in the proper sense is lacking. The area under consideration has been
divided into three zones, here called I, II and III in order of increasing
temperature, as shown in fig. 7 A-I. The temperature increases towards the
southeast over a distance of 50 km, until the terrane is cut by a large fault, called
the Median Tectonic Line. The widespread metamorphism is evidently of a
regional type. The distribution of granitic masses is independent of the
distribution of the temperature of metamorphism.
The progressive mineral changes are summarized in fig. 7 A-2. Metapelites are
dominant and metabasites are scarce. Hence, the metamorphic facies names
shown in this figure are tentative. The individual zones show the following
characteristics (Ono, 1969b).

Zone I (chlorite-biotite zone). Chlorite is abundant in the metapelites of this

zone, but a small amount of brown biotite ({3 = 1.635-1.644) also occurs.
Recrystallization is incomplete, and the common assemblages of recrystallized
minerals are as follows:
1. Quartz + plagioclase + chlorite + muscovite + sphene + ilmenite + graphite,
2. Quartz + albite + sodic plagioclase + chlorite + muscovite + biotite +
sphene + ilmenite + graphite.
In conventional geologic mapping, these rocks have been regarded as unmeta-
morphosed slates. Most metapelites of this zone contain two plagioclases with 7
per cent An and 14 per cent An. Ono has ascribed this to the effect of the
peristerite solvus. The metapelites do not contain K-feldspar, but some metap-
sammites do, though it is not clear whether the mineral is stable or relict.
Muscovite is very fine-grained and sometimes greenish, suggesting a phengitic
composition.
170/METAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

QShiojiri

+
+ +
+
+ +
+
+ + + +
+ +
0I 10 km
I +

I::~::::':::::::':I Cenozoic sediments and volcanics

1+ + + I Granitic rocks
Metamorphic rocks (zones I, II & III)

Fig.7A-l Zones of progressive regional metamorphism in the Shiojiri-

Takato area of the Ryoke belt, Japan (Ono, 1969a, b). The location of this
area is shown in fig. lS-8.
 7A-2 LOW-PRESSURE METAPELITES IN THE RYOKE BELT/171

Shiojiri-Takato Area

Tentative metamorphic Greenschist Ep. amph. : Amphibolite

facies facies facies? : facies
Mineral zoning I II III

Chlorite -- --
Muscovite --
Biotite ---------
'" Almandine ----
.~
a; Andalusite --- --
0.
....'"
'"
:;;
Sillimanite
Cordierite
---
Plagioclase - 7 % A n - - - ---
20% 25% An
Plagioclase !---14%An
Orthoclase --
Quartz

Fig. 7 A-2 Progressive mineral changes in the shiojiri-Takato area (Ono,

1969b).

Zone I! (biotite-andalusite zone). This zone is characterized by the disappear-

ance of chlorite. Biotite becomes abundant. Andalusite occurs in the high-
temperature part of this zone. The observed mineral assemblages are:
1. Quartz + sodic plagioclase + muscovite + biotite
2. Quartz + oligoclase + orthoclase + muscovite + biotite
3. Quartz + oligoclase + orthoclase + muscovite + biotite + andalusite.
Assemblages (2) and (3) are confined to the higher-temperature part of this
zone. The rocks are slates and schists.
Plagioclase usually contains 14-20 per cent An. In the lower temperature part
of this zone, the metapelites do not contain K-feldspar but some metapsammites
do. In the highest temperature part, the metapelites commonly contain a small
amount of orthoclase, which shows monoclinic symmetry in X-ray powder
pattern, and can coexist with andalusite. Ono considered that this orthoclase is
produced by decomposition of the phengite component in muscovite. Biotite is
brown in the low-temperature part, but becomes slightly reddish brown in the
high-temperature part.

Zone II! (sillimanite zone). The entrance to this zone is marked by the
appearance of sillimanite. Though andalusite still occurs in some rocks, it is
regarded as a relict mineral. The observed assemblages are:
1. Quartz + oligoclase + orthoclase + muscovite + biotite + andalusite +
sillimanite.
172/METAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

2. Quartz + oligoclase + orthoclase + biotite + sillimanite,

3. Quartz + oligoclase + orthoclase + biotite + sillimanite + cordierite,
4. Quartz + plagioclase + biotite + garnet + cordierite,
5. Quartz + oligoclase + orthoclase + biotite + garnet + sillimanite +
cordierite.
Assemblage (5) is confined to the highest temperature part of this zone. The
amounts of muscovite are around 25 per cent in zone II, but become 8 per cent
or less in zone III. Muscovite in zone III has been interpreted as probably being
formed by retrogressive reaction. Most rocks of this zone are gneisses.
The plagioclase contains 20-30 per cent An. Orthoclase (2 V = 60-65°) is
widespread and abundant. The muscovites are low in iron content (FeO + Fe203
= 0·6-1·2 per cent) and high in Al 20 3 (33-34 per cent). The biotites are reddish
brown. In passing from zone II to III, their refractive index (3 increases from
1,637-1,645 to 1,645-1,652 and the basal spacing d(OOl) changes from
10,06-10'11 to 10,04-10,08 A. The Fe203 content ofbiotites decreases from
7-2 per cent in zones I and II to 1·0-0·0 per cent in zone III. The Ti0 2 content
increases from 2-3 per cent in zones I and II to 3-4 per cent in zone III.
Andalusite in zone II is confined to rocks exceptionally high in Al 20 3 just as
in other parts of the Ryoke belt, whereas sillimanite in zone III occurs in rocks
of a much wider range of composition. Therefore, sillimanite forms partly by the
transformation of andalusite, but largely by the breakdown of muscovite at the
threshold to zone III by a reaction such as muscovite + quartz = sillimanite +
orthoclase + H2 O. Orthoclase forms as a by-product. Usually the Na2 0/K2 0
ratio of orthoclase is higher than that of the associated muscovite, and hence the
resultant orthoclase absorbs Na20 from the coexisting plagioclase, leading to an
increase of the An percentage of plagioclase in zone III, as described above.
Almandine (24-28 per cent FeO, 3-6 per cent MnO) and cordie rite occur in a
high-temperature part of zone III in association with orthoclase. In the highest
temperature part, sillimanite decreases and, instead, cordierite increases, possibly
by some reaction such as sillimanite + biotite + quartz = cordierite + orthoclase +
H2 0.
The H20 content of the metapelites has decreased and Fe203 has been
reduced in progressive metamorphism. There was no metasomatism on a regional
scale.

Granitic Rocks and Pegmatites. The granitic rocks in this area are mostly
granodiorite or quartz diorite. They occur in all zones ranging from unmeta-
morphosed rocks to sillimanite gneisses, and show no relationship to the
distribution of the temperature of the regional metamorphism. They form
discordant intrusions with hornfelsic aureoles. The age of intrusion would be
much less than that of the regional metamorphism. The K-Ar and Rb-Sr ages
of the granites are around 60-70 million years.
 7A-2 LOW-PRESSURE METAPELITES IN THE RYOKE BELT/173

Pegmatitic bodies containing K-feldspar (orthoclase) occur in zone III, and

are especially abundant in its high-temperature part. Thinner bodies, usually
containing biotite, garnet, cordierite and sillimanite as well as quartz, plagioclase
(oligoclase) and orthoclase, have similar mineral assemblages to the surrounding
gneissic rocks. These bodies appear to be approximately in chemical equilibrium
with the surrounding gneisses, and probably have been formed by some local
migration of materials. However, it is doubtful whether the mechanism is partial
melting and squeezing, since their orthoclase content is commonly as high as
50-65 per cent.
On the other hand, pegmatitic bodies more than 50 cm thick are commonly
discordant to the surrounding rocks and show a distinctly different mineral
composition from them. These bodies appear to be certainly of melt origin
(Ono, 1970).

Tukuba Area
The Tukuba (Tsukuba) area is about 60 km northeast of Tokyo, and could
represent an isolated exposure of buried metamorphic complexes connecting the
Ryoke belt with the Abukuma Plateau. It is a classical area studied by Sugi
(1930), who claimed that large-scale injection of granitic magma took place into
gneisses. Since then, the idea of large-scale migration of materials in the Ryoke
metamorphic complexes either by mechanical injection of magma or by
chemical diffusion, has been supported by most Japanese geologists.
The metamorphic complex is mainly composed of metapelites. Andalusite
occurs in the medium-temperature part, whereas sillimanite, cordie rite and rarely
garnet occur in the high-temperature part. Uno (1961) has shown that the
metapelites show no systematic changes in chemical composition with increasing
metamorphism, except for a decrease of H 2 0 content, in spite of their
transformation from biotite slates to sillimanite gneisses. There is no evidence
for large-scale migration of materials.

Central Abukuma Plateau

In the central Abukuma Plateau, 150-190 km NNE of Tokyo (fig. 15-8), the
unmetamorphosed area and the transitional zone into the metamorphosed area
are covered by Tertiary sediments. Miyashiro (1958) and Shido (1958) have
divided the exposed metamorphic terrane into three zones A, B, and C in terms
of progressive mineral changes in metabasites, which are especially abundant in
the eastern half of the area (figs. 3-5 and 8A -1).

Zone A (greenschist facies). Biotite-chlorite-quartz schist is common.

Zone B (amphibolite facies). Biotite-plagioclase-quartz schist is common.

Andalusite occurs rarely.
174/METAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

Zone C (amphibolite facies). Biotite-K feldspar-plagioclase-quartz gneiss with

some garnet and/or sillimanite is common. Andalusite occurs in the lower
temperature part of this zone, and sillimanite in the higher part. The reaction
between muscovite and quartz to produce sillimanite and K-feldspar appears to
begin in the middle of this zone.
Highly aluminous metapelites are rare. The transformation temperature from
andalusite to sillimanite lies within the low-temperature part of zone C.
Recently, the rare occurrence of staurolite and kyanite has been discovered,
and it has been suggested that these formed by a metamorphic event older than
that now under consideration (Hara et al. 1969). However, the evidence for this
is trifling. Staurolite is common in many low-pressure regional metamorphic
terranes of the world as reviewed below, and does not need a peculiar sequence
of events for its formation. Kyanite is extremely rare here, being confined to a
zone close to the isograd representing the transformation of andalusite to
sillimanite. The P- T curve at some stage in the metamorphism could have passed
through, or situated in the close vicinity of, the triple point of the Ah SiO s
system. This situation is similar to that in the southern Abukuma Plateau (Tagiri,
1971) and the Mount Lofty Range in Australia.

7A-3 LOW-PRESSURE METAPELITES IN AUSTRALIA

Cooma and Wantabadgery Areas, New South Wales.

Low-pressure regional metamorphic rocks, mostly of pelitic composition, are
widely exposed in the Paleozoic orogenic belts of eastern Australia (Vallance,
1967). Detailed studies were made in the Cooma area (Joplin, 1942, 1943;
Pidgeon and Compton, 1965) and in the Wantabadgery-Tumbarumba area
(Vallance, 1953, 1960, 1967), both in New South Wales. Abundant granitic
rocks are exposed.
These two areas are near the southeast coast of Australia and are similar to
each other in character. The follOWing sequence of progressive metamorphic
zones was found in the metapelites:
1. Chlorite zone, containing slates and phyllites with chlorite and muscovite.
The Cooma and Wantabadgery areas differ from the Ryoke belt of Japan
in having a wide chlorite zone, which however may not belong to the same
metamorphic event and facies series as the higher temperature zones, as
mention in § 4-1.
2. Biotite zone, containing schists with biotite and muscovite.
3. Andalusite-cordierite zone.
4. High-temperature zone, containing andalusite, sillimanite, cordierite and
K-feldspar.
Garnet and staurolite are very rare in these areas.
 7A-4 LOW-PRESSURE METAPELITES IN FRANCE AND SPAIN /175

Mount Lofty Ranges, South Australia.

The early Palaeozoic orogenic belt in the Mount Lofty Ranges to the east of
Adelaide, South Australia, has a low-pressure metamorphic belt which somewhat
resembles a medium-pressure one (Offler and Fleming, 1968; A. J. R. White,
1966; White, Compston and Kleeman, 1967). The temperature of meta-
morphism increases toward the axial zone over a distance of about 40 km. At
least four mineral zones have been distinguished in pelitic schists:
1. Chlorite zone
2. Biotite zone.
3. Andalusite-staurolite zone. Andalusite begins to occur before staurolite
does in one part of the zone, and the reverse sequence has been observed
in another part.
4. Sillimanite zone. In the higher temperature part of this zone, sillimanite
can coexist with orthoclase.
Kyanite was found to occur sporadically within the andalusite-staurolite and
sillimanite zones. The P- T curve at some stage of metamorphism appears to have
been situated in the vicinity of the triple point of the A12 SiO s system. Granitic
and migmatitic rocks occur in the axial zone.

East Kimberley Area.

Gemuts (1965) has published a preliminary petrographic note on a Precambrian
metamorphic complex composed of pelitic, basic and calcareous rocks and
ranging from the greens«hist to the low granulite facies in the East Kimberley
area near the northeastern corner of Western Australia. The temperature of
metamorphism tends to increase northwards over a distance of 150 km. Though
all the three polymorphs of A12 SiO s occur in metapelites, there is a tendency
for kyanite, andalusite and sillimanite to occur in this order, each in turn being
the most abundant form as the temperature rises. Hence, the geothermal curve
would pass the triple point on the low-pressure side. Chloritoid, staurolite and
cordierite occur in the low-, medium- and high-temperature parts respectively.
A brief general account of metamorphic belts of Australia will be given in
§ 16-2.

7A-4 LOW-PRESSURE METAPELITES IN FRANCE AND SPAIN

Regional metamorphic rocks of the low-pressure type are widespread in the

Hercynian terranes of France and Spain. Andalusite and sillimanite occur there.
Staurolite also occurs in some areas. Small parts of the terranes may belong to
the medium-pressure type, as suggested by the occurrence of kyanite (Capdevila,
1968). There are associated granitic masses, especially common in high-
temperature zones in the metamorphic terranes. Three areas will be reviewed
below.
176/METAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

Aracena Area, Southwest Spain

Metamorphic rocks ranging from the greenschist to the granulite facies occur in
the Aracena area, southwest Spain, near the border with Portugal (Bard, 1969).
No staurolite was found.
Silurian and older formations exposed there are composed of nearly equal
amounts of pelitic, basic volcanic and calcareous rocks with subordinate acidic
volcanics. The metamorphic belt has a thermal axis (fig. 7A -3).
The progressive mineral changes in metapelites, metabasites and limestones
are correlated in fig. 7A-4. The lowest temperature zone of the area is
characterized by the pelitic assemblage chlorite + muscovite + quartz possibly
with occasional pyrophyllite. With a rise of temperature, chloritoid and
andalusite begin to occur in some rocks, and then biotite. With a further rise in
temperature, cordierite and sillimanite begin to occur commonly in association
with K-feldspar and muscovite. The highest temperature zone is characterized by
the decomposition of muscovite, the change of microcline to orthoclase, and the
common occurrence of almandine. Some rocks in this zone contain ortho-
pyroxene, but biotite still remains.
The distance from the biotite isograd to the thermal axis is 3-10 km. Granitic
rocks are distributed in all zones, but granodiorites with cordie rite and garnet or
cordierite and orthopyroxene occur in and near the highest temperature zone.
Guitard and Raguin (1958) and Fonteilles and Guitard (1968) have described
a low-pressure metamorphic complex without staurolite in the Agly area,
eastern Pyrenees. The highest temperature zone in this area has reached the
granulite facies.

Bosost Area, Pyrenees

The Bosost area in the central Pyrenees was studied by Zwart (1962, 1963).
Early and middle Palaeozoic pelites have undergone deformation and meta-
morphism. Four successive phases of deformation occurred throughout a large
part of the Pyrenees. Each of them is characterized by folds having a rather
constant attitute for both axial planes and axes. The first phase formed the
major structure of this and surrounding areas with E-W trending fold axes. The
deformation phases were separated by periods of tectonic inactivity (fig. 7A-5).
The Palaeozoic rocks of the region are widely metamorphosed at the chlorite
grade. In addition, the following four mineral zones indicating progressive rise of
temperature were distinguished:
I. Biotite zone,
II. Staurolite-andalusite-cordierite zone,
III Andalusite-cordierite zone,
IV. Cordierite-sillimanite zone.
Zone IV forms the high-temperature core of this area, around which zone III
and then zone II and zone I form shells. Zone I is absent in the southern part of
 ~ril Grarrtic rocks o Zone A
-...J
Granodiorites WIth >
co<d.erite and garnet • Zone B .J:,.
(or hypersthene)
[~::~
.... Zone C t'"
o
............ BlOme I sograd ~
fo< rnetapehtes
.;"
::a
t%j
[ll
[ll
~
::at%j
s::t%j
'"'l
>
"C
t%j
t'"
::j
t%j
[ll

Z
"1
::a
N
z>
(j
t:<l
>
Z
r t:I
5km {Il
,
o. "C
>
....
Fig.7A-3 Zones of progressive regional metamorphism in the Aracena area, Southwest Spaih. Zones A, B z
and C are characterized by metabasites with actinolite, blue-green or green hornblende, and greenish brown
-...J
or brown hornblende respectively, as shown in fig. 7A-4. (Simplified from Bard, 1969.) -...J
---
178/METAPELITES: DIVERSITY IN PROGRE~SIVE METAMORPHISM

Aracena Area

Metamorphic facies Greenschist Amphibolite facies

facies

Zone A B' s2 C
Plagioclase (%An) <12 12-25 25-45 >40
Chlorite - -
Biotite --
Epidote --
. --
III
~

.....
'0; Actinolite
blue-green green greenish brown or brown
.J:J Hornblende -
G>
:2
Cummingtonite
Anthophyllite
--
? -r--?
--
Clinopyroxene r-
Orthopyroxene -
Sphene -- ---
Zone I II I III IV V VI
I
Chlorite I
--
Muscovite
I
---
Chloritoid -- r-+- I

III
Biotite
Pyrophyllite I-- --
-- I

.~
--
.....
Qj Cordierite
c.
Andalusite --
G>
:2 Sillimanite
Garnet
Sp;;' 15%
---
';;;6%
Gedrite
Hypersthene
--
Plagioclase (%An) <25 >25
I
K-feldspar I microcline orthoclase
: I I
Calcite
Dolomite
III
Phlogopite
--
Q)
c
0
t;
Chlorite
Q)

E
Tremolite -----
:J Diopside
Grossularite -
Forsterite
Wollastonite ~--

Fig.7A-4. Progressive mineral changes in the Aracena area, southwest

Spain (Bard, 1969).
 7A-4 LOW-PRESSURE METAPELITES IN FRANCE AND SPAIN/179

Deformation phases

1 2 :3 4

, Biotite
Muscovite

-
..
Biotite
c: Muscovite
'c
0
Staurolite
> Andalusite
Q)
C Cordierite
Garnet

.
Q)
c:
II
Biotite
0
N
u
'E
0
Muscovite
:cc. 6 Staurolite
...0
~
E
Q)
U
..
II Andalusite
E Cordierite
Garnet
E

.
> I--
Q)

';;;
Biotite
~
Cl
...
0
Muscovite - --
"'
Q.. Staurolite
Andalusite
Cordierite
Sillimanite - - --
I--
Biotite
Muscovite - --
'V Staurolite
Andalusite
Cordierite
Sillimanite

Fig, 7 A-S Relations between mineral formation and deformation in

metapelites of the Bosost area, Central Pyrenees, The four phases of
deformation are separated by periods of tectonic inactivity (Zwart, 1963).
the area, where zone II is in direct contact with chlorite-grade phyllites. The
metamorphic zones cut across the structural and stratigraphic boundaries, and
zone IV does not occur in the lowest rocks stratigraphically, which are exposed
in the area.
By combined structural and petrologic analyses, Zwart has concluded that the
four zones represent a succession in time. First the rocks passed through the
stage of zone I, then part of the rocks went through the stage of zone n, next
most of these rocks passed through the stage of zone III, and finally some
through the stage of zone IV. The depth of formation of zone II was estimated
at 3,5-4,0 km.
180/METAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

It is considered that the biotite schists of zone I were formed during

deformation phase 1, when schistosity was produced in the infrastructure and
slaty cleavage in the suprastructure. In the static period following phase 1,
porphyroblasts of staurolite, andalusite, cordierite, biotite and garnet began to
form.
Granitic rocks and pegmatites occur in zones II, III and especially in zone IV.
No granites were influenced by deformation phase 1, and hence they must
postdate this phase. The granites are presumed to have formed by metasomatic
transformation of metasediments. Pegmatite began to form in phase 2. The
temperature was not high enough to cause partial melting.

Cevennes, France
Metapelites containing widespread pyrophyllite have been described on the River
Beaume in the Cevennes near the southeastern border of the Massif Central
(Palm, 1958; Tobschall, 1969). The metapelites contain 16-20 per cent of
Al 2 0 3. The following zones have been observed.

Ia. Chlorite zone. This is characterized by the occurrence of albite (0-8 per
cent An), quartz, muscovite chlorite and clinozoisite. Pyrophyllite
optically resembles muscovite, but was identified by the X-ray powder method
on concentrated samples. It was estimated to constitute 2-5 per cent of the
rock.
lb. Biotite zone. Biotite begins to occur. This zone resembles the preceding one
in other respects. Pyrophyllite is widespread.
II. Zone of manganiferous garnet. This was distinguished from the preceding
zone by the occurrence of manganiferous garnet, an analysis of which showed a
composition of AlmSOSP39Pytt. The occurrence of garnet in this zone may
have been controlled mainly by a compositional variation of the metapelites.

III. Andalusite zone. A further increase in temperature of metamorphism caused

decomposition of pyrophyllite to produce andalusite. In addition to andalusite,
oligoclase (12-18 per cent An), quartz, muscovite, biotite and garnet occur.
Primary chlorite and clinozoisite are no longer present, and part of the andalusite
is assumed to have formed by the reaction of chlorite with muscovite.

7A-5 LOW·PRESSURE METAPELITES IN NORTH AMERICA

Northern Appalachians
In the northern Appalachians, a wide region including southeastern New York,
Connecticut, Massachusetts, and Vermont was subjected to medium-pressure
 7A-5 LOW-PRESSURE METAPELITES IN NORTH AMERICA /181
regional metamorphism in Palaeozoic time, whereas to the east and northeast,
another region including northern and eastern New Hampshire and Maine was
subjected to low-pressure regional metamorphism (§ 13-3). The boundary
between the two regions is shown in fig. 7 A-6. It is possible that the

Montreal
o
Canada
-------
U.S.A

A tlantic Ocean

o
LiJ
Triassic formatlo~s
(post·metamorphlc)

O&
o New York City
Un metamorphosed rocks,

o
Chlorite zone (c)

Biotite and Garnet zones

o 100
t _ _. ._ _. .
~
200 km
t _ _ _ _ _ _ _'

~ Staurolite, Sillimanite, &

~ Sillimanite·K Feldspar zones
A- - - - A Boundary between medium·
& low-pressure regions

Fig. 7 A-6 Palaeozoic regional metamorphic zones in the northern

Appalachians. Granitic rocks are omitted. The regions on the southwest
and the northeast side of the line A-A are the medium-pressure and the
low-pressure regions respectively. Locations: 1, Waterville, 2, Errol, 3,
Bryant, 4, Dutchess County,S, Woodsville. (Modified from Warner, Doyle
and Hussey, 1967; Doll, 1961 and Thompson and Norton, 1968.)
182jMETAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM
rock-pressure of metamorphism tends to become lower toward the northeast in
the latter region.
Prehnite-pumpellyite facies rocks occur in the northeast corner of the
metamorphic region in Maine (§ 6A-2) and along the west margin of the region
in New York State (E-an Zen and Yotaro Seki, pers. comm. 1970).
Two areas in the metamorphic region will be reviewed below.

Waterville- Vassalboro Area, Central Maine. This area is distant from the
medium-pressure region (fig. 7 A-6). Andalusite and staurolite are common
(Osberg, 1968, 1971).
The progressive mineral changes are summarized in fig. 7A-7. The almandine
is probably more or less manganiferous. Cordierite begins to occur at a higher
temperature than andalusite but at a lower temperature than sillimanite.
Staurolite does not persist into the sillimanite zone.

Waterville·Vassalboro Area

Chlorite Staurolite-
and Almandine andalusite Cordierite Sillimanite
biotite zone zone zone zone
zones

Chlorite
Muscovite
Biotite
Almandine
Staurolite
Andalusite
Cordierite
Sillimanite

Fig.7A-7 Progressive mineral changes in metapelites of the Waterville-

Vassalboro area, central Maine. (Osberg, 1968.)

E"ol-Bryant Pond Area, New Hampshire and Maine. This area is closer to the
medium-pressure region of Vermont than the preceding area (fig. 7A-6).
Metapelites recrystallized at relatively low temperatures have been studied in the
Errol quadrangle (Green, 1963), as summarized in fig. 7A-8.
The lowest temperature part is represented by the biotite zone, where
chlorite and muscovite are common. The aluminous minerals pyrophyllite and
paragonite were looked for without success. Only manganiferous beds contain
pyralspite garnet, probably rich in MnO.
With a slight increase in temperature, almandine begins to occur. With a
further rise of temperature, staurolite and andalusite begin to occur. Chlorite
appears to be still stable at this stage.
 7A-5 LOW-PRESSURE METAPELITES IN NORTH AMERICA/183

The highest metamorphic temperature in the Errol quadrangle is represented

by the sillimanite zone. Sillimanite occurs in fine needles or prismatic crystals
usually enclosed in muscovite, biotite, chlorite, garnet or quartz. There is no
evidence for direct polymorphic transformation of andalusite to sillimanite.
A sillimanite-bearing area, east of the Errol quadrangle, has been divided into
the lower and upper sillimanite zones (Guidotti, 1968, 1970). The former is
characterized by the coexistence of staurolite with sillimanite, whereas the latter
has no staurolite. Within the lower sillimanite zone, staurolite has been partially
replaced by muscovite from the crystal margins. In the higher sillimanite zone,

Errol Quadrangle

~ zoning Biotite Garnet Staurolite Sillimanite

Chlorite -- ---
Muscovite
Biotite
Almandine
Staurolite ----
Andalusite - - - - --
Sillimanite

Fig. 7 A-8 Progressive mineral changes in metapelites of the Errol Quad-

rangle, New Hampshire and Maine. (Green, 1963.) "

aggregates of muscovite in pseudomorphs after staurolite tend to recrystallize

into single large plates, up to 2 cm across and commonly lying at high angles to
the schistosity. It is claimed that the muscovite has formed by a progressive
reaction and not by a retrogressive one.
The mineral parageneses in a sillimanite-bearing area have been analysed in
the Bryant Pond quadrangle (Guidotti, 1963; Evans and Guidotti, 1966). Here,
the pelitic gneisses are very coarse-grained, being mainly composed of quartz,
oligoclase, orthoclase, biotite, muscovite, garnet and sillimanite. Quartz and
biotite are present in all the rocks. The so-called sillimanite-K feldspar isograd
can be drawn on the geologic map. Sillimanite and orthoclase occur individually
on the low-temperature side of this isograd, while they can coexist on the
high-temperature side.

Northern Michigan
The middle Precambrian (Huronian) formations in northern Michigan on the
south shore of Lake Superior have undergone low-pressure metamorphism.
James (1955) has studied the area with special reference to the mineral changes
of iron formations. The metamorphism took place in pre-Keweenawan time.
184/METAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

The metamorphic terrane has been divided into five zones characterized by
chlorite, biotite, almandine, staurolite and sillimanite (fig. 7 A-9). A large part of
the Huronian is in the chlorite zone defined by the assemblages in the
metapelites. There are four thermal domes in which the metamorphism increases
up to the sillimanite grade. These are not elongated in shape nor aligned in a
belt.

Metamorphic facies
Greenschist
fOes
aCI
I
Northern Michigan

Epi~ote:
amphibolite
facies
I Amphibolite facies

Mineral zoning Chlorite Biotite Almandine Staurolite Sillimanite

Albite -- ---
Albite-oligoclase - -_.
Oligoclase-andesine
Andesine
....'"Ql -
- - Blue-green
Epidote
°uo
co Actinolite Green
Green and brown
-
.0
~ Hornblende

I
Ql
::2: Chlorite
Calcite -- - - - -Green-brown Brown
Biotite --
Muscovite - -- -- -- --- -- - -- - - --
Quartz - - - - - - - - - - - - - - - - -- -----
Chlorite - --
Muscovite
Biotite
'"
Ql
o~
Almandine
- - - --
Qj
Co Staurolite
....Qlco
::2:
Andalusite
Sillimanite
---
Chloritoid
Cla~i; - Oligoclase
Plagioclase
Quartz

Fig. 7 A -9 Progressive mineral changes in northern Michigan (James,

1955)0

The chlorite zone is characterized by chlorite-muscovite-quartz slates. Some

original clastic grains remain, especially in derivatives of coarse-grained gray-
wackes. Chloritoid has been found at a locality in this zone. In the almandine
zone, the textural reconstruction is almost complete. Though a staurolite zone
was set up in this area, staurolite is very rare. Andalusite occurs in association
with it. The graywackes show almost no change in outward appearance with
increasing metamorphism. Under the microscope, however, they are seen to be
 7A-6 METAPELITES IN THE SCOTTISH HIGHLANDS /185

recrystallized to an interlocking mosaic of quartz and oligoclase. The sillimanite

zone is confined to the central part of some thermal domes.
The associated iron formations were metamorphosed to produce minneso-
taite, stilpnomelane, grunerite, blue-green hornblende and garnet.

7A-6 MEDIUM-PRESSURE METAPELITES IN THE SCOTTISH HIGH-

LANDS

As shown in fig. 7 A-10, the Dalradian and Moine metamorphic terranes of the
Scottish Highlands are divided into two regions. The major part shows a
medium-pressure facies series, including the classical area studied by Barrow,
whereas a smaller southeastern part shows a low-pressure series. The two regions
are commonly called the Barrovian and Buchan regions respectively, as will be
discussed in § 14.3.
The progressive metamorphism of pelitic rocks in the Barrovian region will be
described below on the basis of Harker's (1932) and later work. A synopsis of
mineral changes has been given in fig. 3-2.

Chlorite zone. Common pelitic rocks in the low-temperature part of the chlorite
zone are slates mainly composed of muscovite, chlorite and quartz in varying
proportions. Sometimes albite and graphite-like matter are also present. Quartz
may preserve the outlines of original detrital grains or may have lost them by
recrystallization. With a slight increase in temperature, the grain sizes increase,
leading to the formation of chlorite-muscovite-albite-quartz phyllite or schist.
In some rocks, albite occurs as porphyroblasts.
The muscovites are usually phengite. The chlorites are usually ripidolite.
Paragonite was found to occur only sporadically in metapelites (McNamara,
1965). Stilpnomelane occurs rarely in metapsamrnites. Pyrophyllite has not been
found (Chinner, 1967).

Biotite zone. Typical pelitic rocks in this zone are biotite-chlorite-muscovite-

albite-quartz schist. The amount of biotite may be very small. Many pelitic
rocks are still rich in chlorite and are devoid of biotite. Chloritoid occurs in
rocks high in Al 2 0 3 and PeO.

Almandine zone (garnet zone). With an increase of temperature, almandine

begins to occur. The almandines on the almandine isograd contain about 26 per
cent PeO, 4-5 per cent MnO, 1-2 per cent MgO and 6-8 per cent CaO
(Atherton, 1964). Almandine sometimes occurs in porphyroblasts with
numerous inclusions of quartz and other minerals, and shows marked zonal
structure having a decreasing MnO content toward the margin (Atherton, 1968).
186/METAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

a
!
100, km

Barrovian region Buchan region

~ Chlorite zone Ch lori Ie zone

D Biotite zone

D
... Almandine zone

fililj
,,"II" Kyanite zon£. Andalusite zone

.
""/,,

[Zj
~
< ,
Sillimanite zone Andalusile·sillimanile zone

Fig. 7 A -10 Zones of progressive regional metamorphism in the Scottish

Highlands. The Barrovian and the Buchan region are to the west and to the
east of the line B respectively. The Moine thrust (M) and Highland
Boundary fault (H) represent the northwestern and the southeastern limit
of the metamorphic regions. The Great Glen fault (G) is a strike-slip fault
with a great sinistral displacement. S represents the Southern Upland fault
and A the Ballantrae area. (Modified from Johnson, 1963 and Chinner,
19600.)
 7A-6 MET APE LITES IN THE SCOTTISH HIGHLANDS/187
Typical pelitic rocks of this zone are almandine-biotite-muscovite-albite-
quartz schists commonly with compositional layering. Chlorite may still be
present. Chloritoid occurs in some rocks. Almandine persists into higher zones.
The FeO and MgO contents of almandine tend to increase, whereas the MnO and
Cao tend to decrease with rising temperature (Sturt, 1962; Atherton, 1968).

Staurolite and kyanite zones. With a further increase of temperature, staurolite

and kyanite begin to occur, commonly with a porphyroblastic habit. Barrow
(1893) showed that staurolite begins to form at a slightly lower temperature
than kyanite. Hence, the staurolite zone was distinguished from the kyanite.
Staurolite persists into the kyanite zone. The significance of the staurolite and
kyanite zones in progressive metamorphism, however, is still very obscure.
Staurolite and kyanite are high in Al2 0 3 , If such aluminous hydrous
minerals as pyrophyllite, chloritoid or paragonite were common in pelitic rocks
of the almandine zone, the breakdown of these minerals should produce
staurolite and kyanite. In the low-temperature parts of this area, however,
pyrophyllite has not been found, and paragonite is rare. Chloritoid occurs only
in a region near the east coast. Accordingly, the nature of the reactions which
produce staurolite and kyanite is not known with certainty in most cases
(Chinner, 1967). The occurrence of the staurolite and kyanite zones, or the
separation of the former from the latter may largely depend on variations in the
chemical composition of the predominant pelitic rocks rather than mineral
reactions with rising temperature.
Chinner (1967) suggested the possibility of the migration of K on a regional
scale to produce the staurolite and kyanite zones.
Typical rocks in the staurolite and kyanite zones are staurolite-almandine-
biotite-muscovite -plagioclase-quartz schists and kyanite-almandine-biotite-
muscovite-plagioclase-quartz schists respectively.

Sillimanite zone. The sillimanite isograd represents the boundary of the stability
fields of kyanite and sillimanite. However, the real mechanism of sillimanite
generation does not appear to be the direct transformation of kyanite to
sillimanite (§ 7B-13).
In the highest temperature part of the sillimanite zone, muscovite reacts with
quartz to produce sillimanite and K-feldspar. This marks the first appearance of
K-feldspar in metapelites in this region. The frequency and amount of sillimanite
increase remarkably. Typical pelitic rocks in this zone are sillimanite-almandine
- biotite-muscovite-K-feldspar-plagioclase-quartz gneisses.
Most metapelites in the Scottish Highlands seem to contain a small amount of
graphite and other carbonaceous matter, which should keep such rocks in
relatively reduced states during progressive metamorphism. Some pelitic rocks,
however, are devoid of such substances, and as a result show highly variable
188jMETAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM
oxidation states. Chinner (1960) described a group of such pelitic gneisses from
lowest temperature part of the sillimanite zone. With increase in the degree of
oxidation, biotite reacts with garnet to produce muscovite and iron oxides, as
may be schematically shown by the following equation:

Fe 2 + eastonite (biotite) + almandine + oxygen

=muscovite + iron oxides + quartz.
With such an increase in oxidation, the biotite changes in colour from reddish
brown to green, and becomes higher in MgOjFeO ratio while the garnet becomes
higher in MnOjFeO. The less-oxidized rocks contain graphite as well as ilmenite
and magnetite, whereas more strongly oxidized rocks contain no graphite and do
contain magnetite and haematite.
The variation in the oxidation state of the graphite-free gneisses appears to
result mainly from the compositional differences of pre-metamorphic sedi-
mentary rocks.

7A-7 MEDIUM-PRESSURE METAPELITES IN THE NORTHERN

APPALACHIANS
In the southwestern parts of the northern Appalachians, the pelitic metamorphic
rocks commonly contain kyanite and staurolite (fig. 7A-6). Barth (1936) made a
pioneer petrologiC study of such rocks in Dutchess County in southeastern New
York State. He described a sequence from a muscovite slate zone, through a
kyanite-almandine-biotite schist zone to a sillimanite-almandine-biotite gneiss
zone. Almandine, staurolite and kyanite begin to appear at nearly the same
temperature, and therefore the zones characterized by each of these minerals
were not distinguished.
In Massachusetts, Vermont and New Hampshire, a large number of areas have
been mapped petrographically by workers from universities in New England.
Zones of chlorite, biotite, garnet (almandine), staurolite-kyanite, sillimanite,
and sillimanite-K feldspar have been delineated (Thompson and Norton, 1968;
Albee, 1968).
The chlorite zone is wide and stilpnomelane occurs in some slates of the zone
(Zen, 1960). In the biotite zone, chloritoid occurs in highly aluminous rocks,
and staurolite and kyanite occur in almost the same zone. In the highly
aluminous rocks, kyanite appears to precede staurolite. Much sillimanite forms
by reactions other than the phase transformation of kyanite. K-feldspar begins
to coexist with sillimanite at a higher temperature than that of the first
formation of sillimanite; hence the sillimanite and the sillimanite-K-feldspar
zone.
 7A-8 HIGH-PRESSURE METAPELITES IN THE ALPS/189

The metamorphic facies corresponding to these zones were determined from

the mineral assemblages of the associated metabasites. The chlorite and biotite
zones have been assigned to the greenschist facies, the garnet zone to the
epidote-amphibolite facies, the staurolite-kyanite and sillimanite zones to the
amphibolite facies, and the sillimanite-K-feldspar zone to the lower part of the
granulite facies.
The quartz diorite, quartz monzonite and granite of the New Hampshire
plutonic series occur in the high-temperature parts of the terrane, and appear to
be genetically related to the regional metamorphism.

7A-8 HIGH-PRESSURE METAPELITES IN THE ALPS

Orogeny and metamorphism appear to have taken place repeatedly at widely

different times in the Alps. Alpine, Hercynian and other older metamorphic
complexes are exposed there. Older metamorphic rocks are usually modified by
more recent metamorphic events.
Alpine metamorphism is most clearly observed in post-Hercynian rocks. In
these, the lowest temperature zone which is exposed in the northernmost part of
the metamorphic terrane in the Swiss Alps is characterized by the occurrence of
prehnite, pumpellyite and stilpnomelane. In a wider zone representing higher
temperatures and exposed on the south side, glaucophane and chloritoid occur
in rocks of appropriate compositions (Niggli and Niggli, 1965; Niggli, 1970; van
der Plas, 1959). These rocks may belong to a series of the prehnite-pumpellyite
and the glaucophane-schist facies. Apparently another series of metamorphic
rocks up to the amphibolite facies (sillimanite zone) is exposed in the Simplon-
Ticino region, about 100 km wide in an east-west direction, in the southernmost
part of the Pennine Alps (fig. 7A-ll). This region, probably belonging to the
medium-pressure type, would have formed by temperature rise at a stage later
than that of the recrystallization in the prehnite-pumpellyite and the glauco-
phane-schist facies.
Frey (1969a, b) investigated the Upper Triassic (Keuper) red-bed formation,
the metamorphism of which increases generally southward across Switzerland.
The unmetamorphosed pelites exposed in the northern areas contain 1 Md-illite,
1 Md mixed-layer smectite-illite and haematite. On entering the Helvetic zone to
the south, effects of very slight metamorphism appear as follows. The mixed-·
layer smectite-illite decomposes to produce illite, chlorite and quartz, and then
1 Md-illite changes to 2M 1 phengite. A mixed-layer paragonite-phengite
appears.
Between the Aar and Gotthard Massifs (fig. 7 A-ll), the intensity of
metamorphism increases to produce paragonite and chloritoid. On the south side
of the massifs, i.e. in the Pennine nappes, staurolite, kyanite and almandine
begin to appear, though paragonite and chloritoid still occur.
190/METAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

-- ---

•
Lugano

~\:,~ Prehnite-pumpellyite zone AU Chlori!oid isograd

Stilpnomelane zone
(outer and inner limits) TTT Kyanite isograd

m Glaucophane zone
~ Sillimanite zone
b::!@

Fig.7A-ll Distribution of metamorphic minerals in the Swiss Alps.

(Based on Niggli and Niggli, 1965 and Niggli, 1970.)

Chatterjee (1961) investigated metapelites in the Pennine nappes to the south

of the Aar massif. The An content of plagioclase coexisting with epidote
increases toward the southeast. The following zones were distinguished.

Zone la (chlorite zone). Ordinary metapelites are mainly composed of chlorite,

muscovite, epidote, albite and quartz, sometimes with calcite or chloritoid.

Zone lb (biotite zone). Ordinary metapelites are composed of chlorite, musco-

vite, biotite, epidote, albite and quartz. The plagioclase coexisting with epidote
is still albite.

Zone lla. The plagioclase coexisting with epidote is oligoclase in this zone.
Ordinary metapelites are composed of chlorite, muscovite, biotite, almandine,
chloritoid, epidote, plagioclase and quartz.
 7A-9 METAPELITES IN THE SANBAGAWA BELT, .JAPAN /191

Zone lIb. The plagioclase coexisting with epidote is andesine in this zone. The
metapelites commonly contain staurolite and/or kyanite, and rarely sillimanite,
together with muscovite, biotite, almandine, calcite, epidote and quartz.
The boundaries between these zones cross the stratigraphic horizons and
tectonic structures. The temperature rise which caused the progressive meta-
morphic recrystallization appears to have taken place at a later time than the
large-scale folding and thrusting movements. Almost all the porphyroblasts grew
after the penetrative movement had died out.

7A-9 HIGH-PRESSURE METAPELlTES IN THE SANBAGAWA BELT,

.JAPAN

The Sanbagawa metamorphic belt of Japan generally belongs to the high-

pressure type, though glaucophane, lawsonite and jadeite are not very wide-
spread nor abundant. The temperature was not high enough to produce biotite
except in the Bessi and limori areas. Detailed petrographic studies have been
made in many places, but our knowledge on the poorly recrystallized, low-
temperature part of the belt is still insufficient. Fig..15-9 shows the widespread
presence of a zone of prehnite-pumpellyite facies rocks on the low-temperature
side of a zone of rocks of the greenschist and glaucophane-schist facies, though it
has not been well established (§ 15-8).
The characteristics of metamorphism appear to differ in different parts of the
belt. For example, facies series with the jadeite + quartz assemblage were found
only in the eastern half of the belt. An exceptional area with well-developed
zeolite facies rocks in central Kii Peninsula has already been reviewed in § 6A-4.
It is to be noted that glaucophane occurs in metabasites, but not in
metapelites. Two areas are reviewed below (fig. 15-8).

Bessi-Ino Area, Southwest Japan

Banno (1964) has shown that a virtually unmetamorphosed Upper Palaeozoic
terrane in southern Sikoku grades northwards into a zone of the glaucophane-
schist facies, which in turn grades into a relatively narrow zone of the
epidote-amphibolite facies, the highest temperature part of the Sanbagawa belt,
in the vicinity of Bessi (fig. 15-9). The distance from the southern end to the
highest temperature part is about 40 km (fig. 7 A-12).
The common metapelites in the glaucophane-schist facies are chlorite-musco-
vite-calcite-epidote-graphite-albite-quartz schists. Magnetite and mangani-
ferous garnet occur in some rocks (fig. 3-6).
On entering the zone of the epidote-amphibolite facies, biotite and ilmenite
begin to occur and garnet becomes more common. Thus, typical pelitic rocks are
garnet-chlorite-muscovite-calcite-epidote-graphite-albite-quartz schists with
or without biotite. In the high-temperature part of the epidote-amphibolite
192/METAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

D Zone E

•
[ill
-- Zone 0

Zone C
Sanbagawa
belt
[ill Zone B

111111
1111,1 Zone A2

Pacific Ocean
...
[J Zone A ,

0 Ultrabasic rocks

Older metamorphic rocks

0
I
4, 8 12
, 16
, km ~ of the Kuroscgawa zone

Fig. 7 A -12 Zones of progressive regional metamorphism in the San-

bagawa terrane of the Bessi-Ino area, Japan (Banno, 1964). The blank
areas at the southern and northern ends represent the Pacific Ocean and
the Seto Inland Sea respectively. ML, Median Tectonic Line. Refer to figs.
15-8 and 17-1. For explanation of the zones shown in the key see fig. 3-6
and § 8A-8.

facies, biotite becomes more common, and chlorite almost disappears in

metapelites. Typical rocks there are biotite-garnet-muscovite-calcite-epidote-
graphite-albite-quartz schists.
Lawsonite and jadeite are virtually absent in this area. Glaucophane occurs in
metabasites but not in metapelites. Paragonite has not been found in the
metapelites. Probably the pelitic rocks are not aluminous enough to produce
paragonite.
 7A-I0 METACLASTICS IN THE CALIFORNIA COAST RANGES /193
The highly silicic metasediments (probably metacherts) of this area show a
more diversified mineralogy. They contain stilpnomelane, garnet, piemontite,
riebeckite, magnesioriebeckite, magnesioarfvedsonite or aegirine in addition to
quartz, albite, chlorite, muscovite, epidote and opaque minerals.

Kanto Mountains, Central Japan

Seki (1958, 1960, 1961a) has mapped metamorphic zones in the Kanto
Mcuntains, several tens of kilometres northwest of Tokyo, which represents the
eastern end of the exposed part of the Sanbagawa belt. The metamorphism in
this area has produced lawsonite and a definite zone characterized by the jadeite
+ quartz assemblage.
An unrecrystallized Palaeozoic terrane in this area grades into a zone of
glaucophane-schist facies, which with rising temperature grades into a zone of
greenschist facies. The change of mineral assemblages with temperature it slight
(fig. 3-7). The typical pelitic rocks are chlorite-muscovite-albite-quartz schists
in all zones except for zone IV, where lawsonite and jadeite occur. Stilpnomelane
occurs in all zones. Garnet and piemontite occur in the high-temperature part.
The jadeite is so fine·grained that microscopic identification is very difficult.
Concentration by heavy liquids and identification by the X-ray powder method
were successfully used on a large number of rocks.

7A-1O HIGH-PRESSURE METACLASTICS IN THE CALIFORNIA COAST

RANGES
Coleman and Lee's Nomenclature
The Franciscan complex in the Coast Ranges of California represents eugeosyn-
clinal sedimentary accumulations deposited during Late Jurassic to Late Creta-
ceous time. It consists mainly of graywacke with moderate amounts of shale and
mafic volcanics. Radiolarian cherts and serpentinites are commonly associated
with them. Parts of the Franciscan have been subjected to feeble or intense
recrystallization. Structural analysis has not been successful, because the meta-
morphic terrane is strongly deformed and cut by many faults, and moreover
lacks suitable marker beds.
According to the degree of recrystallization, Coleman and Lee (1963) divided
the Franciscan metamorphics in the Cazadero area into four 'types' as follows:
Type I: Unrecrystallized rocks,
Type II: Incipiently recrystallized, non-schistose rocks,
Type III: Well-recrystallized, schistose metamorphic rocks,
Type IV: Well-recrystallized, coarse-grained metamorphic rocks (schistose or
non-schistose ).
This nomenclature would be applicable to most or all of the Franciscan rocks.
194/METAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

The type IV rocks occur commonly as large isolated blocks resting directly
upon and within less intensely metamorphosed terranes of types II and III. The
origin of these enigmatic blocks is not clear. Coleman and Lee considered that
they are concentrated in a band which is roughly concordant with some of the
major faulting, and that they appear to have been tectonically transported
upwards from greater depths. Coleman and Lanphere (1971) have suggested that
these blocks are fragments of the crystalline basement on which the Franciscan
rocks were deposited.
There is a gradational relation between areas of types I and II and between
those of types II and III. The sequence of types I~II~III represents progressive
metamorphism, or at least progressive increase in the degree of recrystallization.
A large part of the eastern half of the Franciscan metamorphic terranes is in the
glaucophane-schist facies. The rocks of this region could be regarded as the most
typical glaucophane schist facies terrane in the world (e.g. Ernst et al. 1970).
Two areas where there is a progressive metamorphic sequence in metaclastics
will be reviewed below.

Panoche Pass Area, Central California

This area, being about 180 km southeast of San Francisco, is in the Diablo
Range (fig. 13-3). Metamorphosed graywackes are predominant, though some
metacherts and metabasites occur in association. Ernst (1965) has shown that
metabasites in a part of the terrane have pumpellyite-bearing greenschist-like
mineral assemblages, while those in the rest of the terrane show glaucophane-
schist facies mineral assemblages (fig. 3-8). The former may be considered type
II rocks in Coleman and Lee's nomenclature, and the latter type III rocks.
The type II metagraywackes and most of the type III metagraywackes
do not contain jadeitic pyroxene. They are usually incompletely recrystallized,
being composed of clastic grains of quartz, sodic plagioclase and rock fragments
set in a mesostasis with chlorite, white mica and stilpnomelane. Lawsonite
occurs within the plagioclase. Both calcite and aragonite have been found.
The rest of the type III metagraywackes contain jadeitic pyroxene. Such
rocks occur in a few separate areas, 1-2 km across, surrounded by an area of the
preceding group of rocks (fig. 7A - 13). They are thoroughly recrystallized.
Glaucophane occurs in more than half of the rocks. Jadeitic pyroxene occurs as
radial sprays and fibrous intergrowths with lawsonite and glaucophane. It
contains about 70-80 per cent of the jadeite component. Chlorite, white mica
and stilpnomelane occur in these metagraywackes as in those of the preceding
group. The primary carbonate is aragonite.
Thus, the most notable progressive reaction in the metagraywackes is the
formation of jadeitic pyroxene from albite. The isograd corresponding to this
reaction has been called the jadeitic pyroxene + lawsonite isograd (Ernst, 1965).
 7A-10 METACLASTICS IN THE CALIFORNIA COAST RANGES /195

••

:~ 4o~
r ........

' ~~I
o, ,
2 km
~,

•o
Alluvium and
~ landsl ide deposits
Upper Cretaceous
to Pleistocene
Serpentinites

Metasediments
} Franciscan Formation
~ Metabasites

:OJ'd",.
-..... Jadeitic pyroxene·bearing areas

Fig. 7A -13 Zones of metamorphism in the Franciscan terrane of the

Panoche Pass area in the Diablo Range, California. (Ernst, 1965.) The
location of the Diablo Range is shown in fig. 13-3.

Pacheco Pass Area, Central California

Pacheco Pass, being about 130 Ian southeast of San Francisco, is also in the
Diablo Range. The Franciscan rocks of this area are metamorphosed in the
typical glaucophane-schist facies. The terrane is composed mainly of metagray-
wackes.
The jadeitic pyroxene + lawsonite isograd runs in a NE-SW direction through
the centre of this area (fig. 7A-14). The rocks on the west and east sides of the
isograd belong respectively to types II and III. West of the isograd, jadeite is
virtually absent, and albite is abundant. Slight recrystallization of the fine-
grained mesostasis of the graywackes produces white mica, chlorite, albite and
quartz. With increasing recrystallization, lawsonite grows as prismatic or tabular
crystals (fig. 7 A-IS).
196/METAPELITES: DIVERSITY IN PROGRESSIVE METAMORPHISM
East of the isograd, recrystallization is more advanced and has produced
jadeite, which initially grows as needles and prisms entirely within albite grains.
Finally the jadeite completely replaces albite usually as slightly radiating bundles
of prismatic crystals. The area to the east of the isograd represents the most
extensive development (c. 200 km 2 ) of jadeitic pyroxene- and aragonite-bearing
metamorphic rocks yet recognized in California. Many of jadeitized sandstones
contain glaucophane, which commonly forms subhedral grains in the matrix,
especially adjacent to chlorite (McKee, 1962a, b; Ernst and Seki, 1967; Ernst et

o. .
5 10.km

o = Unaltered Plagioclase • = Jadeite & Albite

~ = Albite & Lawsonite o (& Lawsonite)
= Jadeite (& Lawsonite)

Fig.7A-14 Zones of metamorphism in the Franciscan terrane of the

Pacheco Pass area in the Diablo Range, California. (McKee, 1962a.)

al. 1970). Stilpnomelane occurs on both sides of the isograd. Associated

metacherts may contain stilpnomelane, garnet, piemontite, riebeckite and
deerite.
Type IV rocks occur as many scattered outcrops and blocks. These rocks
show the follOWing mineral assemblages:
Glaucophane + garnet (+ epidote),
Glaucophane + albite + epidote (+ chlorite),
Glaucophane + lawsonite + epidote (+ chlorite + white mica),
Glaucophane + lawsonite + jadeite,
Actinolite + chlorite (+ talc),
Clinopyroxene + garnet (= eclogite).
 7A-IO METACLASTICS IN THE CALIFORNIA COAST RANGES/I97

I
East of
~
West of
Minerals Isograd Isograd

METAVOLCANICS
Albite ---
Quartz -------------------
Lawsonite - -----
Calcite - --------
Aragonite --------------
Aegirinaugite - -- ----------
Pumpellyite ------
Chlorite -
White Mica --------------------
Stilpnomelane - - ------ -
Sphene
Crossite --
METACLASTICS
Albite -
Quartz
Lawsonite -------
Calcite - -------
Aragonite -------------
Jadeitic Pyroxene
Chlorite --
White Mica
Stilpnomelane - - -
Glaucophane - - ---
METACHERTS
Quartz
Riebeckite - - --
White Mica - - -----------------
Stilpnomelane
Garnet - -- - - -- - - - - - - - ------
Piemontite ? ----------
Deerite -------- --
Fig. 7A -15 Progressive mineral changes in the Pacheco Pass area,
California. (Ernst and Seki, 1967; Ernst et ai. 1970.)

Quartz may be present in any of these assemblages. Pumpellyite, stilpnomelane,

rutile and sphene may occur. The distribution of these rocks shows no clear
relation to the above-mentioned regular increase in metamorphic recrystalliz-
ation nor to the jadeitic pyroxene + lawsonite isograd.
Chapter 7B

Metapelites: Progressive Mineral

Changes

7B-l ORDER OF DISCUSSIONS

In this chapter as in chapter 8B, we will first discuss progressive mineral

reactions in low- and medium-pressure regional metamorphism simultaneously
(§§ 78-1 to 78-17), and will then treat those in high-pressure meta.n1orphism
(§ § 78-19 and 78-20). This order of treatment unnaturally cuts off high-
pressure metamorphic reactions from the other ones. However, it is justified for
the following two reasons.
First, a number of mineral reactions involved in high-pressure metamorphism
have no counterparts in low- and medium-pressure ones.
Secondly, the common clastic sediments observed in typical high-pressure
terranes are graywackes. Petrographic data on aluminous metapelites are abun-
dant for low- and medium-pressure terranes, but are not available for high-
pressure terranes. This enhances the uniqueness of high-pressure metamorphism.
Oxides and sulphides of iron, organic matter, and graphite occur in rocks of
all baric types, and hence will be discussed near the end of chapters 7B and 8B.

7B-2 DISAPPEARANCE OF CLAY MINERALS

Available petrographic data on the change of clay minerals to metamorphic

minerals under low-temperature regional metamorphic conditions is very scarce.
Frey (I969a, 1970) described the structural change of disordered illite into
phengite, and the breakdown of a mixed-layer smectite-illite to produce phengite
and chlorite.
Kaolinite could change to pyrophyllite by dehydration in progressive meta-
morphism. McNamara (1965) has stated that although clay minerals are usually
absent in the chlorite-zone schists in the Dalradian Series, kaolinite occurs rarely
in rocks rich in Al2 0 3 and poor in FeO and MgO. However, Chinner (I 967) has
suggested that kaolinite occurs in the area only as an alteration product.
Pyrophyllite has not been found in the Dalradian. In a detailed petrographic
study of the chlorite-zone and biotite-zone metapelites in Vermont, Zen (1960)
looked for pyrophyllite without success. It can only occur very rarely, if at all.
 78-3 PARAGONITE /199
In the Cevennes, on the other hand, Tobschall (1969) has found that
pyrophyllite is widespread in phyllites and schists of the chlorite and biotite
zones, and is decomposed to produce andalusite with a rise in the temperature of
metamorphism. The occurrence of pyrophyllite depends mainly on the chemical
composition of the predominant metasediments (§ 7A 4).
The transformation of kaolinite to pyrophyllite is represented by the
equation:
(78-1)
kaolinite quartz pyrophyllite

This reaction was experimentally investigated by A. B. Thompson (I970a), as

shown in fig. 78-l.

10
9
... 8
~
"'
.c
7
"0
'5
r{ 6
\I
5
cr
~ 4
:>
!:l
~ 3
c..
2

o 100 200 900 1000

Temperature (Oel
Fig. 7B-1 Experimentally determined equilibrium curves relevant to
metamorphism of pelitic rocks. The dehydration curves and the granite
solidus have been determined in the presence of an aqueous fluid.

7B-3 PARAGONITE

Though paragonite is the sodium analogue of muscovite, it behaves as if it were a

more aluminous mineral than the latter because of its limited stability field. In
low-temperature metapelites with an Al 2 0 3 content just in excess of that
required to form alkali feldspar, muscovite is formed. If the Al 2 0 3 content is
greater, paragonite also forms. Muscovite, including illite and phengite, is stable
from the zeolite to the middle amphibolite facies, whereas paragonite is stable
only over a fraction of this temperature range. The decomposition of paragonite
200/ METAPELITES: PROGRESSIVE MINERAL CHANGES

at a moderate temperature should result in the formation of another highly

aluminous mineral such as kyanite.
Paragonite had been regarded as a rare mineral on earth prior to 1954, when
Eugster and Yoder discovered its common occurrence in metapelites in
Vermont. Paragonite is usually mistaken for muscovite on cursory microscopic
observation.

Low- and High- Temperature Limits

In contrast to the widespread occurrence of muscovite (illite) in unmeta-
morphosed soils and shales, the occurrence of paragonite has been extremely
rarely reported. Under ordinary sedimentary and lowest metamorphic con-
ditions, paragonite seems to be unstable. Zen (1960) has noted that paragonite is
absent in chlorite-zone metapelites (slates), and begins to occur in chloritoid-
bearing phyllite of the biotite zone in the Castleton area, Vermont. He has
ascribed this relation to the existence of a low temperature stability limit for
paragonite, as suggested by the following equation:
NaAlSi 30 s + A1 2 Si 2 0s(OH)4 = NaA1 3 Si 3 0 lO (OHh + 2 Si0 2 + H2 0.
albite kaolinite paragonite quartz (7B-2)
The albite + kaolinite assemblage on the low-temperature side of this equation
was observed, for example, by Coombs (1954, p. 81).
Frey (1969a, 1969b, 1970) found a mixed-layer paragonite-phengite in slates
from the Glarus Alps, and considered that it represents an intermediate step in
the transformation of Na-bearing illite or montmorillonite to paragonite in
metamorphism.
The equilibrium curve for the breakdown of paragonite into albite, corundum
and H2 0 lies at temperatures about 100°C lower than that for the breakdown
of muscovite (Chatterjee, 1970).

Paragonite and Pyrophyllite

In the medium-pressure terrane of the northern Appalachians, paragonite is
widespread in metapelites ranging from the biotite to the kyanite zone, usually
in association with quartz and muscovite and rarely with pyrophyllite. Para-
gonite occurs sporadically in metapelites of the Dalradian of western Scotland
(Chinner, 1967). The mineral was found in many mica schists and glaucophane
schists in the Alps (Harder, 1956).
Thus, the paragonite + quartz assemblage is stable over a wide P-T field of
medium- and high-pressure regional metamorphism. The composition relation of
this assemblage to pyrophyllite is shown by the equation:
NaAl 3 Si 3 0 lO (OHh + 4 Si0 2 = NaAlSi 3 0 s + A1 2 Si40 lO (OHh (7B-3)
paragonite quartz albite pyrophyllite
 7B-4 MUSCOVITE /201
The assemblage albite + pyrophyllite was found in the chlorite and biotite zones
of the low-pressure regional metamorphism in the Cevennes (Tobschall, 1969).
There, this assemblage appears to be stable up to the temperature of breakdown
of pyrophyllite to form andalusite.
The existing data suggest that entropy and volume increase as reaction
(7B-3) proceeds. Hence, the right-hand side should represent higher temperatures
and lower pressures than the left. The occurrence of the albite + pyrophyllite
assemblage in low-pressure metamorphism may be due to this. The plagioclase
associated with epidote tends to have a higher An content in metamorphism
under lower rock-pressures (Table 8B-l), and this may promote the stabilization
of the right-hand side of equation (7B-3).

7B-4 MUSCOVITE

Illite and Muscovite

Illite is a general term for the mica-structure clay minerals including both
dioctahedral and trioctahedral types. Though illite is abundant in unmetamor-
phosed and weakly metamorphosed pelites, its nature and stability are not clear.
Illite in unmetamorphosed sediments commonly appears to be detrital in origin
and unstable in many or most cases. The occurrence of celadonite has been
commonly reported in zeolite-facies rocks.
Sericite or muscovite is said to occur in the high zeolite and higher facies
(Utada, 1965; Seki et al. 1969a). In the greenschist, glaucophane-schist and
higher facies with which we are now dealing, muscovite is very widespread in
metapelites.
Metamorphic muscovites usually show a slight deficiency of alkali ions as
compared with the idealized formula KAl 3Si 30 1 o(OH)2 . A very small part (up
to about 5 per cent) of the alkali sites may be occupied by H30+ or H 20.

Phengite
Complete substitution of (Mg, Fe 2 lSi for AlAl in the idealized muscovite
formula, accompanied by limited substitution of Fe 3 + for AI, gives the formula
K(Mg, Fe 2 +)(AI,Fe 3 lSi 4 0 1 o(OH)2 , which is that of the mineral celadonite.
This is a distinct mineral species occurring in sedimentary and hydrothermal
environments.
Metamorphic muscovites are solid-solutions between idealized muscovite and
celadonite. The observed compositions fall in the range of 0- 50 mol per cent
celadonite. The muscovites near the idealized formula may be called muscovite
proper, and those rich in celadonite component are called phengite. The term
muscovite has been and will be used in this book in a wide sense so as to include
muscovite proper, phengite and sericite.
Muscovites in metapelites of the amphibolite facies are usually close to the
202/METAPELITES: PROGRESSIVE MINERAL CHANGES

idealized muscovite formula in composition. On the other hand, Lambert

(1959), Ernst (1963c), Butler (1967) and others have established that musco-
vites from metapelites of the epidote-amphibolite, greenschist and glaucophane-
schist facies are usually phengites.
Phengite appears to be stable in the low-temperature part of low-, medium-
and high-pressure regional metamorphism. However, abundant data are available
only for the latter two types. The regular compositional variation of muscovite
with temperature or metamorphic facies is shown in fig. 78-2. The idealized

Muscovites from:
o Glaucophane-schist facies
• Chlorite and biotite zones
" Almandine zone
12
.. Staurolite and sillimanite zones

10 Field for chlorite, biotite and almandine zones

6'
adl 6 o
u.
+
o
If 4 Field for 0
glaucophane-sch ist facies
00
2

o 2~0~~22~--~24~~2=6~~2~8--~30~--3~2--~3~4--~3~6--~~~~~40
A~O:l wt_%
Fig.78-2 Composition of muscovites from metapelites. The cross on the
abscissa at 38'4 per cent Alz 03 represents the idealized muscovite
composition. Muscovites from the glaucophane-schist facies tend to have
higher MgO/(FeO + Fez 0 3 ) ratios than chlorite and biotite zone mus-
covites. (Compiled from Butler, 1967; Guidotti, 1969 and others.)

muscovite composition has 38·4 per cent Al 20 3 and 0 per cent (FeO + Fe203)'
Muscovites from the staurolite and higher zones have compositions close to it.
At a lower temperature, muscovite compositions tend to show greater departures
from the idealized formula towards celadonite (Lambert, 1959; Butler, 1967;
Guidotti, 1969).
The content of the celadonite molecule in muscovite changes also with
variation in mineral assemblage at a definite temperature. Among the musco-
vites from the almandine zone of the Scottish Moine schists, for example, those
 784 MUSCOVITE/203
in rocks with the muscovite + plagioclase + K-feldspar + quartz assemblage are
systematically lower in Al 2 0 3 and Na20 than those in rocks with the muscovite
+ plagioclase + quartz assemblage (without K-feldspar). Guidotti (1969) has
suggested that muscovite in the muscovite + plagioclase + K-feldspar + quartz
assemblage should give the highest celadonite contents and be a good indicator
of metamorphic temperature.
Muscovites near the idealized composition crystallize in the 2M 1 structural
type, whereas phengites occur usually in the 2Ml and rarely in the 3T type.
Muscovites near the idealized compositiQll have 'Y = 1·598-1·605 and 2 V (-) =
36-40°. Phengites usually have 'Y = 1·600-1 ·626 and 2 V (-) = 28-35° with a
pale greenish or pinkish colour. Thus, the compositional variation can be
detected to some extent by the measurement of refractive indices (e.g.
Miyashiro, 1958, p. 254) or of2 V(e.g. Iwasaki, 1963, p. 77).
Velde (1965) has made a synthetic study on the stability relations of
phengite. In the presence of an aqueous fluid, the composition range of
muscovite expands to phengite with increasing pressure and decreasing tempera-
ture. This is consistent with its observed modes of occurrence. The breakdown
of phengite produces muscovite + biotite + K-feldspar + quartz. The equilibrium
curve for the breakdown reaction of phengite with a composition of 70 per cent
muscovite + 30 per cent (Mg, Al)-celadonite lies at 4 kbar at 400°C.

Muscovite-Paragonite Relations
There is a wide miscibility gap between muscovite and paragonite. Eugster and
Yoder (1955) and Iiyama (1964) made synthetic experiments to determine the
limit of the gap. It becomes smaller with rising temperature, as shown
schematically in fig. 78-3, thus indicating a possibility of its use as a geologic
thermometer.
The basal spacing of muscovite is around 9·9-10·0 A, whereas that of
paragonite is around 9·6-9·7 A. Thus, two micas can be easily distinguished by
the X-ray powder method. The d 002 values of such coexisting micas are a
function of the miscibility gap, and hence could be used as a geologic
thermometer. Zen and Albee (1964) have shown the validity of this idea, as
shown in fig. 78-4.
After the decompOSition of paragonite near the boundary between the
kyanite and sillimanite zones, the maximum value of Na/(Na + K) in muscovite
rapidly decreases. After muscovite begins to decompose to sillimanite and
K-feldspar, the Na/(Na + K) ratio of muscovite appears to show a further
decrease. According to Evans and Guidotti (1966), the maximum value of the
atomic ratio Na/(Na + K) in muscovite in the sillimanite zone of western Maine
is 0·18 before the beginning of muscovite decomposition, and 0·09 after.
Natural muscovite and paragonite are not simple binary solutions. Muscovite
may be phengitic. Paragonite commonly shows a small extent of the NaSi ~
204/METAPELITES: PROGRESSIVE MINERAL CHANGES

Orthoclase+Plagioclase

Plagioclase

Muscovite+Plagioclase

\ I
\1
~

Paragonite

Muscovite+Paragon

Muscovite Paragonite

Fig. 78-3 Schematic relation of muscovite, paragonite and feldspars in

the presence of quartz. An AlzSiO s mineral is present in all the
feldspar-containing fields. (Evans and Guidotti, 1966.)

• Biotite zone
o Almandine zone
)C Staurolite and Kyanite zones

.« 9·65
)C 8
•

d 002 (muscovite), A
Fig. 78-4 Basal spacing of coexisting muscovite and paragonite in rela-
tion to metamorphic zones. (Zen and Albee, 1964.)
 7B-5 CHLORITOID /205
CaAl replacement. These should cause a change in the miscibility gap and
deviation of d 002 from the value in the idealized case.

7B-5 CHLORITOID

like paragonite, chloritoid (Fe2+, Mg, Mn)Al 2 SiOs(OHh occurs character-

istically in highly aluminous low-temperature metapelites. It has been found in
some contact aureoles and low-pressure regional metamorphic terranes in
association with andalusite and/or cordi.erite as well as in the low-temperature
zones of the Barrovian sequence and the Swiss Alps and other high-pressure
metamorphic terranes. Thus, the range of rock-pressure for the formation of
chloritoid is very wide. Chloritoid has monoclinic and triclinic forms.
Chloritoid occurs only in rocks with an Al 2 0 3 content in excess of the
quantity present in micas and epidote. The rocks must also have high Fe 2 +/Mg
ratios (Halferdahl, 1961). Only a small portion of metapelites fulfill these
conditions. Chloritoids usually have Fe 2 +/(Mg + Fe 2 l ratios above 0·6. Some
chloritoids have a considerable MnO content.
It is not clear what are the low-temperature equivalents of chloritoid.
Possible reactions to form chloritoid with rising temperature are as follows:

5 haematite + chlorite
= 5 magnetite + 2 chloritoid + 2 quartz + 4 H2 0 (7B4)
5 kaolinite + chlorite =7 chloritoid + 7 quartz + 2 H 2 0 (7B-5)

5 pyrophyllite + chlorite
= 7 chloritoid + 17 quartz + 4 H2 0 (7B-6)
5 paragonite + chlorite + 3 quartz
= 7 chloritoid + 5 albite + 4 H2 0. (7B-7)

These reactions would be greatly influenced by the Fe 2 +/Mg ratio of the rocks.
The pyrophyllite + chlorite assemblage is widespread but chloritoid is not
present in the Cevennes (Tobschall, 1969).
Chloritoid occurs in the low-temperature parts (probably from the chlorite to
the lower staurolite zone) of the Dalradian metamorphics in the Scottish
Highlands. The localities with chloritoid are within a narrow belt about 40 km
long extending from the east coast toward the southwest. Paragonite has not
been found in this belt. Further to the southwest, chloritoid has not been found,
and instead paragonite has been found at many localities. Equation (7B-7) may
represent the differing stable assemblages in these northeastern and southwestern
areas (Chinner, 1967).
206/METAPELITES: PROGRESSIVE MtNERAL CHANGES
The breakdown of chloritoid with rising temperature usually gives rise to
staurolite. In some cases of low-pressure metamorphism, however, it may
produce cordierite. Since chloritoid is a hydrous mineral high in FeO, its
breakdown reactions depend not oruy on the rock-pressure, but als,? on PH, 0
and Po,. Halferdahl (1961), Hoschek (1967), Ganguly (1968), Ganguly aild
Newton (1968), and Richardson (1968) have made experimental investigations
on the breakdown reactions of pure Fe-chloritoid. According to these studies,
chloritoid by itself could be stable over all the pressure range (Ps = PH, 0) from
near zero to the highest investigated vaiue (about 20 kbar) and up to about
500-700 °c depending on Po,. In natural rocks, reactions with associated
minerals could decrease the breakdown temperature.

7s-6 CHLORITE
Chlorites may be regarded as solid solutions composed of the following two
components and their iron analogues as shown in fig. 7s-5.

antigorite Mg 6 Si40 lO (OH)s

amesite Mg 4Al 4Si20 lO(OH)s

Thus, the main substitution is Si(Mg, Fe21 ¢ AI Al. The structures of natural
chlorite are made up by the stacking of unit layers, about 14 A thick. In
hydrothermal experiments, not only chlorite with the 14 A structure but also
another mineral of similar composition with a 7 A structure is syntheSized. Such
7 A minerals are called septechlorites. The natural minerals antigorite and
amesite are septechlorites. All the 'chlorites' in pelitic and basic schists so far
examined have been found to have a 14 A structure (Banno, 1964; McNamara,
1965).
The range of the Si(Mg, Fe 21 ¢ Al Al substitution in chlorites from meta-
pelites, however, is very limited. The atomic proportion of Si on the basis of
18(0, OH), or on the anhydrous basis of 140, is usually in the range of 2·4-2·9
as shown in fig. 7s-5. Thus, an idealized formula for such chlorites is
(Mg, Fe2+)4·sAl3Si2·S010(OH)s·
Metamorphic chlorites with Si contents in the range described above, and
with a wide variation in the Fe2 +/Mg ratio, belong mostly to ripidolite in Hey's
(1954) classification. The Fe 2+/(Fe 2+ + Mg) ratio of chlorites in metapelites is
usually in the range 0·2-0·8 (Albee, 1962). The chlorites of me tapelites tend to
show a decrease in the Fe 2 +/Mgratio with rising metamorphic temperature. This
tendency is noticeable in the Barrovian zones as well as in high-pressure terranes
(fig. 7s-5). It has been detected, for example, by a decrease of the refractive
indices with increasing metamorphic temperature (Banno, 1964, p. 259-60).
 7B-6 CHLORITE /207

Muscovite (including phengite) has a lower Fel +/Mg ratio than the associated
chlorite, and hence its breakdown causes a decrease of this ratio in chlorite.
Biotite and garnet have higher Fel +/Mg ratios than the associated chlorite, and
hence their formation also results in a decrease of the ratio in chlorite (e.g.
Ernst, 1964; Atherton, 1968).

• o•,••
...... ••• 0
)CA
o
r:/'
)C

Mg2+
A

.. Si

• Chlorite from chlorite zone

Barrovian zones in the Scottish
o Chlorite from biotite zone } Highlands and Appalachians
)C Chlorite from almandine zone
... Chlorite from chlorite zone Sanbagawa high-pressure
} metamorphic belt, Japan
A Chlorite from biotite-almandine
zone
Fi~. 7B-S Composition of chlorites from metapelites. Fe 3 + is added to
Fe +

Mg-chlorite by itself is stable up to about 800°C in the presence of an

aqueous fluid (Fawcett and Yoder, 1966). In metamorphic rocks, however, it
decomposes usually at a much lower temperature by reaction with associated
minerals such as quartz and muscovite. In metapelites and metapsarnmites,
chlorite is always associated with quartz. The chlorite + quartz assemblage is said
208/ METAPELITES: PROGRESSIVE MINERAL CHANGES

to be stable up to about 650°C, above which chlorite reacts with quartz to

produce cordierite + talc at low pressures, yoderite + talc + quartz at high
pressures, and kyanite + talc at very high pressures (Schreyer, 1968).

7B-7 FORMATION OF BIOTITE

In some low-pressure metamorphic terranes, the biotite isograd for metapelites

roughly coincides with the beginning of recrystallization, whereas in other
low-pressure terranes and the Barrovian series there is a chlorite zone on the
low-temperature side of the isograd. Most of the high-pressure terranes have not
reached a temperature high enough to form biotite.
Prior to the middle 1950s, a false idea was popular that muscovite was nearly
constant in composition. Accordingly, the formation of biotite was regarded as
resulting in AI-enrichment of chlorite as represented by the following relation:

6 muscovite + 3 antigorite component of chlorite

= 6 biotite + 3 amesite component of chlorite + 14 quartz + 8H 2 0
(7B-8)

However, this relation is at variance with recent mineralogical data. The

chlorites in metapelites do not become richer in AI 2 0 3 with rising temperature
(fig. 7B-5). As stated before, muscovites in low-temperature metapelites have a
wide range of solid solution. The biotite-producing reactions were discussed
from the new viewpoint by Ernst (1963c), Velde (1965), Mather (1970), and
Brown (1971).
Figure 7B-6 represents three progressive stages of biotite formation. In stage
(a), biotite forms in rocks to the right of the microcline-chlorite join. Then, a
reaction such as micro cline + chlorite -+ biotite + white mica + quartz + H2 0
would take place in a particular range of temperature, above which biotite forms
in rocks to the right of microcline-phengite-chlorite join as shown in (b). With a
further rise of temperature, the composition range of muscovite solid solutions
becomes smaller through a reaction such as phengite + chlorite -+ idealized
muscovite + biotite + quartz + H2 0. Consequently, biotite forms in rocks of a
composition range that becomes wider with rising temperature, as shown in (c).
Thus, the temperature of the first appearance of biotite depends largely on
the chemical composition of rocks in the terranes. In the Scottish Highlands, for
example, biotite begins to occur at a lower temperature in metagraywackes than
in metapelites (Mather, 1970). Since the metapelites of an area usually show a
considerable variation in composition, a biotite isograd cannot appear as a sharp
line on a map. The progressive metamorphism of pelitic rocks usually causes
partial reduction of Fe 3 + to Fe 2 +. The Fe 2 + thus produced is incorporated in
biotite and other minerals.
 70-7 FORMATION OF OIOTITE/209

It has been established that the Fe 2 +/Mg ratio of muscovite and that of the
coexisting biotite in metapelites show a regular and sympathetic variation (Evans
and Guidotti, 1966; Butler, 1967). This indicates that the coexisting muscovite
and biotite are usually in close approach to chemical equilibrium with each
other.

K2 0·R2 0 3
Mierocline

Kaolinite
Pyrophy

itetl-----=========
2R 2 0 3
(a)

~~~_=======l~~~Tale
Pyrophyllite Chlorite 2RO Pyrophyllite Chlorite 2RO
(b) (e)

Fig. 70-6 Progressive changes of phase relations involving muscovite and

biotite in metapelites.

Muscovite and chlorite have lower Fe2+/Mg ratios than the associated biotite.
Hence, the progress of biotite-producing reactions would tend to cause a decrease
in the Fe 2 +/Mg ratio of biotite. A further rise in temperature would tend to
produce almandine in medium- and high-pressure metamorphism. This would
promote a further decrease of the Fe 2 +/Mg ratio in biotite. If cordierite forms in
low-pressure metapelites, however, the Fe 2 +/Mg ratio of the associated biotite
should increase.
210/METAPELITES: PROGRESSIVE MINERAL CHANGES

The compositional change of biotite in metapelites with rising temperature

has been discussed by a number of authors. Decrease of Fe 2 +/Mg ratio and
increase of Ti0 2 content have been widely observed (Miyashiro, 1958; Lambert,
1959; Engel and Engel, 1960; Oki, 1961b; Binns, 1969).

7B-8 PLAGIOCLASE

In the prehnite-pumpellyite and greenschist facies, the stable plagioclase is albite

regardless of the CaO content of the host rock. If CaO is in excess, it is
contained in calcite, epidote and/or other minerals.
At higher temperatures such as the amphibolite facies, the stable plagioclase
ranges from albite to a more calcic composition depending on the composition
of the host rocks. The CaO content of typical pelites is usually very low, leading
to the formation of albite or oligoclase. Semipelitic rocks and graywackes may
higher in CaO content, producing andesine.
Calcite and epidote have a lower A1 2 0 3 /CaO ratio than the anorthite
component of plagioclase. Accordingly, the formation of more calcic plagioclase
from albite should involve the participation of not only calcite or epidote but
also some aluminous minerals such as muscovite and/or chlorite, as exemplified
by:

albite + calcite + quartz + muscovite + chlorite

~more calcic plagioclase + biotite + CO 2 + H 2 0. (7B-9)

Detailed X-ray studies have shown that ordered plagioclase in a composition

range of about 2-20 per cent An is an intergrowth of two 'phases' with
compositions of 0-1 per cent An and 25-28 per cent An (Laves, 1954, p.
409-411; Brown, 1960). This phenomenon is called peristerite unmixing. If it
may be regarded as unmixing in the ordinary thermodynamic sense, the change of
the plagioclase composition towards a more calcic one in progressive meta-
morphism would be discontinuous, and two plagioclases with different compo-
sitions could coexist in stable equilibrium under some conditions.
An apparently discontinuous change of plagioclase composition from about 5
or 10 per cent to about 20 per cent An with increasing metamorphism has been
reported from some areas, and this may be a result of the peristerite unmixing
(Lyons, 1955; Brown, 1962; Wenk, 1962; Crawford, 1966). According to
Crawford (1966), plagioclase in semipelites of the chlorite zone in Vermont
contains usually less than 1·0 per cent An, and that of the lower biotite zone
contains a few to about 5 per cent An. In the higher part of the biotite zone and
in the lower part of the almandine zone, two plagioclases coexist, one with a few
or several per cent of An and the other with about 20-30 per cent An. The sodic
plagioclase disappears in the staurolite zone, where the temperature of meta-
 7B-9 CORDIERITE /211
morphism may have been above that of the top of the peristerite solvus. In the
Alpine schists of New Zealand, the observed relation is similar, but the sodic
member of the two coexisting plagioclases contains less than 2 per cent An.
When two plagioclases coexist in the same rock, the more calcic one may occur
as a rim around the more sodic core. The boundary between them is sharp.
The coexistence of plagioclases with different compositions was reported
from various other metamorphic rocks, including intermediate and calcic
plagioclases in granulite facies rocks (e.g. Tsuji, 1967; Ono, 1969b; Mall and
Singh, 1972). In any of these descriptions, however, sufficient proof for
equilibrium of the coexisting phases has not been given.

1B-9 CORDIERITE

Cordierite and almandine are common in metapelites in low- and medium-

pressure regional metamorphic terranes, respectively. They are fairly similar in
chemical composition: i.e. cordie rite (Mg, Fe)2Al4SiS018 and almandine
(Fe, MghAl 2Si 3 0 12 • They begin to occur on the high-temperature side of the
biotite isograd, usually at a considerable distance from the biotite isograd.
However, metamorphic cordierites usually have an atomic ratio Fe 2 +/(Mg +
Fe 2 110wer than 0'6, whereas almandines have a higher ratio and may contain
significant amounts of MnO and CaO.
Cordierite is a low-pressure mineral with a relatively low density. Figure 7B-7
shows that the high pressure stability limit of Mg-cordierite is close to the
boundary between the sillimanite and kyanite fields (Schreyer, 1968; Seifert
and Schreyer, 1970). The association of cordierite with silimanite or andalusite
is common in metapelites, whereas that of cordie rite with kyanite is very rare.
Hietanen (1956) described a rock showing the coexistence of cordierite,
andalusite, sillimanite and kyanite. The high pressure stability limit of
Fe-cordierite, being about 3·5 kbar, is much lower than that of Mg-cordierite
(Richardson, 1968). Cordierite shows structural variations and hydration
(Miyashiro, 1957a; Iiyama, 1960; Schreyer, 1964-6; Gibbs, 1966).
Almandine is much more widespread in medium-pressure metamorphic
terranes than in low-pressure ones. However, almandine by itself is not a
high-pressure mineral. It is stable on its own composition over the entire pressure
range on the low temperature side of a line representing the high temperature
limit, which varies markedly with oxidation condition (Hsu, 1968). Hensen
(1971) has clearly shown that cordierite is stable in many low-pressure
assemblages, whereas almandine is stable in many high-pressure ones.
A petrogenetic grid for cordierite and related minerals characteristic of
metapelites is shown in fig. 3-9 (Hess, 1969 and pers. comm. 1970). It suggests
that at low pressures cordierite becomes stable at about 450-500 °c in rocks
with very low Fe 2+/Mg ratios and in association with quartz, muscovite, biotite
212/METAPELITES: PROGRESSIVE MINERAL CHANGES

and chlorite. In rocks with higher Fe 2 +/Mg ratios, garnet and staurolite occur
instead. At 2-3 kbar, chlorite reacts with andalusite with a rise of temperature
to form cordierite and staurolite. The composition field of the rocks which form
cordierite becomes much wider by this reaction. At 4 kbar, on the other hand,

12 C
i
I
I

10
Crystals+Liquid

.
....
.c
~
8

0
N

ct
II 6
,,
.,....
~
,,
.
Mg-Cordierite
::l
,, , I::!
~
a.
4 " §
A'" t0
~
.!?
2
..,.6
~
~
~

0L--4~0-0-----6~0~0-----8~OO~--~1~0700~---1~200

Temperature (DC)

Fig.7B-7 Stability fields of Mg-cordierite, almandine and pyrope, each

by itself (Seifert and Schreyer, 1970; Hsu, 1968: Boyd and England,
1959). The stability field of almandine is for the iron-magnetite and
iron-wiistite buffers. Lines AB and Be represent the field boundaries
between andalusite and kyanite, and between sillimanite and kyanite
respectively.

chlorite reacts with garnet with a rise of temperature to form staurolite and
biotite. Generally speaking, the composition field of the rocks which form
cordierite becomes narrower with increasing pressure.
In this grid, the composition of each solid-solution mineral is assumed to be
constant in order to simplify the calculations. Chinner (1962) has discussed
 78-9 CORDIERITEj213

possible paragenetic relations of metapelites with special reference to the

variation of the composition ranges of solid solution minerals. The schematic
AFM diagram of fig. 78-8a represents an assumption that under a very low
rock-pressure, a complete Fe-Mg solid-solution series of cordierite is stable,
whereas garnet can occur only in rocks with unusually high Fe2+jMg ratios and

(a) Very low Ps and high T

A A
Andalusite Kyanite

(b) Low Ps and medium T (e) Medium Ps and medium T

Fig.78-8 Schematic AFM diagrams representing variations of para-

genetic relations of garnet and cordierite in metapelites with temperature
and pressure. A = Al z 0 3 - (K z 0 + Naz 0), F = FeO, M = MgO. (Modified
from Chinner, 1962.)
relatively low Al 2 0 3 j(FeO + MgO) ratios. This would correspond to high-
temperature, low-pressure contact aureoles.
With increasing rock-pressure and decreasing temperature, the Fe 2 +jMg range
of cordierite becomes narrower, resulting in the enlargement of the rock
composition field favourable for the formation of garnet (fig. 78-8b). The
cordierite-gamet join would be replaced by the andalusite-biotite join, which
214/ METAPELITES: PROGRESSIVE MINERAL CHANGES

conceivably corresponds to the medium-temperature zones of low-pressure

regional metamorphism .
. With a further rise in pressure, the Fe 2 +/Mg range of cordierite dwindles and
then the mineral becomes physically unstable (fig. 7B-8c). The composition field
of metapelites which form garnet would become wider. This situation would
correspond to a medium-temperature zone of the medium-pressure regional
metamorphism.

7B-IO PYRALSPITE GARNETS

Pyralspite garnets are mainly composed of the first three of the following five
common components of the garnets: spessartine (Mn3AI2Si3012), almandine
(Fe3Al2Si3012), pyrope (Mg3Al2Si3012), grossular (Ca3Al2Si3012), and
andradite (Ca3Fe2Si3012)' There is a continuous series of solid solution
between spessartine and almandine, and between almandine and pyrope, but not
between spessartine and pyrope owing to a large difference in ionic radius
between Mn 2+ and Mg.
Pyralspites are widespread in metapelites and metapsammites in medium- and
high-pressure regional metamorphic areas. Though they occur in some rocks of
low-pressure regional and some contact metamorphic terranes, higher rock-
pressures evidently increase the frequency of their occurrence and their average
amount in ordinary rocks. Almandine is the most common garnet in such rocks.
The Fe2+/Mg ratio of coexisting minerals increases generally in the following
order: cordierite ~ chlorite ~ biotite ~ staurolite ~ almandine garnet
(Thompson, 1957).

Stability Relations
Spessartine and almandine are stable under a wide range of rock-pressure down
to zero. The stability fields of these minerals change with the oxidation
conditions. Figure 7B-9 gives the stability field of almandine on its own
composition in the presence of an aqueous fluid under two different oxidation
states. The temperature range of stability of almandine becomes narrower and
disappears with increasing P02 as shown in fig. 7B-I0. Relatively reducing
conditions are essential for the formation of the mineral (Hsu, 1968).
In synthetic experiments on the spessartine composition with excess H2 0,
hydrated and unhydrated spessartines form respectively below and above about
600°C. The low temperature limit of stability of spessartine is much lower than
that of almandine and is nearly independent of Po 2 within the Po 2 range
commonly realized in metamorphism. Manganese changes its oxidation state less
readily than iron (Hsu, 1968).
Pyrope is stable only at pressures above 10 kbar at 400°C as shown in fig.
7B-7 (Boyd and England, 1959). Garnets rich in the pyrope component are not
stable in regional metamorphism except for the granulite and eclogite facies.
 7B-10 PYRALSPITE GARNETS /215

A
I A' B'
Sp I
3
I
1
1 B
I I

$1 I
't J I
~I I
~I I
2 $1 lIE
.§ J (f.
~I '"
c:
'0 It
~I c:
co If
t.1 E ~/&
~, « 'O/~
01 , c: III
£1tJ..
~/'~
I

I
I 1/
18
I
I Ii
I
/ I
I

O~--~4~OO~---=50~O~---6~O~O~--~7~O~O----~8~OO~--~90~O~--~10~OO

Temperature (OC)
Fig. 7B-9 Stability relations for almandine bulk composition in the
presence of an aqueous fluid. Curves A and B are with the iron-magnetite
and iron-wustite buffers, whereas curves A' and B' are for a higher PO z
with the fayalite-magnetite-quartz buffer. Curves A and A' have the same
mineral assemblages on their low-temperature side as well as on their
high-temperature one. Curve Sp represents the low-temperature limit of the
stability field of spessartine on its own composition in the presence of an
aqueous fluid. (Hsu, 1968.)

Compositional Changes in Progressive Metamorphism and Zoning

The progressive compositional change of garnet has been investigated in many
areas. In the high-pressure terrane of the Bessi-Ino area, the MnO content of
garnet in metapelites is as high as 9-13 per cent in the lowest temperature part
up to the biotite isograd, and rapidly decreases to about 3 per cent with further
rise in temperature (Banno, 1964).
In the Barrovian region of the Scottish Highlands, Sturt (1962) and Atherton
(1968) have studied the compositional variation of pyralspites. On the alman-
dine isograd, almandine contains about 5-6 per cent MnO and 6-9 per cent CaO.
With rise in temperature, the MnO and CaO contents decrease rapidly, whereas
the FeO and MgO contents increase. Almandine in metapelites of the high-
216/ METAPELITES: PROGRESSIVE MINERAL CHANGES

Cl
o
...J

Pfluid = 2'0 (kbar)

Temperature (0 C)

Fig.7B-IO Poz-Tdiagram for almandine bulk composition + excess H2 0

at 2 kbar fluid pressure (Hsu, 1968). Field boundaries are presented as
thick lines. Oxygen buffer curves are presented as thin solid lines.
Abbreviations: CH = Fe-chlorite, Fa = fayalite, Hc = hercynite, Mt =
magnetite, Mu = mullite, Py = pyrophyllite, Qz = quartz, S = sillimanite.

temperature part of the almandine zone commonly contain about 0·6 per cent
MnO, 28 per cent FeO, 3-4 per cent MgO and several per cent CaO. In both the
Bessi-Ino area and Barrovian region, the almandine isograd lies very close to the
threshold to the epidote-amphibolite facies zone.
Muller and Schneider (1971) re-investigated the compositional variation of
garnets in metapelites of the Stavanger area, Norway, which had formerly been
described by Goldschmidt (1921). The garnets of the chlorite zone showed 14-5
per cent MnO. The MnO content decreases with rising temperature down to
values around 1 per cent in the amphibolite facies. The garnets in metamor-
phosed arkoses and pyroclastics did not show such a regular variation in
composition.
A remarkable decrease of the MnO content of metapelite garnets was found
in the low-pressure metamorphic terrane of the central Abukuma Plateau
(Miyashiro, 19 53b, 1958). Here, the garnet of a biotite-bearing greenschist facies
schist gave 19·74 per cent MnO. With a rise of temperature, the MnO content
decreases to about 16 per cent at the beginning of the amphibolite facies and to
about 3-5 per cent in the high-temperature part of the same facies.
 7B-1O PYRALSPITE GARNETs/217
Thus, the trend of compositional variation in garnet with rising temperature is
similar in all the baric types: The MnO content decreases and the FeO content
increases. With further rise in temperature, the MgO content increases.
Garnets in metapelites are strongly zoned with decreasing MnO and CaO
contents and an increasing MgO content toward the rim (Atherton, 1968,
Brown, 1969). This trend is the same as that of the above-summarized
progressive changes.

Garnet-Chlorite and Garnet-Biotite Relations

When garnet begins to occur at a relatively low temperature, the necessary
materials would be supplied mainly by gradual decomposition of chlorite. The
(Mg + Fe2+ + Mn2+)/Al ratio of metapelite chlorites is roughly in the range of
1·2-2·5 and that of pyralspite is similar, being 1·5 or a little less. Thus, the
formation of garnet from chlorite in metapelites would be approximately
represented by the following sliding eqUilibrium:
2 (Mg, Fe, Mn)4.sAI3Si2.S01O(OH)g + 4 Si0 2
chlorite quartz
(7B-10)
garnet

After chlorite is used up, garnet continues to coexist with biotite in

metapelites. Garnets always have much greater Mn/Fe 2 + and Fe 2 +/Mg ratios
than the associated chlorites and biotites. In chlorite, biotite and most other
ferromagnesian minerals, Mg2 +, Fe 2 + and Mn 2 + ions are in six-fold coordina-
tion, whereas in garnet they are in eight-fold coordination together with Ca 2 +.
This exceptional structural feature of garnet should control the equilibrium
distribution of Mg, Fe and Mn between garnet and associated ferromagnesian
minerals. The sites in eight-fold coordination in garnet are so large that Ca 2 + and
Mn 2 + ions can be properly accommodated, but small ions of Mg2 + would be
concentrated in associated ferromagnesian minerals (Miyashiro, 1953b;
Zermann, 1962).
A number of workers calculated the Mn-Fe and Mn-(Fe, Mg) distribution
coefficients between garnet and chlorite, and between garnet and biotite. The
values obtained showed irregular variations without a clear relationship to the
temperature of metamorphism. This irregularity would come largely from the
strong zoning of garnets.
On the other hand, the Fe-Mg distribution constant between garnet and
biotite, here denoted as K, shows a regular decrease with increasing temperature
of metamorphism (e.g. Albee, 1965b; Lyons and Morse, 1970):
Mg(in biotite) + Fe 2+(in garnet)
= Fe 2+(in biotite) + Mg(in garnet) (7B-11)
218/ METAPELITES: PROGRESSIVE MINERAL CHANGES

(Fe/Mg) in garnet =K (7B-12)

(Fe/Mg) in biotite

At a temperature within the amphibolite facies, the reaction of muscovite and

biotite may begin to produce almandine, as follows:

muscovite biotite quartz

(7B-13)
almandine K-feldspar

Garnets have much higher Mn/Pe 2 + and Pe2'~/Mg ratios than coexisting biotites.
Above this temperature garnet can coexist with K-feldspar.

Complete Fractionation Model for the Progressive Decrease of the MnO Content
of Garnet
The progressive decrease of the MnO content in garnet will be discussed below
with the help of two models representing two extreme cases (Miyashiro and Shido,
1973). Hollister (1966) and Atherton (1968) have demonstrated that the
variation in the MnO content of some zoned garnets can be semiquantitatively
explained under the following assumptions: (a) There is no diffusion in garnet.
Only the outermost layer of garnet crystals is in equilibrium with the sur-
roundings, and once crystallized the garnet is immediately removed from further
reaction with the system. (b) Diffusion in the surroundings is complete. (c) The
fractionation factor of MnO between the outermost layer of garnet crystals and
the surroundings is constant. These assumptions give a complete fractionation
model for garnet growth.
In this case, MnO is highly concentrated in the early formed gamet, which is
then removed from the active diffusional system, with a resultant decrease in the
MnO content of the system. Hence, the subsequently formed garnet becomes
rapidly poorer in MnO. The following relation holds:

(7B-14)

where
x weight per cent of MnO in the outermost layer of gamet,
g weight fraction of all the already crystallized garnet in all the ferro-
magnesian minerals of the rock,
S = (weight per cent of MnO in a garnet rim)/(average weight per cent of
MnO in the associated ferromagnesian minerals), and
C average weight per cent of MnO in all the ferromagnesian minerals of
the rock (including garnet).
 7B-1O PYRALSPITE GARNETS/2l9

Here, the 'rock' should be understood to mean materials in the volume within
the effective range of diffusion around the garnet.
If S remains constant, the average MnO content (Xav) of zoned garnet
crystals is calculated from the above equation as follows:
C
Xav=- {l-(1-gl} =C{S-!g[S/(S-l)]+ . .. } (7B-lS)
g

Therefore, when g -+ 0, Xav -+ CS. For S > 10, the value of {I - (1 - g)S} is
close to unity in the range of g > 0'2; hence the following approximate relation
holds for this range:
C
Xav =-. (7B-16)
g
The change of Xav with g is illustrated in fig. 7B-ll.
The actual growth of garnet may take place with a practically constant value
of S, or may take place with changing S with rising temperature.

I
I
10 ~C/g
I
I

?f<.
0 6
c
2'

°0~.0----~0~'~1----~0~·2~--~0~.~3----~0~'4~----0~.~5
g

Fig. 7B-11 The MnO contents (Xav and x) of garnet as a function of the
weight fraction of garnet (g) for complete fractionation and complete
equilibrium models, respectively (Miyashiro and Shido, 1973). It is
assumed that C = 0'2, and S = S' = 50.
220/ METAPELITES: PROGRESSIVE MINERAL CHANGES

Complete Equilibrium Model for the Progressive Decrease of the MnO Content
of Garnet
As an alternative case, let us consider a garnet-bearing metapelite in which garnet
crystallizes in complete equilibrium. In this model, the garnet crystals are
assumed to be homogeneous and to have been in equilibrium with the associated
minerals during their growth.
The following relations hold:
g+a = 1.00 }
!..=s' (7B-17)
y
gx+ay=C.
where
x = weight per cent of MnO in garnet,
y = average weight per cent of MnO in the associated ferromagnesian
minerals,
g = weight fraction of garnet in all the ferromagnesian minerals,
a = weight fraction of the associated minerals in all the ferromagnesian
minerals,
S' = fractionation factor, and
C = average weight per cent of MnO in all the ferromagnesian minerals of
the rock (including garnet).
Hence, we have

CS' - 1) }
x = 1 + g(S'
(7B-18)

y = 1+ g(~' - 1) .

In metapelites of low-temperature zones, garnet is usually accompanied by

chlorite and/or biotite. In progressive metamorphism, these minerals are gradu-
ally decomposed with advancing dehydration to produce garnet, though the
actual relations should be very diverse and complicated. Since S' is much larger
than unity, the value of {I + g(S' - I)} increases rapidly with increasing amount
of garnet. As a result, both x and y should decrease progressively. The change of
x is illustrated in fig. 7B-ll.
Though equations (7B-15) and (7B-18) are apparently quite different, the
values and trends of the MnO content of garnet calculated by these equations are
similar, as shown in fig. 7B-l1.
Thus, the progressive decrease of the MnO content of garnet with rising
temperature could be mainly a result of the gradual increase in the amount of
this mineral (Miyashiro and Shido, 1973). The high values of the fractionation
 7B-1O PYRALSPITE GARNETS/221

factors (S and S') are essential to this decrease. Change in their values with
temperature may enhance the progressive compositional change of garnet.
However, such a change of fractionation factor is not necessary to effect the
progressive decrease of MnO in both models. The effective removal of the
previously crystallized garnet from the active diffusional system also is not a
necessary condition for the progressive decrease of MnO.

Effect of Rock-Pressure on the MnO Content of Gamet

It appears that the frequency of occurrence and the amount of garnet in
metapelites tend to increase with increasing rock-pressure, Ps . This is partly due
to the higher density of garnet in comparison with other ferro magnesian
silicates. Higher PHz 0 in high-pressure metamorphism may counteract the effect
ofhighPs ·
The increase of garnet with rock-pressure should have the same effect on the
MnO content of the mineral as the increase of garnet with temperature. We may
expect that the MnO content of garnet tends to be lower in metamorphism
under high rock-pressure, if not greatly disturbed by the effect of PH 2 O.

Effect of Bulk Chemical Composition on the MnO Content of Gamet

In both models, the MnO content of garnet varies in proportion to the MnO
content of the host rock. It is likely that in most metamorphic terranes the
observed apparent progressive compositional change of garnet is enhanced by a
regular variation of the bulk chemical composition of the garnet-bearing rocks
selected for investigation. Usually, the bulk chemical composition of metapelites
exposed in an area should vary within some range. Garnet should be more easily
formed in some rocks than in others. In low-temperature zones, garnet could
form only in rocks with favorable compositions, presumably those with higher
MnO and CaO contents than those of typical pelites. Garnets formed in such
rocks could be expected to have higher MnO and CaO contents, at least partly
owing to the effect of the bulk chemical composition as indicated by equations
(7B-IS) and (7B-18). On the other hand, in high-temperature zones, most or all
metapelites may contain garnet. If so, the average MnO and CaO contents of the
garnet-bearing metapelites should become lower in high-temperature zones.
Thus, apparently randomly selected garnet-bearing metapelites may actually
have statistically decreasing MnO contents with increase in the temperature of
recrystallization.
The progressive increase of MgO content in garnet can be explained in a
similar way. The conversion of biotite to garnet causes an increase of the MgO
content of garnet. In this case, the MgO fractionation factor = (MgO per cent in
garnet)/(MgO per cent in biotite) increases regularly with temperature, and this
should enhance the increase of the MgO content of garnet (Miyashiro and Shido,
1973).
222/ METAPELITES: PROGRESSIVE MINERAL CHANGES

70-11 STAUROLITE

In the Scottish Highlands, staurolite begins to occur at a higher temperature than

almandine but at a lower temperature than kyanite. In the northern Appala-
chians, however, kyanite begins to occur nearly at the same temperature as, or
even at a lower temperature than, staurolite (Thompson and Norton, 1968).
There are a number of possible reactions leading to the formation of staurolite
and of kyanite. Each of them may be valid for rocks of limited compositions and
physical conditions.
Staurolite forms over a wide range of rock-pressure. It occurs in the
high-pressure regional metamorphic terrane of the Alps, in the medium-pressure
ones of various regions, and in the low-pressure ones of the Pyrenees and the
Appalachians. It was found in some contact aureoles also.
In the presence of an aqueous fluid and under Po, of the fayalite-magnetite-
quartz buffer, the staurolite + quartz assemblage is stable roughly between 530
and 700°C above 2 kbar, below which cordierite-bearing assemblages form
instead (Richardson, 1968).
Though there is some doubt about the chemical formula of staurolite, it
resembles chloritoid in that they are both FeO-rich aluminous minerals and
could be stable over a wide range of pressure. Natural occurrences and synthetic
experiments suggest that in progressive metamorphism of aluminous pelites,
chloritoid breaks down roughly at the beginning of the amphibolite facies to
produce staurolite-bearing assemblages. However, the actual reactions for stauro-
lite formation appear to depend on the associated minerals, PH, 0 and P02 '
Reactions to produce staurolite from chloritoid in metapelites would be such as:

31 chloritoid + 5 muscovite + quartz =8 staurolite + 5 biotite + 27

H2 0. (70-19)

Hoschek (1967) has pointed out that the composition range of staurolite-
bearing rocks is considerably wider than that of the chloritoid-bearing ones.
Staurolite in some metapelites, therefore, should form by a reaction not
involving chloritoid. Possible reactions to form staurolite in chloritoid-free rocks
would be such as:

31 chlorite + 5 muscovite
= 26 staurolite + 55 biotite + 20 quartz + 173 H2 0. (70-20)

A reaction of this type was found to take place at 540°C under 4 kbar and at
565°C under 7 kbar in the presence of an aqueous fluid and a fayalite-
magnetite-quartz buffer (Hoschek, 1969).
Some crystals of staurolite in contact metamorphic rocks have been found to
 7B-12 FORMATION OF ANDALUSITE AND KYANITE/223

be composed of sectors with slightly different MgO, Al2 0 3 and Ti0 2 contents.
The compositions within anyone sector are relatively uniform. The composi-
tional difference results from growth in different crystallographic directions
(Hollister and Bence, 1967; Hollister, 1970). It is not known whether such a
difference is common in staurolites of regional metamorphic origin.

7B-12 FORMATION OF ANDALUSITE AND KYANITE

In low-pressure regional metamorphism, andalusite begins to occur usually in a

low-temperature part of the amphibolite facies, if highly aluminous rocks are
present. In medium-pressure regional metamorphism, kyanite begins to occur at
the kyanite isograd usually in the lowest amphibolite facies. In the northern
Appalachians and Scottish Highlands (figs. 7 A-6 and 7 A-I0), a low-pressure
metamorphic region grades into a medium-pressure one. The andalusite-bearing
zone in the low-pressure region lies alongside the kyanite-bearing zone in the
medium-pressure one. The reactions which form the Al2 SiOs minerals may be
similar in both regior.s apart from the polymorphism.
However, a number of different reactions could lead to the generation of the
Al 2 SiO s minerals, probably at different temperatures. The meaning of the
andalusite and kyanite isograds, therefore, is not clear until the relevant reaction
is specified.
Possible reactions are as follows:

(7B-21)
pyrophyllite quartz

3 chlorite + 7 muscovite + quartz

= 13 Al2 SiO s + 7 biotite + 18 H2 0 (7B-22)

6 staurolite + 4 muscovite + 7 quartz

=31 Al2 SiO s + 4 biotite + 3 H2 0 (7B-23)

6 staurolite + 11 quartz = 23 AhSiO s + 4 almandine + 3 H2 0. (7B-24)

Andalusite and kyanite are usually associated with quartz in metapelites.

Accordingly, whatever reactions may form them, the P-T condition on the
andalusite and the kyanite isograd should be within the stability field of the
Al 2 Si02 + quartz assemblage. The low temperature limit for this assemblage is
given by the equilibrium curve for equation (7B-21).
Experimental investigations of the breakdown of pyrophyllite have been
reported by several authors in the last twenty years. Most of the early authors
gave eqUilibrium temperatures between 500-580°C in the presence of an
aqueous fluid at 2 kbar. However, a recent experiment by Kerrick (1968) gave
224/ METAPELITES: PROGRESSIVE MINERAL CHANGES

much lower temperatures: 410 °c at 1·8 kbar and 430 °c at 3·9 kbar, as shown
in Fig. 78-1.
In some regions, staurolite disappears in the upper part of the andalusite or
the kyanite zone (e.g. Francis, 1956; Atherton, 1965; Osberg, 1968), whereas in
others staurolite persists to a lower part of the sillimanite zone (Thompson,
1957; Green, 1963; Guidotti, 1968). The difference may be a result of different
rock-pressures as shown in fig. 3-9, if we may neglect the effects of possible
differences in PH 2 0 and Po 2 .
Chinner (1965) has shown in Glen Clova of the Scottish Highlands that
staurolite of the staurolite zone has a wider range of Fe 2 +/Mg than that of the
kyanite zone. Hence, relatively Mg-rich staurolite of the staurolite zone would
become unstable in the kyanite zone, giving rise to a less Mg-rich staurolite and
kyanite with some compositional adjustment by the associated micas.

78-13 PHASE RELATIONS OF ANDALUSITE, KYANITE AND

SILLIMANITE

The stability of Al 2SiO s minerals depends mainly on the rock-pressure and

temperature. Differences in entropy and free energy between the three poly-
morphs are very small, however. It follows that the phase transformations
between them might be somewhat influenced by structural disorder or slight
replacement by other components. In reality Fe 3 + and some other ions can
replace a small part of the Al3+ in these minerals. The replacement by Fe3+
appears to be promoted by an oxidizing environment (Chinner, Smith and
Knowles, 1969). Possible effects of such replacement on the stability relations
have been experimentally demonstrated (Althaus, 1969).
Despite these reservations, we are impressed by the existence of a clear
regularity in the modes of occurrence of the Al2 SiO s polymorphs. It strongly
suggests that the main features of their modes of occurrence are controlled by
their stability relations. Miyashiro (1949) proposed a model of an inverted Y
form for the phase diagram of Al2 SiO s , which since 1963 has been experi-
mentally proved by a number of researchers.
Recent" experiments by Althaus (1967, 1969) and Richardson et al. (1969)
indicate that the triple point is around 600 °c (± about 50 0c) and 6 kbar (±
about 0·5 kbar). The slope of the andalusite-kyanite field boundary is about 1·0
kbar/l00 °C. The triple point is on the high-temperature side of the eqUilibrium
curve for the decomposition reaction of pyrophyllite (78-21) in the presence of
an aqueous fluid (fig. 78-1). Accordingly, even in the presence of an inter-
granular aqueous fluid, andalusite and kyanite could form by dehydration of
pyrophyllite or other reactions as was mentioned in the preceding section, and
with a further rise of temperature, andalusite and kyanite would transform into
sillimanite.
 7B-14 MICROCLINE TO ORTHOCLASE /225
In detailed examination, however, the coexistence of two polymorphs of
Al 2 SiO s within the same rock is not rare, and even the coexistence of all the
three polymorphs was reported from several areas in the world (e.g. Hietanen,
1956; Pitcher, 1965; Chinner, 1966b; White et al. 1967). Since the free energy
differences between the polymorphs are very small, a polymorph may form
metastably near the transformation temperature. Once formed a polymorph
would persist in the P-T field where another polymorph is stable.
In regional metamorphic terranes, it is rare that sillimanite forms by the
direct phase transformation of andalusite or kyanite. It forms more commonly
as small needles (fibrolite) independent of the pre-existing andalusite and
kyanite. In the sillimanite zone of the Scottish Highlands, Chinner (1961)
ascribed the common formation of sillimanite needles on biotite flakes to ease of
nucleation. The material for the growth of sillimanite was thought to have been
derived from kyanite undergoing decomposition within the same rock. Silli-
manite sometimes forms in lenticular or vein-forming aggregates in gneisses,
suggesting chemical migration during metamorphism.

78-14 MICRO CLINE TO OR THOCLASE

Microcline occurs commonly in metamorphosed psammites and semipelites of

the greenschist to the amphibolite facies. It is not so common in typical
metapelites, in which higher contents of Al 2 0 3 , MgO and FeO tend to eliminate
this mineral in the greenschist and epidote-amphibolite facies. Above the
temperatures of the beginning of muscovite breakdown in the amphibolite
facies, however, K-feldspar becomes very widespread in metapelites;
The transformation from micro cline to orthoclase has been observed in the
middle or high amphibolite facies in progressive metamorphism. The continuous
change of symmetry is readily noticed by the X-ray powder method. Corres-
pondingly, the optic angle 2 V becomes smaller continuously. Some grains show
a zonal structure with a rim having a smaller optic angle than the core (Heier,
1957,1961; Shido, 1958).
However, micro cline is known to occur in some high amphibolite and
granulite facies areas (e.g. Eskola, 1952). Presumably, the orthoclase that had
formed at the maximum of metamorphic temperature was later transformed
retrogressively to microcline. In western Maine, K-feldspar in calc-silicate rocks
was found to be micro cline in an area where K-feldspar in pelitic gneisses is
orthoclase (Evans and Guidotti, 1966). In Broken Hill, Australia, K-feldspar in
quartzo-feldspathic gneisses was found to be microcline in a zone where
K-feldspar in pelitic gneisses is orthoclase (Binns, 1964). These would be due to
a difference in composition of K-feldspar in different rocks, or to selective
retrogressive changes.
226/METAPELITES: PROGRESSIVE MINERAL CHANGES

7B-15 BREAKDOWN OF MUSCOVITE

In a part of the sillimanite zone, muscovite reacts with quartz as shown
schematically by the equation:
muscovite + quartz =orthoclase + sillimanite + H2 0. (7B-25)
Muscovite and orthoclase contain some Na, and hence the real reaction is more
complicated. A closer approach to the reaction would be:

muscovite + quartz + plagioclase

=orthoclase + sillimanite + more calcic plagioclase + H2 0. (7B-26)

The isograd corresponding to this reaction is usually called the sillimanite-

K-feldspar (or orthoclase) isograd.
The accurate location of the sillimanite-K-feldspar isograd on the map is
difficult, because the above reaction takes place over a considerable range of
temperature, and moreover muscovite readily forms retrogressively. Evans and
Guidotti (1966) have made a detailed study of this reaction in western Maine.
In the northern Appalachians, the sillimanite-K-feldspar isograd coincides
roughly with the boundary between the amphibolite and granulite facies
(Thompson and Norton, 1968). However, in the Ryoke belt in Japan, the
isograd appears to be within the amohibolite facies (Ono, 1969b).
The reaction muscovite + quartz = K-feldspar + Al 2 SiO s + H2 0 has been
experimentally investigated by Evans (1965) and by Althaus and Nitsch
(Winkler, 1967, p. 74). The result of the latter workers is shown in fig. 7B-1.
If quartz is absent, the breakdown reaction of muscovite would be such as
muscovite = K-feldspar + corundum + H2 0. As was disc)Jssed in §2-5, this
reaction should take place at higher temperature than equation (7B-25). The
equilibrium curve for this reaction was experimentally investigated by Yoder and
Eugster (1955), Evans (1965) and Velde (1966).

7B-16 PARTIAL MELTING

The temperatures for the beginning of melting of granites and pelites in the
presence of an aqueous fluid were determined by Wyllie and Tuttle (1961),
Yoder and Tilley (1962), Luth, Jahns and Tuttle (1964), Boettcher and Wyllie
(1968), and Winkler (1967). They lie between 620 and 700°C for most of the
pressure range within the crust, as shown in figs. I-I and 7B-1. It follows that if
an aqueous fluid is present between mineral grains, pelitic and psammitic rocks
should undergo partial melting in most of the sillimanite zone (amphibolite and
granulite facies), and even in parts of the andalusite-bearing and kyanite-bearing
zones.
 7B-17 BREAKDOWN OF BIOTITE/227
In the sillimanite zone of regional metamorphism, pockets and veins of
pegmatite as well as granitic masses occur very commonly. Some or all of them
may be crystallization products of liquids that formed by partial melting.
Migmatites may be formed by such melts (von Platen, 1965; Lundgren, 1966;
Winkler, 1967, pp. 192-224; Thompson and Norton, 1968).
In the nmthwest Adirondacks, Engel and Engel (1958, 1960) have found that
quartz-feldspar-biotite-muscovite gneisses derived from semi-psamrnitic rocks
are in the amphibolite facies at a distance of about 55 km from the Adirondack
plutonic complex. The metamorphism advances to the lower granulite facies
near the complex, thus causing the decomposition of muscovite and the
formation of garnet. At this high temperature stage, partial melting appears to
have taken place. Silicic liquids so formed would have been removed, leaving
solid residues lower in Si02 and K2 0 and in Fe3+/Fe 2 + ratio, and higher in
°
Al2 3 , FeO, MgO and CaO.
Some granulite-facies terranes were shown to be less silicic than the associated
lower-temperature ones (Ramberg, 1951; Lambert and Heier, 1968). Most
granulite facies terranes may be composed of solid residues of partial melting
(chapter 18).
Some solid residues of partial melting may have rather unusual chemical
compositions. Grant (1968) has discussed the possibility of the generation of
cordie rite-anthophyllite rocks as solid residues of the partial melting of
graywackes.

7B-17 BREAKDOWN OF BIOTITE

Biotite is stable in metapelites up to a higher temperature than muscovite. As
long as the lower temperature limit of the granulite facies is defined by the
formation of two pyroxenes in metabasites, biotite in the associated metapelites
persists into the granulite facies. Ultimately biotite decomposes, but such
high-temperature metamorphism is rare on earth.
The reaction for the breakdown of biotite may be shown schematically as
follows:
K(Mg, FehAlSi 3 0 lO (OHh + 3 Si0 2 =3(Mg, Fe)Si0 3 + KAlSi 3 0 s + H 2 0
biotite quartz orthopyroxene K-fe/dspar. (7B-27)
Heier (1960) described a progressive metamorphic sequence of metasediments
from the amphibolite to the granulite facies on the Island of Langay in northern
Norway. The rocks are so well recrystallized that they look as though they were
igneous. However, interbedded graphite schist and limestone reveal their original
sedimentary nature. Metasedimentary gneisses of the amphibolite facies are
composed of quartz, microcline-perthite, oligoclase, biotite, green hornblende
and sphene. In the zone of the granulite facies, not only hornblende but also
228/METAPELITES: PROGRESSIVE MINERAL CHANGES
biotite is decomposed. Thus, the rocks there are composed of quartz, ortho-
clase-perthite, oligoclase (or andesine), orthopyroxene and clinopyroxene and
opaque minerals.
In the absence of quartz, the breakdown of biotite needs an exceptionally
high temperature. The equilibrium curve for breakdown reaction: pblogopite =
forsterite + leucite + KAlSi0 4 + H2 0 has been determined by Yoder and
Eugster (1954). It lies above 1000 °c in the presence of an aqueous fluid at a
pressure higher than 1 kbar.
The stability relations of iron-bearing biotite have been investigated by
Eugster and Wones (1962) and Wones and Eugster (1965). They are strongly
influenced by Po" Their breakdown temperature tends to decrease with in-
creasing Fe+2 /Mg ratio.

78-18 AFM DIAGRAMS FOR PROGRESSIVE METAMORPHISM OF

PELITES
In order to summarize the progressive changes in paragenetic relations of
metapelites, two series of Thompson AFM diagrams, one for low-pressure and
the other for medium-pressure metamorphism, are shown in figs. 78-12 and
78-13. Since the actually observed characteristics of regional metamorphism are
very diversified, these diagrams are merely intended to show a generalized,
schematic model for progressive mineral changes.

78-19 LAWSONITE AND ARAGONITE

Reactions to form the three most characteristic minerals of metaclastics in
high-pressure terranes, i.e. lawsonite, aragonite and jadeite, will be examined in
this and the succeeding section. Glaucophane is not discussed in this chapter
inasmuch as it does not occur in typical metapelites.
The occurrence of lawsonite is practically confined to rocks of the glauco-
phane-schist facies. The calculated and synthetic curves representing phase
relations of lawsonite are shown in fig. 78-14. Since lawsonite is a hydrous
Ca-bearing mineral, its stability is controlled largely by PH, 0 and Pc 0 2 (§ 4-4).
Hence, it cannot be a good indicator of rock-pressure.
CaC0 3 appears to have at least three stable polymorphs, calcite I (ordinary
calcite), calcite II, and aragonite, with a triple point at about 480°C and 10·0
kbar, as shown in fig. 78-14 (Boettcher and Wyllie, 1967; Goldsmith and
Newton, 1969). The calcite I-aragonite transition pressure at 100°C is said to
be about 4·35 kbar (Crawford and Fyfe, 1964). Hence, aragonite becomes stable
at a conSiderably lower pressure than the pure jadeite + quartz assemblage. It is
conceivable that in many areas the aragonite which had been formed during a
phase of high-pressure metamorphism, was later transformed back to calcite.
 78·19 LAWSONITE AND ARAGONITE/229
A12~

(a) Biotite zone

AI 2 0 3
Sillimanite

(b) Andalusite zone (e) Sillimanite zone

Fig.78-12 Thompson AFM diagrams for low-pressure progressive meta-

morphism.

Calcite is widespread in low-temperature metamorphic rocks of the low-,

medium· and high-pressure types. In some high-pressure metamorphic rocks of
the Franciscan formation in California, however, aragonite was found to occur
(Coleman and Lee, 1962; McKee, 1962b; Ernst, 1965). Such aragonite is
characteristically associated with jadeite and lawsonite, and is regarded as
formed at high rock-pressures.
Vance (1968) has found the occurrence of aragonite limestones in a
prehnite-pumpellyite facies terrane in Washington State (6A-6). It mayor may
not indicate that aragonite commonly forms metastably at a lower pressure than
its stability field. Newton et al. (1969) have pointed out that deformed calcite
would store a considerable amount of plastic strain energy and hence may
become unstable relative to aragonite at a much lower pressure than undeformed
230/ METAPELITES: PROGRESSIVE MINERAL CHANGES

(a) Chlorite zone

(b) Kyanite zone (c) Sillimanite zone

Fig. 7B-13 Thompson AFM diagrams for medium pressure progressive

metamorphism.

calcite. At low temperatures, aragonite growth is a faster process than strain

recovery of calcite. Thin section observations suggest that the aragonite has
selectively replaced highly strained calcite crystals. The Washington aragonite
would have formed from deformed calcite at relatively low pressures.

7B-20 FORMATION OF JADEITE

Idealized reactions to form pure jadeite NaAlSi 2 0 6 are given by equations:

albite = jadeite + quartz (7B-28)
albite + nepheline = 2 jadeite. (7B-29)
The stability fields of jadeite in the presence and the absence of quartz are
shown in figs. 2-1 and 7B-14. Jadeite occurs as a major component in quartz-free
 78-20 FORMATION OF JADEITE /231

/ /
Aragonite
Y
Calcite II
10
\
\
\
\
\
.... \
'"
.0 \
::! \
'""; \
cC \
II
\
~
E \
5
::J \
fg
E \
Q.
\
Calcite I \
\
\
\'J
~\%
c.J."\'"
~
,\~
\
\
\
°0~--~--2~0~0--~--~40~0~~--~6~0~0--~--~BO~0~~--~10==00

Temperature (OC)
Fig. 78-14 Equilibrium curves for reactions relevant to high-pressure
metamorphism. The phase relations of CaC0 3 are after Boettcher and
Wyllie (1967) and Crawford and Fyfe (1964). The high-pressure limits of
an!llcime and laumontite are after Newton and Kennedy (1968) and
Thomp~on 0970b), respectively. The high temperature lirqit oflawsonite
(curve A-B) is after Newton and Kennedy (1963). "
inclusions within serpentinite, and also as a constituent pf metaclastics usually in
association with quartz, both in high-pressure terranes.
Coleman and Oark (1968) have emphasized that jadeites occurring in
serpentinit~ ar~ commonly of pure jadeite composition, wJ:rile jadeites in
quartzose metamorphic rocks always contain more than 10 per cent of other
pyroxene components (diQPside, hedenbergite and aegirine). Hence, the jadeite-
forming reactions in metamorphic rocks would be of the type: sodic plagioclase
+ minor ferromagnesian components (in chlorite for instance) -+ jadeitic
pyroxene + quartz + lawsonite (Ernst, 1965). The impure composition of jadeite
232/ METAPELITES: PROGRESSIVE MINERAL CHANGES
should reduce the rock-pressure necessary for its formation. Therefore, it is
conceivable that most of jadeite-forming metamorphism took place at lower
pressures than the stability field of the pure jadeite + quartz assemblage.

7B-2l OXIDES AND SULPHIDES OF IRON

Though oxides and sulphides of iron are common in metamorphic rocks, only a
few systematic investigations have been made on them. Banno and Kanehira
(1961) and Kanehira, Banno and Nishida (1964) found that oxides of iron
(except ilmenite) are much less common than sulphides of iron in metapelites.
The frequency of occurrence of oxides and sulphides in metapelites from
high-pressure and low-pressure terranes in Japan is summarized in Table 7B-1.
There is little difference between the high and low-pressure terranes, and no
progressive changes have been noticed in any of them.

Table 7B-l Iron oxides and sulphides in metamorphic rocks

Sanbagawa belt and Omi area Ryoke belt and Abukuma Plateau
(high-pressure type) (low-pressure type)

Metapelites Metabasites Metapelites Metabasites

Haematate Not found Common Not found Not found

Magnetite Rare Rare Rare Common
Ilmenite Rare Rare Common Common
Pyrite Rare Common Rare Common
Pyrrhotite Common Rare Common Common
Chalcopyrite Widespread Widespread Widespread Widespread

Note: After Banno and Kanehira (1961) and Kanehira et al. (1964).

7B-22 ORGANIC MATERIAL AND GRAPHITE

Ordinary pelitic sediments contain a considerable amount of organic material,

which may be either dominantly coaly or dominantly bituminous, or a mixture
of both. In progressive metamorphism, such organic materials suffer a series of
decomposition reactions, ultimately leading to the generation of well-crystallized
graphite. They exert a reducing effect on associated mineral assemblages (§2-7).
The progressive increase in the crystallinity of such materials was investigated
in particular by Landis (1971). In zeolite facies rocks, such materials are
virtually amorphous, or show only a rudimentary graphitic structure. Their
crystallinity increases with rise in temperature. The ordered graphite structure is
realized in the epidote-amphibolite and the amphibolite facies.
 7B-22 ORGANIC MATERIAL AND GRAPHITE/233

Izawa (1968) separated such materials from unmetamorphosed and meta-

morphosed sediments for chemical investigation. They were shown to resemble
various ranks of coal in composition and X-ray diffraction pattern. Materials
from metapelites of the glaucophane-schist and greenschist facies, giving C =
84-94 per cent and H = 1·8-1·0 per cent in complete analysis, were similar to
semianthracite and anthracite, whereas those from epidote-amphibolite facies
rocks, giving C = 93-96 per cent and H = 0·7-0·4 per cent, were found to be
similar to anthracite in composition.
The metamorphism of coal seams has rarely been investigated in relation to
the surrounding rocks. Coal seams in zeolite facies areas are composed of either
bituminous coal or semianthracite. The rank of coal becomes higher with rise in
temperature of metamorphism of the surrounding rocks (Kisch, 1966).
Chapter 8A

Metabasites: Diversity in Progressive

Regional Metamorphism

8A-1 MINERALOGICAL SENSITIVITY OF METABASITES

Metabasites are mineralogically sensitive to variations in temperature and

pressure. This is the reason why Eskola (1920, 1939) based his definitions of
individual metamorphic facies on the mineral assemblages of metabasites. The
main characteristics are synoptically shown in table 8A-I. The mineralogical
sensitivity is especially great under high rock-pressures, as exemplified by the
formation of glaucophane schist and eclogite.
Metabasites derived from pyroclastic rocks are more easily recrystallized and
show more marked schistosity and more complex variation in composition than
those derived from lavas and minor intrusives. The former are commonly
mixtures of volcanic with some sedimentary materials, and more susceptible to
pre-metamorphic chemical changes including weathering and hydrothermal
action. Some altered rocks could produce very unusual mineral assemblages on
metamorphism (Vallance, 1967).

8A-2 LOW-PRESSURE METABASITES IN THE R YOKE BELT AND

ABU KUMA PLATEAU, JAPAN

The Ryoke metamorphic belt of Japan has abundant metapelites and meta-
psammites, but few metabasites. Recently, however, Katada (1965) has been
able to follow progressive mineral changes of metabasites in the Shiojiri area of
the belt. On the other hand, the Higo metamorphic complex and Abukuma
Plateau, the possible western and the eastern extension of the belt respectively,
have abundant metabasites. Two areas will be reviewed.

Shiojiri Area (Northern Kiso) in the Ryoke Belt

The basaltic rocks in this area are more easily recrystallized than the clastic
sediments at low temperatures. The mineral changes in this area (Katada, 1965)
have been summarized in fig. 3-4. A sketch map of metamorphic zones is shown
in fig. ?A-I.
Table 8A-I Baric types of regional metamorphism and characteristic minerals ofmetabasites ~
N
Epidote-amphibolite
Prehnite-pumpellyite and and
Baric type glaucophane-schist facies Greenschist facies amphibolite facies Granulite facies t'"
o
~
Low-pressure Actinolite, Hornblende, Orthopyroxene, .;"
:;0:1
chlorite, clinopyroxene, clinopyroxene t%l
epidote, biotite CIl
CIl
muscovite, c:::
:;0:1
biotite t%l
Intermediate- Actinolite, Hornblende, Orthopyroxene, is:
pressure chlorite, epidote, clinopyroxene ..,t%l
epidote, clinopyroxene, >
\;Il
stilpnomelane, biotite >
CIl
muscovite,
::;j
biotite t%l
CIl
High-pressure Glaucophane, Actinolite, Hornblende,
actinolite, chlorite, barroisite, ..,Z
chlorite, epidote, clinopyroxene, :I:
lawsonite, stilpnomelane, epidote, t%l
pumpellyite, muscovite muscovite, :;0:1
><:
epidote, biotite o
stilpnomelane, :;.:
t%l
muscovite \;Il
(11
t'"
..,
"N
w
VI
236/META8ASITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

In the lowest temperature zone, the original plagioclase is largely preserved.

The recrystallized minerals are quartz, chlorite, calcite, and muscovite with
s~bordinate albite, epidote and green biotite.
With a slight rise of temperature, recrystallization becomes complete, and
actinolite appears. The plagioclase becomes more calcic than 10 per cent An,
leading to the formation of the actinolite + oligoclase assemblage. Brown biotite
becomes common. This zone belongs to the greenschist or a closely related
facies. (The plagioclase is more calcic than in the typical greenschist facies.)
With a further rise in temperature, pale blue-green hornblende appears (zone
Ie in fig. 3-4). When it is in contact with actinolite, the boundary is sharp. The
associated plagioclase is oligoclase or andesine, indicating the onset of the
amphibolite facies.
Advancing metamorphism then begins to form green hornblende and andesine
(zone IIa). Actinolite disappears and calcic clinopyroxene may appear.
In this area, the zones of the greenschist and a closely related facies (zones
Ia,b) thus grade through a narrow transitional zone (Ie) into zones of the
amphibolite facies (zones IIa,b, and III). In the adjacent Shiojiri-Takato area,
however, Ono (1969b) found a few epidote amphibolites in association with
actinolite greenschists in a zone intermediate between the zones of the
greenschist and of the amphibolite facies. Since metabasites are very scarce in
the latter area, it is not clear whether it means the existence of a narrow zone of
the epidote-amphibolite facies.

Central Abukuma Plateau

The metamorphic terranes of the central Abukuma Plateau (fig. SA-I) have
abundant metabasites, as was mentioned in §7A-2. The temperatur~f meta-
morphism increases generally westward. The metabasites range from the green-
schist to the high amphibolite facies, as summarized in fig. 3-5.
The exposed metamorphic terrane has been divided into three zones A, B and
C on the basis of progressive changes in calcic amphibolite in metabasites
(Miyashiro, 1955). Zone A is characterized by actinolite, zone B by blue-green
hornblende, and zone C by green or brown hornblende (without a bluish tint).
The changes in optic constants and chemical composition of the amphiboles will
be described in §S84.
Common metabasites are as follows:
Zone A (greenschist facies). Chlorite-epidote-actinolite schist.
Zone Band C (amphibolite facies). Hornblende-plagioclase schist or amphi-
bolite.
In a zone transitional between zones A and B, actinolite coexists with
blue-green hornblende (§S84). The plagioclase in zone A is usually albite. In
 8A-2 LOW-PRESSURE METABASITES IN THE RYOKE BELT /237

O Tertiary sediments
(post·metamorphic)

:.:.: Plutonic rocks of

••••• the younger group

~::j:~ Granodioritic rocks

-:-0: of the older group

Gabbroic and dioritic

rocks of the older group

o Metamorohic rocks

Zone A

o, 5 10 km

Fig. SA-l Zones of progressive regional metamorphism in the central

Abukuma Plateau. The transitional area from an unmetamorphosed area to
zone A is covered by Tertiary sediments. Plutonic rocks of the older group
have given rise to some thermal effects in the surrounding schists
especially in the eastern part of the region, whereas those of the younger
group produced a hornfelsic contact aureole. (Miyashiro, 1955.)

zones Band C, it is oligoclase, andesine and labradorite. Almandine garnet does

not occur in ordinary metabasites.
In metabasites in the area of Barrovian zones of the Scottish Highlands,
hornblende (blue-green) begins to occur at the almandine isograd, where the
associated plagioclase is still albite. Hence, the albite + epidote + hornblende
assemblage characteristic of the epidote-amphibolite facies is produced. In the
central Abukuma Plateau, the plagioclase of metabasites at the onset of zone B
contains about 20-30 per cent An, thus forming the calcic oligoclase or andesine
238/METABASITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

(+ epidote) + hornblende assemblage, which belongs to the amphibolite facies.

We may well consider that the greenschist facies zone (A) grades directly into the
amphibolite facies zones (B and C).
Amphibolites in zones Band C sometimes contain calcic clinopyroxene.
Cummingtonite is absent in zone A, occurs rarely in metabasites of zone B, but
is very common, though minor in amount, in metabasites of zone C. The
marked increase of the frequency of cummingtonite in zone C appears to be due
to the breakdown of tschermakite component in hornblende (Shido, 1958; Shido
and Miyashiro, 1959).
The metabasites tend to show a decrease of the H20+ content and the
Fe 3 +/Fe 2+ ratio with increasing temperature. The metabasites in the low-
temperature part i.e. in the eastern part of the area are tholeiitic in composition.
The Na20 and K20 contents tend to increase westward, that is, with rising
metamorphic temperature. This may mean that Na20 and K2 0 were introduced
from some source lying to the west, causing metasomatism on a regional scale.
The metabasites in zone C usually occur in small bodies interbedded with
metapelites, whereas those in zones B and A are not so closely interbedded with
metapelites. Thus, there is a possibility that materials containing Na20 and K2 0
may have migrated from the surrounding metapelites into the metabasites.
A third possibility is that this regular variation could represent the original
compositional variation of basaltic rocks across the late Palaeozoic geosyncline
and island arc (Miyashiro, 1967b, p. 440). Quarternary volcanic rocks in the
island arc of northeast Japan show similar compositional variation toward the
west (fig. 15-3).

8A-3 LOW-PRESSURE METABASITES IN SPAIN

Fabries (1963, 1968) studied metabasites interbedded with metapelites in

northeast Seville Province, Spain. The Hercynian metamorphic complex exposed
there is in the low-pressure amphibolite facies, being divided into three zones:
Zone I. Greenish blue hornblende with an acicular habit occurs in the meta-
basites.

Zone II. Green hornblende with an acicular or prismatic habit occurs in the
metabasites. Andalusite occurs in the associated metapelites.
Zone III. Greenish brown hornblende with a prismatic or dumpy habit occurs in
the metabasites. Sillimanite occurs in the associated metapelites.

The amphiboles of zones I and II are high in tschermakite component, while.

those of zone III have higher contents of Na in vacant sites.
Bard (1969, 1970) described a progressive series of metabasites ranging from
 8A-5 METABASITES IN THE SCOTTISH HIGHLANDS/239

the greenschist to the low granulite facies in the Hercynian terrane of the
Aracena area (figs. 7A-3 and 7A-4).
The lowest temperature zone contains actinolite greenschists. With a rise in
temperature, blue-green hornblende with a high tschermakite content begins to
occur in association with actinolite. A further rise causes disappearance of
actinolite and the change of hornblende from blue-green to green colour. At this
stage, the associated plagioclase contains 25-45 per cent An. Then cumming-
tonite and clinopyroxene appear. In the highest temperature zone, cumming-
tonite and sphene disappear and orthopyroxene appears. Green-brown and
brown hornblendes and biotite are still stable (§8B-4).

8A-4 LOW-PRESSURE METABASITES IN MICHIGAN

The middle Precambrian metamorphic terrane of northern Michigan contains

basaltic volcanics, dolerites, and gabbroic dikes and sills (James, 1955). All these
rocks were subjected to metamorphism along with the surrounding sedimentary
ones. The area has been divided into the chlorite, biotite, almandine, staurolite,
and sillimanite zones in terms of mineral changes in the metapelites (fig. 7 A-9).
The chlorite zone metabasites show textural relics, but their mineralogical
reconstitution is virtually complete. The common rocks are actinolite-chlorite-
epidote-albite rocks. In the biotite zone, blue-green hornblende appears. In the
garnet zone, the common metabasites are epidote-amphibolites with blue-green
hornblende and sodic plagioclase (about 15 per cent An). Epidote, biotite and
quartz are common. Gamet is extremely rare. In the staurolite zone, epidote
disappears, and the common metabasites are amphibolites with green hornblende
and andesine. In the sillimanite zone, some amphibolites contain brown
hornblende. Generally, the metabasites of this area lack garnet.

8A-5 MEDIUM-PRESSURE METABASITES IN THE SCOTTISH HIGH-

LANDS AND NORWAY

Metabasites in the Scottish Highlands may be classified into two groups: Green
Beds and epidiorites. The former represent metamorphosed pyroclastic deposits
of broadly basaltic composition occasionally mixed with some clastic sedi-
mentary materials. Epidiorite is an old field term used by British geolOgists, in
particular in the Scottish Highlands, which represents metamorphosed doleritic
intrusions cutting through metasediments. The following description will qe
made with reference to the Barrovian zones (see § 3-1 and fig. 7A -10).

Green Beds of the Scottish Highlands

The Green Beds in the chlorite zone are usually composed of chlorite-albite-
epidote schist with or without calcite and quartz. Very small amounts of white
240/ METABASITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

mica and magnetite are present occasionally. In the middle of the chlorite zone,
green-brown biotite and actinolite begin to occur in some rocks. In the biotite
zone, biotite in Green Beds becomes brown in colour.
With rising temperature, chlorite and epidote disappear and almandine may
appear. Thus, in the almandine zone for metapelites, the rocks of the Green Beds
become hornblende-oligoclase schist with or without biotite or almandine. The
Green Beds in the kyanite and sillimanite zones are composed of coarse-grained
hornblende-andesine schists with or without almandine (Phillips, 1930; Harker,
1932).

Epidiorites of the Scottish Highlands

A detailed description has been given by Wiseman (1934). The epidiorites from
the chlorite and biotite zones as indicated by the metapelites are actinolite-
chlorite-epidote-albite felses or schists with or without biotite. Actinolite is
absent in some rocks. Quartz or calcite are present in some rocks. Some epi-
diorites preserve remnants of the porphyritic and ophitic textures of the original
rocks, but most are schistose. Actinolite occurs as slender needles commonly
intergrown with chlorite, or as big crystals pseudomorphous after igneous
augites. Rarely, original augites remain surrounded by actinolite. Biotites of a
green and a brown colour occur in small amounts in epidiorites of both chlorite
and biotite zones. White mica occurs in some epidiorites from the chlorite and
biotite zones, but not from the higher zones. Stilpnomelane occurs very rarely.
These epidiorites belong to the greenschist facies (fig. 3-2).
On entering the almandine zone of the metapelites, the epidiorites show two
marked mineral changes: the appearance of blue-green hornblende and of
almandine. Near the beginning of the zone, blue-green hornblende occurs
sometimes in parallel growth with actinolite and sometimes as a rim around an
actinolite core.
With a further rise in temperature, actinolite disappears. The blue-green
hornblendes in almandine-bearing epidiorites from the almandine zone have {3 =
1·663-1·685, whereas those in almandine-free ones from the same zone have {J =
1·639-1·675. Rocks with higher FeO/MgO ratios tend to form almandine. The
mineral may be present in one part of an outcrop of epidiorite and absent in
another. Quartz and biotite are present in some rocks and the biotites are
commonly brown but sometimes green. The plagioclase is still albite. Epidote is
widespread, but chlorite is almost absent. Thus, typical epidiorites from the
lower temperature part of the almandine zone are blue-green hornblende-
epidote-albite schists willi or without almandine. In ordinary nomenclature,
they are called epidote amphibolites (fig. 3-2).
With a little rise in temperature, however, epidote reacts with albite to
produce more calcic plagioclase. Thus, the highest temperature part of the
almanine zone contains epidiorites with andesine. Plagioclases may show zoning
 8A-6 METABASITES IN THE NORTHERN APPALACHIANS /241
with increasing An content towards the rim in some cases and with decreasing
An in others.
In the kyanite and sillimanite zones, the epidiorites are usually mainly
composed of green or brown hornblende and plagioclase (usually andesine).
Brown hornblende appears to occur in the highest temperature part of the
sillimanite zone. Typical epidiorites in the kyanite and sillimanite zones are thus
coarse-grained amphibolites with or without almandine or clinopyroxene. In the
vicinity of syn-metamorphic granites, metasomatic effects on the epidiorites are
noticed.
The K2 0 content of the epidiorites ranges from 'trace' to 1·2 per cent
whereas that of the Green Beds ranges from 0·6 to 2·0 per cent. The higher
contents of the latter may be due to the clay admixture in the original materials.
Alternatively, some glass which was contained in original pyroclastics, may have
absorbed K2 0 from sea water during palagonitization, as observed on the
present-day ocean floors.
Sulitjelma Area, Norway
The Sulitjelma area on the boundary between Norway and Sweden was made a
classical metamorphic terrane by Vogt's (1927) study, though a part of his
geologic interpretation was in error (Mason, 1967). The metapelites range from
the chlorite zone to the staurolite-kyanite zone (cf. Henley, 1970). The
associated metabasites range from the amphibolite through the epidote-
amphibolite to the greenschist facies (Mason, 1967). In the lowest temperature
part of the greenschists, the CaO tends to be lost, resulting in the decomposition
of actinolite and then of epidote to produce chlorite.

8A-6 MEDIUM-PRESSURE METABASITES IN THE NORTHERN

APPALACHIANS

Thompson and Norton (1968) give the correlation of mineral zones in the
metapelites with the mineral assemblages of interbedded metabasites in the
northern Appalachians, as follows:
Metapelites Metabasites
Biotite zone Greenschist facies
Almandine zone Epidote-amphibolite facies
Staurolite zone
Sillimanite zone } Amphibolite facies
Sillimanite-K feldspar zone Low granulite facies.

Woodsville Quadrangle, Vermont and New Hampshire

Billings and White (1950) have described the progressive metamorphism of basic
dikes that were intruded into a Palaeozoic geosynclinal pile in the Woodsville
242/METABASITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

quadrangle, Vermont and New Hampshire (fig. 7A-6). The metabasites range
from the greenschist to the amphibolite facies. The surrounding metapelites
range from the chlorite to the sillimanite zone. The most interesting point of this
study is the discovery of a great variety of metamorphic assemblages, which
formed in response to variable Peo ., in the chlorite zone metabasites.
In the chlorite zone, some rocks in dikes preserve their original minerals and
textures, but most have been completely recrystallized and become schistose.
The completely recrystallized metabasites show the following four mineral
assemblages (mineral names in decreasing order):

1. Actinolite+albite+epidote+chlorite+quartz+calcite,
2. Chlorite+albite+epidote+calcite+quartz,
3. Chlorite+albite+calcite+quartz+epidote,
4. Albi te+chlorite+ankerite+quartz+epidote.

(Ankerite is a dolomite-structure mineral in which Mg is partly replaced by Fe

and it shows porphyroblastic development.)
This sequence (1) to (4) indicates that actinolite was decomposed first and
then epidote, with resultant formation first of chlorite, calcite and quartz, and
then of ankerite. Probably this change took place with increasing p co • under
essentially the same temperature and rock-pressure. Ankerite becomes stable at a
higher p co • than calcite. In some large dikes, the carbonate content of the
rocks increases toward the margin. Rocks with higher carbonate contents tend to
show clear schistosity.
It is noted that schistose albite-epidote amphibolites oq:ur in a higher
temperature part of the chlorite zone and in the biotite zone (not in the
almandine zone). Massive or weakly schistose amphibolites occur in the higher
temperature part of the almandine zone and in the staurolite and sillimanite
zones. The plagioclase is andesine or labradorite. Almandine is virtually absent in
the metabasites of this area.

8A-7 METABASITES IN PASSAGE FROM HIE AMPHIBOLITE TO THE

GRANULITE FACIES

The progressive changes of metaba~ites from the amphibolite to the granulite

facies in the Broken Hill area, Australia, and the northwest Adirondack
Mountains in New York State, are well documented. In both cases, it is not clear
whether they belong to the medium-pressure series or to the low-pressure, since
no diagnostic minerals occur. The associated metasediments contain sillimanite
but no andalusHe nor kyanite. The metabasites of the two areas are similar to
each other except that garnet and cummingtonite occur in the former but not in
the latter.
 8A-7 METABASITES IN THE GRANULITE FACIES/243

Broken Hill, Australia

The Broken Hill area is near the western border of New South Wales. The older
Precambrian Willyama complex has been metamorphosed under conditions of
the amphibolite to the granulite facies (Binns, 1964, 1965b).
Three zones have been distinguished in terms of regular mineral changes in
metabasites as follows:

Zone A The metabasites are amphibolites with blue-green or green hornblende

(low amphibolite facies). Gamet and epidote are present in some metabasites.
The associated metapelites are sillimanite-bearing muscovite-biotite schists often
containing garnet. The K-feldspar is orthoclase and is confmed to sillimanite-free
metapelites.

Zone B. The metabasites are amphibolites with brown or green-brown horn-

blende (high amphibolite facies). The colour change of the hornblende is
ascribed to an increasing content of Ti02 • Almandine, cummingtonite and
clinopyroxene occur in some metabasites. The associated metapelites are
orthoclase-bearing silimanite gneisses, where muscovite is no longer stable. Gamet
and cordierite occur in some metapelites.

Zone C. The metabasites are hornblende-orthopyroxene-clinopyroxene

gneisses (low granulite facies). The associated metapelites are nearly the same as
in the preceding zone.
Zones A, Band C are at least 11 km wide, 13 km at its widest, and more than
18 km wide respectively. There are not contemporaneous batholithic intrusions
exposed in this region. Quartzo-feldspathic gneisses occur in all the zones. Some
of them could have been derived from sedimentary rocks and others from
igneous. The K-feldspar in quartzo-feldspathic gneisses is microcline in zone A
and orthoclase in zones Band C.
The limestones in zone A contain wollastonite and diopside.

Northwest Adirondacks, New York State

Progressive metamorphism of Precambrian sedimentary and basic rocks has been
followed along a belt 55 km long that extends across the Grenville Lowlands
into the central massif of the Adirondack Mountains (Engel and Engel, 1958,
1960, 1962a,b; Engel, Engel and Havens, 1964).
The metabasites at the low-temperature end of the belt are amphibolites with
the composition of saturated basalts, essentially composed of hornblende,
andesine and quartz accompanied by minor biotite and sphene. With increasing
temperature, the amounts of hornblende and quartz decrease, whereas that of
plagioclase increases and sphene disappears. Qinopyroxene appears, and then
with a further increase in temperature, orthopyroxene also appears. Thus, near
244/METABASITES: DIVERSITY IN PROGRESSIVE METAMORPHISM

the high-temperature end, the metabasites are hornblende-pyroxene granulites

with olivine basaltic or noritic composition.
The chemical composition of the metabasites changes with increasing tem-
perature: The contents of Si02 , K20, Fe203 and H20 decrease, whereas those
of Cao and MgO increase. The associated metasediments show a similar trend of
compositional change, including a decreasing Si02 content with increasing
temperature. The compositional change in the metasediments has been ascribed
to partial melting and removal of the resultant melt, whereas that of the
metabasites has been ascribed to ionic diffusion or migration of an aqueous
fluid, since the temperature is not considered high enough for partial melting.
The hornblendes are bluish-green in the low-temperature part, and become
brownish with rising temperature. Compositionally, their contents of Ti0 2,
Na20 and K20 increase and those ofMnO and (OH- + F + CI) decrease.
In the Broken Hill area, wollastonite was found in calcareous rocks of the
lower amphibolite facies zone. In the Adirondacks, on the other hand, the
calcite + quartz assemblage is stable to the highest temperature zone.

8A-8 HIGH-PRESSURE METABASITES IN THE SANBAGAWA BELT,

JAPAN

Metabasites are abundant in the Sanbagawa metamorphic belt of Japan. In the

lowest temperature part of the belt, the basaltic volcanics preserve their original
structure. They are mostly pyroclastics mixed with some lava flows. The rocks in
higher temperature zones are well-recrystallized schists. Two well-documented
areas will be reviewed below.

Bessi-Ino Area
The metamorphic terrane of this area has been divided into five progressive
zones A, B, C, D and E (figs. 3-6 and 7A-12; §7A-9). The first two zones belong
to the glaucophane-schist facies, and the last two to the epidote-amphibolite
facies (Banno, 1964).

Zone A. The typical metabasites of this zone are pumpellyite-epidote-chlorite-

albite-quartz schists and glaucophane-epidote-chlorite-albite-quartz schists.
The mafic rocks in the low-temperature part of this zone are poorly recrystall-
ized and are usually non-schistose, whereas those in the high-temperature part
are considerably recrystallized and schistose. Pumpellyite-bearing metabasites are
lower in Fe203 than pumpellyite-free ones. Jadeite was found in two rocks and
lawsonite only in one. Prehnite, paragonite and laumontite were found only very
rarely.
 8A-8 METABASITES IN THE SANBAGAWA BELT, JAPAN /245
Zone B. The typical metabasites are chlorite-epidote-actinolite-albite-quartz
schists with or without glaucophane. Glaucophane with b = Z (Le. crossite) is
much more common than that with b = Y. Glaucophane-bearing metabasites
tend to be higher in Fe203 content than glaucophane-free ones. Correspond-
ingly, the epidotes in glaucophane-bearing metabasites tend to have a higher
Fe2 0 3/ Al2 0 3 ratio than those in glaucophane-free ones.

Zone C. This zone is transitional between the zones of the glaucophane-schist

facies and the epidote-amphibolite facies. The most important feature of the
metabasites is the disappearance of actinolite and appearance of barroisite.
Glaucophane still occurs in some metabasites in coexistence with barroisite.
Under the microscope the barroisite looks like blue-green hornblende. In this
and succeeding zones, albite commonly shows porphyroblastic development.

Zone D. The typical rocks of this zone are blue-green hornblende-epidote-

chlorite-albite-quartz schists with or without garnet. Not only blue-green
hornblende but barroisite occurs, though the latter is rare. Some metabasites
contain biotite.

Zone E. The typical rocks are epidote amphibolites with almandine or calcic
clinopyroxene. Biotite, muscovite and quartz may occur. The hornblende is
blue-green. Kyanite, omphacite, zoisite and paragonite have been found
occasionally. Kyanite was found in a zoisite-kyanite-paragonite-hornblende-
quartz rock.
Hematite occurs in the metabasites of all the zones, whereas ilmenite occurs
only in those of zones D and E. Magnetite is very rare.

Kanto Mountains.
A synoptic description of this area has already been given in fig. 3-7 and § 7 A-9
(Seki, 1958, 1960, 1961a, 1965).
The recrystallization of basaltic rocks begins with the incipient formation of
chlorite, albite and stilpnomelane (zone I in fig. 3-7). Then, actinolite, glauco-
phane, and pumpellyite begin to form (zone II). With a slight rise of tempera-
ture, epidote enters (zone III). At this stage, recrystallization is still incomplete,
and foliation is very weak.
With a further rise of temperature, lawsonite and jadeitic pyroxene begin to
occur (zone IV). These two minerals are confined to a narrow zone. Some
metabasites in this zone show the coexistence of lawsonite, pumpellyite and
epidote. With an additional increase of temperature, lawsonite and jadeitiC
pyroxene disappear (zone V). At this stage, recrystallization becomes fairly
complete and the metabasites are usually schistose.
246/METABASITES: DIVERSITY IN PROGRESSIVE METAMORPHISM
Glaucophane in a broad sense occurs in some of the metabasites ranging from
nearly the stage of incipient recrystallization up to this zone. Glaucophane
rimmed by actinolite and actinolite rimmed by glaucophane are common.
Glaucophane with b = Z occurs in all the above-mentioned range, whereas
glaucophane with b = Y is nearly confmed to the lawsonite-jadeite zone. The
latter is probably closer to the Mg-Al end member of glaucophane.
With an additional rise in temperature, glaucophane and pumpellyite dis-
appear, and thus the metabasites come to have a mineral assemblage character-
istic of greenschist (zone VI): chlorite + actinolite + epidote + albite. Some of
the albites show porphyroblastic development. Some 'actinolites' of this zone
may tum out to be barroisite.

8A-9 HIGH-PRESSURE METABASITES OF THE CALIFORNIA COAST

RANGES
A great variety of metamorphosed pillow lavas and tuff breccias of broadly
basaltic compositions are associated with metaclastics in the Franciscan group
of the Coast Ranges of California. Two areas will be reviewed below.

Cazadero Area
Coleman and his co-workers studied the metabasites in the Cazadero area, about
100 km NW of San Francisco (Coleman and Lee, 1962, 1963; Lee, Coleman and
Erd, 1963; Coleman et aI, 1965; Coleman and Clark 1968; Coleman and Papike,
1968).
The poorly recrystallized metabasalts (type II in Coleman and Lee's nomen-
clature as commented upon in §7A-IO) are non-schistose and resemble unmeta-
morphosed basalts in field appearance. Pillow structure is well pr~served.
Microscopic observation has revealed that features of the original subophitic and
variolitic textures are preserved but mineralogical recrystallization is well
advanced. Glaucophane, lawsonite, muscovite (lM), chlorite, omphacite, sphene
and quartz are common constituents. Rarely, relict calcic plagioclase and
clinopyroxene are present. Albite and paragonite have not been found.
With advancing recrystallization, the metabasalts become schistose (Coleman
and Lee's type III). The pillow structure is conSiderably destroyed. Glaucophane
is abundant. Muscovite (2Md, chlorite, lawsonite, pumpellyite, garnet,
omphacite, aragonite, sphene and quartz are common. Pyrite is ubiquitous,
whereas magnetite, hematite and ilmenite are absent.
Coarsely crystalline metabasalts (type IV) occur as isolated blocks in this
area. In such rocks, epidote and rutile occur, whereas lawsonite and sphene
dwindle.
The stable form of CaC0 3 is aragonite in well-recrystallized glaucophane
schists (type III), whereas it is partly aragonite and partly calcite in poorly
 8A-9 METABASITES OF THE CALIFORNIA COAST RANGES/247

recrystallized metabasites (type II). Garnet does not occur in poorly recrystal-
lized metabasalts, but does occur in well-recrystallized metabasalts of type III in
which it contains spessartine, almandine and grossular components, and in
coarsely crystalline metabasalts of type IV in which it contains almandine and
grossular components. The pyroxenes of the metabasalts are aegirine-jadeite in
poorly recrystallized rocks (type II), and are omphacite with only a small
proportion of the acmite component in well-recrystallized rocks (type III).

Panache Pass Area

Synoptic descriptions of this area (Ernst, 1965) were given in figs 3-8. and 7A-13
and §7A-IO.
The metabasites in the poorly recrystallized part of the metamorphic terrane
(Coleman and Lee's type II) sometimes preserve relict ophitic texture and relict
plagioclase and augite. The plagioclase is heavily altered and ranges from
labradorite (relict) to albite. Minute lawsonite tablets are present in many of the
plagioclases. The augites are peripherally altered to aegirine or to an intergrowth
of actinolite and chlorite. In some rocks, fine-grained plagioclase, stilpnomelane,
chlorite and actinolite consitute an almost irresolvable mesostasis. Pumpellyite is
present in most specimens, generally as irregular stringers and patches. Actinolite
needles commonly exhibit a feathery periphery of crossite. Vesicles are filled
with pumpellyite, chlorite, and rarely calcite and quartz, and/or albite. These
rocks resemble greenschists insofar as chlorite, actinolite and albite are main
products of recrystallization, but differ from them in containing lawsonite and
pumpellyite instead of epidote.
On the other hand, metabasites in the completely recrystallized part of the
metamorphic terrane (type III) are crossite schists mainly composed of crossite,
lawsonite, and chlorite together with minor quartz, white mica, stilpnomelane,
sphene and aragonite (partly altered into calcite). Actinolite is lacking, and albite
has been found very rarely. A minor amount of aegirine is common. These rocks
are in the typical glaucophane-schist facies.
Epidote-albite amphibolitic rocks, garnet-albite amphibolitic rocks and
eclogites have been found in this area as well-recrystallized, coarse-grained rocks
(Coleman and Lee's type IV). The amphibolitic rocks contain blue-green
hornblende commonly rimmed by crossite.
Chapter 8B

Metabasites: Progressive Mineral

Changes

8B-1 CHLORITE

The chlorites of metabasites show a limited range of the Si(Mg,Fe 2 l ~ AlAl

substitution, just like those of metapelites (§ 7B-6). The number of Si atoms on
the basis of 18(0,OH) is in the range 2·5-3,0. Hence, the same formula
(Mg,Fe)4'5Al3Si2.~010(OH)8 may be used for both metabasite and metapelite
chlorites. The Fe +J(Mg + Fe 2 l ratio of metabasite chlorites, however, is
usually in the range 0'1-0'5, tending to be lower than that of metapelite
chlorites.

8B-2 EPIDOTE AND ZOISITE

Epidote and zoisite have the composition HCa2(Al,Fe3+hSi3013' The Fe3+J

(Al + Fe 3+) ratio is 0'0-0,2 (mostly 0'0-0'1) in zoisite (orthorhombic), and
0·0-0·4 (mostly 0,1-0'4) in epidote (monoclinic). Thus, epidote is usually
higher in iron content than zoisite, though their composition ranges overlap. In
the overlapping range, they are in a polymorphic relation. The stability relation
between them, however, is not clear.
Epidote, including both clinozoisite and pistacite, is widespread in low-
temperature metabasites. In many high-pressure terranes, it occurs in association
with prehnite, pumpellyite or lawsonite, but persists to a higher temperature
than any of them. In many medium- and low-pressure terranes such as the
Scottish Highlands and the Ryoke belt, epidote is the predominant calcium-
aluminium silicate in low temperature zones.
Most iron in epidote and zoisite is in the trivalent state. The stability of these
minerals depends not only on temperature and rock-pressure but also on PH 2 0
and P02 ' With rising temperature, these minerals usually dwindle and finally
disappear. This change involves dehydration and reduction. The liberated iron
will mostly be reduced into the divalent state to enter other silicates. The
liberated CaO and Al 2 0 3 mostly enter plagioclase to make anorthite component.
The high-temperature stability limit of zoisite + quartz assemblage was
 88-3 PLAGIOCLASE /249
experimentally determined by Newton (1966a). It is about 670°C at 6 kbar in
the presence of an aqueous fluid.
In progressive metamorphic terranes, zoned epidotes with decreasing iron
content towards the rim are widespread. Probably this means that the ferric iron
in epidote was partly reduced and transferred to other minerals with rising
temperature. In relatively rare areas, a zonal structure with the reverse trend
has been observed, suggesting retrogressive adjustment (Miyashiro and Seki,
1958).
In various parts of the Sanbagawa glaucophanitic belt, progressive enlarge-
ment of the composition field of epidote has been found (Miyashiro and Seki,
1958; Banno, 1964, p. 249). In the lowest temperature zones, the epidotes have
a narrow range of composition near Fe3+/(Al + Fe3~ = 0'3. With rising
temperature, the composition range is expanded in both directions, i.e. towards
lower and higher Fe 3 +/(Al + Fe3~, though enlargement towards lower Fe3+/(Al
+ Fe3~ is more marked. In the contact aureole of the Arisu area, northern
Japan, the composition range is expanded but also shifts toward a lower
Fe 3 +/(AI + Fe 3 ~ ratio with rising temperature (Seki, 1961b).

8B-3 PLAGIOCLASE

In the prehnite-pumpellyite, greenschist and epidote-amphibolite facies, preh-

nite, pumpellyite and epidote occur instead of the anorthite component in
plagioclase, which is therefore albite. Intermediate and calcic plagioclase can
form in the amphibolite and higher facies. The anorthite component usually
forms at the expense of epidote. Gradual increase of the An content of
plagioclase in metabasites with rising temperature has been clearly shown, for
example, in the central Alps (Wenk and Keller, 1969) and the Tanzawa
Mountains (Seki et al. 1969a). In the former area, however, there exists a
compositional gap in the range of calcic albite and sodic oligoclase. For a
discussion of this, refer to §7B-8.
The reaction of sodic plagioclase with epidote should be influenced by T, p s•
PH, O. Po, and by the minerals occurring in association with the albite and
epidote. The rate of compositional variation of plagioclase with rising tempera-
ture should differ in different metamorphic terranes. Petrographic experience,
however, has indicated that the rate of compositional variation is usually gentle
up to 30 per cent An, above which it becomes very fast. This is the reason why a
plagioclase composition of 30 per cent An in equilibrium with epidote is taken
as the boundary between the epidote-amphibolite and amphibolite facies
(Ramberg, 1952).
At present, a reliable scale relating temperature to the compositional change
of plagioclase is not available. However, the change of the stable calcic
amphibole in metabasites from actinolite to hornblende will be tentatively used
250/METABASITES: PROGRESSIVE MINERAL CHANGES

here as a metamorphic grade reference against which plagioclase changes will be

compared. At this grade, the composition of metabasite plagioclase is probably
about 5 per cent An in the high-pressure terrane of the Sanbagawa belt and the
Barrovian region of the Scottish Highlands, but is 20-30 per cent An in the
low-pressure regional metamorphic terranes in Japan, as summarized in table
88-1. While the reference point defined above does not represent a definite
temperature, the relations described suggest the possibility that the formation of
a more calcic plagioclase is promoted by lower rock-pressure.
Table 8B -1 Composition of plagioclase associated with epidote or of most
calcic plagioclase if epidote is not present at a metamorphic grade where
predominant calcic amphibole changes from actinolite to hornblende in
metabasites.
Baric type Area and author Plagioclase (% An)
High-pressure Bessi (S. Banno, pers. comm. 1969) 5
(atypical)
Medium-pressure Northern Appalachians (Billings and Albite
White, 1950)
Scottish Highlands (Wiseman, 1934) 5
Low-pressure Aracena (Bard, 1969) 20
regional Ryoke (Katada, 1965) 30
Central Abukuma (Miyashiro, 1958) 20-30
Low-pressure, Tanzawa (Seki et al. 1969a) 20
small scale
Low-pressure Iritono (Shido, 1958) Labradorite
contact Arisu (Seki, 1961b) Labradorite

Kohler (1941, 1949) has pointed out that the optical properties of plagioclase
vary with its thermal history. Plagioclase in volcanic rocks has what are called
high-temperature optics, whereas that in plutoniC and metamorphic rocks
usually has the different optic properties characteristic of low temperatures. The
difference in optic angle between the two forms is great in relatively sodic
plagioclase (Smith, 1958). There are continuous intermediate states between the
high- and low-temperature optics. The effect of thermal history is also evident in
the X-ray diffraction pattern (Smith and Yoder, 1956; Engel et al. 1964).
Though most metamorphic plagioclases show low-temperature optics, the
granulite-facies metamorphic rocks of the Broken Hill area, Australia, were
found to contain plagioclases showing intermediate optic and X-ray properties
(Binns, 1965b).
Albites in low-temperature metamorphic rocks are mostly water-clear, un-
twinned and unzoned. With rising temperature, the frequency of twinning tends
to increase (e.g. Iwasaki, 1963; Engel and Engel, 1962a). Gorai (1951) has
pointed out that twinning on the albite and pericline laws occurs in both igneous
 8B-4 CALCIC AND SUBCALCIC AMPHIBOLES /251
and metamorphic rocks, whereas that on the Carlsbad and albite-Carlsbad laws
is confined to igneous rocks. However, some observations at variance with this
have been recorded (e.g. Binns, 1965b).
It is generally said that Carlsbad twinning forms at the time of crystal growth,
whereas albite and pericline twinning can also form secondarily, for example, by
deformation (Smith, 1962; Brown, 1962).

8B-4 CALCIC AND SUBCALCIC AMPHIBOLES

The general formula for the calcic and subcalcic amphiboles is:

where Ca = 1·5 - 2·0 in calcic amphiboles, and

Ca = 1·0 - 1·5 in sub calcic amphiboles,
for 24 (0, OH) or 230 on anhydrous basis.

Comprehensive reviews on the chemical composition of these amphiboles

have been given by Hallimond (1943) and Leake (I968). A general review of
their crystal chemistry, phase relations and occurrence has been given by Ernst
(1968).

Calcic Amphiboles
To a first approximation, the composition of calcic amphiboles may be regarded
as being controlled by the following three variables:
1. Degree of the Tschermak substitution: (Mg, Fe 2 +:>Si": AlAl,
2. Total of Ca + Na + K. Since the calcic amphibole structure has large vacant
sites that can accommodate Na + and K + ions, the total number of Ca 2 + +
Na + + K + ions is variable,
3. Degree of the Mg": Fe 2 + substitution.
The first two variables are shown in fig. 8B-1. Each point on the diagram still
represents amphiboles whose composition may vary because of the Mg": Fe 2 +
substitution.
Calcic amphiboles may be divided into two groups: the actinolite group and
the hornblendes. The actinolite group comprises tremolites, actinolites and
ferrotremolites (= ferroactinolites). The rest of the calcic amphiboles are here
collectively called hornblendes. Fig. 8B-2 shows the curve for the frequency of
occurrence of calcic and sub calcic amphiboles. There is a clear minimum in the
curve at about Si = 7·2 for 24(0 + OH). Hence, the Si value of 7·2 is here
regarded as the boundary between the actinolite group and the hornblendes.
The significance of this minimum is in dispute. It may be the result of a
miscibility gap between the two groups of amphiboles under a limited range of
252/METABASITES: PROGRESSIVE MINERAL CHANGES

_ Ca+Na+K

2·0 2·5 3·0

S'OK
Tremolite
I •
Actinolite Richterite
(Ferrotremolite) Na2Ca(Mg,Fe)5 SiaOn(OH)2
Ca2(Mg,Fe)5 SiaOn(OHh

Si

t
Actinolite group

7·0
Edenite
(Ferro-edenite)
NaCa2(Mg,Fe)5 Si7 AIOn(OHb
Hornblendes

6·0~-------------------------------------------4
Tschermakite Pargasite
( Ferro-tschermakite) (H'lstingsite)
Ca2(Mg,FeIaA~Si6AI20n(OH)2 NaC~(Mg, Fe4AISisA~0n(OH)2

Fig. 88-1 Calcic amphiboles. The compositions are plotted in the

trapezoid tremolite-tschermakite-pargasite-edenite. These names represent
Mg-end members. The names of Fe-end members are given in brackets
below the corresponding Mg-end members.

metamorphic conditions (Shldo, 1958; Shido and Miyashiro, 1959; Klein, 1969;
Cooper and Lovering, 1970). The possible gap would not result in zero
frequency for two reasons: (1) The gap may exist only under a limited range of
conditions, especially at relatively low temperatures, and (2) there are a number
of compositional variables, and the frequency diagram for Si does not reveal a
gap within a two- or poly-dimensional composition space.

Modes of Occurrence. In metabasites, actinolite does not form in the zeolite

facies and the lower part of the prehnite-pumpellyite facies, but does form in the
higher part of the latter facies and in the greenschist facies.
 8B-4 CALCIC AND SUBCALCIC AMPHIBOLES /253

<I> Hornblendes
~
(5
<:
'"
'0
....
~
E 100
:l
Z

Ca on the basis of 24 (0, OH, F) Si on the basis of 24 (0, OH, F)

Fig. 8B-2 Frequency distribution of 936 analyses of calcic and sub-calcic

amphiboles. (After Leake, 1968, with a correction in the Si diagram based
on his personal communication, 1970.)
In greenschist facies terranes, the temperature of formation of actinolite
would increase if PH, 0 and Pco , were high, thus delaying the formation of
actinolite in progressive metamorphism. This may be illustrated schematically by
the following equation:
5(Mg, FehSi 20 s (OH)4 + 6 CaC0 3 + 14 Si0 2
antigorite molecule calcite quartz
of chlorite
= 3 Ca2(Mg, Fe)sSi s 0 22 (OH)2 + 6 CO 2 + 7 H20. (8B-l )
actinolite
Actinolite-free greenschists occur mixed with actinolite-bearing ones in some
metamorphic terranes. In the equation given below, the left-hand side would be
stabilized under a very high Peo ,. This was observed in assemblage (4) of the
Woodsville quadrangle given in §8A-6.
5 Ca(Mg, Fe)(C0 3 h + 8 Si0 2 + H20
ankerite quartz

(8 B-2)
actinolite calcite

In the epidote-amphibolite and amphibolite facies, the calcic amphiboles in

ordinary metabasites are hornblendes. It has not been settled whether the
254/METABASITES: PROGRESSIVE MINERAL CHANGES

compositional change from actinolite to hornblende in progressive metamorph-

ism of metabasites is continuous or not. The frequency minimum of calcic
amphiboles at about Si = 7·2 may be regarded as a manifestation of the possible
compositional gap under such metamorphic conditions.
Shido (1958) and Shido and Miyashiro (1959) have considered that the
compositional change from actinolite to hornblende in the central Abukuma
Plateau and probably also in the Scottish Highlands is discontinuous. There is a
transitional zone where actinolite coexists with blue-green hornblende. Horn-
blende occurs sometimes as a rim embracing an actinolite crystal, sometimes in
parallel growth with actinolite, and sometimes as discrete crystals. When
actinolite is in direct contact with hornblende, the boundary between them is
sharp. When the boundary looks to be gradational under the ordinary micro-
scope, examination of the thin section tilted in an appropriate angle on the
universal stage has always revealed that the boundary is actually sharp.
Klein (1969) and Cooper and Lovering (1970) have made electron micro-
probe analysis of actinolites and hornblendes which coexist in such rocks.
Homogeneous actinolite and hornblende were found to be in contact with each
other with a sharp boundary. Exsolution lamellae of hornblende in actinolite
and of actinolite in hornblende were found. These give a strong support for the
existence of a miscibility gap at the relevant low temperatures. At higher
temperatures, the gap may vanish.
Hornblendes in some metabasites not uncommonly have exsolution lamellae
of cummingtonite. The lamellae are parallel to (DOl), (100) and (Iol) (Binns,
1965b; Jaffe, Robinson and Klein, 1968).

Colour Change. The colour of hornblende for Z shows a relatively regular

change. In metabasites recrystallized at low temperatures (usually epidote-
amphibolite facies or low amphibolite facies), the colour for Z is usually
blue-green (greenish-blue or bluish-green). At higher temperatures (usually
amphibolite facies), the colour for Z is green (without a bluish tint) and at still
higher temperature it is brown. There can be complete gradation between these
colours, and in some metamorphic terranes green and brown hornblendes are
known to occur in rocks of different compositions within the same area.
However, the general scheme of colour change for Z of blue-green ~ green ~
brown with rising temperature is widespread.
For this reason, actinolite and hornblendes with variable colours have been
used as index minerals for zonal mapping of progressive metamorphic terrane in
the Abukuma Plateau (Miyashiro, 1958; Shido, 1958), Aracena area (Bard,
1969) and Broken Hill (Binns, 1965b). Similar colour change was observed in
the Ryoke belt (Katada, 1965), Seville province, Spain (Fabries, 1968), Scottish
Highlands (Wiseman, 1934), northern Michigan (James, 1955), northwest
Adirondack, New York (Engel and Engel, 1962a, b), northern Appalachians
 88-4 CALCIC AND SU8CALCIC AMPHI80LES/255

(Billings and White, 1950) and east Kimberley area, Australia (Gemuts, 1965).
Blue-green hornblende tends to have a higher H2 0 content and a higher
Fe 3 +/Fe 2 + ratio than other hornblendes (Seitsaari, 1953). The brown colour has
been ascribed to higher contents of Ti0 2 (Binns, 1965b).
An approximate relationship between colour and optical constants is shown
in fig. 88-3.
100

90

80

~~ 70
\ \
\ \
60
\ \
\ B \
\ \
"- \
50 "- I
"- I
"- .........
---
I
.............. /
40L-____ ~ ______ ~ ____- L____ ~~ ____ ~ ____ ~

1·64 1·65 1·66 1·67 1·68 1·69 1·70

'Y
Fig. 88-3 Optical constants of actinolites (A), Blue-green and green
hornblendes (B) and greenish brown and brown hornblendes (C). The
general trends of optical changes of calcic amphiboles with increasing
temperature are shown by arrows. These trends are similar in high- and
low-pressure metamorphic facies series.

Compositional Change. Since actinolite is poor and hornblende is rich in Al 2 0 3 ,

some authors have suggested that the total Al 2 0 3 content or the tetrahedral Al
content of such amphiboles increases with increasing temperature (e.g. Harry,
1950). However, the change is probably discontinuous as discussed above, and
once hornblende becomes the predominant calcic amphibole, the Al 2 0 3 and
tetrahedral Al contents do not show a regular increase.
256/METABASITES: PROGRESSIVE MINERAL CHANGES

The compositional changes of calcic amphiboles with rising temperature in

the central Abukuma Plateau (Shido, 1958; Shido and Miyashiro, 1959) and
Broken Hill (Binns, 1965b, 1969) are shown as the fields A -+ B -+ C in fig. 8B-4.
The blue-green hornblende is rich in the tschermakite component. With a rise in
temperature, the component tends to react with quartz as follows:

7 tschermakite + 10 quartz
=3 cummingtonite + 14 anorthite + H2 0. (8B-3)

This reaction tends to decrease the tetrahedral and octahedral AI. At the same

- Ca+Na+K

~ M ~
8·01~-"----.-...,.---'--r--.---'---.--.--='

A
Si

t
7·0
Edenite

6·0~---------------------~
Tschermakite Pargasite
Fig. 8B-4 Calcic amphiboles in metabasites of the central Abukuma
Plateau (Shido and Miyashiro, 1959) and Broken Hill (Binns, 1965b). A =
actinolite field, B = blue-green hornblende field, C = field of green and
brown hornblendes in the high amphibolite and granulite facies. Compare
with fig. 8B-1.
 8B-4 CALCIC AND SUBCALCIC AMPHIBOLES/257

time, the Na in vacant sites tends to increase by the replacement Si ~ NaAl,

resulting in a change toward the edenite or pargasite-hastingsite composition.
The two reactions take place in the gradual transition from blue-green to green
and brown hornblendes, and during a further increase in temperature into the
granulite facies. Thus, the net results are a progressive decrease of octahedral AI
and an increase of Na in vacant sites. It should be noted that the trend of
compositional change differs between the transformation of actinolite -+ blue-
green hornblende and that of blue-green hornblende -+ green and brown
hornblende.
In the Haast Schist Group (Cooper and Lovering, 1970), amphibolite facies
hornblendes show only slightly higher contents of (Ca + Na + K) than
epidote-amphibolite facies ones.
In the Seville province (Fabries, 1963, 1968) and Aracena area (Bard, 1969,
1970), a somewhat different trend of change has been noticed. In the
transformation of blue-green to green and brown hornblendes, the Na in vacant
sites increases. In fig. 8B-4, the field of such hornblendes moves simply to the
right in the zone of Si = 6·2-7·0.
Leake (1965, 1968) and Binns (1969) have emphasized that the Ti0 2 content
tends to increase with metamorphic temperature.
The calcic amphiboles of metabasites in the Sanbagawa high-pressure terrane
were found to have slightly higher Si0 2 + Al 20 3 and Na20 contents and lower
Ti0 2 and CaO contents than those showing a very similar mineral assemblage in
medium- and low-pressure terranes (Ernst, 1972b).

Sub calcic amphiboles (Barroisites)

Na-rich subcalcic hornblendes, called barroisite, are rather common in high-
pressure metamorphic terranes (Iwasaki, 1963; Banno, 1964) and in eclogites
(Banno, 1964; Coleman et al. 1965; Binns, 1967).
The status of barroisite in progressive glaucophanitic metamorphism has been
studied by Banno (1964) in the Bessi-Ino area. The amphiboles occurring in
metabasites in three progressive stages are as follows:

Glaucophane-schist facies zone glaucophane, actinolite,

Transitional zone glaucophane, barroisite,
Epidote-amphibolite facies zone barroisite, blue-green hornblende.

Glaucophane can coexist with actinolite or barroisite in the same rock. The
compositional change from actinolite to barroisite appears to be continuous, and
actinolite does not coexist with barroisite. The compositional variation from
barroisite to hornblende also appears to be continuous.
258/META8ASITES: PROGRESSIVE MINERAL CHANGES

88-5 CUMMINGTONITE, ANTHOPHYLLITE AND GEDRITE

The chemical compositions of cummingtonite (monoclinic) and anthophyllite
and gedrite (both orthorhombic) can be approximately represented by the
formula:

Here, cummingtonite (including grunerite) has a low Al 20 3 content and

Fe2+/(Mg + Fe2 ~ = 0·3-1·0; anthophyllite has a low Al 20 3 content and
Fe 2+/(Mg + Fe2~ = 0·0-0·5; gedrite has a high Al 20 3 content and Fe2+/(Mg +
Fe2~ = 0·4-1·0.
It appears that the composition fields of these three amphiboles are separated
from one another by miscibility gaps (Stout, 1971, Robinson, Ross and Jaffe,
1971). However, the compositional difference between cummingtonite and
anthophyllite for Fe2+/(Mg + Fe 2 ~ =about 0·4 is very slight.
Gedrite is apparently a relatively rare mineral and is similar in chemical
composition to the assemblages: anthophyllite + cordierite, cummingtonite +
cordierite, and almandine + cordierite.
Anthophyllite is a rare constituent ofmetabasites. However, it is abundant in
some unusual rocks such as anthophyllite-cordierite rock (e.g. Eskola, 1914).
When anthophyllite coexists with cummingtonite, the former is always higher
in Al 20 3 than the latter. The Fe2+/(Mg + Fe2~ ratio of the former is
sometimes higher and sometimes lower than that of the latter (Klein, 1968).
Cummingtonite is a relatively common but minor constituent of many
low-pressure amphibolites. The frequency of occurrence appears to increase in
the high amphibolite facies, where the breakdown of the tschermakite com-
ponent of hornblende leads to the production of cummingtonite and anorthite
component of plagioclase, as shown by equation (88-3). Cummingtonite is not
so common in amphibolites of medium- and high-pressure metamorphic
terranes, where probably garnet forms instead (Shido, 1958, p. 211-12). There
is a wide miscibility gap between cummingtonite and hornblende (e.g. Klein,
1968). Cummingtonite in some metamorphic rocks contains exsolution lamellae
of hornblende (Jaffe et al. 1968). Some Ca-poor metabasites contain a large
amount of cummingtonite.

88-6 MUSCOVITE AND 8IOTITE

In high-pressure metamorphism, muscovite occurs in metabasites of the glauco-
phane-schist, greenschist and epidote-amphibolite facies. In medium- and low-
pressure metamorphism, muscovite occurs in metabasites of the greenschist
facies, but disappears in those of the epidote-amphibolite and higher facies by
 8B-7 ALMANDINE GARNET /259

reactions with amphibole or other associated minerals as exemplified by the

reaction:

muscovite + calcic amphibole -+ biotite + anorthite + quartz + Hz O. (8B-4)

Muscovite of metabasites is phengitic in low-temperature zones and becomes

higher in Al Z 0 3 , NazO and Ti0 2 contents with increasing temperature (Ernst,
1972b).
Biotite, on the other hand, occurs in metabasites of a wider range of
temperature extending from the low gr~,enschist to the low granulite facies.
Biotites in metabasites of the greenschist or epidote-amphibolite facies are
either green or brown, whereas those of higher facies are brown or reddish
brown (Phillips, 1930; Wiseman, 1934; Katada, 1965). In the Scottish Highlands,
biotite occurs in some chlorite zone metabasites. In other words, biotite begins
to occur at a lower temperature in metabasites than in metapelites (Phillips,
1930; Wiseman, 1934).

8B-7 ALMANDINE GARNET

In medium- and high-pressure metamorphism, epidote amphibolites and amphi-

bolites commonly contain almandine garnet with 6-10 per cent CaO. Most
almandines in metapelites have a much lower CaO content. This difference is
apparently due to a chemical difference of their host rocks. In the Dalradian
metamorphics of Scotland, almandine begins to occur in metabasites roughly at
the same temperature as in metapelites, that is, from the beginning of the
almandine zone for the metapelites (Wiseman, 1934).
On the other hand, almandine is very rare or absent in ordinary metabasites
of the low-pressure series in the Orijiirvi area (Eskola, 1914), central Abukuma
Plateau (Miyashiro, 1958), Ryoke belt (Katada, 1965), and northern Michigan
(James, 1955). Higher rock-pressure would favour the formation of almandine
garnet in metabasites.
The formation of almandine is, however, very sensitive to factors other than
rock-pressure. Almandine does not occur in medium-pressure metabasites of the
Sulitje1ma area (Vogt, 1927), Woodsville quadrangle (Billings and White, 1950),
and Wilmington complex near Philadelphia (Ward, 1959), conceivably owing to
their unfavourable composition.
It is common for some metabasites to contain almandine and others not to do
so even in the same outcrop. Wiseman (1934) and Binns (1965b) have shown
that almandine-bearing metabasites have a higher FeZ +/Mg ratio than the
associated almandine-free ones. A higher content of MnO and a lower content of
CaO would also favour the formation of almandine in metabasites.
260/META8ASITES: PROGRESSIVE MINERAL CHANGES

Hsu (1968) has shown that reducing conditions are essential for the
formation of almandine. This may be the main factor controlling almandine
formation in metabasites. Since metabasites do not contain organic matter or
graphite, their oxidation state is highly variable (§ 78-1 0).

88-8 CALCIC CLINOPYROXENE

In calcareous metasediments, diopside begins to occur roughly at the threshold

of the amphibolite facies. The diQPside usually forms by the reaction of
tremolite, calcite and quartz. Similarly, if a metabasite contains calcite and
quartz together with a calcic amphibole, calcic clinopyroxene (diopside or salite)
would begin to form near the threshold of the amphibolite facies. This is indeed
the case in some metamorphic terranes.
In many other terranes, however, calcic clinopyroxene begins to occur in
metabasites in the middle or a higher part of the amphibolite facies. The
temperature where calcic clinopyroxene begins to occur depends largely on the
bulk chemical composition of the rocks. In the granulite facies, where horn-
blende decomposes, calcic clinopyroxene is a major component of metabasites
(e.g. Binns, 1965b).
When calcic clinopyroxene is associated with hornblende or orthopyroxene,
the CaO content of the former tends to decrease with rising temperature,
conceivably owing to the decreasing width of the miscibility gap between the
associated minerals with rising temperature. In metabasites of the amphibolite
facies, the clinopyroxenes are usually salites, whereas in those of the granulite
facies, they are usually augites (Shiao, 1958, p. 186-8; Binns, 1965b; Bard,
1969).
Binns (1965b) has found that calcic clinopyroxenes in granulite facies
metabasites of the Broken Hill area contain two kinds of exsolution lamellae:
one composed of orthopyroxene parallel to (100) of the host and the other
composed of clinohypersthene (very poor in CaO) parallel to (001).

88-9 ORTHOPYROXENE

In ordinary metabasites, orthopyroxene begins to form at a higher temperature

than calcic clinopyroxene. The first occurrence of orthopyroxene in associadon
with clinopyroxene in metabasites is usually considered to mark the beginning of
the granulite facies. The pyroxenes form by gradual decomposition of horn-
blende.
Orthopyroxene, however, may form by other reactions also. The reaction
temperature should vary with the type of reaction and the Fe2+/Mg ratio of the
 8B-9 ORTHOPYROXENE /261
rock, as well as with P s , PH,o and Po" Figure 8B-S gives experimentally
determined high temperature stability limits for four amphiboles in the presence
of an aqueous fluid. The Mg-end members of calcic amphiboles (tremolite and
pargasite) are stable up to a much higher temperature than the Mg-end member
of anthopyllite. Tremolite is stable up to a much higher temperature than
ferrotremolite. Thus, if an amphibolite contains much anthopyllite or cumming-
tonite, these minerals would decompose to form orthopyroxene at a much lower
temperature than hornblende of ordinary amphibolite. If Fe-rich amphibolites

4 Fe-Tr Anth

Tr
~

'"
.D
3
~
"0
·5
~
II

a::: 2
~
::l
~
~
c.

°40~O~--~5~OO~--~6~OO~--~7~O~O----~80~O~--~90~O~--~100'O

Temperature (0 C)

Fig. 8B-5 High temperature stability limit of anthophyllite (Anth),

tremolite (Tr), ferrotremolite (Fe-Tr), and pargasite (Par) in the presence
of an aqueous fluid. The ferrotremolite curve is for Po, controlled by the
fayalite-magnetite-quartz buffer. (Greenwood, 1963; Boyd, 1959; Ernst,
1966.)

are mixed with Mg-rich ones, the former may produce orthopyroxene at a lower
temperature than the latter.
Some orthopyroxenes in rocks of the granulite facies are very high in Al 2 0 3
content (up to about 10 per cent) and show strong pleochroism (Howie, 1965;
Burns, 1966). Exsolution lamellae composed of augite were found in some
granulite facies orthopyroxenes (Binns, 1965b).
When orthopyroxene coexists with calcic clinopyroxene in metamorphic
rocks, the partition of Mg and Fe 2 + between the two minerals indicates an
262/METABASITES: PROGRESSIVE MINERAL CHANGES
approach to chemical equilibrium. This problem was discussed by a large number
of authors (e.g. Mueller, 1960; Kretz, 1961; Howie, 1965).

8B-1O GLAUCOPHANE
Glaucophane and stilpnomelane, both common in metabasites in high-pressure
terranes, will be treated in this and the next section. Lawsonite, aragonite and
jadeite occur in metabasites of some high-pressure terranes, but will not be
discussed here, because they are controlled by similar factors in metagraywackes,
which have been discussed in § § 7B-19 and 7B-20.
In metamorphic rocks, glaucophane forms characteristically at low tem-
peratures and high rock-pressures. In synthetic experiments, on the other hand,
glaucophane readily forms in charges of its own composition in the presence of
an aqueous fluid at a temperature above 850°C and down to low pressures
(Ernst, 1961, 1968). It follows that the stability of glaucophane should vary
greatly with the bulk chemical composition of the rock. Many glaucophane
schists have a chemical composition practically identical to that of some
greenschists and amphibolites (Washington, 1901; Ernst, 1963a; Coleman and
lee, 1963). In such rocks, the formation of glaucophane appears to require a
low temperature and high pressure.
The idealized composition of glaucophane is chemically equivalent to a
mixture of albite and antigorite:

2 Na2Mg3AI2Sis022(OH)2 + 2 H 20 = 4 NaAISi 30 s + Mg 6 Si 4 0 lO (OH)s.

glaucophane albite antigorite (8B-5)

Therefore, glaucophane could form in rocks with ordinary chemical com-

positions, in which otherwise most or all the Na20 would have entered
plagioclase.
Natural glaucophanes, however, are usually solid solutions between idealized
glaucophane and riebeckite Na2 Fe3 2+Fe2 3+Si s 022(OH)2 (e .g. Miyashiro,
1957b). In a transitional condition between the glaucophane-schist and green-
schist facies, some metabasites have glaucophane-bearing mineral assemblages
and others greenschist assemblages; the former tend to have higher Fe3+/Fe 2 +
ratios (Iwasaki, 1963, p. 32). The P-T field for the stability of glaucophane in
metabasites may become wider by the presence of the riebeckite component.
Synthetic glaucophane has two polymorphs: I and II. Glaucophane I is stable
on the high-temperature and low-pressure side of the field of glaucophane II.
Natural glaucophane belongs to polymorph II (Ernst, 1963b, 1968).
Glaucophane is sometimes stably associated with actinolite or barroisite, or
rarely with hornblende. The existence of wide miscibility gaps between glauco-
 88-11 STILPNOMELANE/263
phane and the latter amphiboles has been demonstrated by Klein (1969) and
Himmelberg and Papike (1969).

88-11 STILPNOMELANE
Stilpnomelane is widespread in, but not confined to, high-pressure metamorphic
rocks. It is rare in medium-pressure and is extremely rare in low-pressure
metamorphic rocks. So far as I am aware, the sole discovery of stilpnomelane in
low-pressure metamorphic terranes has been made in slightly metamorphosed
basaltic pyroclastics in the lowest temperature zone (zone Ia in fig. 3-4) of the
Shiojiri area in the Ryoke belt, Japan (Katada and Sumi, 1966). In medium-
pressure metamorphic terranes, stilpnomelane was found, for example, from an
epidiorite (Wiseman, 1934, p. 376) and metagraywacke (Mather and Atherton,
1965; Mather, 1970), both in a low-temperature part of the Scottish Highlands,
and from slates and phyllites of the northern Appalachians (Zen, 1960).
In high-pressure metamorphic terranes, stilpnomelane is a common mineral in
metabasites, metapelites, metapsammites and metacherts of the glaucophane-
schist, greenschist and epidote-amphibolite facies. Stilpnomelane is more
common in the metapsammites than in the other rocks. In high-pressure facies
series, stilpnomelane appears to form most readily in the glaucophane-schist and
greenschist facies. When the series extend downwards to prehnite-pumpellyite
facies or upwards to the epidote-amphibolite facies, the mineral is less common
or absent in the extended parts of the series (Iwasaki, 1963; Hashimoto, 1968;
Ernst and Seki, 1967; Niggli, 1960, 1970).
The stability of stilpnomelane, however, varies greatly with the chemical
composition of the rock. It is known to occur even in granophyre patches
associated with the marginal olivine gabbro of the shallow gabbroic intrusion of
Skaergaard, Greenland (Wager and Deer, 1939). It has long been known to
occur in iron ores, and especially commonly in weakly metamorphosed iron
formations.
Chemically, an Fe 2 +-rich, dark green variety called ferrostilpnomelane is
distinguished from an Fe3+-rich, reddish brown variety, called ferristil-
pnomelane. The two may occur in different parts of a single hand specimen of
schist (e.g. Yui, 1962). Hutton (1938) assumed that ferristilpnomelane forms by
the oxidation of ferrostilpnomelane accompanied by loss of H, that is, by the
substitution Fe 3 +0 2 - ~ Fe 2 +(OH)-, as in oxyhornblende. However, Hashimoto
(1969) has emphasized that there is no systematic difference in H2 0+ content
between the two varieties and there appears to exist a discontinuous change in
the oxidation state. Brown (1971) has given strong evidence suggesting that the
oxidation of ferrostilpnomelane takes place after metamorphism.
264/METABASITES: PROGRESSIVE MINERAL CliANGES

8B-12 OXIDES AND SULPHIDES OF IRON

The frequency of occurrence of iron oxides and sulphides in metabasites of

high-pressure and low-pressure metamorphic terranes in Japan is summarized in
table 7B-1.
If we compare these minerals in glaucophane-schist facies metabasites with
those in amphibolite facies metabasites, haematite is common in the former but
is absent in the latter, whereas magnetite and ilmenite are rare or absent in the

.~ ________--,.l....IIiIiiIl~~§§~~~~~Haematite
Fe203
Fig. 8B-6 Paragenetic relations of oxides of iron and titanium in some
metamorphic rocks. (Chinner, 1960; Banno and Kanehira, 1961.)

former but are common in the latter. This difference may be regarded as a
progressive reduction with rising temperature.
In the low-temperature part of high-pressure metamorphic terranes, meta-
basites commonly contain haematite, whereas associated metapelites contain no
haematite or magnetite. Thus, metapelites are evidently in a more strongly
reducing condition than the associated metabasites. This is probably due to the
presence of organic matter in the former.
Haematite, magnetite and ilmenite belong to the FeO-Fe203-Ti02 system,
as shown in fig. 8B-6. There is a continuous series of solid solution between
 88-12 OXIDES AND SULPHIDES OF IRON /265
ilmenite and haematite at high temperatures above 1000 °C, but a miscibility
gap exists under metamorphic conditions. If we neglect the effect of this solid
solution, the magnetite-haematite assemblage has a definite value of P02 under
a definite P s and T, as was discussed in § 2-7. If we consider the effect of solid
solution, however, the state of the system is fixed only when the three phases
magnetite, haematite and ilmenite coexist (Crunner, 1960).
The TiO z content of magnetite in some igneous rocks may be as high as 20
per cent, but that of magnetite in metamorphic rocks is usually less than 4 per
cent (Buddington, Fahey and Vlisidis, 1955; Heier, 1956).
For a detailed account of the phase relations in the Fez03-Feo-TiOz
system, refer to Nagata (1961).
Chapter 9

Progressive Regional Metamorphism

of Limestones

9-1 COMPOSITIONS OF LIMESTONES AND FLUIDS

Compositions of Limestones
Most limestones are mainly composed of three components: CaC0 3 , MgC0 3 and
Si0 2 • Since the number of components is so small, the relationship between the
chemical and mineral compositions in metamorphosed limestones is relatively
simple. This greatly facilitates theoretical treatment.
If a limestone is composed of pure CaC0 3 , metamorphism would produce
crystalline limestones (marbles) which tend to become coarser grained with
increasing temperature. In most cases, there will be no other changes at all. The
temperature of metamorphism is not high enough to dissociate calcite into CaO
and CO 2 , (In nature, the mineral lime CaO has only been reported as occurring
in limestone inclusions in a lava of the volcano Vesuvius.) In some high-pressure
metamorphism, however, calcite in limestone may be transformed into aragonite.
If an original limestone contains calcite and dolomite but no quartz,
metamorphism may decompose only the dolomite as follows: dolomite --+ calcite
+ periclase + CO 2 , This decomposition could take place at temperatures several
hundred degrees lower than that of CaC0 3 under otherwise similar conditions,
and the decomposition product periclase is not rare in dolomitic limestones
subjected to con tact metamorphism. Periclase, usually in small octahedral grains,
almost always suffers hydration to form brucite Mg(OHh at some later time.
Brucite-calcite rocks produced in this way are called predazzite.
If an original limestone contains calcite and quartz, contact or regional
metamorphism at relatively high temperatures would produce wollastonite by
the reaction: calcite + quartz = wollastonite + CO 2 , Wollastonite is a common
mineral in limestones in amphibolite and pyroxene-hornfels facies areas. In
some rare cases of contact (and not regional) metamorphism, siliceous lime-
stones may contain tilleyite, spurrite, rankinite and larnite.
As a more general case, an original limestone is assumed to be composed of
calcite, dolomite and quartz. The metamorphism of such a siliceous dolomitic
limestone produces a variety of silicate minerals. In the metamorphism of
 9-2 LOW-PRESSURE REGIONAL METAMORPHISM /267
siliceous dolomitic limestones of the Precambrian terrane in Finland, Eskola
(1922) found three steps characterized by the first appearance of the following
minerals in order of rising temperature: tremolite ~ diopside ~ wollastonite.
Some metamorphosed dolomitic limestones contain forsterite and lack
quartz. In some cases of contact metamorphism, monticellite, akermanite,
tilleyite, spurrite, rankinite, merwinite, and larnite may form, depending on the
temperature, Peo • and other factors.
If an original limestone contains some Al 2 0 3 besides CaO, MgO, and Si0 2 it
may produce epidote, hornblende, grandite (Ca-garnet), vesuvianite, and calcic
plagioclase. Epidote is usually stable from the greenschist or even lower facies.
Grandite begins to appear in the epidote-amphibolite facies in some areas and in
the amphibolite facies in others.
Some metamorphic rocks derived from impure calcareous sediments are
mainly composed of plagioclase and hornblende. This indicates the possibility
that some ordinary amphibolites may also be metamorphic derivatives of
calcareous sediments. When metamorphism is accompanied by a degree of
metasomatism or metamorphic differentiation, the chemical composition of the
resultant amphibolites may be similar to that of amphibolites of basic igneous
origin. This problem was discussed in detail by Orville (1969) and Vidale (1969).
If K2 0 is present, muscovite, phlogopite and/or K-feldspar would be
produced by metamorphism.

Distribution of Peo •
In progressive metamorphism, limestones are subjected to a series of decarbon-
ation reactions. At the same time, the associated pelitic and psammitic rocks
°
undergo dehydration reactions. Thus, CO 2 and H2 are liberated together.
The temperatures of decarbonation reactions vary with Peo. and the
composition of the fluid phase if any. The composition of the fluid varies
considerably within a short distance in some metamorphic areas, as was clearly
demonstrated by Melson (1966) and Carmichael (1970). In such areas, the
isograds for decarbonation reactions may intersect with isograds for dehydration
reactions.
Our knowledge of the frequency and extent of uneven distribution of CO 2 in
metamorphic areas is still very meagre. However, uneven distribution is con-
ceivably very common (§ 4-4). Therefore, decarbonation reactions cannot be
used as a reliable geothermometer.

9-2 LIMESTONES IN LOW-PRESSURE REGIONAL METAMORPHISM

Ryoke Belt
Metamorphosed limestones in the Shiojiri area of the Ryoke belt, Japan, were
described by Katada (1965). The lowest temperature limestones contain calcite,
268/PROGRESSIVE REGIONAL METAMORPHISM OF LIMESTONES

dolomite, quartz, and plagioclase with subordinate white mica and tourmaline.
With rising temperature, tremolite and diopside begin to occur at nearly the
same temperature. This is at variance with observations in many other areas of
the world where tremolite begins to form at a lower temperature than diopside.
Grandite begins to form at a higher temperature than diopside, and wollastonite
and scapolite at a still higher temperature (fig. 3-4).
The grandite + quartz assemblage persists to the highest temperature zone,
though it is accompanied by wollastonite in some rocks. Wollastonite is
accompanied by calcite and quartz.

Central Abukuma Plateau

Most limestones in this area are rather pure CaC0 3 rocks and show only a
progressive increase in grain-size with rising temperature. In low-temperature
zones, most grains of calcite measure 0·05-0·1 mm in diameter, whereas in
higher temperature zones they measure 1-10 mm.
Small parts of the limestones, however, are impure. Further, the metabasites
in this area frequently contain calcic bands (Miyashiro, 1953a, 1958). These
rocks show the following order of the first appearance of minerals: tremolite -+
diopside -+ wollastonite (fig. 3-5).
In the high-temperature part of the area, the assemblage garnet + quartz +
wollastonite occurs in some rocks, and the assemblage quartz + calcite + calcic
plagioclase + wollastonite occurs in others. This may mean that the garnet-
quartz reaction is beginning to take place as follows: grossular + quartz = 2
wollastonite + anorthite. This reaction will be discussed near the end of this
chapter.
In some quartz-free limestones of the same zone, forsterite occurs in
association with calcite, dolomite, tremolite, diopside and phiogopite.

Pyrenees
In some low-pressure regional metamorphic areas in the Pyrenees and Iberian
Peninsula, more or less similar sequences of progressive mineral formation have
been observed. For example, Bard (1969) has found in the Aracena area in Spain
the following sequence with rising temperature: tremolite -+ diopside -+
forsterite -+ grandite -+ wollastonite (fig. 7A-4).

9-3 LIMESTONES IN MEDIUM-PRESSURE REGIONAL METAMORPHISM

Scottish Highlands
Harker (1932, p. 252-9) gives a cursory description of limestones in the Scottish
Highlands, though information from other areas is indistinguishably mixed. In the
biotite zone, limestones contain tremolite, zoisite, muscovite, biotite, quartz and
often albite. In the almandine zone, limestones contain grandite, vesuvianite,
 9-4 HIGH-PRESSURE REGIONAL METAMORPHISM /269
diopside, hornblende, forsterite and microcline. Skarns are formed at the
boundary between a limestone and kyanite schist at Glen Urquhart. They
contain epidote, zoisite and calcic amphiboles, and occasionally prehnite and
pectolite (Francis, 1958).
In the sillimanite zone, limestones contain diopside, grossular, plagioclase and
orthoclase, but no wollastonite. Harker considered that wollastonite does not
normally form in regional metamorphism. Indeed, in another medium-pressure
metamorphic region (northern Appalachians), wollastonite does not occur even
in the low granulite facies (Thompson and Norton, 1968).
However, the absence of wollastonite is not a general feature of regional
metamorphism, or a general feature of medium-pressure regional metamorphism.
Examples of low-pressure metamorphism, producing wollastonite, have already
been given. An example of wollastonite-forming medium-pressure regional
metamorphism will be described below.

Nanga Parbat
In Nanga Parbat, northwest Himalayas, calcareous metasediments occur in a
terrane mainly composed of metapelites which contain kyanite and sillimanite.
They show unusually advanced decarbonation for medium-pressure meta-
morphism (Misch, 1964). In an amphibolite facies area, three progressive
metamorphic zones in terms of calc-silicate assemblages have been distinguished:
1. Calcite-quartz-grossular zone, which is characterized by the presence of
grossular and of the calcite + quartz assemblage instead of wollastonite,
2. Wollastonite-grossular-quartz zone,
3. Wollastonite-anorthite zone, in which the reaction of grossular with
quartz (Le. grossular + quartz = 2 wollastonite + anorthite) has been
completed. Wollastonite occurs there in contact with anorthite or by-
townite.

9-4 LIMESTONES IN HIGH-PRESSURE REGIONAL METAMORPHISM

The Alps
Trommsdorff (1966) has described metamorphosed limestones in the Lepontine
and Bergell Alps near the border between Switzerland and Italy. The tem-
perature of metamorphism was lowest in the northwestern part of the areas. The
limestones exposed there contain talc and tremolite in such assemblages as talc +
calcite + dolomite, talc + calcite + tremolite, tremolite + quartz + calcite, and
tremolite + calcite + dolomite. Talc appears to begin to form at a little lower
temperature than tremolite.
The temperature rises toward the southeast and soon diopside begins to form,
and then forsterite. However, the distribution of forsterite is capricious.
Tremolite is still stable in some limestones. In the highest temperature part of
270/PROGRESSIVE REGIONAL METAMORPHISM OF LIMESTONES

the area where the Bergell granite is intruded, wollastonite forms in siliceous
limestones. Probably this is a result of contact effects and not of regional
metamorphism. Scapolite occurs over the whole area from the tremolite zone to
the highest temperature one (see § 7 A-8).

Sanbagawa Belt
In the Bessi-Ino area, the lowest-temperature zone, belonging to the glauco-
phane-schist facies, contains limestone with chlorite and quartz besides calcite.
In a higher temperature zone of the lower epidote-amphibolite facies, lime-
stones contain tremolite, muscovite, albite and quartz besides calcite. Lime-
stones of the higher epidote-amphibolite facies contain epidote, grossular,
diopside and hornblende (Banno, 1964, p. 290).

Franciscan Formation
In well-recrystallized Franciscan metamorphics of the typical glaucophane-schist
facies (type III in Coleman and Lee's nomenclature) in the Cazadero area,
California, metamorphosed limestones occur as small beds and lenses composed
of aragonite. Here, the pressure was high enough to cause transformation of
calcite to aragonite. The rock contains, besides carbonate, pyralspite garnet,
green amphibole, stilpnomelane, clinozoisite and quartz (Coleman and Lee,
1962, 1963).

9-5 PROGRESSIVE METAMORPHIC REACTIONS IN SILICEOUS

DOLOMITIC LIMESTONES

Introductory
Bowen (I940) made an elegant theoretical analysis of the progressive decarbon-
ation of siliceous dolomitic limestones. A number of thermochemical and
synthetic studies have been made on individual decarbonation reactions which
take place in the presence of a pure CO 2 fluid (e.g. Weeks, 1956; Danielsson,
1950; Harker and Tuttle, 1956). In recent years, theoretical and experimental
works on systems involving a fluid containing both CO 2 and H2 0 have been made
by Greenwood (1962), Winkler (I967), Metz and Winkler (I964), and others.
These works ,have made important contributions to our understanding of natural
decarbonation processes.
Winkler (I967) and Metz and Trommsdorff (I968) give elaborate systematic
accounts on this problem. The follOWing discussion is based largely on theirs.
The presence of a fluid composed of CO 2 and H2 0 is assumed below.
Limestones composed of calcite, dolomite and quartz and their metamorphic
derivatives may be regarded as belonging to the five-component system CaQ-
MgO-Si0 2-C0 2-H 20. Disregarding CO 2 and H20, the relevant rocks can be
represen ted in a triangular diagram CaQ- MgO-Si0 2 . The compositions of the
siliceous dolomitic limestones which we will now consider fall on the left half of
the diagrams in Fig. 9-1.
 9-5 PROGRESSIVE METAMORPHIC REACTIONS /271
Si0.2
A\artz

CaO~MgO
Calcite Dolomite (Magnesite)
(a) Original dolomitic limestone
/ (ruled area) \

1ti
Quartz

lA
Quartz
Calcite Dolomite
(b) Greenschist facies
after reaction (9-1 )
...

Bs
Quartz

Calcite Dolomite
(b') Greenschist facies

Calcite Dolomite
"'~T"I9~J
(c) Greenschist facies after

•
reaction (9-2)

Quartz Quartz

Dio~r
~orsterite
Calcite Dolomite Calcite Dolomite
(d) Low amphibolite facies (d') Low amphibolite facies
after reaction (9-5) ""-.. / after reaction (9-6)
"'-.Quartz

Calcite Dolomite
(e) Hi~h amphibolite facies

Fig. 9-1 Progressive changes of paragenetic relations in siliceous dolo-

mitic limestones in the presence of a mixture of CO 2 and H 2 0. AbbreviLl-
lions: Diop = diopside, Tr = tremolite. Only the rocks plotted on the left
half of the triangles are considered. If the fluid is of virtually pure CO 2
composition, hydrous minerals such as talc and tremolite would not form_
Such an extreme case is not considered here.
272/ PROGRESSIVE REGIONAL METAMORPHISM OF LIMESTONES

Formation of Talc and Tremolite

In regional metamorphism of the greenschist facies, tremolite is a widespread
mineral in limestones. Talc is also known to occur in some cases. (Conceivably
talc may have been mistaken for white mica in some petrographic descriptions.)
Under most CO 2 /H 2 0 ratios « M in fig. 9-2), talc is the first mineral to form in
metamorphosed siliceous dolomitic limestones:

3 dolomite + 4 quartz + H2 0 = talc + 3 calcite + 3 CO 2 , (9-1)

In the same range of CO 2 /H2 0, the next reaction with rising temperature will
form tremolite as follows:

5 talc + 6 calcite + 4 quartz = 3 tremolite + 6 CO 2 + 2 H2 0. (9-2)

The experimentally determined equilibrium curves for reactions (9-1) and (9-2)
are shown in fig. 9-2 (Metz and Winkler, 1963; Metz and Trommsdorff, 1968;
Metz and Puhan, 1970). The paragenetic relations in siliceous dolomitic
limestones should change in the order (a) -+ (b) -+ (c) in fig. 9-1.
However, Metz and Trommsdorff (1968) have shown that in the presence of a
fluid with very high CO 2 /H 2 0 ratios (range M-N in fig. 9-2), the first reaction to
take place in siliceous dolomitic limestones directly forms tremolite as follows:

5 dolomite + 8 quartz + H2 0 = tremolite + 3 calcite + 7 CO 2 , (9-3)

In this case, the paragenetic relations should change in the order (a) -+ (b').

Formation of Diopside and Forsterite

With an additional rise in temperature, diopside should begin to form roughly at
the threshold to the amphibolite facies. The reaction would be as follows for the
above two ranges of CO 2 /H 2 0:

tremolite + 3 calcite + 2 quartz = 5 diopside + 3 CO 2 + H2 0. (9-4)

The equilibrium curve for this reaction (Metz and Winkler, 1964) is also shown
in fig. 9-2.
A further rise in temperature causes the formation of forsterite, conceivably
owing to the following reactions (9-5) or (9-6) depending on the CO 2 /H 2 0
ratio:

talc + 5 dolomite =4 forsterite + 5 calcite + 5 CO 2 + H2 0. (9-5)

This reaction takes place for most values of CO 2 /H 2 0 in the fluid, and the
parage netic relations after this reaction are shown in fig. 9-1(d).
 9-5 PROGRESSIVE METAMORPHIC REACTIONSJ273

900

800

U 700
e
..,'"
~
<l>
0.

~ 600
I-

500

400

0·0 0·6 '·0

Mole ratio CO:!/(H2 0+CO:!)

Fig. 9-2 Isobaric equilibrium curves for progressive metamorphic re-

actions in limestones in the presence of a fluid consisting of CO 2 and H2 0
at a total pressure of 1 kbar. Figures in parentheses represent the numbers
of equations in the text. (Based on Winkler, 1967; Metz and Winkler 1963,
1964; Metz and Trommsdorff, 1968, and Greenwood, 1962.)

tremolite + 11 dolomite = 8 forsterite + 13 calcite + 9 CO 2 + H2 0. (9-6)

This reaction takes place only for very high CO 2 /H 2 0 ratios, and the
paragenetic relations after this reaction are shown in fig. 9-1(d').
The equilibrium temperatures for (9-2), (9-4) and (9-5) as shown in fig. 9-2
are very close to each other, especially at low CO 2 /H 2 0 ratios. In other words,
tremolite, diopside and forsterite begin to form nearly at the same temperature.
(This is at variance with our ordinary petrographic experience on regional
metamorphic limestones.)
274/PROGRESSIVE REGIONAL METAMORPHISM OF LIMESTONES

With a slight rise in temperature, the tremolite + calcite assemblage would

become unstable as follows:

tremolite calcite
= 11 CaMgSi 2 0 6 + 2 Mg 2 Si0 4 + 5 CO 2 + 3 H2 0. (9-7)
diopside forsterite
There is no experimental work on this reaction. Judging from the mode of
occurrence of tremolite and theoretical consideration (Metz and Trommsdorff,
1968), this reaction takes place at a considerably higher temperature than
reactions (9-5) and (9-6).
In the case where the fluid is almost entirely composed of CO 2 , talc and
tremolite, both being hydrous, cannot form, and the reaction of dolomite with
quartz would directly form diopside as follows:
dolomite + 2 quartz = diopside + 2 CO 2 . (9-8)
This appears unusual for regional metamorphism. At all events, a further rise in
temperature in this case would cause the following reaction:

diopside + 3 dolomite = 2 forsterite + 4 calcite + 2 CO 2 • (9-9)

Formation of Wollastonite
The formation of wollastonite is represented by the equation:
CaC0 3 + Si0 2 = CaSi0 3 + CO 2 • (9-10)
calcite quartz wollastonite
The equilibrium curve for this reaction determined by Greenwood (I962, 1967)
is shown in fig. 9-2. The temperature is about 580-680°C at 1 kbar and
610-750 °c at 2 kbar for a wide range ofC0 2 /H 2 0 in the coexisting fluid.
The paragenetic relations in such high amphibolite facies conditions are
shown in fig. 9-1(e). Anthophyllite is shown here instead of talc (Greenwood,
1963; Winkler, 1967, p. 39).

9-6 METAMORPHIC REACTIONS IN ALUMINA-CONTAINING SILICEOUS

LIMESTONES

The high-temperature stability limits for grossular and the grossular + quartz
assemblage are marked by the following reactions:

(9-11)
grossular wollastonite anorthite gehlenite

(9-12)
grossular quartz wollastonite anorthite
 9-6 ALUMINA-CONTAINING LIMESTONES/275

....i\;
6 1 .e2
~I
.\2/ ..::::-
I~

.I~
/ \
.I/ \
" \11>.
,/ ~\~
+-~" 0.\
\
,," ~
,,~~
\
...co 4
/
"" \

= ...
.c \
~
\
\ 0)
~
f \ '"
::l \
:a \
f
Q..
\
\
\
\
2 \
\
\
\
\
\
\
\
\

0
500 700 900
Temperature (OC)

Fig. 9-3 High-temperature stability limits of grossular and the grossular +

quartz assemblage under low PC0 2 • (Based on Newton, 1966a; Storre,
1970, and thermochemical calculations from Robie and Waldbaum's, 1968,
constants.) The broken lines represent the stability fields of Alz SiO s
minerals.

As shown in fig. 9-3, the equilibrium curve for (9-11) lies at temperatures about
300°C higher than that for (9-12). Many grossular-bearing limestones contain no
quartz. The reaction of grossular with quartz with rising temperature was most
clearly demonstrated in the Nanga Parbat area (§ 9-3).
The grossular + quartz assemblage is stable only under a relatively low Pco ,.
Increase in Pco , or in the CO 2 /H 2 0 ratio of the intergranular fluid would
change the assemblage into calcite and less calcic minerals such as zoisite or
anorthite.
276!PROGRESSIVE REGIONAL METAMORPHISM OF LIMESTONES

Progressive metamorphic reactions which take place in Al2 0 3 -bearing siliceous

limestones involve zoisite, clinozoisite, anorthite and grossular. The stability
relations of relevant mineral assemblages in such reactions were discussed by
Storre (1970) and Kerrick (1970).
Chapter 10

Progressive Contact Metamorphism

and Pyrometamorphism

10-1 CONTACT METAMORPHISM OF PELITES TO PRODUCE ANDA-

LUSITE AS THE ONLY Al.SiO, MINERAL

The physical conditions of contact metamorphism are diverse (Miyashiro,

1961b, p. 306-7; Reverdatto, Sharapov and Melamed, 1970). A classification of
contact metamorphism in terms of the prevailing rock-pressure was given in
§4-S. The first three sections of this chapter deal with descriptions of pelitic
contact metamorphic rocks on the basis of this classification.
The early days of metamorphic petrology had two monumental works on
contact metamorphism by Rosenbusch (1877) and Goldschmidt (1911). They
both treated the most common type of contact metamorphism, i.e. that
characterized by andalusite and cordie rite (without sillimanite, kyanite and
staurolite ).

Ba"-Andlau Area, France

Rosenbusch (1877) investigated the contact aureole around a Hercynian granite
mass of the Barr-Andlau area in Vosges, Alsace. This area was recently re-studied
by Bosma (1964). In the southern part of the aureole, about 0'5-1'5 km wide,
the early Palaeozoic Steige slates show three zones of advancing recrystallization
inwards as follows:
1. Spotted slates or phyllites
2. Spotted schists
3. Hornfelses.
The Steige slates are highly aluminous (A1 2 0 3 = 20-27 per cent). The first
sign of metamorphism is the generation of black spots, resulting in spotted slates
or phyllites. The rocks are mainly composed of quartz and white mica, with
subordinate plagioclase, chlorite, tourmaline, haematite and magnetite.
At first, the spots are very small, but soon become bigger up to about 3 mm
in size. Rosenbusch considered the dark colour of spots as being due to local
concentration of opaque minerals. Some students have considered the possibility
278/PROGRESSIVE CONTACT METAMORPHISM

that originally the spots had been porphyroblasts of cordierite which later
altered into other minerals. X-ray investigation by Bosma has demonstrated the
presence in the spots of quartz, muscovite and haematite, occasionally with
feldspar and minerals showing 7'1 A and 14·5 A reflections. No cordie rite has
been found. Nonetheless, he considers that the spots were originally cordie rite
porphyroblasts containing inclusions of quartz, feldspar and iron ore, and that
later alteration of the cordierite produced muscovite and the 7·1 A and 14' 5 A
minerals.
The spotted schists are transitional from spotted slates (or phyllites) to
hornfelses. Though they have spots, the schists are much less fissile than spotted
slates. Biotite and andalusite begin to occur. The latter mineral always forms
outside the spots.
The hornfelses have neither fissility nor spots. They all contain andalusite
porphyroblasts and a few contain cordie rite porphyroblasts also. Muscovite,
biotite, quartz, feldspars, tourmaline and opaque minerals occur. Bosma con-
siders that the hornfelses of the inner aureole probably did not pass through the
stage of spot formation owing to their rapid heating by the intrusive granite
which afforded no time for recrystallization at low temperature.

Oslo Area, Norway

Permian volcanic and plutonic rocks are exposed in the Oslo graben cutting
through Precambrian terranes in Norway. The plutonic masses, including alkali
syenite, nepheline syenite and alkali granite, were intruded in Palaeozoic
sedimentary rocks within the graben, giving rise to contact metamorphic effects
(Barth, 1945; Oftedahl, 1959).
Goldschmidt (1911) studied well-recrystallized hornfelses in the inner
aureoles of the Oslo (Christiania) area. Though the extent of the aureoles
depends largely on the size of the relevant intrusions, the metamorphism in the
inner aureoles were found to be in the pyroxene-hornfels facies regardless of the
size and rock types of the intrusions. The outer aureoles are poorly recrystallized
and have not been well studied, but could belong to the amphibolite facies.
It has been shown that the mineral compositions of hornfelses in the inner
aureole vary regularly with their chemical compositions. As shown in fig.
11-1(c), Goldschmidt classified hornfelses ranging from pelitic to calcareous
compositions into ten classes:
Class 1: Andalusite-cordierite hornfels.
Class 2: Plagioclase-andalusite-cordierite hornfels
Oass 3: Plagioclase-cordierite hornfels
Class 4: Plagioclase-orthopyroxene-cordierite hornfels
Class 5: Plagioclase-orthopyroxene hornfels
Class 6: Plagioclase-diopside-orthopyroxene hornfels
10-2 PRODUCING SILLIMANITE WITH OR WITHOUT ANDALUSITE /279
Class 7: Plagioclase-diopside hornfels
Class 8: Grossular-plagioclase-diopside hornfels
Class 9: Grossular-diopside hornfels
Class 10: Grossular-wollastonite-diopside hornfels.
He considered that quartz and orthoclase could occur in any of these types.
However, it is likely that grossular is unstable in the presence of quartz, judging
from the recent experimental data on the reaction: grossular + quartz = 2
wollastonite + anorthite (§9-6; fig. 9-3). Biotite occurs in classes 1-7. Muscovite
does not occur in any class. Hornfelses of class 10 may contain vesuvianite.
Pelitic hornfelses belong to classes 1- 5, whereas classes 7- 10 represent
calcareous rocks.

10-2 CONTACT METAMORPHISM OF PELITES TO PRODUCE SILLIMAN-

ITE WITH OR WITHOUT ANDALUSITE

Sillimanite is not as common in contact aureoles as in low-pressure regional

terranes. This is probably because many instances of contact metamorphism
represent a lower rock-pressure than low-pressure regional metamorphism, and
the transformation from andalusite to sillimanite takes place at higher tem-
perature with decreasing rock-pressure (fig. 2-1). If sillimanite occurs it does so in
the inner aureole, whereas andalusite occurs in the outer aureole.
Four areas will be reviewed in order of increasing widespread occurrence of
sillimanite.

Comrie Area, Scottish Highlands

The Carn Chois diorite was intruded into greywackes and slates of the Dalradian
Series in the Comrie area near the southern boundary of the Scottish Highlands.
The mass of diorite, about 3 krn across, is mainly composed of biotite,
hornblende and andesine. The country rocks had been metamorphosed in the
chlorite-zone grade before the intrusion. The slate outside the contact aureole is
mainly composed of muscovite, chlorite, quartz and iron ores.
Three zones of progressive recrystallization have been distinguished (Tilley,
1924).
(a) Zone of spotted slates. The first sign of contact metamorphism in slates is
the development of minute spots and the hardening of the rock. It begins at a
distance of about 400 m from the contact. The spots are less than 1 mm across
and are dark grey or black. Both spots and matrix are composed of the same
minerals as the original slate. The darker colour of the spots is due to higher
proportions of chlorite and muscovite.
(b) Zone of biotite development. The first new mineral produced by the
contact effect is biotite, which occurs as little brown flakes within spots as well
280/ PROGRESSIVE CONTACT METAMORPHISM

as in the matrix. Chlorite, muscovite and biotite show a tendency to parallelism

with the original cleavage. On a closer approach to the contact, the preferred
orientation becomes weaker, and the fissility is lost. The spots disappear and the
grain size increases.
(c) Zone of cordie rite hornfelses. Cordierite begins to occur at a distance of
140 m from the contact, and soon becomes very widespread. In this zone,
recrystallization is complete. Orthoclase and biotite occur in most hornfelses.
Muscovite and chlorite do not occur. Some hornfelses contain quartz, cordierite
and orthopyroxerte, whereas others contain no quartz but have cordierite, spinel
and corundum.
Completely recrystallized pelitic and psammitic hornfelses were divided into
two groups:
1. Quartz-bearing hornfelses, which show nearly the same mineral
assemblages as the Oslo Hornfelses.
2. Quartz-free hornfelses, in which the decrease of silica changes andalusite
into corundum first, and then cordierite into spinel. Formation of olivine
by desilication of orthopyroxene has not taken place.
In this aureole, andalusite is the common Al 2 SiO s mineral. Although
sillimanite was found in some hornfelses near the contact, no distinct sillimanite
zone was noticed.
In the associated grits, the first sign of contact metamorphism is hardening of
the rocks. Spots do not form. As the contact is approached, little biotite flakes
form, and then andalusite and cordie rite appear. Some hornfelses contain spinel
or corundum.
The Dalradian Series in the area contains small masses of metabasite
(epidiorite), which show the actinolite + chlorite + sodic plagioclase + epidote
assemblage outside the contact aureole. Within the aureole, the rocks become
biotite amphibolites, and then hornblende-orthopyroxene-clinopyroxene-
biotite-plagioclase hornfelses. In some rocks, hornblende almost disappears.
Thus, the outer and inner aureoles are in the amphibolite and the pyroxene-
hornfels facies respectively.

Arisu Area, Northeast Japan

The Cretaceous Tono granodiorite was intruded into Late Palaeozoic sedi-
mentary rocks in the Kitakarni Mountains, northern Honshu. The granodiorite
mass, containing hornblende and biotite as the main mafic minerals, measures
about 40 km across. The contact aureole is 3-5 km wide, being situated about
110-130 km NNE of the city of Sendai.
Seki (1957, 1961b) described the progressive metamorphism of the pelites,
limestones and basic volcanics of the Arisu area within the aureole. The
10-2 PRODUCING SILLIMANITE WITH OR WITHOUT ANDALUSITE /281
progressive metamorphic changes are summarized in fig. 10-1. He divided the
aureole into four zones in terms of mineral changes of the metapelites:
(a) Chlorite-muscovite zone. Chloritoid occurs in .some rocks. This zone is
present in the aureole on the western side of the granodiorite, but is absent on
the eastern side. The rocks are phyllitic.
(b) Biotite-andalusite zone. Biotite begins to occur.
(c) Cordierite-almandine zone. Andalusite and gedrite are common.
The gedrites have unusually high Al 2 0 3 and FeO content (Seki and Yamasaki,
1957). Cummingtonite occurs rarely. Spinel and corundum may occur in the
same thin section as quartz, but not in direct contact with it.
(d) Sillimanite zone. Muscovite, biotite, cordierite, garnet and micro cline are
widespread. The garnet contains 82-83 mol per cent of AIm and 9-11 per cent of
Pyr. The host rocks have Fe 2 +/Mg ratios as high as 1'9-2,3 (Onuki, 1968).
The metabasites and limestones in this aureole will be described in § § 10-6
and 10-7.

Steinach Area, Northeast Bavaria, Germany

The contact aureole surrounding the Leuchtenberg granite massif in the Steinach
area was investigated by Okrusch (1969, 1971). The . rocks outside the aureole
are Hercynian banded gneisses and mica schists, which formed by regional
metamorphism in the amphibolite facies. These rocks occasionally contain
garnet, sillimanite and staurolite. The Leuchtenberg granite (Carboniferous) was
intruded into the regional metamorphic rocks with resultant formation of a
contact aureole, about 1 km wide.
The contact metamorphism has decomposed garnet and sillimanite of regional
metamorphic origin, and has produced cordierite and andalusite instead in the
outer aureole. Thus, the characteristic mineral assemblage of the outer aureole is
muscovite + biotite + andalusite + cordierite + quartz + plagioclase.
Andalusite is the only polymorph of Al 2 SiO s in the outermost part of
the outer aureole, while both andalusite and sillimanite occur in the rest of the
aureole. The sillimanite/andalusite ratio generally increases with closer approach
to the granite border. The maximum abundance of andalusite has been observed
at a distance of 200-400 m from the granite border. This means that the
sillimanite isograd passed across the andalusite zone during the contact meta-
morphism. Cordierite and garnet of contact metamorphic origin tend to increase
as the granite border is approached. In the transitional zone from the outer to
the inner aureole, muscovite breaks down by reaction with quartz and biotite to
produce sillimanite, cordierite, garnet and K-feldspar. In the inner aureole,
K-feldspar can coexist with sillimanite, cordierite and garnet. Thus, the charac-
teristic mineral assemblage of hornfelses in the inner aureole is: sillimanite +
cordierite ± almandine ± K-feldspar + biotite + quartz + plagioclase.
The ratio Fe/(Mg + Fe) of cordierite is 0,48-0,49 in the transitional grade
282/ PROGRESSIVE CONTACT METAMORPHISM

from the outer to the inner aureole, and 0'50-0·54 in the inner aureole. The
contact metamorphic garnet at the outermost part of the aureole is a fine-
grained almandine-spessartine nearly free from zoning, containing 14-18 per cent
MnO and 23-25 per cent FeO. In the transitional zone from tht: outer to the
inner aureole, some grains of garnet are zoned with increasing MnO content to
the rim and others are zoned with decreasing MnO content to the rim. In the
inner aureole, the zoning of garnets is very weak. They are usually typical
almandine containing 2-6 per cent MnO and 33-37 per cent FeO.

Glen Clava, Scottish Highlands

The Dalradian sediments in the Glen Clova area were subjected to regional
metamorphism of the sillimantie grade of the Barrovian series. Subsequently,
adamellite, granodiorite, diorite and picrite of the Newer Granite suite were
intruded into them, producing polymetamorphic contact aureoles. In the inner
part of these aureoles, the amphibolite formed by regional metamorphism has
been transformed into two pyroxene-plagioclase hornfelses, and the metapelites
have been changed into orthoclase-cordierite-biotite-hornfelses often with hyper-
sthene, sillimanite, almandine, spinel or corundum. Thus, the inner aureoles
belong to the pyroxene-hornfels facies (Chinner, 1962), if not to the granulite
facies. These aureoles are different from the Oslo hornfels zones in the common
occurrence of sillimanite and almandine.
Almandine occurs commonly in the original regional metamorphic rocks.
When subjected to the contact metamorphism, the original almandine in some
rocks was entirely replaced by aggregates of cordierite, orthoclase, biotite and
spinel, whereas that in other rocks with higher Fe 2 +/Mg ratios partly perSisted in
the contact aureoles.
Almandine by itself is stable under the physical conditions of the pyroxene-
hornfels facies (fig. 7B-9). The composition field of the rocks which form
almandine, however, is more restricted under such conditions than in regional
metamorphism of the medium-pressure type. Hence, partial or complete decom-
position of regional metamorphic almandine takes place in contact aureoles, as
shown in fig. 7B-8.

10-3 CONTACT METAMORPHISM OF PELITES TO PRODUCE STAURO-

LITE AND KY ANITE

Contact metamorphism that produces staurolite and kyanite has been described
in some areas of the world. Since staurolite is stable over a wide range of
rock-pressure, staurolite-bearing aureoles do not necessarily represent particu-
larly high pressures. When kyanite occurs in contact aureoles, it is usually
associated with andalusite and sillimanite. Therefore, P-T conditions near the
triple point of the Al 2 SiO s system probably prevailed in a part of such aureoles.
 10-3 PRODUCING STAUROLITE AND KYANITE/283

The possibility of metastable crystallization of AlzSiO s minerals cannot be

precluded. Even if this is the case, however, the prevailing rock-pressure was
probably relatively high in kyanite-bearing aureoles, as will be discussed below.

Kwoiek Area, British Columbia, Canada

This area, about 130 km northeast of Vancouver, has a contact aureole formed
on the eastern flank of the Coast Range batholith. A roof pendant or screen of
metamorphic rocks, about 10 km wide, is bordered by hornblende-biotite or
biotite quartz diorite. The major rock type within the pendant is a meta-
graywacke. Metabasites containing pillow structure are also present.
The contact aureole along the major batholith border is about 6 km wide, and
the metamorphism ranges from the chlorite to the sillimanite grade (Hollister,
1969a b). Gamet, staurolite, kyanite and sillimanite are common and andalusite
has been found in one locality. The following progressive reactions took place in
aluminous metagraywackes:

muscovite + chlorite + quartz = garnet + biotite + Hz 0

muscovite + chlorite + ilmenite
= staurolite + garnet + biotite + quartz + H2 0
staurolite + quartz = garnet + sillimanite + H2 0.

Andalusite, kyanite and sillimanite can occur singly or in combinations of any

two. The most common pair is kyanite + sillimanite. The textural relations
suggest the existence of two sequences of polymorphic transformations:

(a) Andalusite -+kyanite -+sillimanite

(b) Andalusite -+ fibrolite (fme-grained mats of sillimanite needles).

Hollister (1969b) has considered that both sequences formed with nsmg
temperature at a similar and constant rock-pressure which is higher than the
pressure of the triple point of the Al 2 SiO s system (fig. 2-1). In sequence (a),
andalusite crystallized as a metastable form within the stability field of kyanite,
and subsequently stable kyanite formed, followed by stable sillimanite. In
sequence (b), andalusite crystallized first as a metastable form, and subsequently
fibrolite crystallized as another metastable form, both within the stability field
ofkyanite.
Gamet grains are usually strongly zoned. The MnO content is as high as
6-18 per cent in the grain centre, and decreases outwards to a value as low as 1-3
per cent, but may increase slightly in the outermost few microns, whereas the
FeO content increases outwards. The early formed garnet contains a large part of
the MnO present in the rock. Diffusion of material in garnet is negligible under
the conditions of this aureole, and so the gamet, once crystallized, is removed
284/PROGRESSIVE CONTACT METAMORPHISM

from the reacting system, with resultant depletion of MnO from the reacting
part of the rock. The new garnet layer to form in equilibrium with the
depleted surroundings should be very poor in MnO. Thus, strong zoning could
form even "without any appreciable change in the partition relation of elements
during the growth (Hollister, 1966, 1969a).
Staurolite shows a more complicated pattern of zoning. Sectors of a crystal
formed by growth in different directions have different compositions, and
within each sector the composition changes from centre to margin (Hollister,
1970; Hollister and Bence, 1967).
Biotite and chlorite are more homogeneous, and would have continuously
internally re-equilibrated during their growth with changing external conditions.
A series of Hollister's papers on rocks and minerals of this area represents a
successful use of the electron probe X-ray rnicroanalyser for better under-
standing of the mechanism of metamorphic reactions.

Donegal Granites, Northwestern Ireland

Caledonian granitic masses were intruded into already metamorphosed Pre-
cambrian pelitic rocks in Donegal on the northwest coast of Ireland. The
regionally metamorphosed pelites outside the contact aureoles are mostly
chlorite-muscovite schists, occasionally containing biotite. Some aureoles con-
tain staurolite and kyanite besides andalusite and sillimanite (Pitcher and Read,
1960,1963; Naggar and Atherton, 1970).
The Thorr pluton and another granodioritic mass have produced strong
metamorphic effects on the country rocks. The contact aureole of the Thorr
pluton is about 2 or 3 km wide. The plutoniC masses are heterogeneous. In the
aureole produced by the adamellites, the first mineralogical effect is the
appearance of chlorite porphyroblasts. Then, biotite begins to occur, being
followed by andalusite. In the advanced stage, cordie rite appears, and the rocks
becomes andalusite-cordierite hornfelses. Sillimanite occurs only in the inner-
most zone. In the aureole produced by dioritic rocks, on the other hand,
muscovite flakes and andalusite porphyroblasts are accompanied by cordierite
and some sillimanite in the outer aureole. In the inner aureole, sillimanite is
abundant, and the hornfelses are extensively mobilized.
The Ardara diapiric granodiorite is about 8 km across, consisting of a core of
granodiorite surrounded by a rim of foliated biotite-hornblende tonalite. It was
emplaced by a forceful shouldering aside of the country rocks. The contact
aureole is about 2 km wide. In the outermost part of the aureole, biotite
porphyroblasts appear. On a nearer approach to the intrusion, andalusite and
then sillimanite begin to occur. The narrow innermost zone is rich in cordierite.
The andalusite is often accompanied by staurolite, and very rarely by a little
kyanite. The sillimanite is accompanied by garnet.
 10-4 PARTIAL MELTING IN A CONTACT AUREOLE /285
The Main Donegal granite proper is the largest intrusion in this area, and is
exposed over an area of about 11 by 60 km. It was forcefully intruded by lateral
wedging combined with horizontal stretching. The mass is composed of grano-
diorite showing banding parallel to the steep country-rock walls. The banding is
interpreted as being due to flow of consolidating magma. The aureole is
1·5-2·5 km wide. The intruding magma exerted a marked outward push and
horizontal drag on the aureole rocks, producing folds, strain-slip cleavages and
schistosity in the aureole.
The metapelites in the outer aureole are characterized by porphyroblasts of
chlorite, muscovite and biotite. Choritoid was found at one locality. On a closer
approach to the contact, garnet, staurolite, kyanite, and andalusite begin to
occur. In the innermost part, sillimanite occurs. In this aureole, staurolite schists
are widespread. Kyanite schists are also common. Some schists contain both
kyanite and andalusite, or both andalusite and sillimanite. Cordie rite is virtually
absent.
It may appear that kyanite forms in aureoles subjected to higher mechanical
effects due to intrusion. However, Naggar and Atherton (1970) consider that its
formation depends not on the mechanical effects but mainly on the lower
temperature and the lower FeO/MgO ratios of the metamorphosed rocks. In
rocks with higher ratios, garnet and staurolite takes the place of kyanite.
In addition to the two areas reviewed above, the occurrence of kyanite and
staurolite in contact aureoles has been reported from the Burke area in the
northern Appalachians (Woodland, 1963) and from Ghana (Lobjoit, 1964).
Staurolite, andalusite and sillimanite occur in the contact aureole of the Santa
Rosa Range, Nevada, which has been described in detail by Compton (1960).

10-4 POSSIBLE EXAMPLE OF PARTIAL MELTING IN A CONTACT

AUREOLE

The Caledonian intrusion of the Cashel- Lough Wheelaun area is composed of

basic and ultrabasic rocks and was emplaced into pelitic rocks in Connemara
near the west coast of Ireland. The rocks outside the contact aureole are
Caledonian regional metamorphic rocks of the sillimanite grade. The aureole is
about 1·5 km wide, being composed of hornfelses and partly mobilized pelites.
The regional metamorphic rocks outside the aureole are biotite-andesine-
quartz-sillimanite-garnet schists. Within the aureole and as inclusions in the
intrusion occur cordierite-sillimanite-Iabradorite-magnetite-biotite hornfels
and spinel-magnetite-cordierite hornfels with occasional corundum or ortho-
pyroxene.
The compositional changes accompanying the formation of the hornfelses
were studied by Leake and Skirrow (1969) and by Evans (1964). Unhornfelsed
pelites and hornfelses have the following composition ranges:
286/ PROGRESSIVE CONTACT METAMORPHISM

Unhornfelsed pelites Hornfelses

per cent per cent
42·6-51'6 32-49
0·5-2·5 0·2-2·6
3·7-6·3 1·2·6·0.
It appears that Si, Na, K, Rb, es, ti, Ba and La were removed from hornfelses.
This change was considered due to partial melting of the pelites followed by
removal of the resultant acidic liquid leaving the solid residue behind. The
materials removed in an early stage of partial melting resemble K·rich granite in
composition, whereas those in higher temperatures resemble granodiorite.

10·5 PYROMETAMORPHISM OF PELITIC AND PSAMMITIC ROCKS

Pyrometamorphism takes place commonly in xenolithic inclusions in volcanic

rocks or minor intrusions, and rarely in the country rocks around minor
intrusions of basic or intermediate compositions. Intrusions giving such marked
heat effects to the country rocks could represent a conduit for continuously
rising magma. A continuous flow may have maintained a high temperature for a
considerable period of time.
A pyrometamorphic zone in country rocks is usually less than 20 m wide,
that is, much narrower than ordinary contact aureoles. Though the metamorphic
rocks show intense mineralogical reconstitution, they are commonly far from
chemical equilibrium. Minerals characteristic of the sanidinite facies such as
sanidine, anorthoclase, high-temperature plagioclase, tridymite, cristobalite and
mullite form. Partial melting takes place. Two examples of the progressive
pyrometamorphism of the pelitic and psammitic rocks will be reviewed below.

Sithean Sluaigh, Scotland

In Sithean Sluaigh in Argyllshire, Scotland, a gabbroic plug penetrates through
Dalradian phyllites. The plug, which is thought to be the remnant of a volcanic
pipe, is made up of a core of olivine-augite gabbros surrounded by a ring of
dolerite. The latter has been interpreted as having been intruded between the
gabbro and phyllite. The plug is about 100 m across.
The thermal aureole is about 10 m wide (Smith, 1969). The rocks outside the
aureole are phyllites of pelitic and psammitic compositions, sporadically con-
taining almandine and biotite. At 10-9 m from the contact on the northeastern
side of the plug, the muscovite of the phyllite has broken down to a mat of very
fine needles of sillimanite and mullite set in an amorphous matrix.
At 4-3 m from the contact, all the mica and chlorite have broken down to
give poorly crystallized cordierite and spinel. Sanidine rims form around the
decomposed mica. Almandine begins to break down to form a very fine-grained
 10·5 PYROMETAMORPHISM OF ROCKS/287

aggregate probably of orthopyroxene, cordierite and spinellides. So far, indi-

vidual grains tend to decompose as more or less isolated chemical systems.
At 1·6 m from the contact, the decomposed minerals tend to lose their
original outlines. Rims of sanidine and orthopyroxene surround quartz.
Breakdown products of muscovite are aggregates of spinel, corundum and
cordierite.
At 1-0'5 m from the contact, intergrowths of quartz and feldspar and
prismatic orthopyroxene are widespread. Diopsidic pyroxene begins to appear.
The grain size becomes coarser and granophyre veins occur.
At 15 cm from the contact, plagioclase grains possibly derived from igneous
rocks are rimmed by sanidine. Patches of brown devitrified glass occur inter-
stitially.
In other parts of the aureole close to the contact, mullite-spinel-pseudo-
brookite rocks and spinel-corundum rocks occur. Granophyre veins are
common, and some of them contain quartz paramorphs after tridymite.
The granophyre veins are not thought to be a magmatic differentiate but a
product of partial melting of phyllites. Some phyllites have a compositional
layering, and only the quartzo-feldspathic layers appear to have been melted to
form a granophyric text-ure. When the siliceous melts were removed, the solid
residues were highly desilicated aluminous and ferromagnesian rocks rich in
corundum, spinel and mullite

Karroo Dolerite, South Africa

A sill of the Karroo Dolerite near Heilbron, South Africa, has fused overlying
arkose (Ackermann and Walker, 1960). Such fusion is relatively rare for the
dolerite. The arkose is flat-lying and the dolerite sill is about 15 m thick. The
arkose is mainly composed of quartz, orthoclase and subordinate oligoclase.
Some almandine and ilmenite were found.
The rocks harden towards the contact. Vitrification of arkose is noticed within
a distance of 6 m from the contact. The first glass is formed where grains of
K-feldspar and quartz come into contact with one another. It is light in colour.
With advancing melting, the colour deepens and the refractive index rises. It
contains cordierite microlites. The fusion is accompanied by various mineral-
ogical changes including the transformation of orthoclase to a nearly uniaxial
sanidine and the formation of cordierite, tridymite and pyroxene.
Near the contact, pyroxene appears in the glass, and the original K-feldspar is
completely resorbed, while some of the quartz and plagioclase still remain. The
resultant rock is a green or black buchite containing up to 70 per cent of glass.
Mixing of partly fused sediments with fluid dolerite took place locally in a zone
less than 0·6 m wide.
288/ PROGRESSIVE CONTACT METAMORPHISM

10-6 CONTACT METAMORPHISM OF BASIC ROCKS

The existing data on metabasites in contact aureoles are not so abundant as

those on metapelites. Three aureoles will be reviewed here, two of which have a
metabasite zone with the assemblage actinolite + calcic plagioclase in a
low-temperature part.

lritono Area, Northeast Japan

The Iritono area is in the central Abukuma plateau, about 180 km NNE of
Tokyo. The intrusion is heterogeneous and mainly composed of granodiorite and
diorite. The rocks outside the contact aureole are regionally metamorphosed
basic schists in the greenschist facies in the eastern half of the area and in the
amphibolite facies in the western. The contact aureole, less than I km wide, has
been divided into three progressive zones A', C' and D' (Shido, 1958, pp.
199-201).

Zone A I. On entering the contact aureole, the metabasites are soon recrystallized
nearly completely and lose their original fissility. The outermost zone is
characterized by metabasites with the actinolite-calcic plagioclase assemblage
with or without chlorite, quartz and biotite. No epidote is present. The calcic
plagioclase is labradorite or andesine. In the higher temperature part of this
zone, some metabasites contain a little blue-green hornblende in association
with actinolite, and others contain diopsidic pyroxene. The associated meta-
psammites are biotite hornfelses with occasional muscovite or garnet.

Zone c'. This zone is characterized by metabasites with the hornblende +

plagioclase assemblage, occasionally containing biotite, cummingtonite and
diopsidic pyroxene. The hornblende is brownish green or brown. (Note that in
the regional metamorphism of the Abukuma Plateau a zone with blue-green
hornblende usually occurs between that with actinolite and that with brownish
green hornblende, whereas in this contact aureole a zone characterized by the
predominance of blue-green hornblende does not occur.)
The associated metapsammites in zone C' are biotite hornfelses commonly
with garnet and/or cummingtonite.

Zone D'. Where the dioritic facies of the intrusion is in contact with the country
rocks, orthopyroxene is formed in metabasites up to a distance of several metres
from the contact. This part is deSignated as zone D'. The rocks show the
orthopyroxene + clinopyroxene + hornblende + plagioclase assemblage com-
monly with cummingtonite and biotite. Orthopyroxene does not form in the
metabasites in contact with granodiorites.
 10-6 CONTACT MET AMORPHISM OF BASIC ROCKS /289
Arisu Area, Northeast Japan
As shown in fig. 10-1, three progressive metamorphic zones have been distin-
guished in terms of mineral changes in metabasites (Seki, 1957; especially
1961b, pp. 74-75).
1. Greenschist facies zone. Actinolite, chlorite, sodie plagioclase, muscovite,
biotite and epidote occur.
2. Actinolite-calcic plagioclase zone. This zone resembles zone A' of the
Iritono aureole. The plagioclase is andesine or labradorite. Epidote is rare
and chlorite is present.
3. Amphibolite facies zone. The metabasites are mainly composed of horn-
blende and plagioclase.
Arisu Area

I
Gree nsch i st Actinolite-calcic Amphibolite
Metabasites plagioclase zone facies
facies
Chlorite- I Biotite- Cordierite- Sillimanite
Zoning muscovite andalusite almandine zone
zone zone zone

Chlorite
Muscovite
Chloritoid --
Andalusite --
Biotite
V>
Anthophyllite
l!l
a; Cummingtonite
0-
....'" Cordierite
QJ
::2; Almandine
Corundum
Spinel
Microcline
Sillimanite

Chlorite ----
Muscovite
Epidote
V>
Tremolite
--
Q)
c
0 Hornblende
1;;
QJ
Diopside
E
:.:i Ca-garnet
Vesuvianite
Wollastonite
Scapolite
Microcline

Fig. 10-1 Progressive mineral changes in the Arisu contact aureole,

Kitakami Mountains, Japan. (Seki, 1957, 1961 b.)
290/ PROGRESSIVE CONTACT METAMORPHISM

Tilbuster Area, Australia

A Late Palaeozoic monzonite mass has been intruded into Palaeozoic formations
in the Tilbuster Area, New South Wales, Australia (Spry, 1955; Binns, 1965a).
The monzonite is mainly composed of plagioclase and orthoclase with subordinate
biotite, hornblende and quartz. The contact aureole is only about 0·5 km wide.
The pelitic rocks adjacent to the contact contain andalusite, cordierite and
orthoclase.
Basic lavas occur abundantly in the aureole. Some of them are basaltic and
others spilitic in composition. The first sign of metamorphism in the ordinary
basic lavas of the outer aureole is the formation of a very pale bluish green
hornblende with ragged or fibrous 'actinolite-like' habit by reaction between the
plagioclase and clinopyroxene of original basalts. A chemical analysis has shown
that it is a hornblende (Si =6·50 on 23 0). With advancing recrystallization, the
rock becomes a hornblende-andesine hornfels.
As the contact is approached, the hornblende becomes moderately deep
bluish green in colour and then dark green, its habit less ragged, and its amount
tends to decrease. Finally close to the contact, deep brown or brownish green
hornblende was formed as stout crystals. The Na20 content of the hornblendes
increases with temperature. Such hornfelses in the inner aureole are made up of
hornblende and plagioclase together with subordinate calcic clinopyroxene and
iron ore. No orthopyroxene was found. The plagioclase is usually andesine or
labradorite. Hornfelses derived from spilitic rocks contain sodic plagioclase. The
whole aureole is in the amphibolite facies.
Within a distance of 200 m from the contact many metabasites contain
grossular, which would have formed by reaction of calcite-bearing amygdales.
The aureole and especially its outer part is widely scapolitized. Scapolite occurs
as fillings of joints and veins and as a mineral replacing plagioclase in the
metabasites.

10-7 CONTACT METAMORPHISM OF LIMESTONES

Ordinary Contact Metamorphism of Limestones

Among the contact metamorphic rocks of the Oslo area, Norway (§ 10-1),
classes 7-10 of Goldschmidt (1911) were derived from impure limestones. Many
features are shared with regionally metamorphosed limestones.
Progressive contact metamorphism of limestones has been described in a
number of areas. For example, Seki (1961 b) has described a progressive series in
the Arisu area, Kitakami Mountains. The progressive changes are summarized in
fig. 10-1. As regards limestones, the first zone characterized by tremolite, the
next zone by diopside, and the third by Ca-garnet together with diopside.
Microcline also begins to occur in the third zone. In the innermost fourth zone,
wollastonite and vesuvianite occur. Scapolite is also confined to this zone.
 10-7 CONTACT METAMORPHISM OF LIMESTONES /291
It is interesting that within the second zone, there is a small area where
wollastonite occurs. No intrusion has been found in this area to account for the
appearance of wollastonite. It has been ascribed to a local decrease of PC02 •
The progressive sequence of calc-silicate minerals in the Arisu area is tremolite
~ diopside ~ Ca-garnet ~ wollastonite. This is very similar to that observed in
regional metamorphism, especially of the low-pressure type.

Advanced Decarbonation in Contact Metamorphism

There are a number of contact aure~les where de carbonation reactions have
proceeded to greater degrees than in the above case. In such aureoles, periclase,
monticellite, akermanite, tilleyite, spurrite, rankinite, merwinite, and larnite may
be produced (Bowen, 1940; Tilley, 1951).
Periclase marble or its hydration product (i.e. predazzite) appears to begin to
form in a high-temperature part of the pyroxene-hornfels facies in contact
metamorphism. W. Johannes determined the equilibrium temperature of the
following reaction for a wide range of the COz/Hz 0 ratio for the coexisting fluid
at 1 kbar (Winkler, 1967, p. 42):
dolomite =calcite + periclase + COz. (10-1)
The reaction takes place generally in the range of 700-850°C as shown in fig.
9-2. For more information about the hydration reaction of periclase, see, in
particular, Fyfe (1958).
Decarbonation reactions leading to the formation of such rare minerals as
monticellite, akermanite, spurrite and merwinite, have been experimentally
investigated in the presence of a nearly pure COz phase by Walter (1963, 1965)
and by Tuttle and Harker (1957). The eqUilibrium temperatures are considerably
higher than those for the reactions to form wollastonite and periclase.
Monticellite, akermanite, tilleyite, spurrite, rankinite, merwinite, and larnite
are products of extremely advanced decarbonation reactions. If these minerals
form in the presence of a fluid composed almost entirely of CO z , the
temperatures of formation would be high enough to class them in the
characteristic minerals of the sanidinite facies (Reverdatto, 1970). However,
they occur also in some contact aureoles that do not appear to have reached so
high a temperature. In some contact aureoles, COz could be actively removed
possibly in solution in flowing water. Such a situation would result in an
unusually low PC02 , leading to the generation of above-mentioned minerals.
In contact metamorphic limestones, Pco , and PH,o may be so highly
variable as to produce a great mineralogical diversity (Agrell, 1965).

Crestmore, California
The Crestmore limestone is a well-documented example showing advanced
decarbonation. It is situated about 90 km east of Los Angeles. Monticellite,
292/PROGRESSIVE CONTACT METAMORPHISM

gehienite, tilleyite, spurrite, and merwinite have been found (Burnham, 1959;
Walter, 1965; Carpenter, 1967).
Masses of dolomitic limestones were engulfed and metamorphosed by a
quartz-diorite pluton which is part of the Cretaceous Southern California
batholith. A layer, less than 1 m thick, of contact skarns mainly composed of
diopside, grossular and wollastonite formed between the limestone and quartz
diorite. Later, a small pipe-like body of quartz-monzonite porphyry was
intruded and a thicker skarn zone, up to about 20 m thick, formed between the
porphyry and the already metamorphosed limestone. The quartz-monzonite
porphyry is made up of andesine phenocrysts set in a microcrystalline ground-
mass of andesine, orthoclase and quartz with minor green clinopyroxene,
hornblende and biotite.
The limestones are now composed of alternating layers of coarsely crystalline
calcite marble and brucite marble. The brucite is a hydration product of
periclase.
The thicker skarn zone was formed by the metasomatic introduction of great
amounts of Si02 and Al 2 0 3 from the quartz-monzonite porphyry into the
surrounding limestones. It may be divided into the following three sub zones
listed in order from the contact outward:
1. Garnet subzone, consisting of uniform garnet-wollastonite-diopside rock.
2. Vesuvianite subzone, consisting of vesuvianite, garnet, wollastonite, diop-
side, monticellite, wilke rite , crestmorite, phlogopite, and xanthophyllite.
3. Monticellite subzone, consisting of monticellite, gehienite, tilleyite,
spurrite, merwinite, forsterite, scawtite and dark green spinel.
These sub zones have gradational contacts between one another. The outer
boundary of the monticellite subzone against the crystalline marble is very sharp
and represents the limit of Si and AI metasomatism.
Conceivably, a large amount of water escaping from the crystallizing quartz-
monzonite porphyry carried the Si0 2 and Al 2 0 3 to cause the metasomatism.
The subzones closer to the igneous body have higher Si0 2 and Al 2 0 3 contents.
The differences in mineral composition between subzones probably resulted
from the chemical differences due to metasomatism together with some
differences in Pcoz and temperature.
Chapter 11

Metamorphic Facies and

Facies Series

11-1 DEVELOPMENT OF THE CONCEPT OF METAMORPHIC FACIES

Eskola's Concept
Goldschmidt's (1911) study of the contact metamorphism of the Oslo area and
Eskola's (1914) study of the regional metamorphism of the Orijarvi region
revealed that well-recrystallized metamorphic rocks approached chemical equili-
brium at a stage of their history and that the mineral compositions they acquired
then have usually been preserved over long subsequent periods of time. In a
group of metamorphic rocks that have reached chemical equilibrium under the
same definite P-T conditions, the mineral compositions of the rocks depend only
on their bulk chemical compositions, regardless of their previous history. There
exists a regular relationship between the mineral and chemical compositions of
such metamorphic rocks. Eskola (1915) attempted to visualize the relationship
by his ACF and A'KF diagrams.
Since the metamorphism at Oslo differs in P-T conditions from that at
Orijiirvi, the relationship between the mineral and chemical compositions of the
rocks in the former area differs from that in the latter area. When rocks having
the same chemical compositions are subjected to metamorphism under different
P-T conditions, the mineral compositions of the resultant rocks can differ from
one another.
Eskola (1915) defined a metamorphic Jacies as including all the rocks
recrystallized under the same P-T conditions (range). Therefore, each meta-
morphic facies corresponds to a definite P-T range and is characterized by a
regular relationship between the chemical and mineral compositions of the rocks.
This has led to the establishment of a sound empirical classification of
metamorphic rocks in terms of P- T conditions. Eskola (1915, p. 115) wrote as
follows: 'A metamorphic facies includes rocks which ... may be supposed to
have been metamorphosed under identical conditions. As belonging to a certain
facies we regard rocks which, if having an identical chemical composition, are
composed of the same minerals. It may expressly be pointed out, that this
conception does not postulate any supposition whatsoever concerning the
294/METAMORPHIC FACIES AND FACIES SERIES
genetic, pre-metamorphic relations of the rocks, and that every facies may
include all possible chemical and genetical varieties.'
Eskola (1920) proposed five metamorphic facies named greenschist, amphi-
bolite, hornfels, sanidinite, and eclogite facies. In his final systematic account,
Eskola (l939) added the epidote-amphibolite, glaucophane-schist and granulite
facies, and changed the name of the hornfels facies into the pyroxene-hornfels
facies.
Eskola (1920, 1939) thought that there were igneous rocks having virtually
the same mineral assemblages as some metamorphic rocks. For example, horn-
blende and pyroxene gabbros have virtually the same mineral assemblages as
amphibolites and two pyroxene-bearing basic hornfelses, respectively. Therefore,
he formulated a group of igneous facies, corresponding to the metamorphic
ones, such as the hornblende gabbro facies and the gabbro facies. The term
mineral facies was proposed so as to comprise both metamorphic and igneous
facies (Eskola, 1920, 1939).
Though the main merit of the facies classification comes from the inference
that each facies represents a restricted range of temperature and pressure, Eskola
in his later years did not like to base the definition of metamorphic facies on
such an inference. Thus, Eskola (l939) attempted to give an excessively
positivistic definition which was based only on directly observable chemical and
mineral compositions. This attempt stemmed probably from the positivistic spirit
he had aquired in his youth under the influence of F. W. Ostwald, Ernst Mach
and others (cf. Eskola, 1954).

Metamorphic Facies and Progressive Metamorphism

Eskola's (1920) metamorphic facies was a static scheme of classification of
metamorphic rocks. Each facies was defined on the basis of paragenetic rela tions
observed in different parts of the world. For example, the amphibolite facies was
mainly based on the Orijarvi region, the pyroxene-hornfels facies on the Oslo
area, and the eclogite facies on some areas in southwestern Norway. Therefore,
he did not need to consider the problems related to the transitional relations and
boundaries between facies.
Vogt (1927) and Barth (l936) studied progressive metamorphism in relation
to metamorphic facies. Since the mineralogical changes are generally gradual, the
facies boundaries must be defined more or less arbitrarily. Vogt defined a
number of boundaries and contributed in particular toward the establishment of
the epidote-amphibolite facies between the greenschist and amphibolite facies.
In areas of intermediate or transitional states between two metamorphic
facies, rocks having mineral assemblages apparently characteristic of one and of
the other facies usually occur side by side. These different assemblages usually
result from slightly different chemical compositions. In areas transitional
between the amphibolite and the typical granulite facies, for example, amphi-
 11-1 CONCEPT OF METAMORPHIC FACIES/295
bolites and two-pyroxene-bearing metabasites usually occur intermingled, as
described by de Waard (1965) and Himmelberg and Phinney (1967).
There are wide regions exhibiting intermediate states between the greenschist
and the typical glaucophane-schist facies. Greenschists and glaucophane schists,
both of broadly basic composition, occur intermingled. Iwasaki (1963, pp.
32-3) stu~ied such an area in the Sanbagawa belt, and showed that glaucophane
schists had higher Fe 3 +/Fe 2 + ratios than the associated greenschists.
The greenschist and the amphibolite facies are very widespread. The tran-
sitional states from these to other facies (e.g. the glaucophane-schist or the
granulite facies) are treated as belonging to the latter facies in this book, since
such a definition reduces the sizes of the greenschist and amphibolite facies.
The gradual mineralogical changes are largely due to the effects of solid-
solution minerals in metamorphic rocks. Therefore, the precise definitions of
metamorphic facies boundaries must be based on an understanding of solid-
solution effects. This was emphasized in particular by Ramberg (1952, p. 136),
who defined a metamorphic facies as follows: Rocks formed or recrystallized
within a certain P-T field, limited by the stability of certain critical minerals of
defined composition, belong to the same metamorphic facies.

Mobile Components and Metamorphic Facies

Eskola (1915, 1920, 1939) did not clearly discuss the behaviour of H2 0 in
metamorphism. Presumably he considered that an aqueous fluid was present and
its pressure was equal to that acting on the solids. He had no clear idea of mobile
components.
H2 0, CO 2 and possibly some other components appear to be more or less
mobile in rock complexes undergoing metamorphic recrystallization. When their
mobilities are high, their chemical potentials become externally controlled
variables, just as are temperature and rock-pressure (Thompson, 1955). The
contents of such mobile components in a rock vary in response to external
conditions, whereas those of fixed components are intrinsic to the rock (§ 5-1).
Thus, rocks having the same chemical composition in terms of the fixed
components are recrystallized to different mineral compositions under different
external conditions. Under definite external conditions, a definite relationship
exists between the chemical composition of a rock in terms of the fixed
components and its mineral composition, which can be represented by
composition-paragenesis diagrams (Korzhinskii, 1959).
Thus, we come to the following defmition, essentially given by Korzhinskii
(1959, p. 64): Rocks which formed under such similar external conditions that
the relation between their chemical composition in terms of the fixed com-
ponents and their mineral composition is marked by the same regularities, can be
considered as belonging to the same metamorphic facies.
296/ METAMORPHIC FACIES AND FACIES SERIES

This defmition has real significance only when combined with the knowledge
of what components are highly mobile during metamorphism. If only H2 0 is
highly mobile, a metamorphic facies should represent a volume in the
T-Ps-PH20 space. If H2 0 and CO 2 are highly mobile, a facies should represent
a 'volume' in the T-Ps-PH20-PC02 'space'. (Whether CO2 may be regarded as
being mobile, depends partly on the areal extent of a metamorphic zone which
should be regarded as belonging to a facies, as will be commented upon below.)
Our knowledge on the mobility of CO 2 is very meagre. In this book, a
metamorphic facies is regarded tentatively as representing a volume in the
T-Ps-PH20 space. Figure 3-12 shows a rough projection of such volumes onto
the Ps- T plane.
A strictly defined relationship between the chemical and mineral composi-
tions of rocks holds only under a set of definite values of external conditions. A
metamorphic facies, as we now use it, does not represent such a set of definite
values, but represents a set of definite ranges of external conditions. Usually the
ranges are chosen so that a progressive metamorphic zone representing a
metamorphic facies could be shown on a map. Hence, the composition-
paragenesis diagrams corresponding to a metamorphic facies represent only
approximate or average relationships under a set of definite ranges of external
conditions.

Facies Names
Eskola's (1939) eight metamorphic facies are greenschist, epidote-amphibolite,
amphibolite, pyroxene hornfels, granulite, sanidinite, glaucophane-schist and
eclogite facies. Coombs et al. (1959) and Coombs (1960, 1961) proposed two
additional facies named zeolite and prehnite-pumpellyite facies. (The latter
facies was originally called the prehnite-pumpellyite-metagraywacke facies. We
delete 'metagraywacke' after recent usage.) Thus, we have ten facies names, and
we use only these facies names in this book.
That zone in contact aureoles which is characterized by the actinolite + calcic
plagioclase assemblage in metabasites (§ § 4-5 and 10-6) does not belong to any
of the above ten facies. The available data on it, however, are not sufficient
to establish a new facies. Therefore, such a zone is simply referred to as the
actinolite-calcic plagioclase zone in this book.
Many other facies names were proposed in the past. In certain progressive
metamorphic terranes, several mineral zones have been set up and a set of
metamorphic facies corresponding to the individual zones has been proposed.
However, the mineralogical characteristics of progressive. metamorphism are very
diverse. If we continue to propose a set of facies names for each area, the total
number of names will grow to a terribly large figure. Since we are not very
successful in determining the P- T relations between zones of differen t areas, the
growing number of facies names will simply mean an increasing burden to our
 11-1 CONCEPT OF METAMORPHIC FACIES/297

memory without any essential progress in our knowledge. I will refrain from any
more advanced subdivision of metamorphic facies.
It is very important to note that metamorphic facies have been defined on the
basis of the relationships between the chemical and mineral compositions of a
group of rocks, and not on their geologic settings. Certain zones of regional
metamorphism show virtually the same relationship between the chemical and
mineral compositions as certain zones of contact metamorphism. All such zones
should belong to the same metamorphi~ facies and should have the same facies
name regardless of the difference in their geologic settings.
If one wishes to distinguish a greenschist facies, for instance, of regional
metamorphism from that of contact metamorphism under two different facies
names, one must find a difference in paragenetic relations between the two
greenschist facies. Proposal of such different names is justifiable only when the
new names are defined on the basis of distinct paragenetic relations, and not on
geologic settings.

Classifications of Facies Series

Metamorphic facies series as observed in progressive metamorphic terranes all
over the world have been broadly classified into three baric types (§ 3-3). Facies
series belonging to the same baric type, however, are still diversified, and may
well be further classified into SUbtypes. The sound basis of such subdivisions will
be ultimately found in petrogenetic grids (§ 3-3). However, it is probably too
early to discuss such subdivisions from the viewpoint of a general clas~ification of
metamorphism.
A very tentative fivefold classification mainly based on the field association
of facies is shown in table 11-1, where each of the low-pressure and high-pressure
types is divided into two subtypes. However, the basis of this classification is not
strong enough. Therefore, we will use the threefold classification based on the
baric types in this chapter.
The zeolite, prehnite-pumpellyite, greenschist, epidote-amphibolite and
amphibolite facies represent P-T fields so wide as to extend over two or three
baric types. That part of such large facies which belongs to one baric type will be
treated as a subfacies, and will be deSignated by such names as either the
low-pressure sub facies of the greenschist facies or the low-pressure greenschist
facies. Zwart et al. (1967) and den Tex (1971) adopted the term facies group to
represent what is here regarded as a metamorphic facies extending over two or
three baric types. This terminology may prove to be useful in future.
In the following classification and descriptions of metamorphic facies, it is
assumed beforehand that one progressive sequence of mineral zones belongs to
one baric type. For example, if andalusite occurs in a part of a progressive
sequence of zones, the whole sequence is assumed to belong to the low-pressure
type. A greenschist facies zone within the sequence is regarded as belonging to
Table 11-1 Tentative fivefold classification of metamorphic facies series
N
\0
Name Occurrence Examples Facies series 00

a::
-
Low- (a) Some zeolite-facies (a) Akaishi Mountains Zeolite -+ prehnite-pumpellyite -+ tz1
>-l
pressure terranes (Matsuda and Kuriyagawa, greenschist -+ actinolite calcic >
1965) a::
0
(b) Contact metamorphism (b) Oslo (Goldschmidt, 1911), plagioclase zone -+ amphibolite ::0
at lower Ps Iritono (Shido, 1958), -+ pyroxene hornfels -+ sanidinite ."
::I:
.....
Arisu (Seki, 1961b) (j
(c) Pyrometamorphism (c) Sithean Sluaigh (Smith, 1969) ..."
Low- (a) Some zeolite facies (a), (b) Tanzawa Mountains Zeolite -+ prehnite-pumpellyite-+ >
(j
pressure terranes (Sekietal. 1969a) greenschist -+ amphibolite -+ .....
tz1
II (b) Contact metamorphism (b) Nakoso (Shido, 1958) granulite rn
at higher P s (c) Shiojiri-Takato (Ono, >
Z
1969a, b;Katada, 1965), 0
(c) Low-pressure Aracena (Bard, 1969), ..."
regional metamorphism Uusimaa (Parras, 1958) >
(j
.....
(a) Some zeolite facies (a) Taringatura (Coombs Zeolite -+ prehnite-pumpellyite-+ tz1
Medium- rn
pressure terranes et al., 1959) greenschist -+ epidote-amphibolite rn
(b) Regions of Barrovian (b) Scottish Highlands -+ amphibolite -+ granulite tz1
::0
.....
zones (Wiseman, 1934; Harker, tz1
1932) rn
Northern Appalachians
(Thompson and Norton,
1968)
High- Glaucophanitic terranes Bessi-Ino (Banno, 1964), Prehnite-pumpellyite -+ glaucophane-
pressure without a jadeite-quartz limori (Kanehira, 1967), schist -+ greenschist -+ epidote-
I zone Katsuyama (Hashimoto, 1968) amphibolite -+ amphibolite;
Lepontine Alps (Niggli, 1970) possibly eclogite in some parts
High Glaucophanitic terranes Kanto Mountains (Seki, 1958) (Prehnite-pumpellyite -+) glaucophane-
pressure with a jadeite-quartz Panoche Pass (Ernst, 1965) schist -+ greenschist; possibly
II zone eclogite in some parts
 11-2 METAMORPHIC FACIES OF LOW-PRESSURE TYPE/299

the low-pressure subfacies of the greenschist facies, even though a conclusive

paragenetic criterion may not be availllble for it.
Since this assumption has not been proved, the characterization of many of
the subfacies given below is incomplete from the logical viewpoint. Moreover,
subfacies cannot be regarded as having been established, so long as strict
defmitions are not given on the basis of paragenetic relations. Accordingly, the
classification and description given below are entirely tentative. A great deal of
work is needed in future to prove or disprove the basic assumptions.

11.2 METAMORPHIC FACIES OF THE LOW-PRESSURE TYPE

Zeolite Facies
This is the low-pressure sub facies of the zeolite facies. The zeolite facies rocks of
the Akaishi and Tanzawa Mountains and the Mogami area, all in Japan, probably
belong to this sub facies (§ 6A-2).
We have discussed in § 68-1 that the progressive metamorphism of Ca-zeolite
takes place in four steps, which are characterized by the follOWing zeolites:
(a) Stilbite, chabazite, gismondine
(b) Heulandite, chabazite, thomsonite, (scolecite)
(c) Laumontite, scolecite, thomsonite
(d) Wairakite, thomsonite
In each step, only the first-mentioned zeolite could ceoxist with quartz in the
presence of a fluid of a virtually pure H2 0 composition.
The associated phyllosilicates are smectite, mixed-layer minerals (smectite-
vermiculite-chlorite) and celadonite. Discrete crystals of chlorite can occur
only in the high temperature part.
Our knowledge of the paragenetic relations of minerals in this subfacies is
meagre.

Prehnite-Pumpellyite Facies
The low-pressure sub facies of the prehnite-pumpellyite facies occurs probably in
the Akaishi and Tanzawa Mountains (§ 6A-2). Prehnite, pumpellyite, epidote,
chlorite, muscovite and albite are stable.

Greenschist Facies
In burial metamorphic terranes of the Akaishi and Tanawa Mountains, a zone of
the greenschist facies occurs (§ 6A-2). In contact aureoles, a zone of the
low-pressure greenschist facies has only been observed rarely (§ 4-5). In
low-pressure regional metamorphism, on the other hand, a zone of the green-
schist facies is well developed in many areas, as exemplified by the Shiojiri area
(fig. 3-4) of the Ryoke metamorphic belt.
300/ METAMORPHIC FACIES AND FACIES SERIES
Albite, chlorite, actinolite, muscovite and epidote are widespread. Muscovite
may be phengitic. Biotite occurs throughout the zone of this subfacies in many
areas (e.g. Katada, 1965). In other areas, however, the zone of this subfacies can
be subdivided into a chlorite zone (without biotite) and a biotite zone (e.g.
Tobschall, 1969). Garnets of the almandine-spessartine series may occur in some
manganiferous rocks.

Actinolite-Calcic Plagioclase Zone and Epidote-Amphibolite Facies

In the greenschist facies metabasites, the plagioclase is albite and the calcic
amphibolite is actinolite, whereas in the amphibolites the plagioclase is inter-
mediate or calcic and the calcic amphibole is hornblende.
In many contact aureoles, the change of plagioclase takes place at a lower
temperature than that of calcic amphibole, thus resulting in the formation of a
zone of the actinolite-calcic plagioclase assemblage in the area intermediate
between those of the two facies (§ 4-5). In low-pressure regional metamorphism,
on the other hand, the two mineralogical changes usually take place nearly at the
same temperature, thus resulting in a direct transition from the greenschist to the
amphibolite facies (§ 3-3), whereas in medium-pressure regional metamorphism,
the change of calcic amphibole usually takes place at a lower temperature than
that of plagioclase, thus resulting in the formation of epidote-amphibolite in a
zone intermediate between those of the two facies.
However, the two changes are controlled by a number of factors, and hence
an epidote-amphibolite facies zone could form in some cases of low-pressure
metamorphism.

Amphibolite Facies
The low-pressure sub facies of the amphibolite facies is very widespread in
contact as well as in low-pressure regional metamorphism. Hornblende,
cummingtonite and clinopyroxene occur in metabasites. The colour of horn-
blende changes with increasing temperature from blue-green through green to
brownish green and brown. Cummingtonite and clinopyroxene (usually salite)
are more common in the high-temperature part. The plagioclase is usually
oligoclase, andesine or labradorite.
In metapelites, the stable Al 2 SiO s mineral is andalusite in the low-
temperature part and sillimanite in the high. Muscovite is stable in the lower part
of this subfacies, and biotite is present throughout. Cordierite and almandine are
common only in metapelites. The stable K-feldspar is micro cline in the
low-temperature part, but appears to be orthoclase in the high.
Calcareous rocks may contain wollastonite, grossularite, diopside and some-
times vesuvianite.
 11-2 METAMORPHIC FACIES OF LOW-PRESSURE TYPE/301

C~----------~~~---A·
~ ________~~~~~~.F
Calcite Cummingtonite
(a) Greenschist facies (b) Amphibolite facies

wolg·st...o-n-:-it-e----::-:-"'~-:---::--:----..".
(c) Pyroxene-hornfels facies (d) Granulite facies

(e) Sanidinite facies

Fig. 11-1 ACF diagrams for the low-pressure facies series.

302/METAMORPHIC FACIES AND FACIES SERIES

Pyroxene-Hornfels Facies
This facies is well developed in the Oslo and Comrie contact aureoles. Ortho- and
clinopyroxenes occur in metabasites. Hornblende may accompany them in the
low-temperature part (§ § 10-1 and 10-2).
In metapelites, the stable Al2 SiO s mineral is usually andalusite, but silli-
manite may occur in some cases. Muscovite has decomposed, but biotite is still
stable. The stable K-feldspar is orthoclase. Almandine is usually lacking, but may
occur in some metapelites with high Fe 2 +/Mg ratios, as in the Glen Clova
aureole (§ 10-2).
In calcareous rocks, wollastonite and diopside are stable. Ca-garnet may not
be stable in the presence of quartz, though it is stable by itself (§ 10-1).

Granulite Facies
The granulite facies resembles the pyroxene-hornfels facies in that they are both
characterized by two pyroxenes produced by decomposition of hornblende in
metabasites, but the former represents higher rock-pressures than the latter. It
has usually been assumed that the former belongs to regional metamorphism
whereas the latter to contact. However, since some contact aureoles show the
same facies series as low-pressure regional metamorphism, the above assumption
is not tenable.
In pyroxene-hornfels facies metapelites, cordierite is common and almandine
occurs only rarely, whereas in granulite facies metapelites of low-pressure
regional metamorphism, not only cordie rite but also almandine is widespread.
The pyrope content of almandine, however, is below 30 per cent, that is, lower
than that in the medium-pressure granulite facies. The stable Al 2 SiO s mineral is
sillimanite.
At present our knowledge of the mineral parageneses of rocks of the
pyroxene-hornfels and granulite facies is not precise enough to permit us to
define the boundary between the two facies precisely. In future, the boundary
might well be defined on a mineralogical basis such that all the two-pyroxene-
bearing zones of the low-pressure types of regional metamorphism are included
in the granulite facies. If this is done, contact metamorphism showing the same
facies series as the low-pressure regional metamorphism could produce granulite
facies rocks if the temperature becomes high enough, and contact meta-
morphism under lower pressures could produce pyroxene-hornfels facies rocks.
Most low-pressure regional metamorphic terranes such as the Ryoke belt do
not reach the temperature range of the granulite facies. However, the granulite
zone was well described in the low-pressure regional metamorphic terrane of the
Aracena area, Spain (§ § 7 A-4 and 8A-3). Here biotite and hornblende are stable
up to the highest temperature part. If the temperature of metamorphism had
become still higher, these minerals would have disappeared. The well-
 11-2 METAMORPHIC FACIES OF LOW-PRESSURE TYPE/303
documented granulite-facies rocks of Lapland (Eskola, 1952) and Uusimaa
(Parras, 1958), both in Finland, probably belong to the low-pressure type.
The mutual relation between the sub facies of the granulite facies as well as
their relation to the eclogite facies will be discussed in the next chapter (§ 12-2).

Sanidinite Facies
The sanidinite facies has been considered to represent a P- T range of pyrometa-
morphism, which is nearly the same as that of crystallization of volcanic rocks
(§ 10-5). The range is probably on the high-temperature side of the pyroxene-
hornfels facies. The sanidinite facies has minerals characteristic of volcanic
conditions, such as tridymite, cristobalite, sanidine, anorthoclase and high-
temperature plagioclase. The rocks of this facies occur in inclusions within
volcanic rocks and some minor intrusions as well as in some country rocks
adjacent to the latter (e.g. Thomas, 1922; Agrell and Langley, 1958).
In this facies, the alkali feldspars form a continuous series of solid solution
from the K to the Na end-member. Orthopyroxene and/or pigeonite may occur.
Sillimanite and/or mullite may form. Wollastonite forms a solid solution (Ca,
Fe 2 lSi0 3 • Glass may occur.
The sanidinite facies has usually been considered to represent a very low
pressure. Certainly tridymite and cristobalite are characteristic of low pressures.
However, it is not clear whether all the rocks of this facies were formed at such
low pressures. Pyrometamorphic rocks are mostly out of equilibrium, and hence
the determination of paragenetic relations in this facies is very difficult.
The definition of the low-temperature limit of the sanidinite facies has not
been clearly discussed. Sanidine and the continuous series of alkali-feldspar solid
solutions appear to be stable above about 600-650°C (Tuttle and Bowen,
1958). The temperature of 600-650 °c, however, is probably in the amphibolite
facies (§ 3-5). Accordingly, sanidine and alkali-feldspar solid solution would be
stable in the pyroxene-hornfels and perhaps also in the granulite facies. Some
K-feldspars in rocks of the pyroxene-hornfels facies in the Comrie area (Tilley,
1924) and of the granulite facies in the Wilmington complex (Ward, 1959) have
been observed to show sanidine-like optics. It is conceivable that K-feldspars in
pyroxene-hornfels and granulite facies rocks were sanidine at the peak of
metamorphic temperature, but changed into orthoclase or even into micro cline
at a declining stage of metamorphism. If so, the presence of sanidine simply
indicates that the relevant rocks were more rapidly cooled than some orthoclase-
bearing ones, and not necessarily that they formed at higher temperatures.
For the same reason, high-temperature plagioclase also would not be a safe
indicator of metamorphic facies. The stability relations of pigeonite and mullite
are in dispute. Thus, the precise defmition of the sandinite facies is yet to be
found.
304/METAMORPHIC FACIES AND FACIES SERIES

11-3 METAMORPHIC FACIES OF THE MEDIUM-PRESSURE TYPE

Zeolite Facies
The classical zeolite-facies terrane of Taringatura (§ 6A-4) would belong to the
medium-pressure subfacies, as also would the terrane in the central Kii Peninsula
(§ 6A-4).
It has been shown that the four dehydration steps in Ca-zeolite assemblages
are realized in some low-pressure zeolite facies areas. On the other hand, only
two steps characterized by heulandite and by laumontite were found in
Taringatura, and only one step characterized by laumontite was observed in
Central Kii. The analcime + quartz assemblage is stable in the heulandite step,
while albite occurs instead in the laumontite step.
Smectite and celadonite are associated with zeolites. Chlorite and prehnite
were found in the zeolite facies zone of central Kii, but not of Taringa tura.

Prehnite-Pumpellyite Facies
In the prehnite-pumpellyite facies zone of the Taringatura area, prehnite,
pumpellyite, chlorite and albite are stable, whereas in that of central Kii, epidote
is stable in addition. With rising temperature, prehnite disappears and actinolite
appears instead. A further rise in temperature causes the disappearance of
pumpellyite, thus resulting in a greenschist facies zone.

Greenschist Facies
This is the medium-pressure sub facies of the greenschist facies, as observed in the
central Kii Peninsula and in the chlorite and the biotite zone of the Barrovian
series.
Chlorite, actinolite, epidote, albite and calcite are common in metabasites.
Muscovite and biotite occur in some metabasites and more commonly in
metapelites. The occurrence of biotite in metapelites is confined to the higher
temperature part. Thus, a chlorite-zone and a biotite-zone subfacies may be
distinguished. The muscovite is usually phengitic in composition. Stilpnomelane,
chloritoid, and garnets of the almandine-spessartine series are common.
In Eskola's (1939) defmition, actinolite was erroneously removed from the
list of greenschists-facies minerals. Really, it is common in metabasites of this
facies. Under a high Pcoz ' actinolite is decomposed to form calcite and chlorite.
With a further increase in Peoz ' calcite and chlorite would react to form
dolomite or ankerite, as has been described for the Woodsville quadrangle
(§ SA-6). Therefore, actinolite greenschist represents a relatively low Pco 2 if
all the other factors are the same. The name of the actinolite-greenschist facies
has been used in some cases in order to emphasize the presence of actinolite in
relevant greenschist facies rocks.
 11-3 METAMORPHIC FACIES OF MEDIUM-PRESSURE TYPEj30S

Whether rocks recrystallized at the same T, Ps and PH, 0, but at different

Pco , ' belong to the same metamorphic facies depends mainly on the mobility
of CO 2 , If CO 2 is highly mobile, not only T, Ps and PH , 0 but also Peo ,
should be externally controlled. According to Korzhinskii's defmition (§ 11-1),
variations in these conditions could lead to different metamorphic facies. If CO 2
is not so mobile, it may be regarded as a fixed component. Though this problem
has not been well investigated, it seems that rocks recrystallized at different
Pco , occur intermingled within a mineral zone of an ordinary size in many
cases. If it is the case, they might well be classed in the same metamorphic facies.

Epidote-Amphibolite Facies
The medium-pressure subfacies of the epidote-amphibolite facies roughly
corresponds to the almandine zone of the ordinary Barrovian series. It corres-
ponds, however, to the higher chlorite and the biotite zone in the Woodsville
quadrangle (§ 8A-6).
This facies is characterized by the albite + epidote + hornblende assemblage in
metabasites. The hornblende is usually blue-green. Since the problem of the
peristerite solvus (§ 78-8) has not been settled yet, it is not possible to make an
unequivocal statement concerning the composition of the 'albite'. Almandine
may occur in metabasites with high Fe 2 +jMg ratios.
Since actinolite occurs in greenschist facies metabasites and coexists with
hornblende in the transitional zone to an epidote-amphibolite facies zone, the
boundary between the greenschist and the epidote-amphibolite facies lies at the
temperature where the dominant calcic amphibole changes from actinolite to
hornblende (§ 8B-4).
In Al 2 0 r poor rocks such as limestones, actinolite or tremolite may occur
even in this and the next facies.
Metapelites contain quartz, plagioclase, biotite, muscovite, paragonite,
chlorite, and sometimes chloritoid.

Amphibolite Facies
The medium-pressure subfacies of the amphibolite facies roughly corresponds to
the kyanite and the sillimanite zone in the Barrovian series.
The metabasites of this facies are mainly composed of plagioclase and
hornblende. The plagioclase is usually andesine or labradorite, and the horn-
blende is green or brown. Epidote is absent and salite may occur. Almandine
occurs in amphibolites of this sub facies of some areas. However, the occurrence
of almandine depends on the Fe 2 +jMg ratio, po. and other factors, and the
mineral is absent in amphibolites of this sub facies of other areas (§ 8B-7).
Metapelites in the low-temperature part of the amphibolite facies contain
muscovite, biotite, almandine, kyanite and staurolite. In the high-temperature
part, muscovite is decomposed by reaction with quartz, sillimanite takes the
306/ MET AMO RPHIC FACIES AND FACIES SERIES

(a) Greenschist facies (under low PC02) (b) Epidote-amphibolite facies

c..._-------'....- ~~~ ____~~________~F

C21cite Diopside Orthopyroxene

(c) Amphibolite facies (d) Granulite facies

Fig. 11-2 ACF diagrams for the medium-pressure facies series.

place of kyanite, and staurolite disappears. Biotite and almandine are stable as
far as the highest part.
In most of this sub facies , K-feldspar cannot coexist with kyanite, sillimanite
and almandine, whereas it can in the highest temperature part where the
muscovite + quartz assemblage is no longer stable.
In calcareous rocks, diopside and Ca-garnet occur commonly, and
wollastonite occurs occasionally (§ 9-3).

Granulite Facies
The medium-pressure sub facies of the granulite facies is characterized by
metabasites with ortho- and clinopyroxenes formed by the decomposition of
hornblende. Biotite also tends to decompose to form orthopyroxene. The
decomposition reactions are gradual, and hence usually there is a wide transi-
tional zone where hornblende, biotite, orthopyroxene and clinopyroxene co-
 11-3 METAMORPHIC FACIES OF MEDIUM-PRESSURE TYPE/307
exist. Typical granulite facies areas with no amphibole and mica are rare. It has
therefore been proposed that the granulite facies be divided into two sub facies: a
low-temperature subfacies with hornblende and biotite, and a high-temperature
sub facies without them (Turner and Verhoogen, 1960; de Waard, 1965).
Hornblende and biotite are stable to a higher temperature in quartz-free rocks
than in quartz-bearing ones (fig. 12-2).
In metabasites of the medium-pressure-granulite facies, ortho- and clino-
pyroxenes are strongly pleochroic and are sometimes high in Al 2 0 3 content.
Plagioclase is common, but tends to decrease in amount in garnetiferous
varieties, since garnet forms partly at the expense of plagioclase.
Almandine-pyrope garnet is common and sometimes abundant in metabasites
as well as in metapelites. Metapelites are mainly composed of quartz, plagioclase,
K-feldspar, garnet, and usually sillimanite and biotite. Probably orthoclase or
sanidine is the stable form of K-feldspar, but microcline is also common,
conceivably owing to retrogressive change.
In the low-pressure sub facies of the granulite facies, cordierite is common and
the Fe 2 +/Mg ratio of almandine is high, whereas in the medium-pressure
subfacies, cordie rite is absent, and the Fe 2 +/Mg ratio of almandine can be lower.
The pyrope content of almandine may be as high as 55 per cent (Eskola, 1957).
The high pressure stability limit of cordierite is close to that of sillimanite (fig.
7B-7). Accordingly, the usual absence of cordierite in the medium-pressure
granulite facies would not be due to the instability of cordierite by itself but
would be due mostly to the dwindling of the rock-composition field over which
cordierite is stable with rising pressure.
The mineral assemblages of calcareous metamorphic rocks are not well
investigated. The calcite + quartz assemblage is stable in some cases, but
wollastonite in others. The grossular + quartz assemblage may be unstable, but
Ca-garnet by itself is stable (fig. 9-3).
Since partial melting in the presence of an aqueous fluid begins to occur in
the amphibolite facies, metamorphism at higher temperatures should have caused
a considerable extent of melting provided that an aqueous fluid was present in
the rocks. Therefore, some granulite-facies rocks should represent solid residues
of partial melting, and others should have recrystallized without melting under
the conditions of PH, 0 ~Ps'
Granulite facies metamorphic rocks occur commonly in Precambrian terranes
in association with anorthosite massifs, which are sometimes of batholithic
dimensions. The genetic significance of the association is not clear, though some
anorthosites may represent solid residues of partial melting.
Sometimes areas of the granulite facies merge into amphibolite facies regions.
More commonly, such areas are in fault contact with other rocks and granulite
facies rocks formed at greater depths could have been brought up into
juxtaposition with overlying rocks. Though most granulite facies rocks occur in
308/ METAMORPHIC FACIES AND FACIES SERIES
shield regions, some are present in Phanerozoic orogenic belts. Many of the
latter, however, appear to be uplifted basement rocks.
Oliver (1969) has given a review of the modes of occurrence of granulite-
facies rocks in the world (fig. 4-3).

11-4 METAMORPHIC FACIES OF THE HIGH-PRESSURE TYPE

Prehnite-Pumpellyite Facies
The high-pressure sub facies of the prehnite-pumpellyite facies has been ob-
served, for example, in the Katsuyama area (§ 6A-S). Prehnite, pumpellyite,
epidote, albite, chlorite, actinolite and quartz occur in metabasites, whereas
muscovite, chlorite, prehnite, epidote, albite and quartz occur in metapelites. This
facies resembles the greenschist facies except for the presence of prehnite and
pumpellyite. A transitional state between this subfacies and the glaucophane-
schist facies is commonly observed in the lowest temperature part of typical
high-pressure metamorphism in California (e.g. fig. 3-8).

Glaucophane-Schist Facies
Lawsonite and the jadeite + quartz assemblage occur in the typical parts of the
glaucophane-schist facies. Glaucophane (including crossite) forms not only in
such typical parts but also in wide transitional states toward the prehnite-
pumpellyite, greenschist and epidote-amphibolite facies. In this book, all such
transitional states are included in the glaucophane-schist facies. Moreover, the
stability of lawsonite is so strongly influenced by PH, 0 and PCO 2 that the
mineral cannot be a reliable indicator of rock-pressure (§ 4-4).
The temperature conditions for the facies are especially variable, ranging
probably from 20(}-SSO°C, thus covering the temperature ranges of the
prehnite-pumpellyite, greenschist and epidote-amphibolite facies, but being on
the high-pressure side. It appears that the formation of lawsonite and the jadeite
+ quartz assemblage is confined to temperatures below 400°C (Taylor and
Coleman, 1968).
In addition, aragonite, chlorite, actinolite, muscovite (phengitic), paragonite,
stilpnomelane, albite, pumpellyite, epidote, chloritoid, spessartine-almandine,
piemontite, sphene, and rutile occur in this facies. Some metamorphic rocks and
especially metacherts contain aegirine, riebeckite, piemontite, or deerite. Many
so-called glaucophane schists are nearly devoid of schistosity.
Nearly all glaucophane schists occur in Palaeozoic and younger orogenic belts.
They are usually associated with serpentinite belts. Van der Plas (1959) has given
a comprehensive review of the localities of glaucophane and lawsonite in the
world (fig. 4-2).
 11-4 METAMORPHIC FACIES OF HIGH-PRESSURE TYPE /309
Greenschist Facies
A zone of the greenschist facies occurs on the high-temperature side of a zone of
the glaucophane-schist facies in many areas. Chlorite, actinolite, calcite, epidote,
albite and quartz occur in metabasites, whereas chlorite, muscovite, albite and
quartz occur in metapelites. It is noted that the common occurrence of
stilpnomelane and the absence of biotite characterize the high-pressure sub facies
of the greenschist facies as distinct from its medium- and low-pressure sub facies.

Epidote-Amphibolite Facies
The high-pressure sub facies of the epidote-amphibolite facies is known, for
example, in the Bessi-Ino area (§ 8A -8) and the limori area (Kanehira, 1967),
both of the Sanbagawa belt.
The assemblage albite + epidote + hornblende occurs in metabasites. These
may be associated with chlorite, diopside, garnet, muscovite and biotite.
Metapelites contain chlorite, biotite, muscovite, almandine, epidote, albite,
calcite and quartz.
This subfacies resembles the medium-pressure subfacies of the same facies.
However, epidote amphibolites of the high-pressure series commonly contain
muscovite and sometimes barroisite, whereas those of the medium-pressure series
do not.

Amphibolite Facies
A wide zone of the amphibolite facies exists in the Pennine nappes of the
Central Alps (§ 7A-8; also Wenk and Keller, 1969). The metapelites there
contain not only biotite and almandine, but also staurolite, kyanite and even
sillimanite. It is conceivable that this zone belongs to the medium-pressure type,
though glaucophane occurs in other parts of the Central Alps. No amphibolite
facies zone has been reported from typical high-pressure terranes.
Chapter 12

Eclogites and the Eclogite Facies

12-1 ECLOGITES AND ESKOLA'S CONCEPT OF THE ECLOGITE FACIES

The rock name eclogite was coined by Haiiy, a founder of crystallography, in

1822. Since the original definition of the name was vague, its meaning was
gradually settled through its usage in later years. According to the generally
established usage in the first half of the twentieth century, it means a rock
mainly composed of omphacite and almandine-pyrope-grossular garnet.
Omphacite is a clinopyroxene containing diopside, hedenbergite and jadeite
components, and quite frequently some acmite. However, no rigid limitations
were given for the composition ranges of these minerals. Thus, some authors
called diopside-almandine rocks eclogite. Small amounts of quartz, kyanite,
olivine, orthopyroxene, amphibole and rutile may be present, but not plagio-
clase.
Eskola (I920) proposed the name eclogite facies on the basis of this wide
definition of eclogite. Since eclogites are basaltic in chemical composition, their
unique mineral composition was ascribed to the unique P-T conditions of their
formation. Not only the eclogites but also other rocks formed under the same
P-T conditions as eclogites were claimed to belong to the eclogite facies. As
examples of the latter rocks, he referred to dunite, enstatitite, pyrope-olivine
rock, chloromelanitite and jadeitite. Even taking all these rocks into considera-
tion, the composition range is very limited. Rocks of other compositions were
unknown in this facies.
Clinopyroxenes high in diopside and hedenbergite components as well as
garnets of the almandine-pyrope series are stable in the granulite facies,
however. The assemblage of these two minerals, which may be called eclogite in
the wide definition, certainly occurs in some granulite-facies areas (e.g. David-
son, 1943). Eskola's idea of the eclogite facies was mainly based on his study of
eclogites in southwestern Norway (Eskola, 1921). Some of these eclogites also
appear to belong to the granulite facies. Therefore, if the eclogite facies is
characterized by traditionally defined eclogite, a part of it overlaps a part of the
granulite facies. Since this is very undesirable, we have to change the definition.
Clinopyroxene-almandine rocks occurring in high-pressure terranes have been
widely referred to as eclogites. Many of them, if not all, appear to have formed
 12-2 THE GRANULITE AND THE ECLOGITE FACIES/311

under the same Ps-T conditions as the associated glaucophane schists. However,
the eclogite facies differs from all other metamorphic facies in that it has not
been found as characterizing a zone in a progressive metamorphic sequence in
any area of the world. Thus, the relations of the eclogite facies to the granulite,
glaucophane-schist and other facies are not easy to solve.
Judging from the diversity of the modes of occurrence of eclogite facies
rocks, the P-T field corresponding to this facies is probably very wide as
compared with other facies. It may be divided into some subfacies.

12-2 RELATION BETWEEN THE GRANULITE AND THE ECLOGITE

FACIES

Green and Ringwood's Experimental Study

Green and Ringwood (1967) made an experimental study on the transformation
of gabbros into eclogites under virtually anhydrous conditions. Plagioclase,
pyroxene and possibly olivine are the main constituents of gabbros at low
pressures. With an increase in pressure, garnet begins to occur and then increases
in amount, whereas plagioclase and olivine decrease and disappear. Thus, a
pyroxene-garnet rock, i.e. eclogite, forms.
The P-T relations of such a transformation for quartz-tholeiitic compositions
are shown in fig. 12-1. For a medium value of the Fe 2 +/(Mg + Fe2 l ratio, the
P-T field below curve B in the figure represents a gamet-free granulite or gabbro
assemblage. In this pressure range, the An component of plagioclase is increas-
ingly incorporated into pyroxenes with increase in pressure. This results in the
formation of highly aluminous pyroxenes containing Tschermak's molecule: (Ca,
Mg) Al 2 Si0 6 • In the pressure range between curves Band 'Eclog', the reaction
between orthopyroxene and plagioclase produces an increasing amount of garnet
having an increasing pyrope/almandine ratio with rise in pressure. At higher
pre.ssures within this range, the Ab component of plagioclase begins to change
into the jadeite component of pyroxene, leading to the disappearance of
plagioclase. The Tschermak's molecule of pyroxene decomposes with resultant
formation of additional garnet. Thus, typical eclogite with magnesian garnet and
jadeite-rich clinopyroxene forms in the P- T field above the curve 'Eclog'.
For undersaturated alkali olivine basalt compositions, the appearance of garnet
and the disappearance of plagioclase take place at pressures a few kilobars lower
than those for quartz-tholeiite compositions. The higher the Ab content of
plagioclase, the higher the pressure necessary for the disappearance of plagioclase
(i.e. the formation of eclogite). The higher the Fe 2 +/(Mg + Fe 2 l ratio of the
rock, t1ullower the pressure necessary for the formation of garnet.
312/ ECLOGITES AND THE ECLOGITE FACIES
30
/
/
/
/
// M
/
/
/d"
/o:f
/
/
/
20 /
/
/
/

N/ '
/
/
/
/
/
~/
/
/
/
/
/

Temperature (OC)
Pig. 12-1 Granulite to eclogite transformation for quartz-tholeiitic com-
positions under a virtually anhydrous condition (Green and Ringwood,
1967). A, B and C: These curves indicate the first appearance of garnet
with rising pressure for quartz-tholeiitic compositions with Pe 2 +/(Mg +
Pe 2 +) = 0·9, 0·4 and 0·1, respectively. Eclog: This curve represents the
disappearance of plagioclase with rising pressure. M-N: approximate
solidus under dry condition. The low-pressure stability limits of pyrope
and almandine end-members are shown by the curves designated as
'pyrope' and 'aIm'. The high-pressure stability limit of anorthite is shown
by the curve designated 'anor'. The stability fields of kyanite, sillimanite
and andalusite and the low-pressure stability limit of the jadeite + quartz
assemblage are shown for comparison.

Granulite and Eclogite Facies in Relation to Baric Types

In the temperature range 900-1200 °e, the following stability fields are
distinguished in terms of mineral changes in rocks of basaltic compositions with
medium Fe 2 +/(Mg + Fe 2 "1 ratios.
 12-2 THE GRANULITE AND THE ECLOGITE FACIES /313
1. Just below the curve S in fig. 12-1. This field represents the low-pressure
sub facies of the granulite facies. The assemblages calcic plagioclase + ortho-
pyroxene. and calcic plagioclase + olivine are stable. Garnet does not occur in
metabasites oversaturated or undersaturated with Si0 2 • Almandine and
cordierite may occur in metapelites. The granulites of Lapland (Eskola, 1952)
and Uusimaa (parras, 1958) belong to this field. The ACF diagram for
oversaturated rocks and an analo_gous one for undersaturated rocks are shown in
the upper part of fig. 12-2.
2. Between the curves Sand B in fig. 12-1. This field represents a
low-pressure part of the medium-pressure subfacies of the granulite facies. The
calcic plagioclase-orthopyroxene assemblage is stable, but the calcic p1agioclase-
olivine assemblage is unstable. Gamet may occur in undersaturated metabasites,
but does not occur in oversaturated metabasites. Pyroxenes tend to be highly
aluminous. The granulites of Scourie, Scotland (O'Hara, 1961) belong to this
field. The parageneses are shown in the middle part of fig. 12-2.
3. Between the curves Band 'Eclog' in fig. 12-1. This field represents the
high-pressure part of the medium-pressure sub facies of the granulite facies.
Almandine-pyrope occurs in both oversaturated and undersaturated metabasites,
being formed by reaction between calcic plagioclase and orthopyroxene (fig.
12-2, lower part). The granulites and 'eclogites' of South Harris, Outer Hebrides
(Davidson, 1943; Dearnley, 1963) belong to this field.
4. Above the curve 'Eclog' in fig. 12-1. This field represents a part of the
eclogite facies distinct from the granulite facies. Calcic plagioclase no longer
occurs in metabasites and metapelites. In the pressure field immediately above the
curve, anorthite by itself is stable as well as albite by itself. However, they
become unstable at higher pressures. The paragenetic relations are schematically
shown in fig. 12-3.
It must be emphasized that the pressure necessary for reactions to form
eclogites varies with the Fe 2 +/Mg ratio and the plagioclase composition; hence
the transformation should be essentially gradual with a wide transitional P-T
region.

Petrographic Criteria for the Eclogite Facies

The eclogite facies assemblages as defined above are characterized by the
possible presence of quartz or kyanite, as may be schematically shown by the
following equations:
Eclogite Facies Granulite Facies
Mg 3 Al 2 Si 3 0 12 + CaMgSi 2 0 6 + Si0 2 = 4 MgSi0 3 + CaAl 2 Si2 0 s
pyrope diopside quartz enstatite anorthite

diopside kyanite pyrope anorthite quartz

314/ECLOGITES AND THE ECLOGITE FACIES

OVERSATURATED ROCKS UNDERSATURATED ROCKS

C FC~__~~~~______~~.
Wolla·st-o-n':""it-e---::-:-:-......;:-----::O:""rt'":"h-o-""" Wollastonite
Calcite pyroxene Calcite
Low-pressure subfacies of the granulite facies

Almandine
Orthopyroxene
C (Aluminous) F C..._ _ _=----::*"__--::---:-_.. F
Wollastonite Clinopyroxene Wollastonite Clinopyroxene Orthopyroxene
Calcite (Aluminous) Calcite (Aluminous) (Aluminous)
Low-pressure part of the medium-pressure subfacies of the granulite facies Olivine

C~--------~--------~F
Wollastonite Clinopyroxene Orthopyroxene
High-pressure part of the medium-pressure
. subfacies of the granulite facies

Fig. 12-2 ACF diagrams for rocks oversaturated with Si0 2 (left column)
and analogous ones for rocks undersaturated with Si0 2 (right column) in
the granulite facies.
 12-2 THE GRANULITE AND THE ECLOGITE FACIES /315
A
Kyanite

pyrope

C__------------------~------------------~F
Wollastonite Clinopyroxene Orthopyroxene

Fig. 12-3 Schematic A CF diagram for the eclogite facies. Since the P- T
range of the eclogite facies is very wide, various modifications of
paragenetic relations are conceivable.

Thus, the occurrence of quartz- or kyanite-eclogite may be used as a petro-

graphic definition of the eclogite facies, if appropriate limits are fixed for the
Fe 2 +/(Mg + Fe 2 l and the Na/Ca ratios of the rocks.
The AlzSiO s mineral of the eclogite facies is kyanite, whereas that of the
granulite facies is commonly sillimanite but rarely kyanite. This observation is
consistent with the experimentally determined phase relations of fig. 12-1.
The eclogite facies is characterized only by a group of eclogites with basaltic
compositions in which plagioclase is lacking, and clinopyroxene has high ratios
of jadeite/Tschermak's component, and occasionally quartz or kyanite occurs.
Eclogites in the broad sense cited at the beginning of this chapter could form
under any part of the P- T region above curve S in fig. 12-1. However, Green and
Ringwood (1967) have proposed that the use of the rock name eclogite be
confmed to the eclogites of the eclogite facies. If this is accepted, the so-called
eclogites of the granulite facies should be called garnet pyroxenites. Perhaps, this
nomenclature will become common in future.
316/ECLOGITES AND THE ECLOGITE FACIES

12-3 RELATION BETWEEN ECLOGITE AND HYDROUS METAMORPHIC

FACIES
Comparison of figs. 12-1 and 3-12 suggests that the low-temperature part of the
eclogite facies overlaps the amphibolite, epidote-amphibolite, greenschist, and
glaucophane-schist facies as regards T and Ps ' but is distinguished from them by
lower PH2 o. This is consistent with our tentative definition of metamorphic
facies (§ II-I) as representing a volume in the T-Ps-PH2 o.

Formation of Eclogite in Continental Crust

Figure 3-3 suggests that the condition of 10 kbar and 700°C is realized in
regional metamorphism intermediate between the medium- and high-pressure
types. Thus, the high-pressure type of regional metamorphism would be able to
form eclogites of the eclogite facies. This inference is consistent with geologic
data. However, eclogite is not very common in high-pressure metamorphic
terranes. In most such cases, hydrated rocks of the epidote-amphibolite,
greenschist, or glaucophane-schist facies may form instead of eclogites, which
may form only under exceptional, anhydrous conditions (e.g. Fry and Fyfe,
1969).
The base of a stable continental crust, 35 km thick, is probably near 10 kbar
and at 300-700°C. Accordingly, most of the stable crust will be on the
high-pressure side of the curve 'Eclog', that is, within the Ps-T range of the
eclogite facies. Virtually no dehydration reactions would take place in
non-orogenic crust, and hence PH 2 0 would be much lower than Ps • If this is
true, it may be expected that eclogite is widespread within continental crust.
However, the occurrence of eclogite on the surface of continents is not very
common. In stable continental crust, there are no penetrative movements, and
the rate of recrystallization may be too slow to form eclogite. Alternatively,
eclogites once formed in stable lower crust may not be uplifted so as to be
exposed on the surface.
Banno (1966) and Green and Ringwood (1967) have given critical reviews of
the modes of occurrence of eclogites and related rocks in areas of the granulite,
amphibolite, epidote-amphibolite, and glaucophane-schist facies.
The compositions of clinopyroxenes and garnets in eclogites of various modes
of occurrence are shown in figs. 124 and 12-5.

Eclogites in Gneiss Areas Seemingly of the Amphibolite Facies

Eclogites of the eclogite facies have been reported to occur in some gneiss areas
seemingly of the amphibolite facies. Well-documented examples are the eclogites
in the Nordfjord area, Norway (Eskola, 1921; Lappin, 1960; O'Hara and Mercy,
1963; Bryhni et al. 1970) and those near Glenelg in the Scottish Precambrian
(Alderman, 1936).
 12-3 THE ECLOGITE AND HYDROUS METAMORPHIC FACIES /317
Di+He

•••
• 0
30 • •
000
~o •
40 ~ ... •
......... 14. ... .....
... 0
.... 0
o
o

• Eclogite from amphibolite or granulite facies areas

... Eclogites from high-pressure metamorphic terranes
o Eclogite inclusion in kimberlite
)( 'Eclogite' or granulite inclusion in basaltic rock

Fig. 12-4 Composition of clinopyroxenes from eclogites of various en-

vironments, shown in terms of diopside + hedenbergite, jadeite, and
Tschermak's component (Ca, Mg)AI2 Si0 6 • Jadeite is calculated as the
atomic proportion of Na-Fe 3 +, and Tschermak's component is taken as
the atomic proportion of Al in four-fold co-ordination_ The dotted line
represents the 1: 2 jadeite to Tschermak's component ratio that may be
used to distinguish eclogite- from granulite-facies clino-pyroxenes. (White,
1964; Lovering and White, 1969_)
In such cases it has usually been assumed that the eclogites represent
tectonically transported masses and that there is no direct relation between the
eclogites and the surrounding metamorphic rocks. This is possible, but has not
been proved. It appears largely to have stemmed from the traditional idea that
eclogites should have formed at greater depth than ordinary metamorphic rocks.
Some geologists have thought that some such eclogites were intruded and
crystallized as igneous masses independent of the surrounding rocks. During the
crystallization, conditions of the eclogite facies would have prevailed, but the
surrounding rocks would not have been recrystallized possibly owing to the
short duration of the conditions.
318jECLOGITES AND THE ECLOGITE FACIES

K= 17
/
/
/
/
10 /
A/
/
~ A /
K =6·5

.
/
AA /
/
." /
A A/ /

,/
A ~
/
/
/ • • / K= 2·7

../
/
/
/
• / //
/ /
/ ./ /
/
~/ /
/e

.
/ /0 /
1·0 / •• / X K = 1·0
/ 60/ /
/

/ / 0 /
/' : / 0 /~x /
/
/
.~
/ • /
cfx /
/ /
/ / /)( X /
/
/ / / X X /
/ / XX /
/ //00 / /
/ / / /
/ / / /
/ 0 0 / /
/ / /
/ /,/
0·1/ 1·0
/
Fe2 +/Mg -Pyroxene

Fig. 12-5 Log-scale distribution of Fe 2 + jMg between co-existing garnet

and clinopyroxene in eclogites from various environments (Banno and
Matsui, 1965; Lovering and White, 1969). Here, Fe 2 + jMg is in atomic
ratio, and the symbols are the same as in Fig. 12-4. K represents the
distribution coefficient for the reaction: Fe-garnet + Mg-pyroxene =
Mg-garnet + Fe-pyroxene (§ 2-5).

A third possibility is that not only such eclogites but also the surrounding
metamorphic rocks underwent metamorphism in the eclogite facies. The
mineralogical differences in quartzo-feldspathic gneisses between the eclogite
and the amphibolite facies are slight, and might well be obscured by the
superposition of later metamorphism in the amphibolite facies (e.g. Bryhni et al.
1970).
A fourth possibility is that PH 2 0 may have been lower in the eclogites than
in the surrounding rocks. This would result in the formation of eclogites within
true amphibolite facies terranes, keeping all the rocks in practically the same
 12-3 THE ECLOGITE AND HYDROUS METAMORPHIC FACIES/319
Ps-T conditions. Eclogite masses frequently show a transformation into amphi-
bolites towards their margins presumably owing to permeating HzO. However, if
the PH20 is so uneven, PH2 0 cannot be an externally controlled variable. It
becomes desirable to treat Hz 0 as a fixed component in the definition of
metamorphic facies, and to regard such eclogites as belonging to the same facies
as the surrounding rocks (e.g. amphibolite facies).
At present, we have no decisive evidence for choosing among the above
possibilities.

Eclogites in High-Pressure Metamorphic Te"anes

Eclogite occurs as isolated blocks (Coleman and Lee's type IV) in metamor-
phosed areas of the Franciscan formation of California (e.g. § 7A-I0). They are
usually composed of clinopyroxene, gamet, barroisite, clinozoisite, rutile and
sphene. The areas are in the glaucophane-schist facies, and the eclogites are
commonly interlayered with glaucophane schists. Lawsonite occurs in some of
them as a late-stage mineral. Usually quartz is not present (Coleman and Lee,
1963; Coleman et al. 1965). The eclogites appear to have formed under the same
Ps- T conditions as the interlayered glaucophane schists.
In the Bessi area of the Sanbagawa belt, Banno (1964, 1966) described the
occurrence of eclogite within an epidote-amphibolite mass in the highest
temperature part of th~ area. The rock contains omphacite, almandine, horn-
blende (or barroisite), quartz and rutile.
The compOSitional differences of pyroxene and garnet from eclogites of
high-pressure terranes and of gneiss areas are shown in figs. 124 and 12-5. Some
eclogites in high-pressure terranes contain quartz, and so have the omphacite +
garnet + quartz assemblage, characteristic of the eclogite facies, although they
have relatively high FeZ +/Mg ratios.
In metamorphosed pillow lavas of the Pennine nappes of the Alps, the cores
of some pillows have been converted into eclogites, whereas the matrix between
the pillows is of glaucophane schist (Bearth, 1959). It might be thought that
during metamorphism PH2 0 was lower in the core than in the matrix, resulting
in two different mineral compositions forming under the same Ps-T condition.
However, the chemical composition of the pillow core could differ from that of
the matrix in terms of components other than Hz O. The chemical difference
may be mainly responsible for the mineralogical difference.
In the Franciscan and Bessi eclogites, the chemical similarity of the eclogites
with the associated metamorphic rocks has not been established. For example,
Coleman et al. (1965) have noted that the Franciscan eclogites have normative
nepheline, whereas the unmetamorphosed Franciscan basalts are not so under-
saturated, and that the Franciscan eclogites have higher (FeO + Fez03}/MgO
ratios than most basalts. If the mineralogical differences are due to such
chemical differences, the eclogites should belong to the glaucophane-schist or
320/ ECLOGITES AND THE ECLOGITE FACIES
the epidote-amphibolite facies, but not to the eclogite facies. In so far as the
relationship between the chemical and mineral compositions in the glauco-
phane-schist facies has not been well established, the possibility of the formation
of eclogite in this facies cannot be precluded.

124 ECLOGITE INCLUSIONS

Eclogites occur as inclusions in basaltic rocks, peridotites and kimberlites. They

may be deep-seated crystallization products of their host magmas, or may
represent xenoliths derived from the lower crust or the upper mantle. Whatever
their origin may be, some of them appear to be in the granulite facies, and others
in the eclogite facies.
For a better understanding of the mineral assemblages of such inclusions, the
high-pressure phase relations for the composition of 1 anorthite + 2 forsterite are
shown in fig. 12-6 (Kushiro and Yoder, 1966). The garnet + forsterite
assemblage is stable in the P-T field of the eclogite and high-pressure granulite
facies. At lower pressures (part of the granulite facies), spinel appears in
association with ortho- and clinopyroxenes.

Eclogite Inclusions in Basaltic Rocks

Well-documented examples of eclogite inclusions in basaltic rocks occur at Salt
Lake Crater in Oahu, Hawaii, and at Delegate, in southern New South Wales,
Australia. They appear to be eclogites in the granulite facies.
Ejected blocks of eclogite, pyroxenite, peridotite and dunite occur in the
olivine-nephelinite tuff at Salt Lake Crater, Hawaii. The pyroxenite, eclogite
and olivine eclogite show mutually gradational relations. The eclogite is com-
posed of orthopyroxene with exsolution lamellae of clinopyroxene, clinopy-
roxene with lamellae of orthopyroxene, opaque spinel, pyrope and pargasitic
amphibole. The pyroxenes are high in Tschermak's component, suggesting that
the rocks belong to the granulite facies (White, 1964; Lovering and White, 1969;
White, 1966; Kuno, 1969).
Green (1966) has accounted for the origin of the eclogite by assuming that
nearly monomineralic masses of aluminous clinopyroxene with or without minor
quantities of spinel and olivine were formed by accumulation of primary crystals
from alkali olivine basalt or basanite magma at a pressure of approximately
13-18 kbar at 1350-1400 °c, and that they were cooled to about 1000 °c at
nearly constant pressure to cause exsolution of garnet from the original
pyroxene and reaction of spinel and pyroxenes to yield garnet and olivine. All
these processes took place on the low-pressure side of curve 'Eclog' but on both
sides of curve SY in fig. 12-6, that is, in the high-pressure part of the granulite
facies conditions.
 u
12-4 ECLOGITE INCLUSIONS/321
30

,
/,1
/
Garnet+Forsterite
/
,,
20
/ + :
ot:>"
(Clinopyroxene)
/ .
,
VJ/
G'
/
/
/ A
/ /
/ / Clinopyroxene+
", Orthopyroxene+
B// Forsterite+Spinel
y
R

X Anorthite+Forsterite

Q
°20~0--~--L--5~0-0--L-~---L--~-1-0~OO--~--~~~-L---1500

Temperature (OC)

Fig. 12-6 Phase relations for the composition 1 anorthite + 2 forsterite.

The line Q-R-S represents the solidus (Kushiro and Yoder, 1966). For
comparison, the curves M-N, Band Eclog of Fig. 12-1 are reproduced by
broken lines. Possible poT fields of eclogites of gneiss areas (X), of
high-pressure metamorphic terranes (Y), and of kimberlites (Z) are shown.
(c!. Green, 1966.)

Two pipes ftlled by basanite, olivine nephelinite and their breccias at

Delegate, Australia, contain a great variety of inclusions comprising two
pyroxene-plagioclase rock, gamet-clinopyroxene-plagioclase rock, eclogite
(fassaite eclogite), spinel-gamet-two pyroxene-olivine rock, spinel-gamet-two
pyroxene-scapolite-plagioclase rock, and peridotite. The eclogite is mainly
composed of clinopyroxene (fassaite) high in Tschermak's component but low in
jadeite, pyrope-rich gamet, and brown pargasite (Lovering and White, 1969).
The mineral assemblages suggest that the eclogite and most other inclusions
were crystallized or recrystallized probably at 10-15 kbar and 1000-1200 °C,
322/ECLOGITES AND THE ECLOGITE FACIES

that is, under P- T conditions of the granulite facies. They appear to have formed
in the uppermost part of the mantle or in the lowermost part of the crust.

Eclogite Inclusions in Peridotites

Eclogites occur as inclusions in peridotite in the Nordfjord area of Norway, the
Bessi area of Japan, and others.
The Nordfjord area is underlain mainly by gneisses, which contain masses of
peridotite, anorthosite and eclogite. The eclogite which occurs as independent
masses has already been commented upon as an eclogite in gneiss areas seemingly
of the amphibolite facies. In addition, eclogite occurs as inclusions in peridotite
and anorthosite (Eskola, 1921). The peridotite contains olivine, ortho- and
clinopyroxenes and occasionally garnet, and shows flow structure but no contact
metamorphism. The eclogite included in the peridotite is composed of garnet,
diopside, orthopyroxene and olivine. The assemblage garnet + olivine + clino-
pyroxene indicates conditions on the high-pressure side of curve SY in fig. 12-6,
that is, the eclogite facies field or the high-pressure part of the granulite facies
field.
O'Hara and Mercy (1963) considered that the eclogite and the associated
garnet peridotite were formed in the upper mantle and moved upward in the
solid state to the present level. However, it is also possible that both the eclogite
inclusions in the peridotite and the independent masses of eclogite in the vicinity
were formed simultaneously by the eclogite facies metamorphism that formed
the gneisses of this region.
A concordant mass, a few km across, of fresh peridotite occurs in the highest
temperature zone (epidote-amphibolite facies) at Higasiakaisi in the Bessi area
of the Sanbagawa high-pressure belt. Eclogite layers up to 30 m thick occur in it.
The eclogites are composed of diopside and pyrope with accessory hornblende,
olivine, rutile and ilmenite. Some of the eclogites, however, contain omphacite
and almandine with a considerable proportion of the pyrope component (Shido,
1959; Banno, 1966; Banno and Yoshino, 1965). Neither quartz nor kyanite
occurs. The olivines of the surrounding peridotite show strong preferred
orientation. These eclogites could have originally been formed in the upper
mantle and subsequently been caught up by the solid intrusion of peridotite and
then subjected to slight modification by metamorphism at the present level.

Eclogite Inclusions in Kimberlites

Kimberlite is well known as a source of diamonds. It is a rare ultrabasic rock and
occurs as small intrusive bodies (pipes and dikes) in continental regions.
Petrographically it IS serpentinized and carbonated mica-peridotite of porphy-
ritic texture. The phenocrysts are olivine (about 90 per cent Fo), garnet
(almandine-pyrope), pyroxenes (ensta ti te and diopside), mica (phlogopite-
biotite), amphiboles, and picro-ilmenite. Diamond, when present, is sparsely
 12-4 ECLOGITE INCLUSIONS /323

distributed in the groundmass or rarely contained in inclusions of garnet

peridotite and eclogite.
The presence of diamond in the groundmass of kimberlite and in inclusions of
garnet peridotite and eclogite, strongly suggests that all these rocks came from
great depths in the mantle. If we assume that the diamond crystallized in its
stability field, their depths of origin must be more than 100 km below the
surface. It is likely that they represent the highest pressure part of the eclogite
facies conditions.
Ringwood and Lovering (1970) have given experimental evidence suggesting
that inclusions of pyroxene-ilmenite intergrowths in African kimberlites were
originally homogeneous garnet solid solutions formed at pressures in excess of
100 kbar. If this is accepted, kimberlite magma and some of the xenoliths
contained in it could have come up from depths of more than 300 km.
Kimberlite contains inclusions of gneisses, granulites, garnet peridotites,
eclogites and so on (Dawson, 1967). The inclusions of gneisses and granulites
were probably derived from the walls of the intrusions within the crust. The
origin of the garnet peridotite and eclogite inclusions, however, has been a
matter of dispute. Many geologists consider that they are xenoliths derived from
the mantle (e.g. O'Hara and Mercy, 1963). Others regard them as segregations
(cognate inclusions) crystallized out of kimberlite magma at great depths
(Williams, 1932).
Kushiro and Aoki (1968) have discussed the possibility that the upper mantle
is largely composed of garnet peridotite, which might undergo partial melting
followed by crystallization to produce pockets of eclogite. These might be
subsequently broken up and brought to the surface by ascending kimberlite
magma. The compositional variation in the eclogite inclusions would largely
depend on the degree of partial melting and/or crystallization differentiation at
high pressures.
Though kimberlite occurs in all the continents, the diamond-bearing type is
abundant only in the southern half of Africa and the Yakutia area of Siberia.
Extensive mineralogical studies by Russian geologists have greatly con tributed to
our knowledge.
In the Siberian kimberlites, a peculiar rock called grospydite has been rarely
found as inclusions. It is mainly composed of garnet, omphacite and kyanite
together with corundum. This rock resembles kyanite eclogite, but differs in the
composition of the garnets, which contain 50-80 per cent of the grossular
component. The garnets of grospydite and the associated kyanite eclogite form a
nearly continuous solid solution series between pyrope and grossular. (The name
grospydite has come from an abbreviation of grossular-pyroxene-disthene
rock.) There is a regular relationship between the grossular content of the garnet
and the jadeite content of the associated clinopyroxene (Sobolev, Kuznetsova
and Zyuzin, 1968; Banno, 1967).
324/ECLOGITES AND THE ECLOGITE FACIES

12-5 GRANITES AND METAPELITES IN THE ECLOGITE FACIES

What are the mineral assemblages of granitic, psammitic and pelitic rocks which
have recrystallized in the eclogite facies? It is possible that such rocks resemble
those in the granulite, amphibolite, and epidote-amphibolite facies, and are not
easily distinguishable from them.
Green and Lambert (1965) have carried out a high pressure experiment at
950 and 1100 °c on samples of anhydrous adamellite composition. At 950°C
and in the pressure range of the granulite facies (13-15 kbar) and the
low-pressure part of the eclogite facies (17-22 kbar), the recrystallized samples
are composed of plagioclase, alkali feldspar of the high albite-sanidine series,
quartz, pyroxene and garnet. The amount of plagioclase decreases, whereas that
of garnet increases with pressure. The pyroxene is entirely or mainly ortho-
pyroxene in the pressure range of the granulite facies, but becomes clino-
pyroxene at higher pressures. With a further increase of pressure (30-35 kbar),
plagioclase disappears, the KINa ratio of the alkali feldspar increases, and the
pyroxene becomes more abundant and higher in jadeite content. Coesite appears
at pressures above 32 kbar. At 1100 °c, plagioclase is abundant in the pressure
range of both the granulite and the eclogite facies. In other respects, the
products are similar to those at 950 °c. The alkali feldspar becomes nearly pure
K-end member at 40 kbar. The Na present is contained in pyroxene mainly as the
jadeite component. -
Thus, the anhydrous assemblages of granitic rocks in the granulite facies
resemble those in the low-pressure part of the eclogite facies. They are both
similar to assemblages at lower pressures except for the appearance of garnet by
the reaction between orthopyroxene and An component.
When PH 2 0 is high, biotite and/or hornblende may form instead of garnet
and pyroxene. Granitic rocks recrystallized under such conditions could not be
distinguished from ordinary granitic rocks, unless a detailed study of solid-
solution minerals is made. A relatively high jadeite content of pyroxene and a
low Fe 2 +/Mg ratio of gamet, if any, would be diagnostic of the eclogite facies.
More or less similar relations should hold in psammitic and pelitic rocks in the
eclogite facies. It is likely, therefore, that some quartzose gneissic rocks
associated with eclogite but seemingly in the granulite, amphibolite and epidote-
amphibolite facies are actually in the eclogite facies.
III Metamorphism and
Crustal Evolution

Outline
Regional and ocean-floor metamorphism occur respectively in orogenic belts and
mid-oceanic ridges. Since the P- T distribution during the metamorphism is
recorded in the mineral assemblages of metamorphic rocks, petrologic investiga-
tion could give a clue to a better understanding of the tectonic processes taking
place there. This aspect of the study of metamorphism will be emphaSized here
in Part III.
This part begins with chapters 13 and 14 where the role of regional
metamorphism in the structure of continents will be discussed with reference to
North America and Europe respectively, since relatively abundant data are
available on these continents. In the following two chapters (chapters 15 and
16), the role of regional metamorphism in the structure of island arcs will be
discussed with reference to Japan and other arcs in the Western Pacific Ocean. A
long chapter is devot~d to the Japanese Islands, partly because paired meta-
morphic belts, characteristic of the circum-Pacific region, are typically developed
there, partly because very detailed petrolOgiC, geologic and geophysical data are
available, and partly because I am farnilar with the Islands.
Then, a summary of main features of metamorphic belts will be given and
their tectonic Significance will be discussed in chapter 17, with special emphasis
on the problem of paired metamorphic belts.
In chapters 13-17,. the lateral relations of areas of continental crust will be
discussed, whereas the vertical zonation is treated in chapter 18.
In the last two chapters (chapters 19 and 20), our present knowledge of
ocean-floor and transform-fault metamorphism will be outlined.
Chapter 13

Metamorphic Belts of
North America

13-1 CONSTITUTION OF NORTH AMERICA

There is a wide Precambrian craton in the northern and central parts of North
America. Precambrian granitic and metamorphic rocks are exposed in the
northern part, which is called the Canadian Shield, whereas these rocks are
covered by relatively thin, undisturbed, unmetamorphosed Phanerozoic sedi-
ments in the central part, which is usually called the Interior Lowlands (fig.
13-1 ).
The craton is bordered on the east by the Appalachian orogenic belt, where
Palaeozoic orogenies caused deformation, regional metamorphism and plutonic
intrusions.
The Cordilleran orogenic belt lies on the west side of the Shield and Interior
Lowlands. There, orogenies took place mainly in Mesozoie and Cenozoic time.
Late Mesozoic granitic intrusions took place on a gigantic scale. In both
Appalachian and Cordilleran orogenic belts, geosynclinal volcanism involves not
only basaltic but also no less abundant andesitic and rhyolitic activity (Butler
and Ragland, 1969; Gilluly, 1965; Dickinson, 1962).
The Canadian Shield is bordered on the north by the Palaeozoic Innuitian
orogenic belt.
Thus, the arrangement of orogenic belt~ suggests continental growth by
accretion of younger orogenic belts. The oldest part of the Canadian Shield is in
its central part. This suggests that continental growth occurred in Precambrian
time also.

13-2 METAMORPHISM IN THE CANADIAN SHIELD

Orogenies and Provinces

Four major orogenies have been recognized through the radiometric dating of
the Canadian Shield, and are summarized in table 13-1. The Shield may be
subdivided into well-defmed tectonic and radiometric provinces, each of which
underwent a definite major orogeny (fig. 13-1). Such a major orogeny
would include a number of events, each of which is comparable to one cycle of
328/METAMORPHIC BELTS OF NORTH AMERICA

[Tj Canadian Shield o Miogeosyncline

Phanerozoic Orogenic
} belts
Interior Lowlands Eugeosyncline

Fig. 13-1 Constitution of North America. Only three of the provinces of

the Canadian Shield (Superior, Churchill and Grenville) are shown.
(Modified from Kay, 1951, King, 1959, 1969, and others.)

Phanerozoic orogeny. The Grenville province, however, contains not only the
metamorphic rocks derived from its own geosynclinal pile but also a great
proportion of older crystalline rocks which were more or less modified by the
Grenville orogeny. Most of the Appalachian region is underlain by a Precambrian
basement consisting of extensions of the Grenville province. Hence, there may
have been as great an area of continental crust in Precambrian time as there is at
present. Conceivably, the radiometric age pattern may simply show the order of
stabilization of provinces from the centre outwards, as claimed by Wynne-
Edwards (1969).
 13-2 METAMORPHISM IN THE CANADIAN SHIELD /329
Table 13-1 Orogenies in the Canadian Shield

Orogeny Range (m.y.) Provinces

Grenville orogeny 880-1000 Grenville

Elsonian orogeny 1280-1550 Nain
Hudsonian orogeny 1640-1820 Churchill, Southern,
(called Penokean in Bear, Nain
Minnesota)
Kenoran orogeny 2390-2600 Superior, Slave
(called Algoman in
Minnesota)

Note: Based on Stockwell (1964) and King (1969).

The oldest part of the Shield is the Superior province. However, there is some
evidence suggesting the former existence of a continental crust that was
subjected to orogeny at an earlier time than the Superior province. Graywackes
in the Superior province contain quartz, K-feldspar, and other minerals which
must have been derived from older granite-gneiss terranes. Moreover, zircons that
may have crystallized 3400-3600 m.y. ago have been found in some Pre-
cambrian sediments. Hence, continent-forming processes appear to have begun at
least 3600 m.y. ago in North America as in Europe and Africa. In North
America, such old terranes appear to have been destroyed by later orogenies
(Condie, 1967).

The Grenville and Superior Provinces

In spite of the great complexity of the Shield, petrographic studies have been
made only in very small parts of it. It is impossible to give a balanced account of
the metamorphism of the whole Shield.
The southern half of the Grenville province is relatively well populated and
well investigated. It is characterized by the abundance of anorthosites and
granulite facies metamorphic rocks, though amphibolite facies rocks are also
widely exposed. Almandine is widespread. Areas of the amphibolite and the
granulite facies are irregularly intermingled. Wynne-Edwards (1969) has ascribed
this variation to metamorphism at similar temperature but varying PH z O'
The Adirondack Mountains in northern New York State are a southern
protrusion of the Grenville exposures. The anorthosites and metamorphic rocks
there are well documented (Buddington, 1939; Engel and Engel, 1958,1960,
1962a, b; Isachsen, 1969).
It is to be noted that the Grenville province generally shows the highest
degree of metamorphism, although it is the youngest of the provinces of the
Shield (§ § 8A-7 and 18-2).
330/METAMORPHIC BELTS OF NORTH AMERICA
The Superior province appears to have been metamorphosed mainly in the
greenschist to the low amphibolite facies. The original rocks of this province are
mostly poorly sorted graywacke, conglomerate and volcanics of eugeosynclinal
character in contrast to the Grenville province where pelite, quartzite and
limestones are abundant.
The distribution of baric types in the Shield is not clear, except that no rocks
of the glaucophane-schist facies have been found.

13-3 THE METAMORPHIC BELT OF THE APPALACHIANS

Orogenies
The Appalachian geosyncline contains sedimentary sequences ranging from the
late Precambrian through the Palaeozoic. Its westernmost zone fringing the
eastern margin of the Precambrian craton is miogeosynclinal, and only part of it
was metamorphosed during the Palaeozoic orogenies. The region to the east is
eugeosynclinal and has been widely metamorphosed, in places to a high degree.
The miogeosync1inal zone at least has a Precambrian basement which is a
southeastern extension of the Grenville province.
Probably there was a proto-Atlantic ocean between the North American and
African continents in late Precambrian time. The two continents approached and
then collided with each other to result in the Taconian and Acadian orogenic
episodes in Palaeozoic time. The subsequent separation and drifting of the two
continents created the present Atlantic Ocean (Wilson, 1966; Dietz and Holden,
1966).
The ages of the three main orogenic episodes in the Appalachians (Rodgers,
1967) are:
Taconian Middle and Late Ordovician (450-500 m.y.)
Acadian Middle Devonian (360-400 m.y.)
Alleghanian Carboniferous and Permian (230-260 m.v.).
The most important phase of regional metamorphism and granite emplacement
is the Acadian.

Northern Appalachians
The Appalachians are bisected roughly at the site of New York City into the
northern and southern divisions. The structural zones of the northern Appala-
chians may be summarized as follows (Dietz and Holden, 1966; Zen, 1968;
Rodgers, 1968, 1970; Cady, 1968; Bird and Dewey, 1970).

1. Western miogeosynclinal zone. This is the westernmost zone mainly

composed of deformed and weakly metamorphosed lower Palaeozoic carbonate
rocks (dolomite and limestone) accompanied by orthoquartzite. Within it occur
 13-3 THE METAMORPHIC BELT OF THE APPALACHIANS /331
the Taconic and other allochthons consisting of lower Palaeozoic rocks that
probably slid as submarine sheets from the east. Near the eastern limit of this
zone, the Precambrian basement of Grenville age is exposed as a number of small
massifs such as that of the Green Mountains.
The miogeosynclinal carbonate rocks probably represent a great carbonate
bank on the Palaeozoic continental shelf somewhat resembling today's Bahama
banks.

2. Eugeosynclinal zone. This lies to the east of the preceding zone, and is
mainly composed of deformed and metamorphosed pelites and graywackes (late
Precambrian to Devonian) accompanied by ultrabasic, basic and acidic rocks. It
is conceivable that this zone was derived from the sedimentary piles of two
continental rises. The western one would have formed in early Palaeozoic time
on the east side of the contemporaneous shelf now represented by the western
miogeosync1inal zone, whereas the eastern one would have formed in early and
middle Palaeozoic time on the west side (oceanic side) of the then stable zone
now represented by the very late Precambrian zone (see below).

3. Very late Precambrian zone. The easternmost terrane in the Appalachians

was formed by deposition followed by orogeny in very late Precambrian time. It
is now exposed in areas along the Atlantic coast in Rhode Island, Nova Scotia
and Newfoundland. The associated granitic rocks give radiometric ages of about
580m.y.
Thus, the Appalachian structure was conceivably formed by repeated expan-
sion and contraction of an ocean (Proto-Atlantic Ocean) between the North
American and the African plate in a period ranging from very late Precambrian to
the end of Palaeozoic time. The very late Precambrian zone, mentioned above, is
the oldest part, and was probably formed originally along the continental margin
of Africa. In the next stage (Le. Taconian time), the western miogeosync1inal
zone and the western part of the eugeosynclinal zone, mentioned above, appear
to have become the main theatre of orogeny and metamorphism. The Acadian
orogeny would have affected most of the region, though the metamorphism of a
new continental rise deposit in the eastern part of the eugeosync1inal zone is
particularly important.
The metamorphic rocks of the northern Appalachians have been well
documented (e.g. Barth, 1936, Zen, 1960; Thompson and Norton, 1968; Albee,
1968). As was shown in fig. 7A-6, there are three thermal axes parallel to the
elongation of the orogenic belt, and the eastern part of the metamorphic terrane
belongs to the low-pressure type, whereas the western part belongs to the
medium-pressure type. These two regions mayor may not represent extensions
of the Caledonian and Hercynian belts in Europe (§ 14-6).
332/METAMORPHIC BELTS OF NORTH AMERICA

The high-temperature zones are closely associated with masses of the

Devonian New Hampshire plutonic series, which may have given strong thermal
effects.
Small parts of the metamorphic terrane are younger than Acadian. For
example, the regional metamorphism in the Narragansett basin around Provi-
dence, Rhode Island, affected Pennsylvanian rocks.

13-4 METAMORPHIC BELTS IN THE CORDILLERAN MOUNTAINS

General Features
The Cordilleran Mountains, extending from the Rocky Mountains to the Pacific
Coast, border the craton of the Canadian Shield and Interior Lowlands on the
west side. Being as wide as 1500 km, the region includes units of different
tectonic and petrographic characteristics. Geosynclinal deposition ranged from
late Precambrian to late Mesozoic time with an orogenic culmination in late
Mesozoic. The easternmost zone (Rocky Mountains and Colorado Plateau)
represents a reactivated Precambrian craton and was not in the pertinent
geosyncline. The eastern limit of the geosyncline lies on the western edge of the
Colorado Plateau. The geosyncline may be divided into a miogeosynclinal and a
eugeosynclinal zone (fig. 13-1).
The Nevadian and Coast Range orogenies, Jurassic to Cretaceous in age,
caused deformation, metamorphism and plutonism in the eugeosynclinal zone,
whereas the Laramide orogeny, Cretaceous to Tertiary, deformed the miogeo-
synclinal zone.
At a first glance, the Cordilleran and the Appalachian Mountains are
symmetrical with regard to the craton between them. However, the western part
of the Cordilleran System has many features that are not present in the
Appalachians but are commonly present in circum-Pacific regions. From the
viewpoint of metamorphic petrology, the most marked feature is the existence
of paired metamorphic belts and of extensive glaucophane-schist terranes in the
western Cordillera.
Two belts of metamorphism occur there in a pair. The eastern belt is
characterized by gigantic batholiths of granitic rocks in the Sierra Nevada and
other places, as shown in fig. 13-2. The western belt is characterized by
glaucophane schists of the Franciscan group in the Coast Ranges of California.
In most of the circum-Pacific region, regional metamorphism and intense
acidic magmatism took place in late Mesozoic time. The magmatism involved not
only the formation of granitic batholiths but also acidic volcanism. The
emplacement of gigantic batholiths and associated volcanism in western North
America are probably a part of this circum-Pacific event.
It is to be noted that glaucophane schists in western North America are not
confined to the Franciscan. In northern Washington State, a belt of Shuksan
 13-4 METAMORPHIC BELTS IN THE CORDILLERAN MOUNTAINS/333

metamorphic rocks occasionally containing glaucophane extends southeastward

from near Bellingham for about 200 km, ultimately disappearing under the
Tertiary volcanics of the Cascade Range. The metamorphism appears to be latest

Fig. 13-2 Zone of Mesozoic granitic batholiths in western North

America. The names of batholiths are shown. Line QD- QD represents the
quartz-diorite line of Moore et al. (1961).

Palaeozoic or earliest Mesozoic, i.e. older than the Franciscan. Moreover, this
high-pressure belt is in a pair with a metamorphic terrane accompanied by
granitic rocks on the east (Misch, 1966). A few more localities of glaucophane
schists are known in the Canadian part of the Cordillera (Monger and Hutchinson,
1971).
334/METAMORPHIC BELTS OF NORTH AMERICA

Mesozoic Granitic Batholiths

In the western United States, andesitic and rhyolitic volcanism was widespread
in Palaeozoic and Triassic time, but few granitic plutons were emplaced in this
period. Extensive plutonism began in the Jurassic and culminated in the
Cretaceous producing the greatest group of batholiths in the world. In Tertiary
time, acidic volcanism associated with weak plutonism took place extensively
not only in the eugeosync1inal but also in the miogeosync1inal zone to the east
(Gilluly, 1965; Dickinson, 1962).
Each of the gigantic batholiths is made up of a great number of smaller
plutons, ultrabasic, basic, intermediate and acidic in composition, which were
intruded over a considerable length of time (Buddington, 1959). For example,
the Boulder batholith, about 70 km across, appears to have been emplaced 78-69
m.y. ago, that is, in a time span of 9 million years (Tilling, Klepper and
Obradovich, 1968). The intrusion of the Sierra Nevada batholith occurred over a
much longer time span, ranging from 210 m.y. ago (Triassic) to 80 m.y. ago
(Cretaceous). Five epochs of granitic activity have been distinguished in
California and western Nevada: 210-195, 180-160, 148-132, 121-104, and
90-79 m.y. (Bateman and Eaton, 1967; Evernden and Kistler, 1970).
In these batholiths, quartz diorite, granodiorite and quartz monzonite are
most abundant. Granite in the common petrographic sense, which is more
potassic than the rocks mentioned above, is present in a subordinate amount
(Larsen, 1948; Bateman et al. 1963; Bateman and Dodge, 1970; Anderson,
1952).
In figs. 13-2, the quartz-diorite line proposed by Moore (1959) and Moore et
al. (1961) is shown. On the west side of it, the most abundant granitic rocks are
quartz diorite [(K-feldspar + perthite)/all feldspars = 0·0-0·1]. On the east side
of the line, more potassic granitic rocks (i.e. granodiorite, quartz monzonite and
granite proper) are predominant. In other words, the plutonic rocks tend to be
more potassic and more acidic toward the interior of the continent. The crust is
as thick as 50 km beneath the Sierra Nevada.
Wide regions in the zone of granitic batholiths had been subjected to low- or
medium-temperature regional metamorphism before the intrusion of plutons.
Stilpnomelane, pumpellyite, staurolite, kyanite, andalusite and sillimanite
formed in some schists (e.g. Eric, Stromquist and Swinney, 1955; Compton,
1955; Hietanen, 1961, 1962; Schwarcz, 1969). The contact effects of the
plutons were superposed on, and in some areas are indistinguishable from,
regional metamorphism.
In the Sierra Nevada, the ages of the deposition of wall rocks range at least
from early Palaeozoic to Jurassic. Regional metamorphism is usually in the
greenschist facies. Erosion has not advanced enough to expose high temperature
regional metamorphic complexes. The presence of great batholiths does not
mean the exposure of deep crustal levels. The granitic bodies do not necessarily
 134 METAMORPHIC BELTS IN THE CORDILLERAN MOUNTAINS /335
become bigger with increasing depth. Seismic and gravity data suggest that
granitic rocks now exposed on the Sierra Nevada appear to continue only to a
depth of about 8 km, below which denser rocks such as gneisses and amphi-
bolites are present (Hamilton and Myers, 1967).
Great zones of serpentinite occur in the Sierra Nevada.

Coast Ranges of California

The Franciscan rocks are eugeosynclinal deposits exposed in the Coast Ranges of
California (fig. 13-3). They were probably deposited on the basaltic crust of a
deep ocean floor from Late Jurassic to Late Cretaceous time. The total thickness
of the accumulations would be 15- 20 km. The predominant rock type is
graywacke, though basic volcanics, shale, chert and limestone are also included.
Abundant serpentinite occurs, but there are no granitic rocks (Bailey et al.
1964).
The Coast Ranges are separated from the Sierra Nevada by the Great Valley.
A thick sedimentary pile, called the Great Valley sequence, is well exposed on
the western part of the Valley, though not confined there. It was deposited on
the continental shelf and slope at the same time as the Franciscan. In northern
California, the Franciscan terrane is thrust partly under the granitic and
associated metamorphic terrane of the Klamath Mountains and partly under the
Great Valley sequence to the east. The thrust fault, called the Coast Range
thrust, is commonly at a high angle, dipping eastward (fig. 13-3). Serpentinite
masses occur along it (Ernst, 1970; Bailey, Blake and Jones, 1970). The Great
Valley sequence is mostly unmetamorphosed but its lower parts show incipient
burial metamorphism (Dickinson, Ojakangas and Stewart, 1969).
In southern California, the distribution of rocks has been strongly disturbed
by the San Andreas and other strike-slip faults (§ 20-2). Granitic rocks of the
zone of batholiths have been cut by the San Andreas and the Nacimiento fault,
and the intervening area has been moved northwestward into the Coast Ranges
(fig. 13-3).
In the Franciscan, recrystallized and unrecrystallized areas seem to be
intermingled. Probably the P-T conditions of metamorphism prevailed on a
regional scale, but because of the low temperature, recrystallization took place
only in favourable areas. The northwestern zone of the Franciscan is more or less
metamorphosed in the zeolite facies with laumontite, whereas the eastern zone is
metamorphosed in the glaucophane-schist facies (Bailey et al. 1964; Blake, Irwin
and Coleman, 1969; Ernst, 1971a). The rocks are commonly non-schistose,
preserving their original clastic texture.
During the deposition of the Franciscan, the granitic rocks in the Sierra
Nevada and Klamath Mountains were continually intruded and some plutons
were exposed. K-feldspar and other minerals in the granitic rocks were trans-
ported and deposited to enter the sediments of the Great Valley sequences and
336jMETAMORPHIC BELTS OF NORTH AMERICA

,,
""
""-
"
Pacific Ocean

o 200 km

Franciscan group
""" Laumontite·bearing Los Angeles
""" area
n
~

•
Pumpellyite-bearing
area
Lawsonite·and/or
o Late Mesozoic and
Cenozoic cover
r.o;:J Granitic and associated
jadeite-bearing area ~ metamorphic rocks
SA San Andreas fault
NAC Nacimiento fault
CR Coast Range thrust
Db Diablo Range

Fig. 13-3 Franciscan rocks in California. The tentative three-fold division

shown is mostly based on Bailey et al. (1964), Blake et al. (1969) and
Coleman and Lanphere (1971).
13-4 METAMORPHIC BELTS IN THE CORDILLERAN MOUNTAINS/337

the Franciscan. The western part of the Franciscan is generally younger than the
eastern.
Suppe (1969) has shown the existence of two different metamorphic events
about 150 m.y. ago (Late Jurassic) and 127-104 m.y. ago (Cretaceous), together
with a still younger undated one. This suggests that an early phase of
metamorphism took place just after the deposition of older parts of the
Franciscan. Then deposition occurred in other parts of the Franciscan and a later
phase of metamorphism took place there. Such a sequence of events took place
three times or more. It is not clear whether these metamorphic events have a
direct genetic connection with the five epochs of granitic activity in California
and Nevada mentioned above.
Chapter 14

Metamorphic Belts of Europe

14-1 CONSTITUTION OF EUROPE

There is a wide Precambrian craton, called the East European Platform (or
Fennosarmatia), extending from the Ural Mountains to central England
(Bogdanoff, Mouratov and Schatsky, 1964). Precambrian granitic and associated
metamorphic rocks are widely exposed in the north-western part of the
Platform, called the Baltic Shield (or Fennoscandia). Smaller exposures occur in
the Ukrainian massif (fig. 14-1). The rest of the Platform is covered by a nearly
horizontal layer, commonly 1-3 km thick, of unmetamorphosed sediments
ranging from late Precambrian to Cenozoic age. This region is called the Russian
Plate.
The East European Platform is surrounded on all sides by Phanerozoic
orogenic belts. It is bordered on the northwest side by the Caledonian orogenic
belt which runs from the British Isles to Norway and west Sweden, and on the
south side by the Hercynian orogenic belt and further to the south by the Alpine
orogenic belt. The Ural Mountains are on the east side of the Platform.

14-2 METAMORPHISM IN THE BALTIC SHIELD

Age Relations
In the Baltic Shield, the eastern part in the Kola Peninsula and on the west coast
of the White Sea (KSB in fig. 14-1) is the oldest terrane, including the
Katarchean (3500-3000 m.y.), Saamides (2900-2100 m.y.) and Belomorides
(2100-1950 m.y.). On the west side of this region are younger terranes called
the Karelides and the Svecofennides (both 1900-1600 m.y.). The youngest
terranes, called the Gothides (1400-1200 m.y.) and the Ripheides (1100-665
m.y.), are further west (fig. 14-1). Thus, the exposed Precambrian becomes
younger westwards.
In the East European Platform on the whole, however, the older Precambrian
terranes (more than 1600 m.y. old) extend from the eastern three-quarters of
the Baltic Shield to the Ukrainian massif. Younger Precambrian terranes are
arranged generally surrounding the older (Polkanov and Gerling, 1960), and
Phanerozoic orogenic belts surround the whole Platform. Thus, the age pattern is
 14-2 MET AMO RPHISM IN THE BALTIC SHIELD /339

Fig. 14-1 Constitution of Europe. The Baltic Shield and Ukrainian Massif
are shown in white, where KSB = Katarchean, Saamides and Belomorides.
Abbreviations: K = Karelides, Sf = Svecofennides, G = Gothides, R =
Ripheides, U = Ukrainian Massif.

more or less concentric, suggesting the possibility of continental growth by

accretion.

Svecofennides
The Svecofennides lie in central Sweden and southwestern Finland. The terrane
is metamorphosed mostly at relatively high temperatures (amphibolite facies)
and is accompanied by an enormous amount of granitic rocks (Geijer, 1963;
Eskola, 1963). Because of these characteristics, it was long regarded erroneously
as a part of the oldest rocks on earth. Geologic investigations in various areas of
the terrane played important roles in the progress of Precambrian geology and
geology in general. Above all, Sederholm's (1907, 1923, 1926, 1932) studies of
granites and migmatites, and Eskola's (1914, 1915, 1932) studies of the Orijarvi
region and granites are monumental in our science (§ 4-1).
Though amphibolite facies rocks are the most widespread, there are some
greenschist facies and granulite facies rocks (e.g. Parras, 1958). The meta-
morphism belongs to the low-pressure facies series. Andalusite occurs in lower
340/METAMORPHIC BELTS OF EUROPE
temperature parts and sillimanite in higher, while cordie rite and almandine are
common (Simonen, 1953, 1960b).
The Karelides also appear to belong to the low-pressure type.

14-3 METAMORPHIC BELTS IN THE CALEDONIDES

Metamorphic areas in the Caledonides of Scotland and Norway, including the
Scottish Highlands (Barrow, 1912; Harker, 1932), Trondheim (Goldschmidt,
1915), Stavanger (Goldschmidt, 1921), Sulitjelma (Vogt, 1927), and the Sparag-
mite Formation (Barth, 1938), played very important roles during the establish-
ment of metamorphic petrology before World War II. The classical metamorphic
zones in the Trondheim area and the Scottish Highlands have already been
shown in figs. 1-2 and 3-1. In the past decade, the Scottish Highlands have again
become a centre of a trend of modernized field studies through a renewed wave
of intensive work by many British geologists.

Tectonic and Age Relations

The Caledonian orogenic belt in northwest Europe is in contact with the Baltic
Shield on its southeastern side, and with another Precambrian craton on its
northwestern side along the Moine thrust in northwest Scotland (figs. 14-1 and
7A-I 0). The latter craton is made up of the Lewisian complex, which before the
Mesozoic continental drift was probably in contact with the Precambrian of
Greenland, representing a northeastern part of the original Canadian Shield (fig.
14-4). Three orogenic-metamorphic events were distinguished in the Lewisian
complex: Scourian (2600-2460 m.y. ago), Inverian (2200-2000 m.y. ago), and
Laxfordian (1650-1450 m.y. ago). The rocks exposed are mostly granites and
gneisses commonly in the amphibolite and the granulite facies (Sutton and
Watson, 1969; Bowes, 1968; O'Hara, 1961).
The northern two-thirds of Great Britain belongs to the Caledonides. The
region to the north of the Highland Boundary fault (H in fig. 7 A-I0) is the
metamorphosed Scottish Highlands, which is bisected by a big strike-slip fault
called the Great Glen fault (G in fig. 7 A-10). The region between the Highland
Boundary and Great Glen faults is called the Grampian Highlands where the
metamorphosed Dalradian and Moine Series are exposed. The latter extends to
the north of the Great Glen fault up to the Moine thrust, and is composed of
metamorphosed psammites and pelites deposited in late Precambrian time. The
Dalradian Series is composed of metamorphosed pelites and psammites, accom-
panied by limestone and basic volcanics, which were deposited in late Pre-
cambrian and Cambrian time.
The Highland Boundary fault was possibly initiated in late Cambrian time and
movement continued along it throughout the Caledonian orogeny. The Great
 14-3 METAMORPHIC BELTS IN THE CALEDONIDES/341
Glen fault is said to involve a strike-slip movement of the order of 100 km,
which offsets the metamorphic zones on the opposite sides (Kennedy, 1948).
The Caledonian deformation and regional metamorphism in the Scottish
Highlands and in the northernmost part of the Southern Uplands of Scotland
(down to the Southern Uplands fault) took place probably between 510 and 490
m.y. ago (mainly Arenigian, i.e. early Ordovician), whereas the southern
Caledonides (south of the Southern Uplands fault) were deformed in late
Silurian and early Devonian time, though only slightly metamorphosed. Post-
climactic slow uplift and erosion took place in the Scottish Highlands through-
out late Ordovician, Silurian and early Devonian times. The K-Ar mica ages
from the Scottish Highlands range generally from 500 to 400 m.y., and probably
represent the time of cooling due to the slow uplift and erosion (Dewey and
Pankhurst, 1970).

Metamorphism in the Scottish Highlands

Most of the Caledonian metamorphic belt in Scotland and Norway appears to
belong to the medium-pressure facies series. In the Grampian Highlands,
however, a low-pressure metamorphic region, called the Buchan region, has been
clearly distinguished from a medium-pressure region, called the Barrovian region,
as was shown in fig. 7A-10. There is no marked tectonic line between the two
regions, but the difference is simply a result of difference in rock-pressure
(Johnson, 1963; Chinner, 1966a). The area investigated by Barrow (§ 3-1) is
near the southeastern comer of the Barrovian region.
The distribution of progressive metamorphic zones in the Grampian Highlands
is complicated. The sillimanite zone is confined to a relatively small area
straddling the Barrovian and Buchan regions. Migmatites are abundant there.
Chinner (1966a) has suggested that the generation of sillimanite and migI'natite is
due to a localized temperature rise which took place at a stage later than the
formation of all the other progressive metamorphic zones due to depth-
controlled general P-T distribution.
Harte and Johnson (1969) determined the time relation between the phases
of deformation and the formation of metamorphic minerals in Barrow's classical
area. During the lst and 2nd phases of deformation, the temperature of
metamorphism was low. Then, the temperature increased gradually, and this led
to the formation of gamet, staurolite and kyanite between the 2nd and 3rd
phases. The highest temperature was reached in a static period between the 3rd
and 4th phases of deformation, leading to the generation of sillimanite. Then the
temperature decreased to cause some retrogressive changes.
Two groups of granitic rocks, Older and Newer, have been distinguished in
the Scottish Caledonides (§ 4-1). The Older Granites were broadly syntectonic
and simultaneous with the above regional metamorphism that occurred about
490-500 m.y. ago (Ordovician). They appear to be of metamorphic origin. The
342/METAMORPHIC BELTS OF EUROPE
Newer Granites are post-tectonic and 415-390 m.y. old (Late Silurian-Early
Devonian).

Southern Uplands o[ Scotland

Basaltic magmatism of the Caledonian geosyncline was more intense in the
Southern Uplands of Scotland (fig. 7 A-10) than in the Scottish Highlands
(Wiseman, 1934; Read, 1961). Most of the Southern Uplands were only slightly
recrystallized, but Ordovician rocks in the Ballantrae area have been found to be
metamorphosed in the glaucophane-schist facies. There are associated eclogites
and serpentinites (Bloxam and Allen, 1960).
The medium-pressure metamorphic terrane of the Scottish Highlands and the
high-pressure terrane of the Southern Uplands could be an example of paired
metamorphic belts, though recrystallization in the high-pressure terrane is very
limited (Miyashiro, 1967b, p. 427; Dewey and Pankhurst, 1970, p. 362). If this
is the case, the Benioff zone in early Ordovician (Arenigian) time would have
plunged northwards into the mantle from a trench zone in the position of the
Southern Uplands. In later Ordovician time, a new Benioff zone plunging
southwards from the trench zone in the same position would have formed,
because the middle and late Ordovician volcanism in the Lake district and Wales
to the south of the Southern Uplands is mainly of andesitic and rhyolitic
compositions and strongly suggests the existence of an island arc (Fitton and
Hughes, 1970).

14-4 METAMORPHIC BELTS IN THE WEST EUROPEAN HERCYN IDES

The Hercynides in West Europe were formed between the East European
Platform and the African Platform (fig. 14-1). Regional metamorphism took
place in wide areas in association with abundant granitic intrusions (table 4-1).
As shown in fig. 14-2, they are widely exposed in the so-called Moldanubian
zone (including the Bohemian massif, Black Forest, Vosges, and Massif Central),
the Pyrenees, and the Ibero-Hesperian zone (western Spain and Portugal).
The regional metamorphism belongs to the low-pressure series in most regions
(Zwart, 1967a). Ophiolites are rare and serpentinite is nearly lacking. The
radiometric age of metamorphism is about 360-260 m.y. (Late Devonian to Early
Permian).
A large number of areas in the Pyrenees and the Ibero-Hesperian zone were
described in recent years by Dutch and French geologists (§ 7A-4 and 8A-3).
Most of the Hercynian orogenic belt appears to have older (possibly
Precambrian) granitic and metamorphic basements. Hercynian and pre-
Hercynian metamorphic rocks are exposed more or less intermingled. The
pre-Hercynian basements sometimes contain kyanite, suggesting that they belong
to the medium-pressure facies series. In the basements in some areas, eclogites
 14-5 METAMORPHIC BELTS IN THE WEST EUROPEAN ALPIDES /343
have been found, some of which at least belong to the strictly defined eclogite
facies.
L'Ile de Groix off the south coast of Brittany is exceptional in having
glaucophane schists belonging to the Hercynides (Cogne, Jeanette and Ruhland,
1966). The significance of this occurrence is not clear.

HerCYrMan intrusIve
granote o.... 500km
I ....................t...-........I
Hercynoan regoonal
metamo<phlsm
• AlPIne regIonal
metamorphsm

Fig. 14-2 Hercynian and Alpine metamorphic belts in Western Europe.

(Zwart, 1967 b.)

14-5 METAMORPHIC BELTS IN THE WEST EUROPEAN ALPIDES

The Alpine orogenic belt in West Europe was formed generally on the south
side of, but largely superposed on, the Hercynides. It shows a sinuous pattern,
forming several arcs, for example, in the western Alps. Because of its arcuate
pattern, it covers a considerable area, but the individual constituent belts are
relatively narrow in contrast to the very wide Hercynides (table 4-1).
Regional metamorphic rocks are widely exposed in the Betic Cordillera in
southern Spain (de Roever and Nijhuis, 1964), eastern Corsica (Egeler, 1956),
Western and Pennine Alps (Bearth, 1962; Niggli and Niggli, 1965; Niggli, 1970;
344/METAMORPHIC BELTS OF EUROPE

Bocquet, 1971), western Balkan Peninsula and islands in the Aegean Sea (van der
Plas, 1959). They are of the high-pressure series and mostly in the glaucophane
and the greenschist facies (fig. 14-3). In the Swiss Alps, however, the tempera-
ture of metamorphism increases southwards, reaching a grade high enough to
produce sillimanite (amphibolite facies) in the Lepontine Alps straddling
Switzerland and Italy (§ 7A -8 and fig. 7A-II). There is a zone of the
prehnite-pumpellyite facies in the lowest temperature part of the Swiss Alps.
Not only pumpellyite but also laumontite occurs in the lowest temperature part
of the western Alps east of Grenoble.

o Glaucophane
)( Lawsonite
.. Alpine ophiolites
Go Alpine granites

Fig. 14-3 Localities of glaucophane, lawsonite, ophiolites, and granites in

the Alpine orogenic belt in Western Europe. (Zwart, 1967b.)

The Rb-Sr and K-Ar ages of Alpine metamorphic rocks from the high-
temperature part of the Swiss and Italian Alps range from 30 to 11 m.y.
depending on their cooling history. The climax of the last phase of the Alpine
metamorphism occurred at least 30 m.y. ago, i.e. in Oligocene time (Armstrong,
Jaeger and Eberhardt, 1966).
The Alpine belt has older granitic and metamorphic basements. In the Swiss
Alps, the Hercynian basement is exposed in wide areas of the Aiguilles-Rouges,
Aar, and Gotthard massifs (fig. 7A-II).
The Alpine metamorphic terranes have abundant ophiolites including serpen-
tinite. The Alps are connected eastwards to the Carpathians, though high-
pressure metamorphic rocks are not exposed in the latter. A belt oflate Tertiary
and Quaternary andesite and rhyolite exists in Hungary, southern Slovakia and
Transylvania on the south side of the Alpine-Carpathian zone (pant6, 1968;
Karolus, Forgie and Konecny, 1968). This belt is accompanied by late Tertiary
 14-6 METAMORPHIC BELTS IN THE ATLANTIC REGIONS/345

granitic masses, and has unusually high heat-flow values (Boldizsar, 1964; Lee
and Uyeda, 1965). The Alpine granites of Bergell and Adamello in southern
Switzerland and northern Italy probably belong to this belt, though the former
has been intruded into the southernmost part of the Pennine zone. It is likely
that this belt of granitic plutons and andesitic volcanism has a hidden low-
pressure metamorphic complex in a depth, which is paired with the high-pressure
metamorphic complex now exposed in the Pennine zone of the Alps (cf. § 17-2;
Miyashiro, 1972b). The descending slab at that time should have been inclined
to the south. The recrystallization temperature in the Pennine zone increases
southwards, Le. towards the associated belt of granitic plutons, in the same way
as in the Sanbagawa and Franciscan terranes (Ernst, 1971b).

14-6 CORRELATION AND BASEMENTS OF METAMORPHIC BELTS IN

THE ATLANTIC REGIONS

Pre-Drift Reconstruction
The orogenic belts in North America and those in Europe show a roughly
concentric pattern of radiometric ages, suggesting accretional growth of each
continent. Since the present Atlantic Ocean has been formed by continental
drift in Mesozoic and Cenozoic time, however, we have to discuss the relation-
ship between the older orogenic-metamorphic belts of North America and
Europe.
The pre-drift reconstruction of orogenic belts across the North Atlantic
Ocean has been attempted in recent years by Bullard, Everett and Smith (1965)
on the basis of submarine topography, by Miller (1965), Fitch (1965) and
Wynne-Edwards and Hasan (1970) largely on the basis of radiometric age data,
and by Dewey and Kay (1968) and Kay (1969) with special reference to the
geologic similarity of Newfoundland with the British Isles. There is general
agreement in major points of their interpretations. A tentative reconstruction
based on the age relations and metamorphism is shown in fig. 14-4.
The original shapes and mutual relations of orogenic belts older than the
Grenville are quite unknown, since they have been too strongly disturbed by
later orogenies. The Grenville province in Canada is here correlated tentatively
with the Gothides and Ripheides in the Baltic Shield because of the similarity of
their ages. The Taconian (Ordovician) and Acadian (Devonian) of the Appala-
chians are correlated respectively with the Caledonian and Hercynian belts in
Europe (§ 13-3).
The Grenville- Riphian, Caledonian-Taconian, and Hercynian-Acadian
orogenic belts run across the north Atlantic, but the Caledonian-Taconian belt
cuts across the Grenville- Riphian.
It appears that Phanerozoic orogenic belts in the north Atlantic region were
formed by mutual movements of the Precambrian continental plates of North
346/METAMORPHIC BELTS OF EUROPE

America, Europe and Africa. The Appalachians would have formed by approach
and collision of the North American with the African plate. The Caledonian belt
of Europe was probably formed by approach and collision of the European with
the North American plate (e.g. Dewey, 1969).

250· 400m y (Hercyn",n & 17.':l1200-2000m y (Hudsonoan &

Acad",n) ::::::J Svecofennoan )

1llIll 400- 600m.y. (Caledonian

Taconian)
& D >2000my (Kenoran&Qlderi
r=:::I aoo·1200m y (Grenville &
§ Riphean)

Fig_ 14-4 Tentative pre-drift reconstruction of orogenic belts in the

North Atlantic region_ The fits of continents are based on Bullard et al.
(1965).

Old Acidic Basements

Large parts of the orogenic belts of the Atlantic regions appear to have formed
on older acidic basements, which are exposed, for example, along the west coast
of Norway in the Caledonides, in many parts of the Hercynides (§ 14-4), and
along the Green Mountains in the Appalachians. Such evidence was regarded by
many authors as being against the hypothesis of continental growth (e.g.
Wynne-Edwards and Hasan, 1970; Semenenko, 1970). It was claimed that
pre-existing acidic materials were eroded and then deposited on other parts of
pre-existing basements, and metamorphosed or melted, and that the total area of
 14-6 METAMORPHIC BELTS IN THE ATLANTIC REGIONS /347
continents and the total amount of acidic materials on the surface of the earth
did not increase by orogenies.
However, such evidence is not necessarily against a secular increase in the
amount of acidic material on the surface of the earth. Acidic crusts may have
become generally thicker with geologic time. It is likely that parts of orogenic
belts in the Atlantic and circum-Pacific regions do not have older acidic
basements, and acidic crusts have been enlarged in such places. It appears that
the relations observed in the orogenic belts in the Atlantic regions are not valid
for all other orogenic belts of the world (Miyashiro, 1967a).
Probably the rate of plate motion in Phanerozoic time was much slower in
the Atlantic regions than in the Pacific. This could be the main cause for the
rarity of glaucophane schists and paired metamorphic belts in the Atlantic
regions (§ 17-2).
Chapter 15

Metamorphic Belts in Japan

and its Environs

15-1 PRESENT-DAY ARCS OF THE JAPANESE ISLANDS

Island Arc Structure

A brief account of the present-day tectonics of the Japanese Islands will be given
here as a preliminary to detailed discussions of their Mesozoic and older metamor-
phic belts. The major topography of the Islands is the result of late Tertiary and
Quaternary earth movements. It appears, however, that the movements were
largely controlled by the thickness of the pre-existing crust which formed by
Mesozoic and older orogenies. The Japanese Islands have a thick continental-
type crust. The highest area of Japan is the central mountains (called the Japan
Alps) of Honshu. Their present height was produced by continual upheaval in
the Tertiary and Quaternary. The Quaternary upheaval amounts to 1200 m. On
the other hand, this area has the thickest crust in Japan, and the upheaval was
controlled probably by the buoyancy of the crust formed by the older
orogenies. Therefore, the general shapes of the present-day arcs reflect the
features of older orogenic and metamorphic belts.
Japan is composed of four big islands, called, from northeast to southwest,
Hokkaido, Honshu (main island), Shikoku and Kyushu, together with a great
number of surrounding small islands. Fig. 15-1 shows the present-day island arcs
of Japan and its vicinity.
The main arc of Japan may be called the Honshu Arc, the trend of which
generally represents the direction of Mesozoic orogenic belts. As regards the
present-day arc structure, however, the Honshu Arc must be divided into the
Southwest Japan Arc and the Northeast Japan Arc, as shown in the figure. Only
the latter is an active island arc at the present day. The Southwest Japan Arc was
active during the Cretaceous and older orogenies. Along the axis of this Arc,
there is a great fault, called the Median Tectonic Line (MTL in fig. 15-1).
The Northeast Japan Arc is in contact on its southern end with another
presently active arc, called the Izu-Bonin Arc, which protrudes towards the SSE
into the Pacific Ocean. There is a great fault between the presently active regions
(Northeast Japan and Izu-Bonin Arcs) on the east and the more stabilized
 15-1 PRESENT-DAY ARCS OF THE JAPANESE ISLANDS/349

Pacific Ocean

Philippine Sea
'I

Fig. 15-1 Present-day island arcs in and around the Japanese Islands.
Slant ruling indicates trenches, and vertical ruling indicates deep basins in
marginal seas. Abbreviations: MTL = Median Tectonic Line, F- F =
Itoigawa-Shizuoka Line, H = Hokkaido, T = Kitakami Mountains, A =
Abukuma Plateau, K = Kii Peninsula, S = Shikoku, Y = Kyushu. The main
island containing T, A and K is called Honshu.

Southwest Japan Arc on the west (F-F in fig. 15-1). This fault runs in a NNW
direction through the city of Shizuoka across Honshu, and is called the
Itoigawa-Shizuoka Line after the names of the two cities at both ends. The
region immediately east of the Line is called the Fossa Magna (fig. 15-8).
The southwestern part of Hokkaido is an extension of the Northeast Japan
Are, whereas the rest of Hokkaido is made up of the Sakhalin and KurU Arcs.
The Ryukyu Arc connects Kyushu to Taiwan (Formosa).
350/METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS

There is one more arcuate structure which protrudes from Kyushu south-
eastwards to the Palau Islands. It is mostly made up of aseismic submarine ridges
running through the Philippine Sea. It may be the remnant of an extinct island
arc or a mid-oceanic ridge.
The nature of the Median Tectonic line will be discussed in § § 15-10 and
17-6. The structure of the Northeast Japan Arc will be described in detail in
§ 17-7.

Present-day Orogeny in Japan

Volcanoes, Trenches and Deep Earthquakes. There are more than 200
Quaternary volcanoes in Japan. About 40 of them have erupted in historical
time. Sugimura (1960, 1968) emphasized that the Quaternary volcanoes are
distributed in two broad belts along the presently active arcs, as shown in fig.
15-2. He named them the East Japan volcanic belt and the West Japan volcanic
belt. The former stretches along the Kuril, Northeast Japan and Izu-Bonin Arcs
on the west side of the Kuril, Japan and Izu-Bonin Trenches. The latter stretches
along the Ryukyu Arc on the west side of the Ryukyu Trench.
The intermediate region, i.e. the Southwest Japan Arc, is nearly devoid of
Quaternary volcanoes owing to its relatively stabilized condition.
There is a sharp eastern limit for the distribution of volcanoes in each
volcanic belt. The limit is called a volcanic front (Sugimura, 1968). Volcanoes
are especially crowded in a zone immediately west of the front. The distribution
pattern of Quaternary volcanoes is similar to that of late Tertiary ones in Japan.
This means that the present-day orogeny is a continuation of a late Tertiary one,
beginning in Miocene time (Sugimura, 1968; Sugimura et al. 1963, Matsuda,
1964; Matsuda, Nakamura and Sugimura, 1967).
The chemical characteristics of late Tertiary and Quaternary volcanic rocks
show a close relationship to the arc structure. As shown in fig. 15-3, tholeiitic
and calc-alkali rocks occur near the volcanic front (i.e. on the trench side) within
the East Japan volcanic belt, whereas tholeiitic and calc-alkali rocks with more
alkalic compositions together with some alkali rocks occur on the west side
within the belt (Sugimura, 1960, 1968; Kuno, 1959, 1968; Katsui, 1961;Aoki
and Oji, 1966; Miyashiro, 1972b). A similar westward increase of alkalic
character was observed in the West Japan volcanic belt also. Volcanic rocks of
islands in the Japan Sea and Korea are mostly alkalic.
Deep-focus earthquakes occur along a Benioff zone that dips to the west from
the trench. The generation of magmas for volcanism and their compositions
may be controlled by the plunging lithospheric slab suggested by the Benioff
zone(§ 4-2).
It is to be noted that a regular chemical zonation across an island arc occurs
in granitic rocks as well. Though most of the granitic rocks of Japan were
 15·1 PRESENT·DA Y ARCS OF THE JAPANESE ISLANDS /351

• Active volcano
() Other Quaternary volcano

Fig. 15-2 Distribution of Quaternary volcanoes in Japan. The heavy

arcuate lines represent the volcanic fronts for the East Japan and the West
Japan volcanic belt. (Sugimura, 1960, 1968.)
formed in relation to the Mesozoic and older orogenies, their chemical zonation
may be referred to here for comparison. Granitic rocks on the continental side
of the Honshu Arc tend to be less calcic than those on the oceanic side, just as
the Quaternary volcanic rocks do (Taneda, 1965).

Crust and Upper Mantle. The relatively stable Southwest Japan Arc appears to
have a crustal thickness of the order of 30-40 km, whereas the crust of the
active Northeast Japan Arc is about 30 km or thinner (e.g. Hashizume et al.
1968). Besides near-surface sediments, the crust is generally layered with an
acidic upper crust (Vp ::: c. 6·0 km sec-I) and a more basic lower crust (Vp ::: c.
6·6 km sec-I).
352/METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS

Asiatic Continent

Japan Sea

East China Sea

Fig. 15-3 Petrographic zonation of the Quaternary volcanic rocks of

Japan (Miyashiro, 1972b, with some revision). The full lines represent
volcanic fronts. 1 = tholeiitic and calc-alkali series with low alkali contents,
2 = tholeiitic and calc-alkali series with higher alkali contents, 3 =
calc-alkali series with higher alkali contents and alkali series, 4 = alkali series.

The compressional wave velocity in the uppermost mantle is about 7·7 km

sec- 1 beneath the active Northeast Japan Arc (Research Group for Explosion
Seismology, 1966). Such an anomalously low velocity in the uppermost mantle
has been observed in other active regions; e.g. Kamchatka and the Cordilleran
orogenic region of the western United States (Fedotov, 1968; Herrin, 1969).

Terrestrial Heat Flow. Fig. 15-4 shows that observed heat-flow values are
generally below 40 mW m -2 (1 tIcal cm -2 sec -1) in the Kuril and Japan
Trenches and the ocean-side rim of the active Northeast Japan and Kuril Arcs,
and that the values increase very rapidly toward the Asiatic continent, reaching
the highest values around 100 mW m- 2 (2·5 tIcal cm- 2 sec- 1 ) in the deep basins
of the Japan and the Okhotsk Sea (Uyeda and Vacquier, 1968). Thus, there is a
marked contrast between a low and a high heat-flow zone respectively on the
oceanic and continental sides of the active arcs.
 15-1 PRESENT·DA Y ARCS OF THE JAPANESE ISLANDS /353

"""
40'/'
,,
, "
I
\ ,. , , "
' .... _-,.,

,."---40

,,
I
I

,i " I
,I
"
1\ " ,,
\
', \

,,
I \

,,
\ \
\ \ \
'.
\
\
\
\
\
,
, ,
I

'.......... ~'
I

,,
I I
I --'
I I
I ..... - ..... I I ,-,
I / I \ I '

'f , I

\\ \\ \\\ '40\
\' \

"" I
"
\ \
\,
,,
,
I
I , \'
','," ",
I ,
\'
I I

80 100 ""
".... " ....'60

Fig. 15-4 Heat flow distribution in and around the Japanese Islands in
mW m- 2 (Uyeda and Vacquier, 1968).1/1 cal cm- 2 sec- 1 = 42 mW m- 2 •

A low heat-flow region should generally show lower temperatures than a high
heat-flow region at the same depth. Accordingly, a low- and a high-temperature
zone run in parallel beneath an active island arc. This contrast would be related
to the generation of paired metamorphic belts in the circum-Pacific region, as
will be discussed later.
The existence of the contrasted thermal zones has been confirmed also from
studies on electrical conductivity anomalies (e.g. Uyeda and Rikitake, 1970).
354/ METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS

15-2 THE ASIATIC CONTINENT AND THE JAPANESE ISLANDS

The Asiatic Continent and the Japan Sea

There is a wide Precambrian craton, usually called the Angaran Shield, in north
Siberia, as shown in fig. 15-5. There are a few smaller Precambrian cratons in
China. Some parts of them are more than 2000 m.y. old (Vinogradov and
Tugarinov, 1962).

~ Precambrian Craton

i D Palaeozoic folding

},\
~A~ o Mesozoic·Cenozoic
l:2J
~.:~.
Wtr4~~
folding
,. ~

A","'I;'"~
Shield~ jp
I:l .A
6T

Fig. 15-5 Generalized structure of Asia, Australia and the Western

Pacific.

It is widely believed that the Asiatic Continent has shown accretional growth
around these cratons, and especially around the Angaran Shield. The Himalayan
Mountains were probablY formed by collision of the Indian plate against the
Asiatic one, whereas the Japanese Islands represent a front of continental growth
toward the Pacific Ocean.
 15-2 THE ASIATIC CONTINENT AND THE JAPANESE ISLANDS/355

Small Precambrian terranes are exposed in Korea, and a part of the Japanese
Islands certainly has a Precambrian basement. It is likely that these masses were
formerly in the coast regions of the then Asiatic Continent, and that the old
sialic mass of Japan drifted away from the continent in a relatively young
geologic time (e.g. Matsuda and Uyeda, 1971).
A wide northwestern part of the Japan Sea is more than 3000 m deep, where
unconsolidated sediments are only a few kilometres thick, and the underlying
hard crust (Vp = 6·7 km sec-l) is of the oceanic type. The origin of the Japan
Sea is an important unsolved problem, like that of other mar"ginal seas on the
continental side ofisland arcs (e.g. Menard, 1967).
As regards the Tasman Sea between Australia and New Zealand, van der
linden (1969) has suggested that possible sea-floor spreading from a small
mid-oceanic ridge within the Sea might have resulted in the eastern movement of
New Zealand. This may be applicable to the Japan Sea, which, however, has no
topographic feature suggesting a mid-ocean ridge. Alternatively, the floor of the
Japan Sea may have been created not in a definite mid-ocean ridge but at
innumerable points therein owing to the materials rising through the mantle
beneath the whole area.
There is a remarkable zone of late Mesozoic and early Tertiary acidic-inter-
mediate volcanism and plutonism along the east coast of the Asiatic Continent
from the Gulf of Tonkin to the Bering Strait (fig. 15-6). It was called the East
Asiatic volcano-plutonic belt (Ustiyev, 1965). The late Mesozoic and early
Tertiary volcanism and plutonism in Japan show some resemblance to those of
the belt (Yamada, 1966; Matsumoto, 1968). It is possible that Japan was on the
east coast of the Asiatic continent in early Mesozoic and older time, and that the
Japan Sea was formed by the drift of Japan in late Cretaceous and/or Tertiary
time.

Precambrian and Palaeozoic Rocks of Japan

Whether a Precambrian basement exists beneath the Japanese Islands has long been
a matter of dispute (e.g. Matsumoto et al. 1968). Recently, Precambrian Rb-Sr
ages (600 '" 1200 m.y.) have been reported for certain gneissic rocks from the
Hida Plateau in central Honshu (Sato, 1968). K-Ar ages of 1600-1640 m.y.
have been obtained for biotite and muscovite samples that were separated from
metamorphic rock cobbles in Palaeozoic conglomerates in the same region. The
cobbles had probably been derived from a metamorphic complex of the Hida
Plateau (Shibata, Adachi and Mizutani, 1971). Thus, the existence of Pre-
cambrian metamorphic rocks in that region has been established. Radiometric
dating has further suggested that a metamorphic event took place in a
considerable part of Japan probably around Ordovician time (§ 15-5).
Though the available evidence is still meagre, it appears that phases of
geosynclinal sedimentation and/or orogeny took place continually from middle
356/METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS

o 500
,
1000 km
1

(~ Volcanism & Plutonism

B.:1H ainly Pl utonism

Fig. 15-6 East Asiatic volcano-plutonic belt, where magmatic activity

occurred from the Jurassic to the early Tertiary.

Precambrian time to Ordovician on and in the crust that later became the mass
of Japan. The width and the thickness of the sialic crust of that time are
unknown. The presently available palaeontological record in the Japanese Islands
begins with the Silurian.
There was a big geosyncline, called the Chichibu or Honshu geosyncline,
probably along the coast of the Asiatic Continent from Silurian to Permian time.
The resultant sedimentary pile is now well exposed in Honshu, Shikoku and
Kyushu. The bulk of the terrane exposed at present is Carboniferous and
Permian. Regional metamorphism took place repeatedly in this pile.

15-3 PAIRED METAMORPHIC BELTS

The Paired Metamorphic Belts of the Honshu Arc. Detailed petrographic

investigation has been made on metamorphic rocks of Japan, as summarized for
example in the recently published Metamorphic Facies Map of Japan at a scale
of 1: 2 000 000 (Hashimoto et al. 1970).
 15-3 PAIRED METAMORPHIC BELTS/357
The metamorphic events in the Jurassic and Cretaceous produced a pair of
regional metamorphic belts in the sedimentary pile of the Honshu geosyncline.
The pair is composed of the Ryoke (pronounced Ry6ke) belt of the low-pressure
type on the continental side and the Sanbagawa belt of the high-pressure type on
the Pacific Ocean side as has already been shown in fig. 3-10. In this book, the
continental side (usually the concave side of an arc) is sometimes called the inner
side, and the oceanic side (usually the convex side) the outer side.
Another pair of metamorphic belts is present on the north side of the
Jurassic-Cretaceous pair in Honshu. It is composed of a part of the Hida
metamorphic complex on the inner side and the Sangun belt of western Honshu
on the outer. A large part of the westward extension of the former lies beneath
the Japan Sea (table 15-1). The whole of the Hida complex and its extension are
shown as the Hida metamorphic terrane in fig. 3-10. The metamorphism of this
pair is probably Permian to Jurassic (table 15-1).
The above-mentioned two pairs were recognized and were called the younger
and older pairs by Myashiro (1961a), but recent radiometric dating has revealed
that probably there was a still older pair in Honshu. It is made up of another
part of the Hida metamorphic complex on the inner side and the circum-Hida
metamorphic zone on the outer, both metamorphosed probably in Carboni-
ferous time (table 15-1; fig. 15-8).

Possible Deformations of Arcs. The Jurassic-Cretaceous belts shows a sharp

turn of trend at the Itoigawa-Shizuoka Line and Fossa Magna. It is unknown
whether this turn is a primary or a secondary feature of the metamorphic belts.
If it is primary, Southwest Japan and Northeast Japan should have been two
distinct arcs in Jurassic-Cretaceous time. The Izu-Bonin arc mayor may not
have existed at that time. The oldest known rocks of the Izu- Bonin Arc off
Honshu are early Tertiary pyroclastics in the Bonin Islands.
If it is due to secondary bending, the deformation may have taken place by
the tectonic movement along the Izu-Bonin Arc during Tertiary time, and/or
by that which took place during the drift of the Japanese Islands away from the
Continent in late Cretaceous and Tertiary time.
In the Northeast Japan Are, Cretaceous granitic rocks of the Kitakami
Mountains (110-120 m.y.) show remanent magnetization generally in the
direction of N 60° _30 0 W, and those of the Abukuma Plateau (90-100 m.y.) in
the direction of N 500 _0 0 W, as schematically shown in fig. 15-7. On the other
hand, Cretaceous granitic and volcanic rocks of the Southwest Japan Arc are
generally magnetized in the direction of N 30°-60° E. These data suggest that
Northeast Japan made successive counterclockwise rotations with reference to
Southwest Japan in Cretaceous and later time (Kawai, Ito and Kume, 1961; N.
Kawai, pers. comm. 1970).
 w
VI
()O

;:
-
ttl
...,
:>
;:
Table 15-1 Tentative chronology of regional metamorphic events in the Honshu Arc of Japan o
~
."
Broadly simultaneously :I:
metamorphosed area of low- ()
High-pressure metamorphic area pressure type with granites I:tI
ttl
Event No. Geologic age (radiometric age) (radiometric age) t"'
...,
en
6 Miocene Not exposed yet (?) Northeast Japan (3-25 m.y.) ....
Z
to the present .....
5 Jurassic to Sanbagawa belt (110 m.y.); Ryoke belt (c. 105 m.y.); :>
."
Cretaceous Part of Shimanto terrane Abukuma Plateau (100-120 m.y.) :>
Z
4 Permian to Sangun belt of western Hida IV (180-240 m.y.) :>
Jurassic Honshu (160 m.y.) z
o
3 Carboniferous Circum-Hida zone including Hida III (320 m.y.);
Omi area (320 m.y.)
...,....
Maizuru zone (330 m.y.) en
2 Ordovician Kiyama (450 m.y.) Hida II (500 m.y., Cambro- ttl
Z
Ordovician); <:
Kurosegawa zone (430 m.y.) ....
~
Precambrian Hida I (1,640 m.y.) o
Z
en
 15-3 PAIRED METAMORPHIC BELTS /359
Accordingly, the shape of the Northeast Japan Arc during metamorphism
would be restored probably by a clockwise rotation of the Abukuma Plateau by
about 60°. This rotation gives the Jurassic-Cretaceous pair of metamorphic belts
a more natural-looking shape.

110-120 m.y.

Fig. 15-7 Mean directions of remanent magnetization of Cretaceous

granitic and volcanic rocks in three parts of the Honshu Arc. Dots
represent the localities of measured specimens. (After N. Kawai and T.
Nakajima; Kawai, pers. comm. 1970.)

Thermal Structure and Associated Magmatism. In the two metamorphic belts of

the Jurassic-Cretaceous pair, the original thermal structure is fairly well
preserved. The two belts are in contact with each other at the Median Tectonic
Line. The thermal axis of each belt is near the Line (fig. 15-13).
In the older metamorphic belts, the original thermal structure is not well
known because of the later destruction and less advanced investigation.
The relationship between the baric types of regional metamorphism and the
characteristics of the associated magmatism as outlined in § 4-1 holds good in
each belt of all the pairs. Ophiolites including serpentinites are abundant in
high-pressure metamorphic belts and regions on the outer side, whereas granitic
bodies are confined to low-pressure metamorplj.ic belts and regions on the inner
side. Thus, there are three remarkable belts of ophiolites associated with the
three high-pressure metamorphic belts of southwest Japan.

Paired Metamorphic Belts in Hokkaido. There is one more pair of belts in

Hokkaido, as shown in fig. 3-10. It is composed of the Hidaka belt of the
360/METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS

low-pressure type on the east and the Kamuikotan belt of the high-pressure type
on the west. The original sedimentary rocks here are probably Mesozoic and
their regional metamorphism is probably Cretaceous and Tertiary.
The main part of Hokkaido, where this pair is exposed, belongs to the
Sakhalin Arc situated on the west side of the Okhotsk Sea. The paired belts were
formed probably along the ancient Sakhalin Arc with the associated trench on
the west side. Since the west side is generally the continental side, the Arc may
be regarded as a reversed island arc similar to the present-day arc of the New
Hebrides in the southwest Pacific Ocean (§ 17-1).

15-4 HlDA METAMORPHIC COMPLEX

Distribution and Petrographic Characteristics. Metamorphic and granitic rocks

are exposed in wide areas of the Hida Plateau on the Japan Sea of central
Honshu (figs. 3-10 and 15-8). This metamorphic terrane is mainly composed of
quartzo-feldspathic gneisses, amphibolites and crystalline limestones. Sillimanite
and pyralspite garnet are widespread, and wollastonite is common in limestones.

_Low-pressure metamorphic terrane

~Hida. Ryoke and Abukuma)

F~High-pressure metamorphic terrane

~-~~ Sangun and Sanbagawa)

Oki

Po·

o. .
400km

Fig. 15-8 Metamorphic belts and well-documented metamorphic areas in

the main part of Japan. Abbreviations: CH = circum-Hida metamorphics,
F-F = Itoigawa-Shizuoka Line, MTL = Median Tectonic Line, KS =
Kurosegawa zone, MZ = Maizuru zone.
 15-5 OLDER HIGH-PRESSURE MET AMORPHIC TERRANES/361
Most of the rocks belong to the epidote-amphibolite, amphibolite and
granulite facies (Sato, 1968; Suzuki and Kojima, 1970). The distribution of
metamorphic temperature in the complex, however, has not been clarified, and
is presumably irregular. The metamorphism is of the low-pressure type and
possibly partly of the medium-pressure type. Andalusite, kyanite, sillimanite and
staurolite occur in the Kurobegawa area near the northeastern end of the
complex. No ultramafic rocks occur.
Similar metamorphic and granitic rocks occur in the Oki Islands in the Japan
Sea and also within a faulted area, called the Maizuru zone, west of Kyoto (fig.
15-8). Therefore, the Hida metamorphic complex is considered to extend
westward in a wide zone, most of which is now covered by the Japan Sea and
Palaeozoic and younger sediments.

Age Group. The Hida complex appears to contain metamorphic and plutonic
rocks of a number of different ages, ranging from the middle Precambrian to the
Jurassic. Radiometric dating made in several recent years has greatly contributed
to the elucidation of the long, complicated history, though the age and the
nature of individual events have not been established yet.
The Precambrian events are represented by K- Ar ages of about 1640 m.y.
(Shibata et al. 1971) and a Rb-Sr age of about 1200 m.y. (Sato, 1968). As the
details of the events are unknown, they are shown as Hida I in table 15-1. There
are four Phanerozoic age clusters around 500 m.y. (Cambro-Ordovician), 320
m.y. (Carboniferous), 240 m.y. (Permian) and 180 m.y. (Jurassic). Tentatively,
the complexes corresponding to these clusters are shown as Hida II-IV in table
15-1.
The metamorphic complex of the Hida IV event is here regarded as
representing the low-pressure metamorphic belt paired with the Sangun belt. The
belt of Hida IV was probably formed on the basement of older metamorphic
belts, a large part of which was probably reactivated during the event. The age
cluster around 180 m.y. is most marked and widespread allover the Hida
Plateau (e.g. Nozawa, 1968). This appears to indicate the age of the consolida-
tion of the Funatsu granite and related regional metamorphism.

15-5 HIGH-PRESSURE METAMORPHIC TERRANES OLDER THAN THE

SANGUN BELT
In various parts of Honshu and Kyushu, there are a number of small areas
composed of low-temperature metamorphic rocks occasionally containing
glaucophane besides the Sangun metamorphic belt to be described later. Recent
radiometric dating has revealed that at least some of the small areas were
metamorphosed much earlier than the Sangun (table 15-1). Two age groups will
be commented upon below.
362/METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS
Kiyama Group (Ordovician)
In central Kyushu, a glaucophane-bearing metamorphic complex is exposed in
the Kiyama area (fig. 15-8). It has given an Ordovician age (about 450 m.y.) by
the Rb-Sr method (Yamamoto, 1964; Karakida et al. 1969).
It is noted that the metamorphic event around 500 m.y. in the Hida complex
(i.e. Hida II) is relatively close in age to the Kiyama event. A similar age has
been reported on some rocks from the Kurosegawa tectonic zone, which is a line
of fault blocks uplifted from the basement within the Paleozoic terrane on the
Pacific Ocean side of the Southwest Japan Arc (Matsumoto et al. 1968; Hayase
and Nohda, 1969).
These data suggest widespread regional metamorphism in and around
Ordovician time in Japan. The metamorphism was partly of the high·pressure
series and partly associated with granitic activity. However, the shapes and
relations of metamorphic belts at that time are not known.

Omi Group (Carboni/erous)

There are several small metamorphic areas along an arcuate line on the south and
east sides of the Hida metamorphic complex. This group is sometimes called the
circum-Hida metamorphics (fig. 15-8). They are commonly in the greenschist
facies and rarely in the glaucophane·schist and the epidote-amphibolite facies
(e.g. Seki, 1959).
The Omi area is at the northeastern end of the arcuate line (Banno, 1958).
Metamorphic rocks of this area have given Rb-Sr and K- Ar ages of about 320
m.y., i.e. Carboniferous (Yamaguchi and Yanagi, 1970).
A cluster of isotopic ages in the Hida complex occurs at the same age (i.e.
Hida III). Some rocks in the Maizuru zone also give similar ages (Hayase and
Ishizaka, 1967). Thus, it is conceivable that widespread regional metamorphism
took place in the Carboniferous, and that paired metamorphic belts composed of
a Hida and the circum-Hida zone were produced.
The metamorphic rocks of the Chikugo area (fig. 15-8) give a Rb-Sr age of
about 270 m.y. (Yamaguchi and Yanagi, 1970). It is not clear whether they
belong to the Omi group.
Several separate metamorphic areas, occasionally containing glaucophane,
occur on the east side of the Abukuma Plateau and in the Kitakami Mountains;
e.g. Yaguki area (fig. 15·8). Their metamorphism was claimed to be pre-Late
Devonian (Kanisawa, 1964), but a part of their original rocks was recently found
to be Permian. The K- Ar age of a metamorphic rock from one of the areas was
300 m.y. (Carboniferous). Thus, these areas would belong to two or more age
groups.

15-6 SANGUN METAMORPHIC BELT

Distribution. A few tens of small areas of schists and phyllites are exposed
within a broad zone with an ENE trend in westem Honshu (fig. 15-8). The
 15-7 RYOKE METAMORPHIC BELT AND ABUKUMA PLATEAU/363
Palaeozoic terranes lying between these areas also show incipient metamorphism,
though they have usually been called 'unmetamorphosed' in the literature. They
all are regarded as constituting one and the same metamorphic belt, called the
Sangun. Glaucophane occurs occasionally. Serpentinites occur in many places in
the zone. The metamorphic rocks are intruded and affected by later granitic
rocks, which are genetically related to the Ryoke metamorphism and not to the
Sangun.
The age of deposition of the oriWnal sediments is at least partly Permian and
Carboniferous. On the other hand, it has been claimed that metamorphic rocks
in a certain area are unconformably covered by unmetamorphosed Middle
Triassic sediments. Thus, the age of Sangun metamorphism appears to be very
late Permian or early Triassic, that is, immediately after the deposition of
the relevant sediments. A recent Rb-Sr and K-Ar dating of the belt, however,
has given a Jurassic age of about 170 m.y. (Shibata and Igi, 1969).
The name Sangun was derived from a mountain in northern Kyushu, where
unfortunately the Sangun rocks were mostly subjected to con tact meta-
morphism by Cretaceous granites. Therefore, the type areas of Sangun meta-
morphic rocks are considered to be in western Honshu in this book without
regard to the etymology.
Characteristics. The Sangun metamorphic belt is composed of patchy areas of
schistose metamorphic rocks surrounded by poorly recrystallized non-schistose
Palaeozoic terranes. The patchy areas, commonly tens of kilometres across, are
composed of schists and phyllites in the glaucophane-schist and the greenschist
facies (§ 6A-5).Lawsonite is very rare and jadeite does not occur in association
with quartz.
The pelitic rocks in the surrounding Paleozoic terranes are usually very poorly
recrystallized, but the intercalated basaltic rocks usually show more advanced
recrystallization in the prehnite-pumpellyite or the greenschist facies (Nureki,
1969; Hashimoto and Saito, 1970).
The Sangun belt appears to have a number of subparallel thermal axes,
though available data are very meagre. The associated serpentinites have a
chemical composition similar to ordinary alpine-type masses, and contain relict
olivine with 6-10 per cent Fa (Research Group of Peridotite Intrusion, 1967).

15-7 RYOKE METAMORPHIC BELT AND ABUKUMA PLATEAU

Ryoke Metamorphic Belt
The Ryoke metamorphic belt extends from the Shiojiri-Takato area immedi-
ately west of the Itoigawa-Shizuoka Line, westward through the Southwest
Japan Arc, to central Kyushu (figs. 3-10,15-8, and 15-9).
A great amount of related granitic rocks occurs in the metamorphic terranes
and in a wide region to the north. Most or all of them appear to be late- and
post-metamorphic intrusions. The large-scale distribution of metamorphic tem-
 w
~
3:
--
t!j
....,
>
3:
o!
l00km
I
o
:;.::I
."
::t:
c:;
tl:!
t!j
t"'
....,
til

Z
.....
>
."
. >
Z
• M8S\lZOlc - CenoZOIC
[2]
• granite
Ryoke >
Z
o Amphlbohte tacteS ) bell
o
Greenschist factes ::ltil
t!j
Z
m:_:
._Greenschist taCIt's ~~:te - amphibolite) <
.....
~ Akaoshl :;.::I
~ Preh -pump
U
&
htgher zeoli te faCies
lMountains ~ Greenschist &
IS&:l glaucophane - schIst
Sanbagawa
belt
o
Z
R::l facies til
~ Preh-pump facIeS

Fig_ 15-9 Distribution of metamorphic facies in the Ryoke and San-

bagawa belts and the Akaishi Mountains. (Hashimoto et al. 1970.)
 15-7 RYOKE METAMORPHIC BELT AND ABUKUMA PLATEAU/365
perature is independent of individual granitic intrlisions. Contact metamorphism
has been noticed around some intrusions. No peridotite occurs.
The metamorphic rocks are mostly of pelitic and psammitic compositions
accompanied by subordinate quartzite and limestone. Metabasites are rare. The
metapelites are usually high in KINa ratio, and in excess Al 20 3 content, like the
virtually unmetamorphosed, late Palaeozoic pelites exposed to the north of the
belt.
Most of the well-recrystallized parts of the belt belong to the amphibolite
facies, and the rest to the greenschist facies. The belt is of the low-pressure facies
series.
In the eastern end of the belt (Le. Shiojiri-Takato area), the gradual
transition from virtually unmetamorphosed slates exposed in the north into
schists and gneisses exposed in the south has been well documented, and has
already been reviewed in § § 7 A-2 and 8A-2, and fig. 3-4. Andalusite occurs in a
medium-temperature zone, and sillimanite in a high-temperature zone. Cordierite
is common, but pyralspite garnet is relatively rare.
In the eastern end and many other parts, the temperature of metamorphism
increases southward or southeastward across the belt to the Median Tectonic
line, which marks the southern limit of the belt. In the western parts, however,
a thermal axis appears to exist within the belt as shown in fig. 15-13 (Miyashiro,
1959, 1961a;Suwa, 1961).
The well-documented Higo metamorphic complex in central Kyushu could
represent the western end of the Ryoke belt (Ueta, 1961; Yamamoto, 1962;
Tsuji, 1967).
Most of the granitic rocks in southwest Japan were formed in relation to the
Ryoke metamorphism (fig. 15-10). The 206Pbj 238 U age of zircons from
granites in this belt is about 90 m.y. (Ishizaka, 1969). The Rb-Sr ages of granites
are 105-60 m.y., i.e. Cretaceous to early Tertiary (Hayase and Ishizaka, 1967;
Yamaguchi and Yanagi, 1970; also Shibata, 1968).
Figure 15-11 shows that the distribution of the K-Ar ages of granitic rocks
shows the existence of regional age zones (Kawano and Ueda, 1967a, b). Most
granitic rocks in the Ryoke belt belong to a 75-95 m.y. zone (Cretaceous),
whereas those in the eastern part and to the north of the belt belong to a 50--65
m.y. zone (early Tertiary). This could mean that granite magmas were formed by
events which occurred on a regional scale, and that the theatre of the events
moved gradually northwards. Alternatively, these age zones may simply repre-
sent the times of their uplift which led to the cooling of granitic bodies.
It is to be noted that all these radiometric data are concerned with granitic
rocks only. The true age of regional metamorphism is difficult to establish,
because most of the metamorphic rocks may have been more or less affected by
post-metamorphic granitic intrusions, and the age of cooling may have been
considerably different from that of metamorphic climax.
366/METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS

o 100 200km
!

~ Triassic & Jurassic

• Cretaceous & Early Tertiary

Late Tertiary

Fig.15-10 Distribution of granitic rocks in Japan. F-F = Itoigawa-

Shizuoka Line. The region to the south of the Butsuzo Line (B- B) is the
Shimanto terrane. Triassic and Jurassic granitic rocks are related to a Hida
metamorphism (event 4 in Table 15-1), and Cretaceous and early Tertiary
ones to the Ryoke. Late Tertiary granitic rocks in the Southwest Japan
and Northeast Japan Arcs are related respectively to the Shimanto and
Mizuho orogenies.

Thus, the age of the Ryoke metamorphism could be considerably more than
100 m.y. old.

Abukuma Plateau
The eastward extension of the Ryoke belt beyond the Itoigawa-Shizuoka Line
and Fossa Magna could be represented by small exposures of granitic rocks to
the north of the Kanto Mountains and the Tsukuba metamorphic area, and a
great granitic and metamorphic terrane in the Abukuma Plateau (figs. 15-8 and
15-13).
Andalusite and sillimanite occur in some metapelites in these terranes. The
temperature of metamorphism increases generally toward the west in Tsukuba
and Abukuma (§ § 7 A-2 and 8A-2).
15-7 RYOKE METAMORPHIC BELT AND ABU KUMA PLATEAU /367

C,.,~,
• 75·95

m 90·100 }
~ 110·120

170·180 Jurassic

Fig. 15-11 K-Ar ages of granitic rocks in Japan. (Kawano and Ueda,
1967a, b.)
368/METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS

The K-Ar ages of granitic rocks from the Abukuma Plateau are 90-100 m.y.
(middle Cretaceous), whereas those from a region to the west (i.e. on the
continental side) are generally 50-65 m.y. (Kawano and Ueda, 1967a, b).
The metamorphic rocks in the southern part of Abukuma have given a K- Ar
age of 120-100 m.y., slightly older than that of the associated granitic ones
(Ueda et al. 1969). In all these respects, the Tsukuba and Abukuma meta-
morphics resemble the Ryoke terrane.
On the other hand, the metamorphic terrane of the Abukuma Plateau shows
some characteristics different from those of Ryoke. For example, metabasites
are abundant, in contrast to their scarcity in Ryoke. Though andalusite and
sillimanite occur in both Ryoke and Abukuma, chloritoid, staurolite and kyanite
were found to occur rarely in Abukuma, but not in Ryoke. Thus, the Abukuma
Plateau differs from the Ryoke belt in the composition of the original rocks and
in the P-T condition of recrystallization (Tagiri, 1971).
There are several possible explanations for this. One explanation could be
that the Median Tectonic Line cut the Ryoke belt lengthwise so that only a half
of the belt on the continental side is now observed. On the other hand, only half
the belt on the Pacific Ocean side is exposed in the Abukuma Plateau (fig.
15-13). Therefore, the abundance of metabasites in Abukuma as compared with
Ryoke may simply represent the increase of such rocks toward the Ocean. This
may be comparable to the tendency found in geosynclines of North America
(§ 13-4).
The present crust beneath the Abukuma Plateau appears to be about 10 krn
thinner than that beneath the Ryoke belt. The Abukuma Plateau has been a
region of continual elevation and erosion since Late Tertiary time owing to a
tectonic effect of a new orogeny along the Japan Trench. This may have caused
a great extent of denudation there. The present exposure in Abukuma may
represent a much deeper level in the original structure than that in the Ryoke
belt. Another possibility is that the Abilkuma Plateau includes some areas of
pre-Ryoke crystalline basement rocks which were re-metamorphosed at the same
time as the rocks of the Ryoke belt, though there is no positive evidence for it.
Gabbroic rocks and cortlandtite occur in both Ryoke and Abukuma, while
serpentinite is virtually confined to the latter area.

Granite-Rhyolite Association
The close association of a vast quantity of rhyolitic volcanics with granitic rocks
genetically related to the Ryoke metamorphism (fig. 15-12) deserves special
attention.
The volcanics are mostly welded tuffs and other pyroclastic rocks of rhyolitic
composition. In some cases, there is a smaller amount of dacitic and andesitic
rocks. A large mass in central Japan is called the Nohi rhyolites, covering an area
of about 5000 krn 2 with an average thickness presumably of about 2 km.
 15-S SANBAGAWA METAMORPHIC BELT/369
The extrusion of the rhyolitic rocks appears to have occurred in late
Cretaceous time. The rhyolitic rocks are cut and metamorphosed by some
granitic intrusions, but unconformably cover other ones (Yamada, 1966;
Yamada and Nakai, 1969; Yamada, Kawada and Morohashi, 1971).
The close spatial association, the broad contemporaneity and the chemical
similarity of the granitic and rhyolitic rocks suggest the existence of a genetic
connection between them (Miyashiro, 1965; Yamada et al. 1971). These rocks
could represent an ancient volcanic chain (§ §4-2 and 15-2).

p;;;;j] Mainly Rhyolitic Volcanics

• Mainly Andesitic Volcanics

o
-'---'---'
100 200 km

Fig. 15-12 Distribution of Cretaceous volcanic rocks in Japan.

15-S SANBAGAWA METAMORPHIC BELT

Characteristics. The Sanbagawa (Sambagawa) metamorphic belt runs mostly on

the Pacific Ocean side of the Ryoke belt with the Median Tectonic Line between
them. The eastward extension of the Sanbagawa belt beyond the Itoigawa-
Shizuoka Line is exposed in the Kanto Mountains (figs. 3-10 and 15-S).
Glaucophane occurs occasionally but not abundantly. Jadeite and lawsonite
have mostly been found in the eastern half of the belt. Most of the belt belongs
to the glaucophane-schist and the greenschist facies. Zones of the epidote-
amphibolite facies occur in a few areas in the western half of the belt. Biotite
occurs only in rocks of the epidote-amphibolite facies (§ § 7A-9 and SA-S).
The metapelites in the Sanbagawa belt are lower in KINa ratio and in excess
Al 2 0 3 than those of the Ryoke belt. The absence of paragonite, chloritoid and
staurolite in metapelites is probably due to the low Al 2 0 3 content and other
370/METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS

compositional characteristics of the rocks. Paragonite was found only in some

metabasites. Piemontite is a common mineral in metachert. Piemontite schists
make good marker horizons for structural analysis of the terrane (e.g. figs. 4 and
lOin Iwasaki, 1963).
The temperature of metamorphism increases generally northward to the
Median Tectonic Line. In the opposite direction, the metamorphic terrane
usually grades into virtually unmetamorphosed Palaeozoic formations. In many
areas, the temperature increases down the apparent stratigraphic succession, but
in some others it increases up the apparent succession. Generally, the tempera-
ture distribution shows no relation to igneous rocks.
An exceptional situation exists in the Bessi area in Shikoku. There, the
highest temperature zone is associated with large basic and ultrabasic masses,
which are situated about 6 km to the south of the Median Tectonic Line.
The stratigraphy and structure have been worked out in Shikoku (Kojima,
1963). Large-scale inversion of the succession has been found by Kawachi
(1968). There are a few great fault zones trending parallel to the elongation of
the belt (fig. 17-1). They may represent the position of Benioff zones at some
stages of the orogeny (§ 17-6).
The Rb-Sr ages of Sanbagawa metamorphic rocks are 110-85 m.y. (Hayase
and Ishizaka, 1967; Yamaguchi and Yanagi, 1970), and the K-Ar ages are
102-82 m.y., i.e. late Cretaceous (Banno and Miller, 1965). However, the
geologic evidence suggests an early Cretaceous and Jurassic age (e .g. Ernst et al.
1970). The above radiometric ages could represent the time of cooling. As most
of their original rocks were deposited in late Palaeozoic time, the metamorphic
events did not immediately follow the sedimentation. Geosynclinal sedimenta-
tion concurrent with the metamorphic events took place in the Shimanto terrane
on the south side of these metamorphic belts, as will be described later.

Ophiolites. The Sanbagawa belt is accompanied by a large amount of ophiolites

in great diversity. Many of them are of the non-sequence type (as defined in
§4-1), but some ophiolites along the Mikabu Tectonic Line in the belt show
tendencies toward regular sequences. A downward sequence, basaltic pyro-
clastics --+ dolerite --+ gabbro (Suzuki, 1964, 1967), and a symmetrical sequence of
basalt --+ dolerite --+ gabbro --+ dolerite --+ basalt (Suzuki, Sugisaki and Tanaka,
1971) have been observed.
The peridotites and serpentinites along the Mikabu Tectonic Line have higher
(Fe 2+ + Fe~/Mg ratios than those of the typical non-sequence type ophiolites in
other parts of the Sanbagawa belt and the common alpine type ultramafics of
the world. Olivine relics of Mikabu serpentinites usually contain 12- 20 per cent
Fa (Research Group of Peridotite Intrusion, 1967; Hayashi, 1968; Iwasaki,
1969; Miyashiro, 1972b).
 15-9 METAMORPHISM OF THE SHIMANTO TERRANE /371
It is conceivable that the Jurassic-Cretaceous metamorphism in the
Sanbagawa terrane may represent only the last major phase of recrystallization
in this belt, and have effaced all or most of the mineralogical characteristics in
the earlier phases. The abundant extrusion of basaltic rocks of the non-sequence
type ophiolites in this belt is simultaneous with the late Palaeozoic sedimenta-
tion in the geosyncline. Since ophiolites are characteristic of high-pressure
metamorphic terranes, the geosynclinal area of the Sanbagawa belt should have
had features characteristic of a high-pressure metamorphic terrane already in late
Palaeozoic time. This may mean that underthrusting of a lithospheric slab began
in late Palaeozoic time in this zone. If this is the case, the late Palaeozoic
geosyncline of Japan should have had two subparallel underthrusting zones
corresponding to the Sangun and Sanbagawa belts.

15-9 METAMORPHISM OF THE SHIMANTO TERRANE AND THE

MIZUHO OROGENIC BELT

Shimanto Geosyncline and Metamorphism

In early Mesozoic time, the Palaeozoic Honshu geosyncline disappeared, and
large parts of the geosynclinal zone were elevated above sea level. Instead, a new
big geosyncline was created in middle Mesozoic time on the Pacific Ocean side of
the site of the Honshu geosyncline. The thick accumulation in the new
geosyncline, called the Shimanto geosyncline, is now exposed in the so-called
Shimanto terrane, which includes the southern parts of the Kii Peninsula,
Shikoku and Kyushu, in the Southwest Japan Arc (figs. 15-10 and 17-1). This
geosyncline perSisted through Palaeogene Tertiary time. The eastward extension
of the geosyncline is exposed in an area to the south of the Kanto Mountains,
where the Mizuho orogeny overprinted it.
Low-temperature regional metamorphism in the prehnite-pumpellyite and
greenschist facies took place in large parts of the terrane. Pumpellyite is
widespread, and stilpnome1ane is rare. However, it is likely that the meta-
morphism in this terrane had a number of distinct phases with very different
geothermal gradients. Different parts probably belong to different baric types.
The early phase of metamorphism is presumably late Jurassic and Cretaceous
and simultaneous with the Ryoke and Sanbagawa metamorphism. It is likely
that the main phase of the Sanbagawa metamorphism and the early phase of the
Shimanto metamorphism are one and the same event. In the central Kii
Peninsula, there is a well-documented metamorphic area that is continuous from
Palaeozoic rocks in the north into Jurassic ones in the south. A zonal
arrangement of greenschist, prehnite-pumpellyite, and zeolite facies rocks is
clear generally with lower temperatures of metamorphism in younger rocks
(§6A-4). This area appears to represent the early phase of the Shimanto
metamorphism, giving a Rb-Sr age of 110 m.y. (Yamaguchi and Yanagi, 1970).
372/METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS
Another metamorphosed area of Shimanto, which is situated in the Akaishi
Mountains (§6A-2), may represent a higher geothermal gradient and a younger
age. More than ten small granitic masses were intruded into the Shimanto terrane
in the Southwest Japan Arc. Their K- Ar ages cluster around two values
corresponding to the Miocene: 14 and 21 m.y. (Kawano and Ueda, 1967a, b;
Shibata, 1968). It is conceivable that related low-pressure metamorphism took
place in Miocene time in some part of Shimanto. It is a matter of definition
whether this phase of metamorphism is regarded as belonging to the Shimanto
events or to the Mizuho orogeny to be discussed below.

Mizuho Orogeny
A series of orogenic events began in early Miocene time along the Kuril,
Northeast Japan, and Izu-Bonin Arcs. Intense volcanism began along these arcs
and geosynclinal accumulation with strong folding occurred in northeast Japan.
The volcanism has continued through the Pliocene to the present. This new
orogeny is called the Mizuho orogeny (Sugimura et al. 1963; Matsuda et al.
1967).
Volcanism, earthquakes, trench formation and other present-day tectonic
activity in and around the Northeast Japan Arc as summarized in § 15-1 may be
regarded as manifesta tions of this orogeny.
Miocene geosynclinal accumulations, a few kilometres thick, were formed on
the basements of Mesozoic and older metamorphic and granitic rocks on the
Japan Sea side of northeast Japan. Among the products of the first intense
Miocene phase of volcanism, rhyolitic and dacitic rocks were more abundant
than basaltic and andesitic rocks. Subsequently, the proportion of andesitic
rocks increased remarkably, and the increase has gone on to the present.
To the east of the geosynclinal zone, the Abukuma and Kitakami Mountains
continued to rise as blocks from the Miocene to the present. This region is
characterized by great positive Bouguer anomalies. The western limit of the
region, called the Monoka-Shirakawa Line, runs through the cities of Morioka
and Shirakawa. Probably the Line represents a big fault, possibly cutting through
the whole thickness of the sialic crust (§ § 17-6 and 17-7).
Considerable parts of the Miocene geosynclinal pile in northeast Japan were
metamorphosed in the zeolite, prehnite-pumpellyite, and greenschist facies.
Examples of the Mogami area and the Tanzawa Mountains were described in
§6A-2. The latter example is exceptional in that metamorphic rocks up to the
amphibolite facies were formed. The calculated geothermal gradients in these
areas were very high.
Several tens of granitic masses occur over the geosynclinal zone of northeast
Japan. Their K-Ar ages are 3-25 m.y. (Miocene and Pliocene). Only the larger
masses are shown in fig. 15-10 under the name of late Tertiary granitic rocks.
These granites are similar in age to those of the Shimanto terrane of southwest
Japan.
 15-10 METAMORPHIC BELTS IN THE HONSHU ARC /373

15-10 SUMMARY OF THE CHRONOLOGY AND DISTRIBUTION OF

METAMORPHIC BELTS IN THE HONSHU ARC

Chronological Summary
A tentative chronology of six major metamorphic events has already been given
in table IS-I.
The Hida metamorphic complex has preserved records of the first four events.
The corresponding terranes are designated as Hida I-IV in the table. In other
words, granite-bearing metamorphic terranes in all these events are exposed
within the Hida Plateau. This suggests that there was no great shift of the site of
successive metamorphic belts in this period, which would be the main reason
why these older metamorphic belts were obscured and destroyed by later events.
The 4th and 5th events showed successive southward shifts of the site of the
resultant metamorphic belts. This resulted in relatively good preservation of the
metamorphic belts recording these events.
There is no evidence for the formation of glaucophane schists in the
Precambrian event. However, the mineral certainly occurs in the Ordovician
Kiyama complex of the 2nd event. In the 3rd event, the Hida and circum-Hida
terranes were perhaps in a pair. The 4th and 5th events formed two conspicuous
pairs of metamorphic belts. The high-pressure belt of the 6th event, if formed, is
not exposed yet.
The fust five events took place along the Honshu Arc, leading to the
formation of a thick continental type crust, whereas the Southwest Japan Arc
has remained outside the main theatre of the 6th event. Presumably the Japanese
Islands drifted away from the Asiatic Continent in the period between the 5th
and 6th events.

Nature of the Median Tectonic Line

The relation between the Median Tectonic Line and the thermal axes of the
Ryoke metamorphic belt is shown in fig. IS-B. It shows that in the eastern part
of the Southwest Japan Arc, most of the outer half (southeastern half) of the
Ryoke belt and the inner half (northwestern half) of the Sanbagawa belt has
been lost by fault movement along the Median Tectonic Line. That the Line is a
sharp southern limit for distribution of Cretaceous and early Tertiary granitic
rocks (fig. 15-10), also suggests that the lost zone is of a considerable width (say,
30 km or more).
Since the fault cuts the metamorphic belts and granites sharply, the main
movement must have taken place after the completion of Jurassic and Creta-
ceous metamorphism and plutonism, though this does not preclude the possibil-
ity that the initiation of the Line took place in an older time, say, in the
Cretaceous. Cataclasis took place on the fault, and some rocks along the eastern
part of the Line have been interpreted by many authors as protoclastic
intrusives.
374/METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS

It is widely believed that movement along the Line took place in a number of
distinct phases showing different patterns of movement. The initial phase,
usually called the Kashio, is widely believed to be thrusting of the Ryoke terrane
over the Sanbagawa, though there is no positive evidence for the existence of
this phase. It is also conceivable that the Line was initiated as a Benioff-zone
faul t along which the Sanbagawa terrane was thrust under the Ryoke terrane
(e.g. Ernst, 1971 b). Another possibility is that the Line was initiated simply as
the boundary fault between the already metamorphosed Ryoke and Sanbagawa

o 400 km
----~----~--~----~!

Fig. 15-13 The Median Tectonic Line (MTL) and the thermal axes (chain
lines) of the Ryoke and Abukuma metamorphic terranes (dotted). The
vertical hatching represents the Upper Cretaceous Izumu Group. F- F =
Itoigawa-Shizuoka Line. Dashed area = Sanbagawa belt.
blocks, and that the Sanbagawa was uplifted along the fault so as to be
ultimately exposed on the surface (§ 17-6). If this is the case, the Median
Tectonic Line is an ancient analogue of the Morioka-Shirakawa Line of
present-day Japan (§15-9 and 17-6).
In the last 100 000 years, a continual right-lateral movement has taken place
along the central part of the Line at an average rate of several millimetres per
year (Kaneko, 1966; Okada, 1970). However, the eastern end, cut by the
Itoigawa-Shizuoka Line, does not show marked off-setting. Hence, the total
amount of strike-slip movement since the formation of the Itoigawa-Shizuoka
Line in late Tertiary time is very limited. It is possible, however, that the Line
was initiated as an early Tertiary strike-slip fault accompanied by some vertical
movement. Miyashiro (I 972b) has shown that the main features of the thermal
structures of the Ryoke and Sanbagawa belts can be explained by this
hypothesis.
 15-12 METAMORPHIC BELTS OF THE RYUKYU ARC/375

15-11 HIDAKA AND KAMUIKOTAN METAMORPHIC BELTS IN

HOKKAIDO

Hidaka Metamorphic Belt

The Hidaka belt is mainly composed of metapelites and metapsammites
accompanied by migmatites, granites and gabbros. An axis of maximum
metamorphic temperature is exposed within it (fig. 15-14). At the western
margin of the belt, a zone of schists is in thrust contact with an unmeta-
morpho sed sedimentary terrane. The belt grades eastwards into an unmeta-
morphosed terrane (e.g. Hunahashi, 1957).
Granitic rocks occur on the eastern side, whereas gabbroic ones are more
abundant on the western side of the axis. A fresh mass of peridotite is present in
the westernmost part (Nagasaki, 1966). The asymmetrical distribution of
plutonic rocks is similar to that in California (§ 13-4).
The original sedimentary rocks are probably Mesozoic, and the meta-
morphism occurred probably in latest Mesozoic or Tertiary time. The K- Ar ages
of granitic rocks are about 36-17 m.y., i.e. Oligocene and Miocene (Shibata,
1968; Kawano and Veda, 1967b).
Some metapelites contain andalusite and sillimanite. Cordierite is common.
The metamorphism is of the low-pressure type.

Kamuikotan Metamorphic Belt

The original rocks of the Kamuikotan belt are mainly clastic sediments and
basaltic volcanics, Mesozoic in age. They are accompanied by a magnificent
serpentinite belt. The age of metamorphism is latest Mesozoic or Tertiary.
Glaucophane, lawsonite and the jadeite + quartz assemblage are widespread,
though petrographic study has been made only in small parts of the belt (Banno
and Hatano, 1963; Tazaki, 1964). Some rocks of the epidote-amphibolite facies
also occur (Shido and Seki, 1959).
The northward extension of the Kamuikotan belt is exposed in Sakhalin
(Sobolev et al. 1967).

15-12 METAMORPHIC BELTS OF THE RYUKYU ARC AND TAIWAN

In a few areas near Nagasaki in the westernmost part of Kyushu, a group of

low-temperature metamorphic rocks occasionally containing glaucophane occurs
in association with serpentinite. It is called the Nagasaki metamorphic complex
(fig. 15-8). The general structural trend of this complex is N-S. On the west side
of these areas, there is a thrust trending N-S. Granitic rocks with cataclastic
structure are exposed on islands to the west of the thrust.
The Rb-Sr and K-Ar ages of the Nagasaki metamorphic complex are 94-60
m.y. (middle Cretaceous-early Tertiary). The K-Ar age of a hornfels related to
the granitic rock to the west is 81 m.y. (Veda and Onuki, 1968; Shibata, 1968).
376/METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS

• Grani tic rocks

fWfJl Basic & ultrabasic plutonics

D Hidaka metamorphics

~ Kamu ikotan metamorphics

o
Sappo 0
o
Obihiro

o 100 km

Fig. 15-14 Metamorphic belts in Hokkaido.

 15-12 METAMORPHIC BELTS OF THE RYUKYU ARC /377
It is thus conceivable that the Nagasaki complex and the granitic complex
represent paired metamorphic belts along the Ryukyu Arc, which connects
Kyushu to Taiwan (Miyashiro, 1965).
Since the islands of the Ryukyu Arc are mostly small, the structure of the
Arc is not well exposed. Konishi (1963) distinguished four structural zones along
the Arc, correlating them with zones in southwest Japan. A zone that is
relatively well exposed in Okinawa Island contains metamorphic rocks of the
greenschist and epidote-amphibolite facies. Another zone exposed in Ishigaki
and the adjacent islands contains glaucophane schists (Foster, 1965; Kuroda and
Miyagi, 1967). There may be some other submerged zones with metamorphic
rocks.
Some metamorphic belts in the Ryukyu Arc could be continuous with the
rocks of Taiwan beneath the sea floor. On the eastern slope of the Central

o Quaternary

~ Mesozoic and
~ Early Tertiary
r::D Tananao schist
o W (western part)
a Tananao schist
§ (eastern part)

Fig. 15-15 Geologic sketch map of Taiwan.

378/METAMORPHIC BELTS IN JAPAN AND ITS ENVIRONS

Mountain Range of Taiwan, a regional metamorphic complex called the Tananao

schists is exposed with a NNE trend (fig. 15.15). A part of the original sediments
is Permian, and K-Ar dating of the associated quartz diorites gave Cretaceous
and Tertiary ages. Most of the Tananao schists are in the greenschist facies, but
a small part near the northern end is in the amphibolite facies (Yen, 1962, 1963;
Yen and Rosenblum, 1964).
Generally speaking, the eastern part of the Tananao schist terrane is mostly in
the chlorite zone with abundant serpentinite and occasionally with glaucophane,
whereas the western part ranges from the chlorite schists to gneisses of the
amphibolite facies. The relation of the western and eastern parts shows some
resemblance to paired metamorphic belts (Yen, 1959, 1963).
Chapter 16

Metamorphic Belts in the

Southwest Pacific

16-1 CHARACTERISTICS OF THE SOUTHWEST PACIFIC REGIONS

We discussed the metamorphic belts in Japan and the surrounding West Pacific
regions in the preceding chapter, and now we are dealing with metamorphic belts
in the southwest Pacific region. They show many features in common. Paired
metamorphic belts occur in New Zealand and Celebes.
The Australian continent is made up of orogenic belts that tend to become
younger toward the east. New Zealand, lying further to the east, tends to be still
younger. This relation appears to suggest accretional growth of a continent.
Australia is treated here from this viewpoint.

16-2 THE AUSTRALIAN CONTINENT

The Precambrian Shield

There is a wide Precambrian shield with an ordinary continental type crust in the
western and central parts of Australia (fig. 16-1). Before the beginning of recent
continental drift, it was probably connected with Precambrian regions of
Antarctica and India. Seven periods of granitic and metamorphic activity have
been provisionally distinguished in the Precambrian from radiometric data. The
oldest one is 3050-2900 m.y. ago, and the youngest one 700-650 m.y. ago
(Compston and Arriens, 1968).
The Precambrian progressive metamorphism in the Broken Hill area in the
western part of New South Wales (§ 8A-7), and that in the East Kimberley area
(§ 7 A-3) are well documented. In both areas, the highest temperature reached
belongs to the granulite facies. Granulite facies rocks occur here and there in the
Precambrian terranes (fig. 4-3 and § 18-3). A systematic description of rock
types in the Precambrian shield has been given by Joplin (1968), with an
introductory statement, 'The areas of Precambrian are vast and the number of
geologists relatively few.'
It appears that a large part of the Precambrian metamorphic rocks belongs to
the low-pressure series, though rocks of the medium-pressure series also occur.
380/METAMORPHIC BELTS IN THE SOUTHWEST PACIFIC

Rise •
Chatha
Is.

Fig. 16-1 Australia, New Zealand and the Southwest Pacific Ocean. The
thick lines are the coast lines and the thin lines 1000 fathom (1830 m)
isobaths. Trenches are indicated by thick broken lines. The Tonga and
Kermadec Trenches are on the east side of the Tonga and Kermadec
Islands, and another trench is on the west side of the Macquarie Ridge.
The New Hebrides Trench is on the continental side of the New Hebrides
Islands.

Adelaide Geosyncline
The Adelaide Geosyncline was formed in late Precambrian to early Palaeozoic
time around the present site of Adelaide in South Australia. Metamorphism and
plutonism took place in early Palaeozoic time (600-400 m.y. ago). The general
trend of the orogenic belt is N-S, running through the Precambrian craton near
its southeastern margin. Its southern end bends toward the west at Kangaroo
Island.
Metamorphic rocks in the Mount Lofty Ranges are well documented (§ 7A-3).
The metamorphism is transitional between the low- and the medium-pressure
type (§ 7 A-3). Relatively small bodies of granitic rocks are intruded into schists
and gneisses along the axis of the metamorphic belt (White, 1966; White et al.
1967).
 16-2 THE AUSTRALIAN CONTINENT/381
Tasman Geosyncline
The highlands in the eastern part of the Australian Continent represent the
former position of the Tasman geosyncline in Palaeozoic time. The geosyncline
contained a large number of subparallel troughs and ridges with an N-S trend.
Some of the troughs and ridges had intense volcanism, whereas others did not.
The geosyncline tends to become younger eastwards. The western part of the
geosyncline (called the Lachlan geosyncline) has sedimentary piles ranging from
Cambrian to Middle Devonian with the maximum deformation in Middle
Devonian time. The eastern part (called the New England geosyncline) has
sedimentary piles ranging from Ordovician to Late Permian with the maximum
deformation in Late Permian time (Crook, 1969). The easternmost part of the
continent near Brisbane has a still younger sedimentary pile extending from the
Permian to Early Cretaceous, after which it was deformed. The deposition in this
region is roughly simultaneous with that in the New Zealand geosyncline.
Volcanism in the Tasman geosyncline is basic and acidic. There are three big
serpentinite belts.
The K- Ar ages of granitic rocks show a regular eastward decrease from
Silurian to Permian in the Melbourne-Canberra-Sydney region of southeastern
Australia (Evernden and Richards, 1962).
The thick sedimentary piles in the Tasman geosyncline were subjected to
widespread low-temperature recrystallization in the zeolite, prehnite-pumpellyite
and greenschist facies (e.g. Packham and Crook, 1960).
The Hill End trough and the Cowra trough run in a N-S direction to the east
and west of Canberra respectively with the Canberra-Molong ridge between
them. They are all within the Lachlan geosyncline. Smith (1969) has described
recrystallization in these zones in an area about 180 km north of Canberra. The
rocks on the ridge are non-schistose and mainly in the prehnite-pumpellyite facies,
whereas those in the Hill End trough are schistose and tend to show a higher
temperature of recrystallization up to the biotite zone (greenschist facies).
In a region south of Canberra, two high-temperature metamorphic belts run
in a N-S direction in the Hill End and Cowra troughs. They are of the
low-pressure type, being associated with granitic rocks. The Cooma area in the
Hill End trough (Joplin, 1942, 1943, 1968; Pidgeon and Compston, 1965), and
the Wantabadgery-Tumbarumba area in the Cowra trough (Vallance, 1953,
1960, 1967) are well documented. The metamorphic rocks in these areas are
mostly schists and gneisses of pelitic and psammitic composition, producing
andalusite, cordierite and sillimanite (§§ 4-1 and 7A-3).
The occurrence of glaucophane has been reported from a few localities in
New South Wales and Queensland in the easternmost part of the Tasman
geosyncline. Especially in Port Macquarie, New South Wales, glaucophane is
accompanied by lawsonite (Joplin, 1968, p. 100). Their geologic relations and
significance are not clear.
382/METAMORPHIC BELTS IN THE SOUTHWEST PACIFIC

16-3 NEW ZEALAND

Island Arc Structure
New Zealand may be a front of eastward growth of the Australian Continent
(fig. 16-1). It stands in a line with the island arc of the Kermadec and Tonga
Islands. The North Island of New Zealand is strongly influenced by the activity
of this arc, as is clear, for example, from the presence of many Quaternary
volcanoes on the island and also from the Hikurangi Trench to the east. Deep
earthquakes take place with a Benioff zone dipping westward. The South Island,
on the other hand, has been more stabilized. It is mainly composed of Mesozoic
and older metamorphic and plutonic rocks. However, the southern part of South
Island has late Tertiary and Quaternary volcanism. This is probably an expres-
sion of the weak orogenic activity of the Macquarie Rise, which is an island arc
extending southwestwards from South Island and is accompanied by a shallow
trench on the west side (Summerhayes, 1967; Miyashiro, 1972b).
The South Island is cut obliquely by a big strike-slip fault, named the Alpine
fault, which offsets the Cretaceous metamorphic belts. The total amount of
right-lateral shift is estimated to be of the order of 480 km. This fault could be a
transform fault connecting the Tonga-Kermadec Trench with the Macquarie
Trench (e.g. Morgan, 1968). Cataclastic metamorphism along this fault will be
reviewed in § 20-3.
The Tasman Sea, which is on the continental side of New Zealand, has a deep
basin with an oceanic type crust, just as the Japan Sea on the continental side of
Japan. On the north of the Tasman Sea, there are two great submarine ridges,
called the Lord Howe Rise and the Norfolk Ridge, which continue right through
New Zealand and extend further toward the southeast possibly to the Campbell
Plateau and the Chatham Rise respectively. These ridges could be surface
manifestations of old orogenic belts (Cullen, 1967; van der Linden, 1967, 1969).
Landis and Coombs (1967) and van der Linden (1969) considered that the ridge
connecting the Lord Howe Rise and the Campbell Plateau represents the
Palaeozoic Buller geosyncline, which will be commented upon later, though
Aronson (1968) has suggested the possibility of its being of Precambrian age.
The Norfolk Ridge-Chatham Rise could represent the late Palaeozoic-Mesozoic
New Zealand geosyncline (van der Linden, 1969).
Detrital zircons from a pre-Devonian sedimentary formation in western South
Island of New Zealand have shown 207Pb/ 206 Pb ages of 1170-1480 m.y. It
suggests the presence of some Precambrian crystalline rocks in and/or near New
Zealand (Aronson, 1968). Evidently, however, Precambrian rocks have not been
found in New Zealand.

Paired metamorphic Belts

Landis and Coombs (1967) have shown the existence of paired metamorphic
 16-3 NEW ZEALAND /383
belts in the South Island. This Island may be divided into western and eastern
provinces, which represent respectively a low-pressure metamorphic belt accom-
panied by abundant granitic rocks and a high-pressure metamorphic belt with
some glaucophane-schist facies rocks. A major high-angle fault, called the Median

o
,MTL
100 ,
200 km
/

Western Province
(Tasman Metamorphic Belt)

~ Granitic rocks and

~ associated metamorphic rocks

iii Zeol ite facies

Preh nite-pumpellyite facies

• Glaucophane-schist facies

IE] Chlorite zone'l Gr~enschist

. .
Biotite zone J facle~

~
Stewart .1.
Island ::. • "'MTL
~ Epidote-amphibolite and
~ amphibolite facies
MTL : Median Tectonic line

Fig. 16-2 Metamorphic belts in New Zealand. (After Landis and Coombs,
1967.)

Tectonic Line, has been found between the two. It is to be noted that the
southernmost part of the fault shows strong bending toward the Campbell
Plateau (fig. 16-2).
The geologic and geophysical similarity of the paired belts with those in
Japan has been emphasized by Hattori (1968).
384/METAMORPHIC BELTS IN THE SOUTHWEST PACIFIC

Western Province
This province is divided into the South and North blocks by the Alpine fault.
The South block is mainly composed of metamorphic and associated granitic
rocks. The metamorphic rocks are gneisses and amphibolites recrystallized at
medium or high temperatures. The North block is composed of granitic rocks
accompanied by unmetamorphosed sediments and low- or medium-temperature
schists. The oldest formation may be Precambrian. Fossiliferous formations
range from the Cambrian to the Permian.
Sedimentation in early Palaeozoic time was of a geosynclinal nature with a
total thickness of not less than 15 km. The name Buller geosyncline has been
proposed.
Radiometric dating by the Rb-Sr method has indicated that at least two
orogenies with plutonism and metamorphism occurred in the Buller geosyncline
(Aronson, 1968): the earlier one about 350-370 m.y. ago (Devonian) and the
later about 100-120 m.y. ago (middle Cretaceous). These are called the Tuhua
and Rangitata orogenies respectively.
The metamorphic terrane formed in the western province by the Rangitata
orogeny is called the Tasman metamorphic belt. It ranges from the greenschist to
the granulite facies. Garnet, chioritoid, staurolite, andalusite, sillimanite and
cordierite occur, suggesting a low-pressure facies series.

Eastern Province
During or immediately after the Tuhua orogeny, the region lying east of the
Tuhua orogenic belt became a basin for intense sedimentation. This is called the
New Zealand geosyncline. Sediment piles, more than 10 km thick, mainly
composed of graywacke, siltstone and pyroclastic materials, were deposited from
Carboniferous to Jurassic time.
Nearly all the rocks in this province are more or less metamorphosed. Landis
and Coombs (1967) call the terrane the Wakatipu metamorphic belt. As is shown
in fig. 16-2, zones of different metamorphic facies tend to run generally along
the Median Tectonic line. Rocks of the zeolite, prehnite-pumpellyite, glauco-
phane-schist, and greenschist facies are widespread. Lawsonite occurs in one
part. A narrow zone along the Alpine fault is in the epidote-amphibolite and
amphibolite facies. The well-recrystallized part of the Wakatipu metamorphic
belt is called the Haast schist group, which includes the Otago schists (Hutton and
Turner, 1936; Hutton, 1940; Brown, 1967) in the southern part of the South
Island and the Alpine schists along the Alpine fault (Reed, 1958; Mason, 1962;
Cooper and Lovering, 1970).
The zeolite and prehnite-pumpellyite facies regions in the southern and
central parts of South Island have become famous through the studies of
Coombs (1954, 1960) and Coombs et al. (1959), as reviewed in § 6A-4.
Stilpnomelane is common and piedmontite occurs rarely. Progressive increase of
 16-3 NEW ZEALAND/385
metamorphic temperature towards the west is clearly noticed in the Alpine
schists, whereas the Otago schists show an axis of maximum metamorphic
temperature.
Gabbro and serpentinite of Permian age occur in the western part of this
province near the Median Tectonic Line. Small amounts of dioritic and granitic
rocks probably of Permian age also occur along the western margin of the
province. There are no plutonic bodies in the eastern part of the New Zealand
geosyncline.
Landis and Coombs (1967) have emphasized that the New Zealand geo-
syncline is auto-cannibalistic. Thus, sedimentation is immediately followed by
burial metamorphism in the zeolite or prehnite-pumpellyite facies. Some of the
rocks formed in this way are exposed to erosion and transported into a
newly-formed adjacent depositional basin. Then the new deposits are subjected
to another phase of burial metamorphism. Sedimentation, burial metamorphism
and erosion proceeded simultaneously.
The K-Ar ages of schists of the prehnite-pumpellyite and greenschist facies
are 140-100 m.y., Le. early to middle Cretaceous (Harper and Landis, 1967).
This corresponds to the Rangitata orogeny. This regional metamorphism took
place after the Late Jurassic completion of sedimentation and the accompanying
burial metamorphism. It is to be noted that metamorphic rocks of the
glaucophane-schist facies have given Cretaceous ages though their age of
deposition is Late Permian. They would have formed by the Rangitata orogeny
rather than by early phases of burial metamorphism.
The metamorphic climax of the Haast schist group is considered to have
occurred about 130-140 m.y. ago. This value is observed in the high prehnite-
pumpellyite facies rocks. With increasing grade of metamorphism the observed
K-Ar and Rb-Sr ages tend to decrease down to 100 m.y., perhaps because the
higher grade rocks maintained a higher temperature for a longer period.
Metamorphic rocks exposed at a distance less than 30 km from the Alpine fault
show anomalously low ages.
The Wakatipu metamorphic belt is obliquely cut by the Alpine fault. Thus, a
zone of glaucophane-schist facies rocks and serpentinites, for example, occurs to
the east of the fault in the southern part of South Island. Its northward
continuation appears to the west of the fault near the northern end of the
Island.
The New Zealand geosyncline extends northward to form the basement of
North Island. The northward extension of the serpentinite belt in the geo-
syncline shows a gradual change of direction toward the northwestern projection
of North Island. The Median Tectonic Line seems to extend roughly along the
west coast of North Island. The extension of the New Zealand geosyncline
appears to continue to the Norfolk Ridge.
Chatham Island, 800 km east of South Island, is known to be composed of
386/METAMORPHIC BELTS IN THE SOUTHWEST PACIFIC

pumpellyite-bearing metamorphic rocks similar to those of South Island. The

New Zealand geosyncline appears to have continued to this Rise.

Metamorphic History of New Zealand and Vicinity

A tentative account on the metamorphic history of New Zealand will be given
below.
A Precambrian crystalline basement may have formed along the Lord Howe
Rise, the South Island of New Zealand and the Campbell Plateau. Probably the
Tuhua orogeny and the accompanying metamorphism took place again roughly
along the same zone during the Devonian. Some granitic and metamorphic
rocks now exposed in the western province of South Island were formed in this
period.
Van der linden (1969) has shown the possible existence of an extinct or
dormant mid-oceanic ridge trending north-south in the Tasman Sea. It is
conceivable that the Lord Howe Rise was formed originally along the east coast
of Australia, and later, possibly in Tertiary time, moved eastward by the
generation of the Tasman Sea.
The New Zealand geosyncline was formed on the northeast side of the old
orogenic belt, that is, along the Norfolk Ridge, the North and South Islands of
New.Zealand and the Chatham Rise. Thick sediment piles were deposited from
the Carboniferous to Jurassic. Burial metamorphism took place broadly at the
same time as deposition and erosion. Metamorphism accompanying the
Rangitata orogeny occurred in both western and eastern provinces probably in
the Late Jurassic to Early Cretaceous, leading to the formation of paired
metamorphic belts.
The trend of the Median Tectonic Line in North Island and in Stewart Island
(south of South Island) suggests a NW-SE trend for the paired metamorphic
belts. The northwest projection of North Island-represents an effect of the island
arc structure formed in this period.
In Late Cretaceous and Tertiary time, the tectonic activity along the arc
connecting North Island of New Zealand to the Kermadec and Tonga Islands
became intense. Great strike-slip faults were formed on the islands and ocean
floors in connection with this activity. The Alpine fault, being one of the faults,
strongly changed not only the rock distribution but also the shape of South
Island so as to conform with the direction of present-day activity along the
North Island-Tonga line.

16-4 CELEBES

The island of Celebes belongs to an orogenic belt of late Mesozoic to Cenozoic

age. The island shows a remarkable four-armed morphology, which is due to a
connected double arc with its concave side towards the Pacific Ocean (fig. 16-3).
 16-4 CELEBES/387

The north arm, the western part of central Celebes, and the south arm together
constitute the inner arc, where late Mesozoic and Tertiary granitic rocks are
widespread in association with biotite-rich schists (van Bemmelen, 1949).
The granitic masses of the inner arc are cut abruptly on the eastern side by a
great fault, called the Median Line, trending north-south. Mylonite has been
formed along the Line.

Southeast
Arm

o, 100
,
km

~ Granite and gneiss

• Peridotite, serpentinite and gabbro

Fig. 16-3 Geologic sketch map of Celebes. (Modified from van

Bemmelen, 1949, and others.)
The east arm, the central and eastern parts of central Celebes, and the
southeast arm, which all lie to the east of the Median Line, together constitute
the outer arc. In this arc, basic and ultrabasic plutonic rocks of Mesozoic and
Cenozoic ages are widely exposed, together with glaucophane-schist facies rocks.
Basic volcanic rocks are abundant in the geosynclinal pile. In some parts jadeite
has been found in association with quartz (de Roever, 1955a). Thus, the inner
and outer arcs constitute paired metamorphic belts.
388/METAMORPHIC BELTS IN THE SOUTHWEST PACIFIC

16-5 NEW CALEDONIA

New Caledonia is elongated in a NW-SE direction, measuring about 400 km in

length (fig. 16-1). It is noted for extensive development of glaucophane schists
and serpentinite. Peridotite-serpentinite masses occur in a line on the southwest
side of the island. Their tectonic setting is in dispute (Avias, 1967).
Metamorphic rocks at the northwestern end of the island have been described
(Coleman, 1967; Brothers, 1970). The original rocks are Cretaceous to Eocene.
The K-Ar ages of the metamorphic rocks are 21-38 m.y. (Oligocene to
Miocene). Three zones have been distinguished. All the zones are due to
high-pressure metamorphism as is clear from the occurrence of lawsonite in all
zones. The metamorphic rocks are usually non-schistose, preserving their original
structure.

16-6 YAP ISLANDS

The Yap Islands are situated in the Pacific about 1300 km east of the Philippine
Islands. They are on an island arc forming a southeastern border of the
Philippine Sea. There is a trench, about 8500 m deep, on the southeast side of
the island.
The island arcs bordering the Philippine Sea (i.e. the Izu-Bonin, Mariana, Yap,
and Palau Arcs) are generally composed of volcanic rocks and associated
sediments from the early Tertiary to the present. For example, the oldest rocks
exposed on Guam and Saipan are Eocene (or possibly slightly older) pyroclastics
(basaltic, andesitic and dacitic in composition). The Yap Islands, however, differ
from all the rest in having a basement of crystalline schists. Our interest is in the
characteristics of regional metamorphic rocks produced in such a place very far
from any continents.
There are four principal islands in the Yaps, close to one another. The
metamorphic rocks, known as the Yap formation, are exposed in an area about
10 km across in the main island and in smaller areas in other islands. Their age is
not known, but they are unconformably covered by Miocene beds.
The metamorphic rocks are actinolite greenschists and amphibolites. They
were derived from basic igneous materials, some of which may have been
abyssal tholeiites. The amphibolites contain blue-green hornblende. Schistosity
is well developed and generally strikes northeastward, roughly parallel to the
elongation of the arc. Relict minerals are rare. The temperature of meta-
morphism increases southeastward. The metamorphic rocks are accompanied by
small masses of serpentinite (Tayama, 1935; Johnson, Alvis and Hetzler, 1960;
Shiraki, 1971).
The Miocene breccias overlying the schists are reported to contain fragments
 16-6 YAP ISLANDS/389

not only of schist, amphibolite and serpentinite, but also of amphibole granite,
amphibole syenite, and gabbro. The sources of the latter rocks are not known.
The breccias are covered by Miocene volcanics made up of basaltic and andesitic
pyroclastics and lavas. Since then, there has been no volcanism in the Yaps in
spite of the presence of a deep trench to the east.
Chapter 17

Tectonics of Regional
Metamorphic Belts

17-1 TECTONIC CLASSIFICATION OF METAMORPHIC BELTS

Three Main Categories

From the viewpoint of plate tectonics, most or all of the regional metamorphic
belts of Phanerozoic time and possibly also of the Precambrian appear to have
formed in the following three tectonic settings (Miyashiro, 1972b).

1. Metamorphic belts formed in continental margins. This type of belt could be

forming now at the Pacific margin of South America, where the oceanic plate of
the Pacific is being underthrust. The Mesozoic paired belts of the Franciscan and
Sierra Nevada (Miyashiro, 1961a; Hamilton, 1969a, b) and the late Palaeozoic
paired belts in Chile (GonzaIez- Bonorino, 1971) could have formed in a similar
way on the western margin of North and South America, respectively.

2. Metamorphic belts formed beneath ordinary island arcs. This type of belt
could be forming now beneath the Northeast Japan and Kurile Arcs, where the
oceanic plate of the Pacific is being underthrust. The Mesozoic paired belts in
New Zealand (chapter 16) and the late Palaeozoic and Mesozoic paired belts in
southwest Japan (chapter 15) may have formed in a similar way beneath
ordinary island arcs, though it is also conceivable that they may have formed on
the margin of the nearby continents of that time and drifted afterwards to form
island arcs.

3. Metamorphic belts formed beneath reversed island arcs. This type of belt
would be forming now beneath the New Hebrides Islands, where the plate of the
adjacent marginal sea is being underthrust (cf. Isacks and Molnar, 1971). The
paired metamorphic belts in Hokkaido (fig. 3-10) could have been formed by the
eastward underthrusting of the marginal sea plate beneath the Sakhalin Arc.
Paired metamorphic belts can form in any of the three categories. All the
metamorphic belts in these categories may be expected to show a tendency to be
paired only if the velocity of plate descent is rapid (§ 17-2).
 17-2 VARIATION IN THE RATE OF PLATE CONVERGENCE/391

Orogenic Belts due to Continental Collision

Some orogenic belts such as the Alps, Urals and Himalayas are believed to have
formed by the collision of two continental plates. In the initial stage of
continental collision, the two plates are separated by an oceanic plate. The two
continental blocks approach each other by the consumption of the intervening
oceanic plate due to underthrusting along one trench zone or more. In this stage,
the tectonic settings and resultant metamorphic belts, some of which may be
paired, should belong to the above-discussed three categories. The alps and Urals
have glaucophane-schist belts, which would have formed in this stage.
In the fmal stage, the intervening ocean is completely lost resulting in a direct
collision of the two continental blocks. The buoyancy of the continental blocks
would counteract and halt the descent of one plate beneath another. Hence, the
characteristics of orogeny in this stage should differ from those of any of the
above-discussed three categories. At present we cannot tell whether meta-
morphism on a large scale takes place at this stage or not. This stage may now be
realized in the Himalayas, where there has been no sign of igneous activity
(GiUuly, 1971). Metamorphic rocks are exposed in the Himalayas, but most of
them appear to be of Precambrian age and formerly belonged probably to the
Precambrian metamorphic basement of the Indian subcontinent (petrushevsky,
1971). There may be some younger metamorphic rocks. .

17-2 VARIATION IN THE RATE OF PLATE CONVERGENCE AND THE

RELATION BETWEEN PAIRED AND UNPAIRED METAMORPHIC
BELTS

Secular Variation in the Baric Type of Regional Metamorphism in Relation to

Plate Motion
Low-pressure regional metamorphic rocks occur in any geological time whereas
high-pressure metamorphic belts and in particular glaucophane schists formed
mostly in the Phanerozoic (§4-3). Most of the typical glaucophane-schist facies
rocks were produced by Mesozoic and Cenozoic orogenies.
As was discussed in §4-3, this relation can be explained by assuming that in
younger geologic time oceanic plates have become thicker, and inclinations of
descending slabs have become steeper, and/or the velocity of plate under-
thrusting has become more rapid. Such changes should cause lower geothermal
gradients in subduction zones with resultant formation of metamorphic rocks of
the high-pressure type (Miyashiro, 1972b; Ernst, 19720).

Contrast in the Rate of Plate Motion between the Pacific and Atlantic
Though the pattern of spreading in the Pacific should have undergone a number
of changes since early Mesozoic time, the rapidity of the plate motion has
apparently continued to the present. The recent rate of spreading in the East
392/TECTONICS OF REGIONAL METAMORPHIC BELTS
Pacific Rise is as high as 2-6 cm/year in contrast to the rate of 1-2 cm/year in
the Mid-Atlantic Ridge, and it has been demonstrated that the contrast has
existed at least since late Mesozoic time (Heirtzler et al. 1968).
Presumably, the plate motion in the Pacific was fairly rapid in the Palaeozoic
so that atypical high-pressure metamorphic belts were formed in many parts of
the circum-Pacific region, and it became very rapid in the Mesozoic so as to
produce typical high-pressure belts. It may be considered that arc-trench systems
of group I in table 4-4 can produce typical high-pressure belts and paired belts.
It appears that the Atlantic Ocean has closed and-opened repeatedly, and this
caused the orogenies of the Appalachian, Caledonian and even older belts
(Wilson, 1966; Dietz and Holden, 1966; Dewey, 1969). We may consider that
usually the velocity of plate descent in the Atlantic regions was not high enough
to produce well-developed belts of glaucophane schists. Metamorphic belts of
the medium-pressure type could have formed instead.
However, the medium- and low-pressure Caledonian metamorphic beIt of the
Scottish Highlands may be regarded as being paired with the poorly developed
glaucophane-schist area of the Southern Uplands of Scotland. This pair was
formed probably in the Early Ordovician. Presumably the Benioff zone at that
time was inclined to the north, but a new Benioff zone inclined to the south
may have been formed in Late Or~ovician time so as to produce a volcanic arc to
the south (Fitton and Hughes, 1970).

Possible Effects of the Descent of Mid-Oceanic Ridges beneath Continental

Plates. Recent studies in marine geophysics have shown that mid-oceanic ridges
descended beneath island arcs and continental margins in the Pacific regions in
Mesozoic and Cenozoic times. This has posed a new problem to metamorphic
geology, because the lithospheric plate in a ridge must be still hot, and the
underthrusting of a hot plate cannot produce a very low geothermal gradient
even if its rate is rapid, but may cause particularly intense or widespread
magmatism in the arcs or continental margins.
Moreover, the rate of underthrusting of oceanic plates may be unusually rapid
over a period preceding the descent of some mid-oceanic ridges beneath island
arcs or continental margins, because the rate should be increased by the rate of
creation of new lithosphere on the ridges. It may cause high-pressure meta-
morphism (Miyashiro, 1973; S. Uyeda and A. Miyashiro in preparation).
Atwater (1970) has shown that the mid-oceanic ridge between the Farallon
and Pacific plates descended beneath the North American plate in middle
Tertiary time. The Franciscan high-pressure metamorphism took place tens of
million years prior to the ridge descent, possibly owing to the rapid under-
thrusting in this period. R. L. Larson and C. G. Chase (in press) have shown that
the mid-oceanic ridge between the Kula and Pacific plates descended beneath
Japan and the East Asiatic Continent in late Cretaceous times. The Sanbagawa
 17-2 VARIATION IN THE RATE OF PLATE CONVERGENCE/393

high-pressure metamorphism took place tens of million years prior to the ridge
descent.
In both cases, high-pressure metamorphism was halted about 30 or 40 million
years prior to the ridge descent. Conceivably, the descending oceanic litho-
spheres in and near the ridges were still hot enough to increase the geothermal
gradients in the subduction zones, and thereby to put an end to high-pressure
metamorphism. For a period around the ridge descent, magmatism took place
over an unusually wide zone_in the Western United States and in East Asia
(Miyashiro, 1973; S. Uyeda and A. Miyashiro, in preparation).

Nature of Apparently Unpaired Metamorphic Belts

Most of the metamorphic belts in the Atlantic regions and Europe are apparently
unpaired. We may expect, however, that if a high-pressure metamorphic belt
occurs, it could be accompanied by a belt of low-pressure metamorphism, since
the descending slab of the former belt should tend to create the latter. A most
remarkable example of a young, apparently unpaired high-pressure belt is in the
Pennine zone of the Alps. It is of interest whether this belt fulfIls the above
expectation.
The Alpine Ranges continue eastwards to the Carpathians. Andesitic and
rhyolitic rocks of late Tertiary and Quaternary ages are abundant in Hungary,
southern Slovakia and Transylvania, which are on the south side of the
Alpine-Carpathian Ranges. Tertiary granite rocks also occur in the andesitic-
rhyolitic belt, though at the western end of the belt the Bergell granite has been
intruded into the Pennine zone. This belt of granite, andesite and rhyolite shows
usually high heat-flow values, and may be a surface manifestation of a hidden
low-pressure metamorphic complex which is paired with the high-pressure belt
exposed in the Pennine zone of the Alps (§ 14-5).
The Hercynian metamorphic belts in Western Europe are mostly of the
low-pressure type (Zwart, 1967a, 1969). They might be regarded as a typical
example of unpaired metamorphic belts. However, a part of the Hercynides may
have paired belts. Hercynian glaucophane schists occur in the lIe de Groix to the
south of Brittany. It is possible that an extension of the lIe de Groix
metamorphic complex forms a Hercynian high-pressure belt which is now
covered mostly by the sea, and is paired with a Hercynian low-pressure belt.
In this connection, it is noteworthy that in some pairs the high-pressure
metamorphic belt is much narrower than the total area of the associated
low-pressure metamorphic and granitic rocks, and hence it may readily be lost
sight of owing to younger sediment covers, submergence beneath the sea, or later
recrystallization at medium- or low-pressure conditions. A good example is the
Jurassic-Cretaceous pair of southwest Japan. The Sanbagawa high-pressure
metamorphic belt of the pair is only 50 km wide. The contemporaneous granitic
activity took place not only in the Ryoke low-pressure belt but also in a wide
394/TECTONICS OF REGIONAL METAMORPHIC BELTS
region to the north. In Jurassic-Cretaceous time, this region of granitic activity
was probably a part of the East Asiatic volcano-plutonic belt (fig. 15-6) on the
east coast of the Asiatic Continent (cf. Lipman et al. 1971).
If the rate of plate underthrusting is too slow to cause high-pressure
metamorphism, the contrasting characteristics of the paired belts will be
obscured with resultant formation of an apparently unpaired belt. If the rise of
magma and aqueous fluid is stopped at a very deep level, a low-pressure
metamorphic belt mllY not form (Miyashiro, 1973).

17-3 SUMMARY OF GEOLOGIC FEATURES OF METAMORPHIC BELTS

High-Pressure Metamorphic Belts
1. High-pressure metamorphic belts are characterized by geothermal gradients as
low as about 10°C km -1 or even lower (§3-4).
The width of a metamorphic belt measured from the thermal axis to the
outer limit of the lowest temperature zone is usually few tens or several tens of
kilometres.
2. Most of the rocks belong to the laumontite grade of the zeolite facies, the
prehnite-pumpellyite, glaucophane-schist and greenschist facies. The character
of metamorphic facies series usually varies from area to area even in a single
metamorphic belt. There is the possibility that the lowest temperature part (Le.
the zones of the zeolite facies and a part of the prehnite-pumpellyite facies) of
apparently progressive metamorphic terranes may actually have been recrystal-
lized at a distinctly older or younger time than the associated higher temperature
part.
3. In the Kanto Mountains of the Sanbagawa belt, the jadeite + quartz
assemblage occurs only in the middle grade part of the terrane, whereas albite is
stable in both lower and higher grade parts (§§3-4 and 7A-9; fig. 3-7). This is
probably due to the upward convex shape of the geothermal curve in fig. 3-3 (§
3-4). The facies series: prehnite-pumpellyite ~ glaucophane schist ~ greenschist
(or epidote-amphibolite) in the Katsuyama area (§6A-5) and in some western
parts of the Sanbagawa belt also suggest an upward convex geothermal curve. In
the Franciscan terrane, albite is present in the poorly recrystallized part, and
jadeite occurs in the well-recrystallized parts, which probably represent higher
temperature zones (§7A-10). The Franciscan, therefore, may be regarded as
corresponding to the lower half of the Kanto Mountains series (zones I-IV in fig.
3-7).
4. The temperature of metamorphic recrystallization commonly tends to
increase toward the major fault which is the boundary between the high-pressure
and the associated low-pressure metamorphic belt; e.g. the Median Tectonic Line
in Southwest Japan, the Alpine fault in New Zealand, and the Coast Range
thrust in California. However, the temperature decreases from a thermal axis to
 17-3 GEOLOGIC FEATURES OF METAMORPHIC BELTS/39S
both sides, for example, in the Katsuyama area (§6A-S; Hashimoto, 1968).
Therefore, the temperature increase toward the major fault probably means that
the original high-pressure metamorphic complexes were cut by the major faults
into two units of roughly comparable sizes, and the unit on the low-pressure
terrane side was lost. Hence, the nature of the major faults is of vital importance
in the history of high-pressure metamorphic terranes.
A high-pressure metamorphic terrane is commonly composed of a number of
smaller areas which were metamorphosed at somewhat different times, as will be
summarized below. It is an unsolved problem why the whole terrane shows a
relatively simple thermal structure in spite of the complex age structure. Our
petrographic investigations may not have been thorough and precise enough to
reveal a really complex thermal structure.
5. The time relation between sedimentation and metamorphism differs in
different regions. Auto-cannibalism is common in the geosynclines pertinent to
high-pressure metamorphism.
In the Franciscan and New Zealand geosynclines, it has been shown that the
apparently single metamorphic terrane is actually composed of a number of
sedimentary piles in different basins, that the sedimentation of individual piles
was immediately followed by their metamorphism, and that the deposition and
metamorphism in one pile were broadly simultaneous with upheaval and erosion
in another. It is thus possible that rocks recrystallized in an earlier phase of
metamorphism are exposed and eroded, and the resultant materials are trans-
ported and deposited in another basin to undergo a later phase of metamorphism
(Suppe, 1969; Landis and Commbs, 1967).
Such burial metamorphism in New Zealand may be distinct from the later
regional metamorphism which produced greenschist and amphibolite facies rocks
during the Rangitata orogeny, as there appears to be a temperature fall between
the two events. However, the latest phase of burial metamorphism may have
merged with the latter event.
In Japan, the Sangun metamorphism in western Honshu may have immedi-
ately followed sedimentation of a Palaeozoic geosynclinal pile, whereas the
Sanbagawa metamorphism (or the main phase of it) would have taken place in
Palaeozoic rocks at a distinctly later time (Juzassic-Cretaceous).
6. Graywackes are abundant, but typical aluminous pelitic rocks are usually
scarce in high-pressure terranes.
7. Usually, high-pressure metamorphic belts are not accompanied by granitic
rocks (e.g. Franciscan and Sanbagawa). The Alpine metamorphic belt in Western
Europe has granitic rocks, which, however, are in a very small amount, and may
actually belong to the associated andesitic belt to the south (§ 17-1).
8. Basaltic volcanics are abundant in geosynclinal piles. Gabbros are also
common. More or less serpentinized peridotite occurs in association with these
basic rocks. Some of the basaltic volcanics occur as thin beds conformable with
396/TECTONICS OF REGIONAL METAMORPHIC BELTS

surrounding graywackes, and could be the result of volcanism simultaneous with

sedimentation, possibly in a geosyncline. On the other hand, there are big
ophiolite complexes showing a regular sequence of basalts, gabbros and serpen-
tinites commonly in a strongly disturbed part of the metamorphic terrane. These
may represent fragments of the oceanic crust conceivably transported from
ocean floors (§ 4-1).
9. The Franciscan group was probably deposited on an oceanic crust, having
no acidic crystalline basement. Atypical high-pressure terranes such as those of
the European Alps and the Sangun belt have an acidic crystalline basement, as
now exposed, for example, in the Gotthard massif and the Maizuru zone
respectively. Such basements may have originally been thinner than the acidic
basements of low-pressure metamorphic terranes.

Low-Pressure Metamorphic Belts

1. Low-pressure metamorphic belts are formed under geothermal gradients
probably higher than 25°C krn-1 (§3-4). As a consequence, the width of a
metamorphic belt measured from the thermal axis to the outer limit of the
lowest temperature zone is usually relatively narrow (commonly up to 15 krn).
Where the whole metamorphic terrane is much wider, it may include two or
more thermal axes, or alternatively it may have an unusually wide axial zone
representing approximately constant temperatures.
2. Zonal mapping is usually easy, and rocks of the amphibolite facies usually
form the widest zone. It appears that partial melting occurs commonly in the
high amphibolite facies and higher zones.
3. The time relation between sedimentation and metamorphism is diverse.
Regional metamorphism immediately followed geosynclinal sedimentation in
eastern Australia, but in the Ryoke region metamorphism took place about 100
m.y. later than the end of sedimentation in the pertinent terrane. In the latter
case, however, a new basin for sedimentation was created on the ocean side of
the inactive geosyncline, and sedimentation continued there during the Ryoke
metamorphism. The sediments of the old geosyncline were involved in the new
tectonic movement, partly as a non-crystalline basement to the new geosyncline.
4. Low-pressure metamorphic belts are always accompanied by abundant
granitic rocks. The emplacement of granitic rocks occurred mostly in a declining
phase or after the end of regional metamorphism. In the Ryoke belt, the age
difference between the regional metamorphism and the youngest granitic rocks
related to it is probably of the order of 50-70 m.y.
Post-metamorphic granites commonly form hornfelsic aureoles. Late-
metamorphic granites are usually surrounded by schistose metamorphic rocks
apparently similar to the regional metamorphic rocks of the area. However, a
slight increase of recrystallization temperature is noticed toward the contact in
some areas, and especially when the intrusion is more basic (e.g. quartz dioritic)
in composition.
 17-3 GEOLOGIC FEATURES OF METAMORPHIC BELTS/397

5. Rhyolitic and/or andesitic volcanic rocks may occur abundantly in

association with granitic rocks related to low-pressure regional metamorphism.
This is particularly common in the circum-Pacific regions; e.g. the Ryoke belt of
Japan, western North America and eastern Australia (§§4-2 and 15-7). Island
arcs with andesitic volcanoes could be the surface manifestation of the
formation of a low-pressure metamorphic complex in a crust.
6. Basaltic magmatism of the geosynclinal stage is usually scarce, and
peridotite and serpentinite are very rare.
7. Pelitic rocks showing a high matuRty are usually abundant in the terranes.

Paired Metamorphic Belts

1. Paired metamorphic belts are composed of two parallel belts of contrast-
ing characters. Many paired belts are composed of a low-pressure belt on the
continental (i.e. inner) side and a high-pressure belt on the oceanic (i.e. outer)
side. In other pairs, the inner belt may be composed of low- and medium-
pressure metamorphic terranes, and the outer belt may be a complex of high-
and medium-pressure metamorphic terranes. Rarely, a high-pressure belt occurs
on the continental side (§17-1).
2. The inner belt of a pair in New Zealand and in California appears to have
been metamorphosed roughly in the same period as the associated outer belt. In
the Alpine pair, the high-pressure metamorphism of the Pennine zone begun tens
of million years earlier than the andesitic-granitic magmatism of the Hungarian
zone. Precise age relations are not known in Palaeozoic pairs.
The age of cooling as revealed in isotopic dating may differ considerably from
that of the preceding thermal climax, and the thermal climaxes of the two belts
in a pair may be nearly simultaneous in many cases.
3. The age of deposition of the original rocks is generally younger in the
high-pressure belt than in the associated low-pressure belt in New Zealand and
California, whereas it appears to be similar in the Jurassic-Cretaceous pair in
southwest Japan.
4. Polyphase recrystallization and auto-cannibalism have been observed in
many high-pressure regional metamorphic belts and in some small-scale low-
pressure terranes (e.g. §6A-2), but not in typical low-pressure regional meta-
morphic terranes. However, the K- Ar ages of granitic rocks in the Sierra Nevada
and Mesozoic-Early Tertiary Japan show several clusters, which suggest the
possibility that low-pressure metamorphic recrystallization and associated
granitic activity also may have distinct phases. It is not clear whether there is any
close genetic connection between the phases of paired belts.
5. The original distribution of rocks has been greatly modified by later faults
in many regions. The two belts of a pair are now in direct contact with each
other on a major fault in the Jurassic-Cretaceous of southwest Japan and in New
Zealand, but have an intervening zone of virtually unmetamorphosed rocks in
Hokkaido.
398/TECTONICS OF REGIONAL METAMORPHIC BELTS

17-4 TECTONIC SIGNIFICANCE OF MEDIUM-PRESSURE REGIONAL

METAMORPHISM

Many metamorphic belts of the medium-pressure type are associated with

granitic rocks, just like those of the low-pressure type. However, the thermal
effect of individual granitic masses is as obscure as that in low-pressure belt, if
not more so.
If the high temperature conditions in low-pressure terranes are generally
controlled by the rise of granitic magma and aqueous fluid, the maximum
temperature reached may be mainly controlled by the temperature of rising
magma and aqueous fluid, being nearly independent of the depths of intrusion.
If this is the case, low-pressure regional metamorphism should take place in
regions where numerous bodies of magmas rise to shallow depths, whereas
medium-pressure metamorphism should take place in regions where the main
level of intrusive bodies is deeper.
It is thus conceivable that the difference between the low-pressure terranes
and such medium-pressure terranes is not intrinsic but comes simply from a
difference in the level of abundant granitic intrusion.
This view is strengthened by the common gradation of a low-pressure area
into a medium-pressure one as observed in the Scottish Highlands and the
northern Appalachians. There is no tectonic demarcation line between such
areas. It need not be considered important whether the inner belt of a pair is
entirely of the low-pressure series or contains low- and medium-pressure series
areas.
This discussion applies only to the medium-pressure terranes accompanied by
granitic rocks. There are other medium-pressure terranes which are not accom-
panied by granitic rocks, and occur as parts of metamorphic belts that contain
rocks of the high-pressure series in other parts. As discussed in § 17-2, this type
of medium-pressure terrane could form where the tectonic Sinking was not rapid
enough to cause high-pressure metamorphism.
If this is the case, the medium-pressure metamorphic terranes may be
classified into those genetically connected with low-pressure terranes and those
genetically connected with high-pressure terranes. Thus, in tectonics, only low-
and high-pressure regional metamorphism may be of prime importance, and
medium-pressure metamorphism may simply be treated as special variants of
low- and high-pressure metamorphism.

17-5 TEMPERATURE DISTRIBUTION IN OROGENIC BELTS

Geothermal Gradients and Tectonic Sinking

The unusually low geothermal gradient of high-pressure metamorphism is
probably caused by the rapid tectonic sinking of a geosynclinal pile due to the
 17-5 TEMPERATURE DISTRIBUTION IN OROGENIC BELTS /399
pull of a descending lithospheric slab. When this is almost simultaneous with
sedimentation as in the Franciscan, rapid sedimentation could cooperate with
the sinking to decrease the geothermal gradient. Where the sinking was not rapid
enough to cause high-pressure metamorphism, a medium-pressure complex
should be produced.
It is not simple to determine when tectonic sinking began, because recrystal-
lization in the last major phase may efface the evidence for the earlier history of
the terrane, as discussed in relation to the Sanbagawa belt (§ 15-8).
If underthrusting halts for some reason and a high-pressure metamorphic
complex is kept at a deftnite depth for a long time, a stationary temperature
distribution will gradually be approached, resulting in an increase of the
geothermal gradient. Thus, the prevailing P-T condition would change toward
that of medium-pressure metamorphism, ultimately leading to the destruction of
the previously-formed high-pressure mineral assemblages. The destruction of
glaucophane observed in many parts of the Alpine metamorphic belt (e.g. de
Roever and Nijhuis, 1964) may have occurred in this way. Therefore, the
occurrence of a high-pressure metamorphic belt means that the terrane was not
only rapidly sunk but also uplifted tectonically so that previously-formed high-
pressure mineral assemblages are preserved.
The time of beginning and the duration of low-pressure metamorphism may
depend mainly on the way of rise of granitic magma and aqueous fluid. If this is
the case, high- and low-pressure metamorphism in a pair of belts may not be
simultaneous, as exemplifted by the Alpine pair (§ 17-3).

Calculation of the Temperature Distribution in a Geosynclinal Pile and in a

Descending Lithosphere
Most metamorphic reactions may be regarded as equilibrium processes to a good
approximation. On the other hand, heat conduction in a big geologic body is a
very slow process, and the time lag of the temperature rise in a sinking
geosynclinal pile or in a descending lithospheric slab may actually control the
highest temperature attained by and geothermal gradients in the pertinent body.
The most recent method for the estimation of the temperature distribution in
the geologic body is to solve the heat conduction equation by a digital computer
using some assumptions on the crustal and sub crustal structures, the moving
velocity of the body, the thermal conductivity (diffusivity) of the material, and
the amount of heat generation in the mantle and crust. Comparing the calculated
heat flow with the observed one, the model is successively revised and improved
until the calculated thermal state becomes compatible with geologic and
geophysical features of the area.
Minear and Toksoz (1970), Oxburgh and Turcotte (1971), and Hasebe, Fujii
and Uyeda (1970) made detailed calculations of the temperature distribution in
a lithospheric slab descending from a trench zone and its neighbourhood along
4OO/TECTONICS OF REGIONAL METAMORPHIC BELTS

the lines mentioned above. According to their results, a very low or even a
negative temperature gradient is expected in the lithosphere during the first
several tens of millions of years of plunging. When the shear-strain heating or
frictional heating is intense along the upper surface of the plunging body, this
part may be heated up so strongly that a large amount of magma is produced.
Since the basic layer of the original oceanic crust would be present along the
upper surface, the resultant magma would be andesitic and its rise would cause
an increase in the volume of the continental crust above.
Another approach, free from any complexity due to the model adopted, is
the use of the thermal relaxation time T, which is one of the simplest measures
of the rate of temperature variation in a body heated from the outside. The
thermal relaxation time of heat conduction for a body of size H and of thermal
diffusivity k, is represented by H2/k, which represents the approximate time
needed to heat the body to lIe or 1/2'72 of the possible maximum heating
upder the given conditions. When we take k = 10-2 cm 2 sec- 1 , which is a
well-accepted order of magnitude for the thermal diffusivity of the material in
the earth's crust, the thermal relaxation time for a crustal or lithospheric layer
10, 30 or 100 km thick is roughly 3, 30 or 300 m.y. respectively (Shimazu,
1961).
The development of a geosyncline is usually a process of the order of 100
m.y. in duration. A geosynclinal pile, 10-20 km thick, for example, will be
heated in accordance with the slow sinking during its development with no
marked time lag. On the other hand, when such a pile descends rapidly within a
period of a few million years, unusually low geothermal gradients will form in it.
This may be the case with individual phases of the Franciscan metamorphism.
If a lithosphere, 100 km thick, plunges in a period ofless than 100 m.y., a
marked time lag in heating should take place, leading to the occurrence of
unusually low geothermal gradients and high-pressure regional metamorphism.

17-6 MA10R FAULTS, SLAB SURFACES, AND THE UPLIFT OF HIGH-

PRESSURE METAMORPHIC COMPLEXES

Three Classes ofMajor Faults

Metamorphic belts are usually cut by a large number of faults. Our interpre-
tation of the origin and history of metamorphic belts depends largely on our
interpretation of the nature of the major faults. The following three classes of
major faults are probably of particular interest in this connection.
1. Transform faults (Wilson, 1965). The San Andreas fault of California and
the Alpine fault of New Zealand appear to belong to this class. If the angle
between the strike of a strike-slip fault and the axis of a metamorphic belt is
small, the strike-slip movement will be difficult to determine. It is possible that
some large strike-slip faults have been overlooked in metamorphic terranes.
 17-6 HIGH-PRESSURE METAMORPHIC COMPLEXES /401
2. Upper surface of a descending lithospheric slab, which would be a large
fault cutting through the crust and upper mantle. The upper surface may
represent a Benioff zone. However, there is some ambiguity in this correlation,
and hence such faults will be referred to by the name slab-surface faults. As
high-pressure metamorphism has been regarded as being due to underthrusting of
a lithospheric slab, slab-surface faults, if exposed, could be discovered in or near
high-pressure metamorphic terranes. Ernst (1970, 1971b) suggested that the
Coast Range thrust of California and the Median Tectonic Line of Japan may be
such faults.
3. Faults, along which crustal blocks that contain high-pressure metamorphic
complexes are uplifted. A high-pressure complex formed at a considerable depth
must be subsequently uplifted to be exposed on the surface.
The western segment of the Median Tectonic Line of southwest Japan (figs.
17-1 and 15-13) is the fault boundary between the Sanbagawa metamorphic
terrane to the south and the upper Cretaceous Izumi Sandstone Group to the
north. The Izumi Group was deposited on an eroded surface of the Ryoke

-_
........ .........50 km
'

Fig. 17-1 Major faults in and around the Sanbagawa high-pressure meta-
morphic terrane in Shikoku, Japan. Approximately the region between the
Median Tectonic Line and the Kurosegawa Tectonic Zone belongs to the
Sanbagawa belt, though the southern limit of recrystallization is obscure.
402/TECTONICS OF REGIONAL METAMORPHIC BELTS

metamorphic terrane which is paired with the Sanbagawa. Hence, this segment
of the Median Tectonic Line was formed distinctly later than the Ryoke and
Sanbagawa metamorphism, and the Sanbagawa high-pressure terrane should have
been uplifted along the Line, though this does not preclude the possibility that
the movement of the Line involved a strike-slip component.
The Northeast Japan Arc (Le. northeast Honshu) has been active since the
Miocene. As is schematically shown in fig. 3-11, it is composed of the outer
non-volcanic arc and the inner volcanic arc with a large tectonic line, called the
Morioka-Shirakawa Line (S in the figure) between them. The Line was detected
by a gravity survey by Tsuboi et al. (1953- 6), though the fault is entirely
covered by younger sediments. The outer arc is characterized by high values of
positive Bouguer anomalies. It is conceivable that the outer arc has been uplifted
since the Miocene along this fault, with resultant approach to the surface of a
Miocene high-pressure metamorphic complex in the crust on the oceanic side of
this fault. Since the isostatic anomalies in the outer arc are of the order of +100
mgal, the uplift cannot be ascribed to the buoyancy of a thickened crust. The
uplift could be due to the continentward push exerted by the Pacific plate on
the prismatic block above the descending slab.

Models for the Mechanism of Uplift of High-Pressure Metamorphic Complexes

Three alternative possible mechanisms for uplift of high-pressure metamorphic
complexes will be discussed below (Miyashiro, 1972b).
1. High-pressure metamorphic recrystallization may take place mainly in the
descending oceanic crust and the overlying sediment mass, as illustrated in fig.
3-11. The resultant metamorphic rocks may be added to the crust on the
continental side, and be uplifted by movement along a high-angle fault (S in the
figure). In this model, the slab-surface fault would occur at or near the
oceanic-side limit of the high-pressure metamorphic terrane.
2. A high pressure metamorphic complex may be formed in the foot wall of
a slab-surface fault and subsequently it may be uplifted (a) along the same fault
after the halt of the descent of the lithospheric slab, or (b) along a newly formed
high-angle fault presumably like the Morioka-Shirakawa Line. In model (a), the
major fault which represents both the slab surface and the boundary fault of the
uplifted block, should occur at the continental-side limit of the high-pressure
metamorphic terrane. This model has been proposed by Ernst (1970, 1971b).
Model (b) means that a high-pressure complex which formed in the foot wall of
a slab-surface fault is subsequently scraped off, added to the hanging wall, and
then uplifted along a newly formed high-angle fault.
3. If the position of the descending slab shifts successively oceanwards, a
high-pressure metamorphic complex formed at an older time should become a
part of the hanging wall of a subsequently formed slab-surface fault and could be
uplifted along a newly formed high-angle fault. In the Franciscan terrane,
 17-6 HIGH-PRESSURE METAMORPHIC COMPLEXES /403
metamorphism involves a number of distinct phases of recrystallization (Suppe,
1969), and the rocks in the western part are generally younger than those in the
eastern. This may be the result of successive oceanward shifts of the descending
slab (Hsii, 1971). A younger high-pressure metamorphic belt tends to form on the
oceanic side of an older high-pressure belt, as exemplified in Japan (fig. 3-10)
and the Kamchatka Peninsula (Dobretsov and Kuroda, 1969). This means that a
new Benioff zone tends to form on the oceanic side of an older one.
In models (1) and (3), a slab-surface fault should appear at or near the
oceanic-side limit of the high-pressure metamorphic terrane. In the Sanbagawa
high-pressure terrane of Shikoku (fig. 17-1), the Kurosegawa Tectonic Zone
(Ichikawa et al. 1956) is a group of large subparallel faults exposed near the
oceanic-side limit of the metamorphic terrane (Banno, 1964), and might be a
slab-surface fault. The Kiyomizu Tectonic Zone, which is exposed about 20 km
north of Kurosegawa, might also represent a slab surface. This zone is not a
simple fault but a zone, up to 1 km wide, of 'papery' schists probably
representing an unusually strong shear movement contemporaneous with the
Sanbagawa metamorphism (Kojima and Suzuki, 1958).

Distance between Paired Metamorphic Belts

In the present-day island-arc regions, there exists a non-volcanic zone, usually
100-250 km wide, between a volcanic belt and the associated trench. Dickinson
(1971) has proposed the name arc-trench gap for such a zone. The arc-trench
gaps of some present-day arcs are occupied by uplifted mountains, whereas those
of others are occupied by troughs where active sedimentation is taking place.
Since high-pressure metamorphism takes place in and near the trench zone, the
resultant metamorphic complex should have originally been situated about
50-200 km from the low-pressure metamorphic complex beneath the associated
volcanic belt.
Among the above-discussed models for the mechanism of uplift of high-
pressure metamorphic complexes, model (3) may apparently reduce the distance
between paired belts. In this case, the descending slab shifts oceanwards, and as a
consequence the, volcanic zone also should shift oceanwards with the resultant
approach of the underlying low-pressure complex to an earlier-formed high-
pressure complex.
The two metamorphic belts of the pairs in Hokkaido and California are
separated from each other by a zone of virtually unmetamorphosed rocks, which
may correspond to the arc-trench gaps. On the other hand, in the Jurassic-
Cretaceous pair of southwest Japan, the Ryoke low-pressure belt is in direct
contact with the Sanbagawa high-pressure belt along the Median Tectonic Line.
During the metamorphism, the trench was situated probably to the south of the
Sanbagawa belt, that is, in the Shimanto terrane (fig. 17-1), and hence the
distance between Ryoke and Shimanto may represent the arc-trench gap
404/TECTONICS OF REGIONAL METAMORPHIC BELTS

(Dickinson, 1971). The Sanbagawa metamorphism is the recrystallization not of

the contemporaneous geosynclinal sediments but of their basement. However, it
is still an important question how the high-pressure complex has come into
direct contact with the low-pressure complex. The direct contact may have been
caused by an oceanward shift of the descending slab, as in the above model (3).
However, it is more likely that the direct contact is due to a large strike-slip
movement along the Median Tectonic Line (fig. 7 in Miyashiro, 1972b).

17-7 TECTONICS OF ISLAND ARCS AND PAIRED METAMORPHIC

BELTS IN THE MAIN PART OF JAPAN

Double Arcs
The Northeast Japan Arc is an example of a presently active double arc. As
shown in figs. 3-11 and 17-2, it is composed of the following three parallel
structural elements from east to west: (a) the Japan Trench on the outermost
side, (b) the Kitakami Mountains and Abukuma Plateau which represent the
non-volcanic outer arc, mainly composed of Palaeozoic sedimentary rocks and
Mesozoic granites, and (c) the chains of Quaternary volcanoes which constitute
the volcanic inner arc. In this case, the double arc structure is slightly obscured
as the valley between the volcanic and the non-volcanic arc is above sea level.
Figure 17-2 shows SOO-meter contours above sea level to reveal the topographic
characteristics of the double arc.
The Ryukyu Islands also are a double arc, in which a trench is on the eastern
side, the non-volcanic outer arc lies in the median zone, including all the bigger
islands· of Ryukyu, and the volcanic inner arc, being on the western side, is
represented by small volcanic islands and submarine volcanoes.
On the other hand, in the double arc of the Kuru Islands, the volcanic inner
arc is better developed than the non-volcanic outer one. The Izu-Bonin Arc also
has a double arc structure, in which the Bonin Islands constitute an outer arc. To
the north of Bonin, an outer arc does not exist. The Kuril and Izu-Bonin Arcs
have deep trenches on the outermost zone.
The volcanic front (§IS-1, fig. IS-2) approximately represents the ocean-side
limit of the volcanic inner arc. The distance between the trench axis and the
volcanic front is about 200-300 km in the above double arcs.
Thus, many of the present-day island arcs are double arcs, and we may
consider that a double arc is a typical state of presently active island arcs. At
an early stage of its development, the non-volcanic outer arc may not be formed
with resultant formation of a single arc.

Northeast Japan Arc

As a well-investigated example of active double arcs, we will review here the
Northeast Japan Arc. The Arc has been active from the Miocene to the present
 17-7 METAMORPHIC BELTS IN THE MAIN PART OF JAPAN /405

/
!
\

Japan Sea

Kitakami
, Mountains

,,
".' Pacific Ocean

o 100 200 km

Fig. 17-2 Topography of the Northeast Japan Arc. The present-day coast
line is shown by dotted lines. Horizontal ruling indicated areas more than
500 metres above sea level. Quaternary volcanoes are shown by solid
circles. The areas more than 7,000 metres deep in the Japan Trench are
stippled.
406jTECTONICS OF REGIONAL METAMORPHIC BELTS

owing to the Mizuho orogeny. A generalized E-W cross-section of the Arc is

shown in figs. 3-11 and 17-3 (cf. Rikitake et al., 1968).

Trench Zone. This zone includes the Japan Trench and a zone on its immediate
west. It is characterized by a negative isostatic anomaly and low heat flow. The
epicenters of shallow earthquakes are highly crowded.
Probably this zone has been subjected to continual subsidence and sediment-
ation from the Miocene to the present, and may be regarded as a geosyncline.
The Pacific plate should be plunging into the depths along this zone. This plunge
probably brought the sediment piles deep into the mantle, and they would have
undergone metamorphism probably of the high-pressure type. The observed low
heat-flow zone should correspond to a low geothermal gradient (fig. 17-3).
The plunging plate reaches a depth of about 700 km beneath the north-
western part of the Japan Sea, though such a great depth is not shown in figs.
3-11 and 17-3.

Non- Volcanic Outer Arc. The outer half of the arc-trench gap is a submarine
slope with negative isostatic anomalies and has already been included in the
above trench zone, whereas the inner half of the gap is an elevated non-volcanic
arc with positive isostatic anomalies. In so far as arcuate ridges above sea level
are concerned, this zone may be separated from other zones and be treated as a
non-volcanic outer arc in contrast to volcanic inner arc. This zone has low heat
flow like the above trench zone.
The non-volcanic arc, consisting of the Abukuma Plateau and the Kitakami
Mountains, underwent continual uplift as rigid blocks, probablY since the
Miocene. The western limit of this region is the Morioka-Shirakawa Line
discussed above (§ 17-6).
Although granitic and metamorphic rocks are widely exposed in this arc, they
are all the result of the Late Mesozoic orogeny and possibly of older ones. The
metamorphic complexes formed since the Miocene, if any, have not been
exposed yet.

Volcanic Inner Arc. The volcanic inner arc is made up of mountain ranges in the
central and the Japan Sea side zone of Honshu. Miocene and Pliocene granitic
masses also occur in this region. Subsidence took place in many areas and small
scale geosynclinal accumulations were formed in the Miocene.
The present heights of volcanoes are the result partly of volcanic accumul-
ations and partly of the tectonic upheaval of their basements. In the geologic
past, the altitudes of this zone changed by the upheaval and subsidence of its
parts. The present high altitude may be rather fortuitous.
 17-7 METAMORPHIC BELTS IN THE MAIN PART OF JAPAN/407

+100 mgal
Isostatic anomaly
----------------------~~r_--~~~~------~O

---
-100

~---------']T:OoO
2

________________________---''--____H_e_at.L-fl_oW
________ mW m-
_

Honshu
Japan Sea I nner arc Outer arc Japan Pacific
• Trench Ocean

-10

-20

I
-30

-40

t -50

Fig. 17-3 E.-W. cross section of the Northeast Japan Arc. For the
topography, see Fig. 17-2.

This zone is characterized by high heat-flow values. Thick Miocene sediments

were metamorphosed in the zeolite, prehnite-pumpellyite and greenschist facies.
The metamorphism occurred generally under very high geothermal gradients (§
6A-2). The high values in heat flow and geothermal gradient should be somehow
related to volcanism and plutonism in this region. The rise of magmas and the
associated aqueous fluid from depth may be partly or entirely responsible for it.
408/ TECTONICS OF REGIONAL METAMORPHIC BELTS

Japan Sea. The Japan Sea region has an abnormally high heat flow, generally
between 80 and 130 mW m- 2 (2-3 ILcal cm- 2 sec-I). The Okhotsk and East
China Seas as well as the region to the immediate west of the Izu-Bonin Arc
show a similar high heat flow. Hence, this is probably an essential feature of
active island arcs.
The high heat flow should be related to the origin of the marginal seas.
Shallow earthquakes are rare in these seas (Barazangi and Dorman, 1969). There
is no feature suggesting the existence of an active mid-oceanic ridge. The
convective rise of-upper mantle materials together with the rise of magmas and
associated aqueous fluids could create the new lithospheres under the marginal
seas (fig. 3-11), and could also cause high heat flow.

Analogy between the Present-day Northeast Japan Arc and the Late Mesozoic
Southwest Japan Arc
The metamorphism of the Ryoke-Sanbagawa pair took place probably in
Jurassic and Cretaceous time. At and immediately after that time, there were
four tectonically distinct zones along the Southwest Japan Arc, which corres-
pond to the above-discussed four zones of the present-day Northeast Japan Arc
(Matsuda and Uyeda, 1971). The zones are as follows from south to north (fig.
17-1):
1. Shimanto geosyncline. This was probably a trench zone with thick
geosynclinal accumulations. The plunge of the oceanic plate from the trench
would have brought not only the new Mesozoic sedimentary pile but also
Palaeozoic formations under it on the north side into great depths.
2. Sanbagawa high-pressure metamorphic belt. This probably represents the
Palaeozoic terrane which was under and on the north side of the above
geosynclinal pile and was involved in the plunge of the oceanic plate. Sub-
sequently, the Palaeozoic rocks were uplifted and exposed at the surface. This
zone probably corresponds to the non-volcanic outer arc of the present-day
double arcs of northeast Japan.
3. Ryoke belt. This zone is characterized by low-pressure, high-temperature
metamorphism and abundant granitic and rhyolitic activity. During the meta-
morphism this zone probably showed high heat-flow values and volcanism just
like the present-day inner arc of northeast Japan. However, the Ryoke meta-
morphic belt is about 50 km wide in reconstruction, and is narrower than the
present-day inner arc, which is about 150 km wide. Therefore, the belt probably
corresponds only to an oceanic-side zone of the inner arc of northeast Japan.
4. Region of granitic activity on the continental side of the Ryoke belt. The
abundance of granitic masses suggests the occurrence of high heat flow at that
time. lienee, this zone may be regarded as corresponding to the continental side
part of the inner arc of northeast Japan, possibly together with a part of the
Japan Sea.
Chapter 18

~etalDorphic Structure
of Continental Crusts

18-1 CONSTITUTION OF CONTINENTS

The North American and the European Continent have a vast Precambrian
craton surrounded by Phanerozoic orogenic belts. On the eastern side of the
Asiatic Continent and of the Australian Continent, the outermost zones of
Phanerozoic orogenic belts form island arcs which are separated from the
continents by marginal seas such as the Japan and the Tasman Sea.
The thickness of the crust, as defined by the depth to the Moho dis-
continuity, is usually 30-50 km for Precambrian cratons, but is more highly
varied and commonly thinner for younger orogenic belts. The rock-pressure is
about 10 and 15 kbar at depths of 35 and 50 km respectively.
The observed seismic velocities in the continental crust increase generally
downwards:
Vp(km sec-I)
Upper crust 5·8-6·3
Lower crust 6·6-7·2
This increase is too great to be accounted for simply by the effect of pressure.
Thus, it is believed that the crusts of shields show general changes in chemical
composition with depth. Just as the exposed surface of shields are made up of
varied rocks, the depths would also be heterogeneous. However, the average
composition could become more basic with increasing depth. The thicknesses of
the more acidic upper part and the more basic lower part for various regions are
given in table 18-1.
There is a substantial difference in heat-flow values between Precambrian
cratons (including shields) and Phanerozoic orogenic belts. Lee and Uyeda
(1965) have given the following average figures:
mWm- 2 Ilcal cm- 2 sec- I
Precambrian shields 38 0·92
Cratons covered by Phanerozoic sediments 64 1·54
Palaeozoic orogenic regions 51 1·23
Mesozoic-Cenozoic orogenic regions 80 1·92.
410/METAMORPHIC STRUCTURE OF CONTINENTAL CRUSTS

Table 18-1 Thicknesses of crust

Upper Lower
crust crust Total
Vp (km sec-I) (5.8-6.3) (6.6-7.2) thickness

(km) (km) (km)

Baltic Shield (Finland) 18 18 36
Russian Plate 18 17 35
Interior Lowlands (N. America) 20 30 50
Appalachians 15 25 40
Alps (West Europe) 20 40 60
Caucasus (Alpine folding) 20 35 55
Sierra Nevada 25 25 50
California Coast Ranges 15 5 20
Northeast Japan (active arc) 15 17 32
Note: After Pakiser and Robinson (1966), Hashizume et al. (1968), Closs (1969),
and Kosminskaya, Belyaevsky and Volvovsky (1969).
Hence, the temperatures at the same depth are usually higher in Phanerozoic
orogenic belts than in Precambrian shields. The present temperature at the Moho
beneath the shields would generally be 400-750 °C.

18-2 PRECAMBRIAN SHIELDS

Upper Crust
Precambrian shields are mainly composed of granitic rocks and quartzo-
feldspathic gneisses, together with other metasediments and metabasites (§ 1-5).
The average chemical composition of the surface exposures of a shield has
been calculated by Sederholm (1925) for the Finnish part of the Baltic Shield
and by Shaw et aZ. (1967) and Eade and Fahrig (1971) for the Canadian Shield,
as given in table 18-2. The compositions are similar to those of granodiorite,
mica gneiss or graywacke in major elements, but differ from granodiorite
markedly in some trace elements.

Lower Crust
Since the lower crust of shields is more basic than the upper, many authors have
considered that the upper and the lower crusts are composed of granitic and
gabbroic (or basaltic) rocks respectively. Poldervaart (1955), Pakiser and
Robinson (1966) and Ronov and Yaroshevsky (1969) calculated the average
chemical composition of the crust on the basis of such models.
Ringwood and Green (1966a, b) have, however, criticized such models. If
PH,o is equal or close to Ps in continental crusts, the Pg-T condition of the
lower crust as discussed in the preceding section should be usually in the
 18-2 PRECAMBRIAN SHIELDS/411

Table 18-2 Average chemical composition of surface exposures of

Precambrian shields
Canada
Finland Canada Bade and Fahrig
Sederholm (1925) Shaw et aZ. (1967) (1971)

SiOz 67·45 64·93 65·1

TiO z 0·41 0·52 0·52
Alz0 3 14·63 14·63 16·0
FeZ03 1·27 1·36 1·4
FeO 3·13 2·75 3·1
MnO 0·04 0·068 0·08
MgO 1·69 2.24 2·2
CaO 3·39 4·12 3·4
NazO 3·06 3·46 3·9
KzO 3·55 3·10 2·89
PzOs 0·11 0·15 0·16
HzO+} 0·79 { 0·79 } 0·8
HzO- 0·13
COz 0·12 1·28 0·20
S 0·06
Cl 0·01
F 0·05
C 0·02
Less 0·04
Total 99·64 99·708 99·75

Note: Average trace element content for the Canadian Shield is (in p.p.m.): Be 1·3,
Ga 14, Cr 99, V 53, Li 22, Ni 23, Co 21, Cu 14, Zr 400, Sr 340, Ba 1070, Rb 118,
U 2.5, Th 10.3 (Shaw etal. 1967; Shaw, 1968); and Cr 77, Ni 20, Sr 380, Ba 730,
U 1.5, Th 10.8 (Eade and Fahrig, 1971).

greenschist, epidote-amphibolite and amphibolite facies, as shown in fig. 3-12.

There is some reason to think that the lower crust of Precambrian shields is
strongly dehydrated (§ 4-3). Hence, PH 2 0 may be much lower than Ps • If this is
the case, the Ps-T there should usually be in the eclogite and granulite facies
(transitional to the eclogite), as shown in fig. 12-6. Accordingly, even if the
lower crust has a basic composition, it cannot be gabbroic or basaltic in mineral
assemblage, but would be greenschist, epidote-amphibolite, amphibolite,
eclogite or garnet granulite.
Table 18-3 shows, however, that such rocks generally have higher seismic
velocities than the real lower crust. Therefore, the lower crust cannot be basic in
composition. Ringwood and Green have concluded that the lower crust should
be either intermediate (e.g. quartz dioritic) in composition or composed of
heterogeneous mixtures of acidic and more basic rocks. Though the upper crust
also is a mixture of acidic and more basic rocks, the ratio of more basic rocks
412/ METAMORPHIC STRUCTURE OF CONTINENTAL CRUSTS

Table 18·3 Compressional wave velocities (Vp) at 2 and 10 kbar in

km sec- 1

At 2 kbar At 10 kbar

Granite 6·1-6·4 6·3-6·5

Gabbro 7·0-7·1 7·2-7·3
Quartzo-fe1dspathic gneiss 6·0-6·3 6·3-6·7
Amphibolite 6·8-7·0 7·1-7·3
Epidote amphibolite 7·3-7·4 7·6-7·8
Eclogite 7·8-8·0 8·0-8·2
Dunite 8·0-8·3 8·3-8-4

Note: Compiled from Clark (1966) and Christensen (1965).

would be higher so as to give an intermediate bulk composition for the lower

crust.
A possible origin of the crustal layering will be discussed in the next section.

18·3 VERTICAL DIFFERENTIATION OF CONTINENTAL CRUSTS

Partial Melting
The present surface of Precambrian shields is commonly in the amphibolite and
granulite facies. During the metamorphism it was probably common for some
extent of partial melting to take place. At the same time, on the other hand, the
deeper levels within the crust would generally have been at higher temperatures.
The lower crust accordingly would have been subjected to a considerable extent
of partial melting. It is natural therefore to ascribe the crustal layering partly to
partial melting, though there should be many other relevant factors of which we
are quite ignorant.
The melts should usually be of granitic composition, and would move
upwards through the crust owing to their lower density. They may join to form
big granitic bodies, or may remain in a more dispersed state. Upward movement
of such melts would increase the acidic character of the upper crust, leaving
generally more basic solid residues in the lower crust.

Granulite·Facies Terranes
Ramberg (1951, 1952) has pointed out the possibility that granulite facies
metamorphic rocks now exposed on earth might once have been in the lower
crust and basified during metamorphism, though he preferred chemical diffusion
in the solid state to partial melting as a possible mechanism of basification.
Small· scale basification in a granulite facies zone was demonstrated by Engel
and Engel (1958, 1962a) for both metasediments and metabasites of the
 18-3 VERTICAL DIFFERENTIATION OF CONTINENTAL CRUSTS /413
Adirondacks (§ 8A-7). On a larger scale, Eade, Fahrig and Maxwell (1966) and
Bade and Fahrig (1971) calculated average chemical compositions of amphi-
bolite- and granulite-facies terranes in a part of the Canadian Shield. Granulite
facies rocks occur in a great number of separate irregular areas surrounded by
amphibolite facies areas. The granulite facies terranes are slightly higher in FeO
and CaO, but slightly lower in Si0 2 and K2 0 than the amphibolite facies ones.
There are a number of wide terranes of granulite facies rocks in the Australian
Shield. They may be classified into the three sub facies representing different
rock-pressures that were shown in fig. 12-2. The majority of the granulite facies
terranes belong to the low-pressure subfacies and are usually dominated by
acidic rocks, though the rest belong to the two higher pressure sub facies and are
generally dominated by moderately acidic, intermediate and basic rocks. The
latter terranes were probably metamorphosed and partially melted in the lower
crust and later tectonically uplifted to their present level. They should show the
chemical characteristics of the lower crust. Thus, Lambert and Heier (1968) have
made a detailed chemical investigation of such rocks. The rocks of the
amphibolite facies are generally similar in chemical composition to those of the
lower-pressure subfacies of the granulite facies. On the other hand, the rocks of
the two higher pressure sub facies of the granulite facies show an effect of
basification: decreases of Si0 2 , K2 0, Rb and U, and increases of total iron, MgO
and CaO. The K/Rb ratio tends to increase in the process.
The 'K/Rb ratio of ordinary continental igneous rocks varies in a range of
200-400. Metamorphic rocks in the· amphibolite and lower facies show a similar
range, whereas those in the granulite facies tend to show higher values. Sighinolfi
(1969, 1971) has demonstrated that both K and Rb contents have decreased and
the K/Rb ratio has increased in granulite facies basements of the Western Alps
and Brazil.

Possible Retrogressive Recrystallization of the Lower Crust

After the formation and modification of a continental crust in orogeny, the
region would be stabilized with a consequent decrease of the geothermal
gradient. In the upper crust, the decrease of temperature probably results in a
general halt of recrystallization, though retrogressive changes usually occur to a
limited extent and over a limited area. On the other hand, in the lower crust, the
temperature after the stabilization is probably still a few or several hundred
degrees Celsius, and hence it is conceivable that retrogressive recrystallization
occurs on a grand scale, as was mentioned in § 1-2.
Under such a condition, the granulite facies metamorphic complex of the
lower crust would be subjected to eclogite facies conditions, since PH, 0 is
probably very low. Such a change may be an extremely slow process which
could only take place over a long geologic time under non-orogenic conditions.
414/ METAMORPHIC STRUCTURE OF CONTINENTAL CRUSTS

18-4 PHANEROZOIC OROGENIC BELTS

Since Phanerozoic orogenic belts give much higher heat-flow values than
Precambrian shields, there must be a chemical difference in their crust and/or
underlying mantle.
The crust of shields could be strongly layered with the lower crust and the
underlying upper mantle impoverished in radioactive elements, and a consider-
able part of the original upper crust containing abundant radioactive elements
may have been eroded away. The more radioactive materials removed from the
upper part of shields would be mainly deposited in the adjacent Phanerozoic
geosynclines. This ultimately contributes to the increase of the radioactive
content of the Phanerozoic belts.
It has been widely believed that the Precambrian shields have been eroded to
much deeper levels than the Phanerozoic terranes. However, the common
occurrence of andalusite in the former casts a serious doubt on this belief. As far
as regional metamorphic terranes are concerned, there should be no great
difference in the depth of denudation between the Precambrian and the
Phanerozoic. Therefore, the common occurrence of granulite facies rocks in the
shields but not in the Phanerozoic belts suggests that the metamorphism was
generally more intense in the Precambrian terranes than in the Phanerozoic. If
this is the case, the crust of the Phanerozoic belts may not have been as strongly
layered as that of the Precambrian, and the erosion of the upper layer would not
have seriously changed the average compositions of the Phanerozoic belts.
On the other hand, compositional change of continental crusts with geologic
age of formation is also possible. Phanerozoic orogenic belts may have differed
in composition from early Precambrian ones at the time of their formation. For
example, Engel (1963) and Engel and Engel (1964a) have claimed that the first
continental crust in the present site of North America was formed on an oceanic
crust, and that the early continental crust was similar to the crust in present-day
island arcs and their vicinities, the composition of successively added new
continental crusts becoming gradually more similar to that of Phanerozoic
orogenic belts. In other words, the crust newly added by accretional growth of
continents becomes poorer in basic volcanics and richer in acidic volcanics, and
the maturity and sorting of sedimentary rocks increases with decreasing age.
Graywacke was dominant among the earliest sediments, whereas shale, quartzite
and carbonate rocks become important in late Precambrian and younger
sediments.
The Franciscan of California, the New Zealand geosyncline and some other
younger terranes are rich in basic volcanics and graywacke, and hence are
inconsistent with this view.
There are, however, strong opponents to the idea of continental growth and
progressive development of acidic crust. Wynne-Edwards and Hansan (1970), for
 18-4 PHANEROZOIC OROGENIC BELTS /415
instance, have insisted that new orogenic belts are always formed on sialic
basements of older belts, and there is little or no evidence for the significant
increase of sial on earth with geologic time, and that the compositional
difference between the continental crust and the uppermost mantle was greatest
in the earliest geologic time and has decreased since then by upward intro-
duction of basic magma into the crust.
Chapter 19

Ocean-floor Metamorphism and

its Significance

19-1 HARD ROCKS OF DEEP OCEAN FLOORS

Deep ocean floors are mostly covered by unconsolidated sediments, commonly a

few to several hundred metres thick, which are underlain usually by basaltic
rocks.
Basalts are widely exposed on mid-ocean ridges and seamounts. Most of the
basalts on the submarine parts of mid-oceanic ridges are olivine tholeiites with
K2 0 contents below 0.3 per cent and extremely low in Ba, Sr, Ph, Th, U and Zr.
Their Na/K and K/Rb ratios are higher than those of most continental basalts.
Such basalts have been called oceanic tholeiites or abyssal tholeiites (Engel and
Engel, 1964b; Engel et al. 1965).
Abyssal tholeites usually show a relatively small range of compositional
variation which is probably partly due to differences in the original magmas and
partly due to crystallization differentiation at various depths (Miyashiro, Shido
and Ewing, 1969a; Shido, Miyashiro and Ewing, 1971). Some tholeiitic magmas
form intrusive bodies, in which crystallization differentiation occurs, sometimes
leading to the formation of peridotites, ferrogabbros and more acidic rocks
(Miyashiro, Shido and Ewing, 1970b; Aumento, 1969).
Oceanic islands are commonly composed of basalts and their differentiates of
the alkali basalt series. Such basalts occur rarely on the submarine parts of
mid-oceanic ridges also.
Peridotites and serpentinites occur on mid-oceanic ridges (Tilley, 1947; Hess,
1962; Melson et al. 1967; Miyashiro, Shido and Ewing, 1969b). They are
especially abundant within transverse fracture zones across the Mid-Atlantic
Ridge.
Peridotites and pargasite-bearing varieties of serpentinite of oceanic regions
are relatively high in CaO and Al 2 0 3 contents and chemically similar to so-called
high-temperature peridotite intrusions, which are now widely believed to have
risen as solid masses from the upper mantle. On the other hand, pargasite-free
serpentinites are lower in CaO and Al 2 0 3 than pargasite-bearing ones. Such a
compositional variation would be due partly to heterogeneity in the upper
 19-2 METABASALTS AND METAGABBROS/417

mantle, and partly to chemical migration during serpentinization (Miyashiro et

al. 1969b).
Many basaltic rocks and most gabbroic rocks are metamorphosed, however.
Ocean-floor metamorphism gives important information on the activity of
mid-oceanic ridges. This problem will be treated in the following sections.

19-2 METABASALTS AND METAGABBROS FROM DEEP OCEAN FLOORS

Discovery and Modes of Occurrence

Until recently, it was widely believed that the oceanic hard crust was an
essentially igneous structure composed of basalts, gabbros, peridotites and
serpentinites. In 1966, however, metamorphosed basalts were reported to have
been dredged on the crest of the Mid-Atlantic Ridge near 22°N by Melson et al.
(I966) and by Melson and van Andel (I966), and on the crest of the Carlsberg
Ridge near SON in the Indian Ocean by Cann and Vine (1966). In the few
succeeding years, metabasalts were found from many other areas on the crest of
the Mid-Atlantic Ridge (Aumento and Loncarevic, 1969; Miyashiro et al. 1970a,
1971).
The metamorphic rocks were all derived from basalts, dolerites and gabbros
except for the serpentinites which might well be regarded as metamorphic
derivatives of peridotites. The metamorphism must have taken place at some
depth, probably beneath the crest of mid-oceanic ridges, since relatively high
temperature prevails there. Metamorphic rocks exposed at a great distance from
the ridge axis could have been laterally transported by ocean-floor spreading.
Metamorphic rocks occur in the Mid-Atlantic Ridge and Carlsberg Ridge
(northern part of the Mid-Indian Ocean Ridge), which both have rough
topography, in contrast to the East Pacific Rise with smooth topography which
has no exposures of metamorphic rocks. Presumably, metamorphic rocks are
present in depth beneath all mid-oceanic ridges, but they are covered by basalts
on the East Pacific Rise, whereas intense fault and other tectonic movements on
the Mid-Atlantic Ridge resulting in rough topography have led to exposure of
metamorphic rocks.
On the Mid-Atlantic Ridge, metamorphic rocks are exposed on the lower
walls of the Median Valley and transverse fracture zones. In the latter case, they
are accompanied by serpentinites, and appear to be inclusions therein. Probably
the serpentinites were intruded in the solid state along the fracture zones.
Metamorphic rocks formed at some depth were probably torn apart and brought
up to the surface. This interpretation is consistent with the view that the
serpentinites were formed by hydration of upper mantle peridotites.

Metabasalts
The majority of metabasalts so far dredged are non-schistose or only weakly
418/oCEAN-FLOOR METAMORPHISM AND ITS SIGNIFICANCE

schistose, preserving their original igneous textures. Recrystallization is incom-

plete in most specimens. The original structures of pillow lavas and breccias are
commonly noticeable. It suggests that basaltic eruptions produced thick volcanic
piles, the lower part of which became sufficiently hot to undergo metamorphic
recrystallization. This may be regarded as a kind of burial metamorphism.
Most of the metabasalts belong to the zeolite and greenschist facies. It is not
known whether rocks of the prehnite-pumpellyite and epidote-amphibolite
facies exist in deep ocean floors. Some metabasalts are in the greenschist facies
transitional to the amphibolite facies.
Some of the greenschist facies metabasalts preserve their original chemical
compositions as abyssal tholeiites except for an increase of H20 content, but the
rest show a marked decrease of CaO content and a variation of Si0 2 content
(Miyashiro et al. 1971). Many metabasalts are spilitic (Melson and van Andel,
1966; Cann, 1969). Virtually all the zeolite facies metabasalts show marked
introduction of Na20 from the outside. Such compositional variations could
have been caused by flow of a hot aqueous fluid through the crust. With intense
decrease of CaO, the amount of chlorite increases, leading to the formation of
chlorite-quartz rocks (Miyashiro et al. 1971).
The detection of chemical migration during metamorphism is relatively easy,
because the composition range of abyssal tholeiites is limited.

Metagabbros
Most metagabbros preserve their original texture. They are in the greenschist and
amphibolite facies. However, some of them have suffered retrogressive changes
in the zeolite facies.
There are unmetamorphosed basalts as well as metabasalts on mid-oceanic
ridges, whereas nearly all gabbroic rocks are more or less metamorphosed. The
metabasalts are usually in the zeolite or greenschist facies, whereas the meta-
gabbros are usually in the greenschist or amphibolite facies. These two facts
suggest the common presence of gabbroic rocks in relatively deep levels beneath
the ridges.
In view of the small thickness of the oceanic crust, the temperature of typical
amphibolite facies would be too high to be realized within the crust even
beneath the crest of mid-oceanic ridges. It follows that some metagabbros were
probably consolidated and then recrystallized in the upper mantle.
Most metagabbros preserve their original chemical composition. However,
some metagabbros underwent chemical migration similar to that in metabasalts.
In other words, introduction of Na20 occurred in the zeolite facies, and a
decrease of CaO and variations of Si0 2 occurred in the greenschist facies
(Miyashiro et al. 1971).
It is to be noted that chemical migration is much more intense in the zeolite
and greenschist facies than in the amphibolite facies. Amphibolite facies rocks
 19·3 SIGNIFICANCE OF OCEAN·FLOOR METAMORPHISM /419
have not undergone the intense metasomatism characteristic of lower tempera·
tures.

19·3 SIGNIFICANCE OF OCEAN-FLOOR METAMORPHISM

Relation to Heat Flow

The crust beneath normal oceanic basins is only about 6 Ian thick. The
temperature at its base would be about 120-180°C. Thus, fissure filling and
incipient metamorphism in the zeolite facies may take place in the lower crust,
but marked metamorphic recrystallization cannot. Metamorphism in the green·
schist and higher facies would take place at higher temperatures usually in the
crust and upper mantle beneath the crest of mid·oceanic ridges.
The observed heat flow on the crest is much higher than in normal oceanic
basins. This suggests a higher geothermal gradient and higher temperatures at
depth beneath the crest than beneath normal oceanic basins. The temperature in
the deeper crust there may reach about 350°C to correspond to greenschist
facies. Amphibolite facies recrystallization could take place in gabbroic pockets
in the uppermost mantle.
The heat·flow values measured on the crest are highly variable. Values as high
as 300 mW m- 2 (8 Ilcal cm-2 sec-I) occur mixed with those as low as 40-80
mW m-2 (1-2Ilcal cm-2 sec-I). If all these values are accepted, the high values
could be due not only to the general heat conduction but also to some more
localized situations. Igneous intrusion is one possible situation of this kind. A
flow of a hot aqueous fluid is another. The hot fluid that caused metasomatism
should transport heat, causing localized high heat flow. The widespread intro-
duction of Na20 into zeolite facies metabasites suggests permeation of an hot
aqueous fluid. Accordingly, even the apparently general increase of heat flow on
the crest may be due not only to heat conduction but also to permeation of a
fluid (Miyashiro et al. 1971; Miyashiro, 1972a).
If a rock recrystallized beneath the crest moves laterally at the same depth
owing to ocean-floor spreading, the temperature of the rock will gradually
decrease, leading possibly to retrogressive changes.

Magnetized layer and Demagnetization

Normal oceanic basins are covered by unconsolidated sediments, whereas the
crest of mid-oceanic ridges usually has almost no sediment cover. At any rate,
ignoring the sediment cover, the surface layer of the oceanic crust is made up of
basalts and dolerites. This layer is responsible for the linear patterns of magnetic
anomalies of the ocean floors.
Recent investigations, however, have revealed that the magnetized layer is as
thin as about 0'5 km to a few km, and that the underlying layer is virtually
demagnetized (Vine, 1966; Heirtzler, 1968). The demagnetization takes place at
420/0CEAN-FLOOR METAMORPHISM AND ITS SIGNIFICANCE

too shallow a depth to be ascribed to heating above the Curie point. The
demagnetized layer of the crust is probably composed of metabasalts, meta-
dolerites and metagabbros. During metamorphic recrystallization, the strong
thermo-remanent magnetism of igneous rocks should disappear and instead
chemical remanent magnetization should appear. The latter, however, is
negligible in strength in comparison with the former (Miyashiro et al. 1971).

Layers 2 and 3
Seismologically, the oceanic crust beneath normal oceanic basins is usually
divided into three layers. Layer 1 represents unconsolidated sediments. Basaltic
volcanics containing a considerable amount of pyroclastic or fragmental
materials show a seismic velocity corresponding to that oflayer 2, and gabbros,
metabasalts and metagabbros show a seismic velocity corresponding to layer 3.
Accordingly, layers 2 and 3 may be regarded as being composed of such
materials in many cases.
However, it is known that layer 3 is absent beneath the crest of Mid-Atlantic
Ridge, and instead layer 2 is thicker there than beneath ordinary oceanic crust
(Talwani, I.e Pichou and Ewing, 1965). This must be accounted for.
The rough topography of the crest of the Mid-Atlantic Ridge is probably a
surface manifestation of strong fault movements, which should cause fracturing
within the crust and the upper mantle. Intensely fractured metabasites show a
marked decrease of seismic velOCity in comparison with unfractured rocks. This
could lead to the disappearance oflayer 3, and the complementary thickening of
layer 2.
The crestal crust would move laterally, eventually entering the region where
no more fracturing takes place. The openings in the fractured metabasites and in
the overlying pyroclastic accumulations would be gradually filled by low-
temperature mineral veins. The filling would occur more easily in the lower part
of the crust, thus resulting in the generation of a coherent crustal plate, which
should be layer 3. Thus, the boundary between layers 2 and 3 in this case is not
the same as the boundary between volcanics containing pyroclastic materials and
the metamorphosed and unmetamorphosed solid rocks.
In the East Pacific Rise, fault movements are not so intense judging from its
smooth topography. There, layer 3 exists beneath the crest and flanks.
Chapter 20

Cataclastic Metamorphism along

Transform Faults

20-1 CLASSIFICATION AND FIELD RELATIONS OF CATACLASTIC

METAMORPHIC ROCKS

Our knowledge of cataclastic metamorphism in general, and of cataclastic

metamorphism along transform faults, is very superficial and scrappy. In view of
the tectonic importance of transform faults, a brief review of cataclastic rocks
and particularly of those related to transform faults will be given in this chapter.
Since the extent of recrystallization is by definition limited in cataclastic
metamorphism, it is usually considered to occur at low temperatures. However,
the maximum possible temperature for such metamorphism is unknown. If the
rate and amount of deformational movement are great and if the length of time
favourable for recrystallization is short, cataclastic rocks may form even at
medium temperatures.

Classification
Weak cataclasis produces tectonic breccias, which are mainly made up of angular
rock-fragments set in a subordinate matrix of more highly crushed materials.
With advanced cataclasis, the proportion of crushed materials increases, and
eventually aphanitic rocks form. Large crystals and large rock-fragments which
have survived crushing in rocks subjected to advanced cataclasis are called
porphyroclasts.
Rocks subjected to advanced cataclasis are classified here into five groups
(Read, 1964; Higgins, 1971), as follows:
1. Incoherent cataclastic rocks are generally believed to form at low con-
fining pressure under near-surface conditions. These include fault breccia, fault
gouge and fault pug in order of increased crushing.
2. Structureless cataclasites are coherent rocks showing a mortar texture
without foliation or preferred orientation. This texture is characterized by the
presence of porphyroclastic minerals set in a matrix of finely-crushed material.
The name cataclasite was originally coined to represent structureless cataclastic
metamorphic rocks by Grubenmann and Niggli (1924), but was used afterwards
422/CATACLASTIC METAMORPHISM ALONG TRANSFORM FAULTS

to represent cataclastic rocks in general by other authors. The term structureless

cataclasite leaves no possibility of confusion.
3. Mylonites are coherent rocks showing flow structure (foliation) of cata-
clastic origin (Lapworth, 1885; Waters and Campbell, 1935). The flow structure
is usually an alternation of materials showing different degrees of cataclasis, or
showing different mineral compositions. Mylonites are believed to have been
produced by grinding in a definite direction. The names protomylonite,
mylonite (in a narrow sense) and ultramylonite have been used to represent the
three subclasses of the mylonites in order of increasing degree of cataclasis.
Most rocks of the above three classes show a small extent of recrystallization,
which commonly produces greenschist facies minerals such as quartz, chlorite,
epidote, and calcite, though minerals of other facies are not rare. If recrystal-
lization is more important, the resultant rocks belong to the following class.
4. Cataclastic rocks apparently subjected to a considerable extent of re-
crystallization usually show a flow structure, and include mylonite gneiss and
blastomylonite in order of increasing mylonitization. The cataclasis and re-
crystallization are said to be concurrent, but the available evidence for this is not
always sufficient.
5. Rocks may be melted by exceptionally intense frictional heating due to
grinding in a mylonite zone. Such a rock has been called pseudotachylite, but
most rocks described as pseudotachylites may have been formed in other ways.
Truly melted rocks appear to be extremely rare (e.g. Scott and Drever, 1953).
Some mylonites subjected to extremely strong grinding, become submicroscopic
in grain size, and look dark between crossed polars (nicols), even though they are
not true glasses (e.g. Waters and Campbell, 1935).
Most mylonites have been derived from quartzo-feldspathic rocks, though
some examples of basic and ultrabasic mylonites are known.
Quartz and micas are easily affected by deformation and cataclasis. Quartz
first becomes elongated and strained, and then shattered at the edges of grains or
throughout, ultimately becoming fme-grained masses, usually around porphyro-
clasts of other minerals. Micas are easily recrystallized but tend to preserve a
record of deformation in their shapes. With advancing cataclasis, micas are also
ground into fmer flakes. Feldspars and especially garnet are resistant to crushing,
and as a result commonly form porphyroclasts in cataclastic rocks. A detailed
and critical glossary of cataclastic rocks has been given by Higgins (1971).

Field Relations
On and near the surface of the earth, block movement produces a sharp fault
plane or a sharply-limited narrow fault zone, which is filled with incoherent
cataclastic rocks. At greater depths, a similar block movement appears to
produce usually an ill-defmed fault zone which is made up of coherent
cataclastic rocks.
 20-2 SAN ANDREAS FAULT ZONE IN CALIFORNIA/423

Higgins (1971) has suggested that within such a fault zone mylonites are
formed in a zone where deformational movement is especially intense, and the
mylonite zone grades into zones of cataclastic rocks accompanied by a
considerable extent of recrystallization (such as blastomylonites and mylonite
gneisses), which in tum grade outwards into ordinary schists and gneisses. In
such a fault zone, structureless cataclasites usually occur in relatively well-
defined zones, which are either parallel or oblique to the zone of mylonites. It is
said that such structureless cataclastic rocks are younger than the mylonites and
other cataclastic rocks showing flow structure.
In many areas, sharp faults accompanied by incoherent cataclastic rocks cut
through the zones of mylonites and structureless cataclasites. These faults appear
to represent movements at still later times. The San Andreas fault along which
the movement of the San Francisco earthquake occurred, for example, is such a
sharp fault cutting through a wide, old zone of mylonites.
A transform fault should extend from the surface to the base of the
lithosphere. With increasing depth, the temperature and confining pressure
should increase with resultant change in the nature of cataclastic metamorphism.
However, we have no clear knowledge of this change.

20-2 THE SAN ANDREAS FAULT ZONE IN CALIFORNIA

The San Andreas fault cuts through California in a SE-NW direction (fig. 13-3).
Its total exposed length is about 1000 km. The cumulative right-lateral
displacement since the Jurassic is estimated at about 550 km (Hill and Dibblee,
1953).
It is widely believed that this fault is now a transform which connects the
Gulf of California segment of the East Pacific Rise with its Northeast Pacific
segment, i.e. the Gorda Ridge (Wilson, 1965; Suppe, 1970). The Pacific plate is
in contact with the North American plate along this fault. Prior to the middle
Tertiary, the whole length of the East Pacific Rise was probably off the west
coast of North America. The plate spreading eastwards from the Rise, called the
Farallon plate, was being subducted in the trench zone along the west coast of
North America, whereas the present Pacific plate was at that time spreading
westwards from the Rise. Thus, the Pacific plate was separated from the North
American plate by the Farallon plate. The San Andreas fault became the
boundary between the Pacific and North American plates in middle Tertiary
time when the Farallon plate was consumed (Atwater, 1970). The nature of the
San Andreas fault prior to middle Tertiary time is not clear.
In detail, the fault is a complex zone, up to several km wide, of cataclastic
rocks cut by a large number of subparallel faults (e.g. Dickinson, 1966). The
most recent great movement along a fault within the fault zone took place
424/CATACLASTIC METAMORPHISM ALONG TRANSFORM FAULTS

during the San Francisco earthquake of 1906. Cataclastic rocks observed at the
present surface were probably formed at various depths.
The mylonites in the fault zone in an area south of San Francisco were
described by Waters and Campbell (1935). The exposed rocks are heterogeneous
breccias, which are composed of angular blocks of a great variety of rocks such
as metabasalt, serpentinite, quartz diorite, and quartzite, embedded in a
dark-coloured aphanitic mylonite matrix. The mylonite matrix grades into the
blocks with gradual changes in colour, grain size and the degree of banding. The
mylonites show well-developed banding (lamination), which produces a· rock
cleavage similar to that of slate. The banding usually strikes parallel with the
strike of the fault zone, and usually dips nearly vertically, though local variations
and contortions appear.
Every stage in the crushing of original coarse-grained rocks to a submicro-
scopic paste can be observed. In general, crushing begins with the fracturing of
the rock into a coherent breccia whose fragments are megascopically visible.
Further movement result!! in the formation of a thin ftlm of crush powder
between the mineral grains forming mortar structure. The next stage is the
development of a closely spaced series of microscopic shears in the individual
mineral grains. Then, fIlms of pulverized material form and grow along the
shears, until finally the uncrushed material between adjacent shears is isolated as
streaks, lenticles and irregular masses in the powdery matrix. Complete milling
of the rock to a submicroscopic powder takes place only rarely. Some bands in
extremely fmely pulverized mylonites are dark between crossed polars.
In all the mylonites, w~ fmd at least some recrystallization in the matrix. The
most common newly formed minerals are quartz and chlorite.

20-3 THE ALPINE FAULT OF NEW ZEALAND

The Alpine fault cuts through the South Island of New Zealand in a NE-SW
direction (fig. 16-2). The r;umulative right-lateral movement since Jurassic time is
estimated at about 480 km as mentioned before. The fault could be a transform
fault connecting the Tongll-Kermadec trench with another trench on the west
side of the Macquarie Rise, though the possibility of other interpretations has
not been precluded.
Cataclastic rocks of ~ the four classes mentioned above, occur in a zone, up
to about 2 km wide, along the fault (Reed, 1964). The incoherent rocks have
been ascribed to near-surface crushing during relatively recent (probably Quater-
nary) movement. Structureless cataclasites and mylonites have been ascribed to
different types of movement at greater depth in older geologic times. A
gradation has been observed from a mylonite zone to the original Haast schists.
Newly formed minerals including chlorite, epidote and carbonate are wide-
spread.
 20-4 CATACLASTIC ROCKS IN OCEAN FLOORS/425

20-4 CATACLASTIC ROCKS ALONG TRANSFORM FAULTS IN OCEAN

FLOORS

Cataclastic metamorphic rocks have been found to occur along fracture zones
across the Mid-Atlantic Ridge. They could have formed by movement along
transform faults (Miyashiro et al. 1971). However, the degree of crushing in.the
examples reported to date is not high.
St Paul's Rocks in the equatorial Atlantic is a group of small islands made up
of well-mylonitized peridotites and related rocks (Tilley, 1947; Melson et al.
1967). The islands are located near the junction of the St Paul's fracture zone
with the crest of the Mid-Atlantic Ridge. The mylonitization may simply be a
result of upward intrusion of the ultrabasic mass in the solid state, or
alternatively it may be related to the transform fault and/or normal faults in the
crest of the Ridge. The banding of the mylonite is vertical and strikes nearly
north-south. In other words, the banding is roughly perpendicular to the trend
of the fracture zone.
Appendix
History of the Study
of Metamorphism

1. VIEWS ON METAMORPHIC ROCKS IN THE LATE EIGHTEENTH AND

THE EARLY NINETEENTH CENTURY

Fragments of geological knowledge began in Classical Greece or even at an older

time. The science of geology as a field of systematic knowledge was initiated in
the middle eighteenth century by J. E. Guettard and Nicolas Desmarest in
France. Toward the end of the century, two great theorists appeared: Werner
and Hutton. A. G. Werner (1749-1817) was an influential professor of geology
and mineralogy (collectively called geognosy by him) at the School of Mines at
Freiberg, Saxony. This school was founded in 1765 as one of the oldest
institutions for education of professional mining engineers and geologists.
Werner and his followers, who were called Neptunists, explained all the rocks as
being of sedimentary origin (chemical and clastic). They had no ideas of igneous
and metamorphic rocks and of tectonic movements.
James Hutton (1726-97) was a leisured gentleman and typical eighteenth-
century intellectual living in Edinburgh, who had active interest in all the
branches of human knowledge including chemistry, meteorology, agriculture and
philosophy. He spend much time in geological studies, and regarded basaltic and
granitic rocks as being formed by consolidation of molten materials (magmas).
The advocates of this idea were calledPlutonists. The concept of metamorphism
is said to have begun with Hutton's theory, a systematic description of which
was given in his book, Theory of the Earth (1795). In Hutton's view, some
sedimentary rocks were brought to depths of the earth where high temperature
and high pressure caused metamorphism of them. The schists and gneisses of the
Scottish Highlands, for instance, were regarded as being metamorphic, though he
did not introduce such a technical term.
According to Johannsen (1931, p. 185), the term 'metamorphism' was
introduced by A. Boue in 1820, while the term 'metamorphic rocks' became
popular through the first edition of Charles Lyell's Principles of Geology (vol. 2,
1833).
As regards the origin of basalt and granite, the Piutonists had secured a
victory by 1820, that is, immediately after the death of Werner. However, the
problem as to whether schists and gneisses are metamorphic rocks, primordial
igneous rocks or chemical sediments that were deposited in primordial oceans,
430/HISTORY OF THE STUDY OF METAMORPHISM

was not settled until the middle of the nineteenth century. According to the
common view at that time, gneisses were formed in the oldest geologic age, and
then schists were formed, being followed by phyllites. Fossiliferous sedimentary
rocks were considered to have been formed at a still later time.
In the meanwhile, transitional relationships between unmetamorphosed
sediments and schists were found at places. Thus, a greater number of geologists
gradually came to think that some or all of the phyllites and schists are
metamorphic in origin.

2. VIEWS ON REGIONAL METAMORPHISM IN THE SECOND HALF OF

THE NINETEENTH CENTURY

The distinction between contact and regional metamorphism is said to have been
first noticed in the middle of the nineteenth century by Elie de Beaumont and
A. Daubre. The term regional metamorphism was proposed by the latter
author.
The geosynclinal theory for mountain building was formulated in the period
1859-1910 by James Hall, J. D. Dana, and E. Haug. The widespread
occurrence of regional metamorphic rocks in orogenic belts attracted the
attention of all these authors. The high temperature, high pressure, and
deformational movement in the depths of geosynclinal piles were considered to
be the cause of metamorphism.
Though I am not familiar with the publications of this period, it appears that
there were two contrasting views on the cause of regional metamorphism. One
school of geologists stressed the importance of high temperature in the depths of
the earth and effects of plutonic masses as the agents of regional metamorphism,
while the other school emphasized the effects of pressure (hydrostatic or
non-hydrostatic) and deformational movement.
The former school included many geologists in Britain and France, some of
whom used the name of 'plutoniC metamorphism' for regional metamorphism.
The effect of water and other materials emanating from the associated plutonic
masses was especially emphasized by French authors.
The latter school who accentuated the effects of pressure and deformational
movements included many German and Swiss geolOgists. They used the name
of 'dynamic' or 'dislocation metamorphism' for regional metamorphism. The
term 'dynamic metamorphism' (or dynamometamorphism) was proposed in
1886 by H. Rosenbusch, who was a great master in descriptive petrography. This
term, and the idea attached to it, were propagated by his students and pervaded
generations of geologists all over the world.
Furthermore, some later authors began to use the names of dynamic and
dislocation metamorphism to denote cataclastic metamorphism (i.e. intense
mechanical deformation of rocks). Many other names were introduced to
 3. GOLDEN AGE OF MICROSCOPIC PETROGRAPHY /431
represent the supposedly dominant agents such ~ thermal and dynamothermal
metamorphism. This increased the confusion of nomenclature.
Metamorphic rocks closely associated with granitic rocks are widely exposed
in Britain and France, whereas the Alpine metamorphic rocks are rarely
accompanied by granitic ones. This difference may have been a part of the
factual basis underlying the contrasting views on regional metlmlorphism.

3. GOLDEN AGE OF MICROSCOPIC PETROGRAPHY

The period of 1870-1900 was the golden age of microscopic petrography.

Microscopic observation of thin sections was a powerful new technique in
geology at that time, before which rocks had been examined ]:>y means of the
magnifying glass and chemical analysis. Reliable determination of rock-forming
minerals and their textural relations became possible for the tlISt time by the
introduction of the microscope. It was then a new finding that plagioclases form
a continuous series of solid solution. Thus, the nature and the extent of diverSity
of the rocks were first clarified. A great number of rock names were coined and
systems of rock classifications were proposed.
F. Zirkel (1838-1912) of Leipzig and H. Rosenbusch (1836-1914) of
Heidelberg were the two greatest masters in this field who attracted students
from all over the world. Since their contributions were so remarkable, micro-
scopic petrography became the major field in the study of rocks. The threefold
classification of rocks into igneo~s, sedimentary and metamorphic was proposed
by B. von Cotta in 1862, and was accepted by Zirkel and Rosenbusch. They
published voluminous books, among which Rosenbusch's Elemente der Gesteins-
lehre (3rd ed., 19~0) is relatively concise and readable.
The last phase of the pre-eminence of microscopic petrography was represent-
ed by the great work Die Kristallinen Schiefer (1904-6) written by U.
Grubenmann (1850-1924) of Zurich. In this book, h~ classified all the regional
metamorphic rocks into twelve groups according to their chemical compositions.
The rocks of each group were divided into three categories according to the
depth-zones, named epi-, meso- and kata-zones iIi order of presumably increasing
depth of their metamorphism. The epi-zone (Le. shallow zone) was assumed to
be characterized by low temperature, low pressure and strong deformation, the
kata-zone by high temperature, high pressure and weak deformations. The
meso-zone was assumed to be intermediate. This idea of depth zones had been
originally formulated by Becke (1903), but became popullp" through Gruben-
mann's book.
All the observed regional metamorphic rocks were assigned to one of the
three zones. For example, phyllite, chlorite schist, and glaucophane schist were
assigned to the epi-zone, mica schist and amphibolite to the meso-zone, and
some of the gneisses, eclogite and jadeite rock to the kata-zone. The ass~gnment
432/HISTORY OF THE STUDY OF METAMORPHISM
was made on the basis of general impressions about the grain sizes and mineral
compositions, since the mineral zonation of progressive metamorphic terranes
was not known at that time. In Zurich, Grubenmann was succeeded by Paul
Niggli, who continued to advocate the basic idea and terminology of his
predecessor. Hence, Grubenmann's doctrine dominated a large part of con-
tinental Europe almost to the present (e.g. Grubenmann and Niggli, 1924).
In about 1920-30, advocates of new approaches of metamorphic geology had
to justify their existence by criticizing the doctrine of Grubenmann and Niggli.
For example, Harker criticized Grubenmann's static attitude (as was already
mentioned in § 3-1). Eskola also repeatedly criticized Grubenmann's depth-
zones in order to show the superiority of his facies classification.

4. DISCOVERY AND SIGNIFICANCE OF PROGRESSIVE METAMORPHIC

ZONES
Progressive Metamorphic Zones
Prior to the beginning of study of the progressive mineral changes, the
progressive textural changes were investigated by Rosenbusch (l877) in the
contact aureole of the Barr-Andlau area in Alsace (§ 10-1). He distinguished two
textural zones on a geologic sketch map. Progressive textural zones were later
mapped in regional metamorphic terranes of New Zealand and other countries.
Progressive mineral zones were mapped for the first time in 1893 by George
Barrow in a part of the regional metamorphic terrane of the Scottish Highlands
(fig. 3-1). He was a self-taught mapping geolOgist (1853-1932) of the British
Geological Survey. His work was completed in 1912, and was a most epoch-
making contribution to metamorphic petrology. However, it did not attract
much attention until the confirmation of his work by Tilley (1925) and Harker
(l932).
Without being aware of Barrow's work, V. M. Goldschmidt (1915) investi-
gated the progressive regional metamorphism in the Trondheim area (then called
Trondhjem) of Norway (fig. 1-2).
The investigation of mineral zoning did not become popular until the
publication of studies by Tilley (1925) and Harker (l932) in the Scottish
Highlands, by Vogt (l927) in the Sulitelma area of Norway, and by Barth
(1936) in Dutchess County, New York State. Even in the late 1930s, the area
and the diversity of the mapped terranes were so small that this method did not
effectively work as the connecting link between petrography and geology.
The first successful attempt to demonstrate the geolOgic significance of
progressive metamorphic zones was made by W. Q. Kennedy (1948), who
showed the distribution of metamorphic zones in the whole Scottish Highlands
and connected it to the structure of the Caledonian orogenic belt.
 4. DISCOVERY OF PROGRESSIVE METAMORPHIC ZONES/433

Alfred Harker (1859-1939)

Harker was not only a renowned preacher of progressive metamorphism, but also
a creator of a unique doctrine of metamorphism. He belonged to St John's
College and Department of Geology (not that of Mineralogy and Petrology) in
the University of Cambridge. In his early years, he played a leading part in the
introduction of microscopic petrography into Britain and was a brilliant igneous
geologist as the author of the Skye Memoir (1904) and the book The Natural
History of Igneous Rocks (1909). After the publication of this book, his major
interest was directed toward the study of metamorphism. His basic ideas on
metamorphism were first formulated in 1918. With an addition of a lot of
descriptive material, he wrote the book Metamorphism (1932). This beautifully
written book exerted a deep influence on those interested in metamorphic rocks
allover the world in the 1930s and 1940s.
In the book he gave good descriptions of progressive contact as well as
regional metamorphism. Regional metamorphism was described mostly on the
basis of his observation on rocks of the Barrovian region (§ 14-3) of the Scottish
Highlands, though the metamorphism in the Buchan region was also briefly
treated. He believed that most cases of regional metamorphism allover the world
would show a great similarity to one another, and that the regional meta-
morphism observed in the Barrovian region represented the 'normal' kind of
regional metamorphism. Probably because of the expected similarity, descriptive
data from the Norwegian Caledonides, the Alps, Brittany and other areas outside
Britain were mixed into the systematic description of rocks of the Scottish
Highlands. His concept of normal regional metamorphism pervaded the thinking
of geologists for thirty years.
His doctrine was based largely on the structural and mineralogical difference
(or contrast) between regional and contact metamorphism. He ascribed it to the
presence of st~ong non-hydrostatic stress in the former and its absence in the
latter. Chlorite, muscovite, almandine, kyanite are more common in or charac-
teristic of regional metamorphic rocks and hence were regarded as stress-
minerals, which were defined as being stable only in the presence of non-
hydrostatic stress. On the other hand, cordie rite and andalusite were considered
to be characteristic of contact metamorphic rocks and were classed in anti-stress
minerals, which were defined as being stable in the absence of non-hydrostatic
stress.
The hypothesis of stress mineI:als provided geologists with too much of a
short-cut explanation, and discouraged the then growing effort under the
leadership of Goldschmidt and Eskola to explain the mineral composition of
metamorphic rocks on the basis of theoretical chemistry. Harker reviewed
Goldschmidt's classification of the Oslo hornfelses, but had reservations about
its significance, and almost entirely ignored Eskola's works. Goldschmidt and
Eskola belonged to a generation more than twenty years younger than Harker.
434/HISTORY OF THE STUDY OF METAMORPHISM

Criticism of the hypothesis of stress minerals was inevitable for the establish-
ment of metamorphic petrology along Eskola's lines. For example, my proposal
of the inverted-Y model for the phase diagram of Al2 SiO s minerals (Miyashiro,
1949) was originally designed to defend Eskola's standpoint against Harker's
under the circumstance of those days.
In the 1950s, most of the minerals characteristic of schists were synthesized
under conditions which did not include non-hydrostatic stress. Thus, Harker's
hypothesis of stress minerals gradually waned and disappeared, though the
problem of the thermodynamic effects of non-hydrostatic stress has not been
completely solved yet.

5. INVESTIGATION OF EQUILIBRIUM MINERAL ASSEMBLAGES

Theoretical chemistry, including the thermodynamic theory of chemical equi-

librium, began to form in the later half of the nineteenth century, and came to
be used in igneous petrogenesis from the 1890s, especially through the
pioneering works of J. H. L. Vogt. The application of this theory to meta-
morphic petrology was successfully attempted first by Goldschmidt and then by
Eskola, leading to the establishment of a sound basis for the study of the mineral
composition of metamorphic rocks.
The two great masters were followed by T. Vogt, T. F. W. Barth, and Hans
Ramberg. All these investigators were born or worked in North Europe, and
constituted the golden age of North European metamorphic geology in the first
half of the twentieth century.

V. M. Goldschmidt (1888-1947)
Victor Moritz Goldschmidt was born in Zurich as a son of a chemistry professor,
but was educated in Oslo (then called Christiania) and became a professor of
crystallography, mineralogy and petrology there.
In 1911, he published a voluminous work on the pyroxene-hornfels facies
contact metamorphism of the Oslo area at the age of 23 (§ 10-1). It contained
the first successful application of the phase rule to the study of metamorphic
mineral assemblages. Then he calculated from thermochemical data the equi-
librium curve for the reaction to form wollastonite under the condition of Peo ,
::; Ps (Goldschmidt, 1912a). This curve gave the first numerical reference scale
for the temperature of metamorphism.
Later he published one of the earliest successful maps of progressive
metamorphic zones in the Trondheim area (Goldschmidt, 1915) as mentioned in
the preceding section, and a laborious study of metasomatism in the Stavanger
area (Goldschmidt, 1921). All these studies were epoch-making in the relevant
branches of metamorphic geology. However, after World War I, his interest was
turned to the distribution of minor elements in the crust, and to crystal
chemistry.
 5. INVESTIGA nON OF EQUILIBRIUM MINERAL ASSEM BLAGESj435
He had a strong influence upon Eskola, Barth and Ramberg. Barth began
his scientific career as a co-worker of Goldschmidt in the 1920s. Victor
Goldschmidt, the famous morphological crystallographer of Heidelberg, was not
blood-related to V. M. Goldschmidt.

Pentti Eskola (I 883-1964)

Eskola majored in chemistry as a student in Helsinki, and then became a
geologist, hoping to apply theoretical chemistry into geology. From 1908 to
1914, he studied Precambrian metamorphic rocks of the Orijiirvi mining area in
southwest Finland. There he found the equilibrium mineral assemblages of the
amphibolite facies. Comparison of his observation with the mineral assemblages
of the Oslo hornfelses led him to the concept of the metamorphic facies (Eskola,
1914,1915,1920; § 11-1).
Eskola became a professor in Helsinki. He studied in Goldschmidt's labora-
tory in Oslo in 1919-20 and in the Geophysical Laboratory, Washington, D.C.,
in 1921-2. In his later years, he was interested especially in the granite problem.
He was an eclectic magmatist (e.g. 1932, 1933). In the first half of the
twentieth century, the Bulletin de la Commission geologique de Finlande was
among the most important journals for metamorphic geologists of the world for
the papers published by Sederholm, Eskola and associated geologists.
It appears that the works of Goldschmidt and Eskola had not been fully
appreciated in the world prior to World War II. This is observed, for example, in
F. Y. Loewinson-Lessing's book A Historical Survey of Petrology (1955), in
which Harker's doctrine of metamorphism was reviewed, but due attention was
not paid to Goldschmidt and Eskola's.
Though Goldschmidt's paper on the Oslo hornfelses was widely circulated
throughout the world, his later papers were not. Eskola's (1920) first compre-
hensive account of the principle of metamorphic facies was published in the
Norsk Geologisk Tidsskrift, then a young journal with limited circulation. The
next systematic description of metamorphic facies (Eskola, 1929) was written in
Finnish. Their works, therefore, were not easily accessible to many geologists of
the world.
The unpopularity of Goldschmidt and Eskola at that time was partly due to
the fact that the concept of chemical equilibrium in metamorphic rocks was
entirely foreign to the then prevalent way of thinking among ordinary geologists.
The familiarity of these two pioneers with theoretical chemistry was due to their
personal circumstances, whereas most other geologists at that time were far from
this field of science. Now that the concept of chemical equilibrium has become
very popular in petrology, present-day petrologists may not understand how
difficult the concept was for many geologists in the past to appreciate.
Eskola made intensive descriptive studies on the rocks of the Orijiirvi area and
the eclogites of southwest Norway. These rocks did not show progressive
relations, and, though he had some personal acquaintance with progressive
436/HISTORY OF THE STUDY OF METAMORPHISM

mineral changes (e.g. Eskola, 1922), the elaboration of the metamorphic facies
with reference to progressive metamorphism was made mainly by T. Vogt (1927)
and Barth (1936). T. Vogt was a son of J. H. L. Vogt. The petrographic data
from the Scottish Highlands were also very useful. Barth's (1936) study on
regional metamorphism of Dutchess County, New York, played an important
role in the propagation of North European concepts of metamorphic geology to
the United States (e.g. see Balk, 1936, p. 690).
At last, Eskola wrote a comprehensive systematic description of his doctrine
of metamorphism as a part of the book Die Entstehung der Gesteine (Berlin,
1939). This book is the monument of the North European metamorphic
geology. However, its circulation outside Germany was hindered by World War II
that began shortly after its publication, and the unsold copies were burned
during the war.
The doctrine of Eskola became popular around 1950 through the textbooks
written by Turner (1948), Turner and Verhoogen (1951), Barth (1952) and
Ramberg (1952).

Hans Ramberg (1917- )

Hans Ramberg was born in Norway and was educated in Oslo by Goldschmidt
and Barth. He stayed in the University of Chicago in 1948-62. For some years
around 1950 he was the most brilliant star in metamorphic petrology. He
attempted to break through the old framework of thinking within a storm of
applause from young generations. His papers were conspicuous for their novelty.
He discussed the thermodynamics of solid solutions, the petrologic effect of
bond types, fantastic applications of thermodynamics to large-scale migration
within the earth's crust (Ramberg, 1944-5, 1948, 1949, 1951, 1952). He also
gave a clear model for the material migration and the behaviour of H2 0 in the
crust.
The origin of granites was the most loudly discussed problem of petrology in
the late 1940s. The time-honoured doctrine of Bowen was no longer attractive
to many young men. Ramberg was a most thorough and ingenious advocate of
the metamorphic origin of granite.

Glaucophane-Schists and the Diversity of Regional Metamorphism

The problem of glaucophane-schists was of vital importance for the understanding
of the diversity of regional metamorphism. Eskola (1929, 1939) displayed keen
insight in showing that glaucophane schists and associated rocks belong to a
separate metamorphic facies. This insight is especially impressive as glaucophane-
schists do not occur in North Europe. This view was accepted by Barth (I 952),
and Dutch geologists, including Brouwer, Egeler and de Roever. However,
Taliaferro (1943), Turner (1948), and Turner and Verhoogen (1951) in
California raised objections to this view, ascribing the formation of glaucophane-
 6. PRESSURE AND TEMPERATURE OF METAMORPH!SM/437

schists to the effects of peculiar pore solutions emitted from associated basic and
ultrabasic rocks. The latter view was more popular than Eskola's in the 19408
and the early 1950s. It was probably a brief discussion written by de Roever
(1955b) and the discovery of jadeite in apparently unmetarnorphosed gray-
wackes of the Franciscan (Bloxam, 1956) that marked a turning point in the
general trend of thinking in favour of Eskola's.
Even among the advocates of Eskola's view, the genetic relation of glauco-
phane-schist facies metamorphism to other facies was not clear. Many students
of glaucophane-schists presumed at that time that the whole terranes had been
metamorphosed under virtually the same P-T conditions. The best way for
establishing the glaucophane-schist facies was to demonstrate the existence of
progressive metamorphism including a zone of the facies. This was made for the
first time by Banno (1958; also Miyashiro and Banno, 1958) in the Omi area
of Japan. Shortly after this, the progressive nature of the Sanbagawa
metamorphism was demonstrated by Seki (1958) in the Kanto Mountains and
Banno in the Bessi area (§ 7A -9).
Ernst (1961) demonstrated that glaucophane can be syntheSized even at a low
pressure and high temperature on its own composition. This finding contributed
greatly to the clarification of the point that the formation of glaucophane is
controlled by the complicated combination of physical conditions and chemical
compositions, whereas jadeite and lawsonite are related more directly to the
externally controlled conditions of the glaucophane-schist facies.
The nature and the extent of the diversity of regional metamorphism have
been clarified around 1960 (Miyashiro, 1961a; § 3-3). An international project for
the cartography of metamorphic belts of the world has been undertaken since
1967 under the leadership ofR. J. Zwart of Leiden (e.g. Zwart et al. 1967).

6. PRESSURE AND TEMPERATURE OF METAMORPHISM

Nature of Metamorphic Reactions

Prior to World War II, virtually all the petrologists believed that an aqueous fluid
existed in the intergranular space during metamorphism, and that metamorphic
reactions and material migration took place through this medium (§ 2-3). Thus,
it was agreed that a high pressure on the solid phase was accompanied by the
same high fluid pressure.
In the 1940s, many of the enthusiastic advocates of the hypothesis of
granitization came to doubt the existence of a fluid phase. A lucid model of
metamorphiC rocks having no fluid phase but with intergranular H2 0 and CO 2
molecules was described by Ramberg (e.g. 1952, p. 174-82). He considered that
the rocks in great depths were recrystallized generally to lose open spaces
between mineral grains, and so they could not contain a fluid phase. Discrete
atoms, ions or molecules can migrate there by diffusion mainly through grain
438/HISTORY OF THE STUDY OF METAMORPHISM

boundaries and mosaic fissures but not by mechanical flow. A truly fluid phase
existed not in grain boundaries but as inclusions in minerals and only occasion-
ally in open spaces within rocks. Thus he distinguished between rock-pressure
and fluid-pressure.
Danielsson (1950) published a strict thermodynamic calculation of the
equilibrium curve of the wollastonite reaction, where Peo , and Ps were
distinguished. This work served well as an example of the treatment of
volatile-liberating reactions. Meanwhile, the thermodynamics of such reactions
were beautifully treated from a much more general standpoint by J. B.
Thompson (1955).
Korzhinskii (1936, 1959) and J. B. Thompson (1955, 1970) formulated the
thermodynamics of open systems, which would give a theoretical basis to the
chemical treatment of metamorphic processes. The usefulness of thermo-
chemical method was demonstrated in the 1950s, especially in combination with
synthetic methods. Detailed work was done, for example, on the stability
relation of jadeite.

Synthesis of Metamorphic Minerals

J. H. L. Vogt and A. Harker, the two great pioneers in theoretical igneous
petrology, inferred the phase relations of some pyrogenic minerals from
metallurgical data and early incomplete experimental studies. The systematic
attempt at the accurate determination of such phase relations began with the
establishment of the Geophysical Laboratory at the Carnegie Institution of
Washington in 1907. The new flow of accurate data from the Geophysical
Laboratory was so overwhelming that Harker gave up his effort in this field and
turned to metamorphic petrology.
There were some attempts at synthesis of metamorphic minerals before World
War II, but abundant data of hydrothermal syntheses began to flow out at about
1950 by the work ofH. S. Yoder and others (e.g. Yoder, 1952, 1959; Yoder and
Eugster, 1955). The breakdown equilibria of mica and other minerals were
determined in the pressure range up to a few kilobars.
From the early days of geology, various phenomena were ascribed to
the effects of high pressures within the earth. Goldschmidt and Eskola believed in
the importance of the effects of pressure as shown, for example, in their ideas on
the origin of eclogite and glaucophane schist. In the pressure range up to a few
kilobars, however, the effect of rock pressure on the phase relations between
solid minerals is small. For this reason, some of the brilliant experimental
workers in the early 1950s were critical of the supposition of the effect of high
pressure by geologists.
In 1953, Coes succeeded in making a pressure vessel to be used at high
temperatures under a pressure of a few tens of kilobars (Coes, 1953). He
synthesized a new form of Si0 2 (later named coesite) and various metamorphic
 6. PRESSURE AND TEMPERATURE OF METAMORPHISM/439
minerals which had not been artificially produced at a few kilobars. This gave
great encouragement to contemporary experimenters as regards the effectiveness
of high pressure. Shortly after it, Birch and his co-workers quantitatively
demonstrated the effects by determination of phase relations involving jadeite
and kyanite under a pressure up to a few tens of kilobars (Robertson, Birch and
McDonald, 1957; Birch and I.e Comte, 1960; Clark, Robertson and Birch,
1957).

Temperature of metamorphism
The relative temperatures of metamorphic facies were easily known from the
study of progressive metamorphism and dehydration reactions. The numerical
values of temperature were much more difficult to obtain.
Harker (1932, p. 209) stated that the recrystallization of muscovite could
take place even in cataclastic metamorphism at 'ordinary temperature' and hence
he appears to consider that the low temperature limit of the chlorite zone in the
Scottish Highlands probably approaches the 'ordinary temperature' (which
probably means the surface temperature).
This view was nearly unanimously supported to the 1950s. Thus, the chlorite
zone of regional metamorphism, for instance, was considered to represent a
temperature range of about 0-250 °c in textbooks published in the 1950s and
early 1960s. The temperatures of other grades were estimated so as to be
consistent with this. Thus, the temperature range of the amphibolite facies was
considered as about 350-600 °c, too low to cause partial melting of meta-
morphic rocks. This strengthened the belief of some geologists in the meta-
morphic origin of granites (see Miyashiro, 19720).
The first strong impact to revise the estimates of metamorphic temperatures
came from the establishment of the zeolite and prehnite-pumpellyite facies by
Coombs (1954, 1960, 1961) combined with experimental works on zeolite
syn thesis. A considerable range of temperature corresponding to these two facies
must be interposed between the ordinary surface temperature and the tempera-
ture of chlorite-zone metamorphism. Thus, the estimated temperature of the
chlorite zone increased by about 300°C.
The second strong impact came from the synthetically determined tempera-
tUre of the triple point of Al2 SiO s . Experiments in the first half of the 1960s
gave a temperature of 300°C or 390 °c for this point. However, R. C. Newton
(1966b) gave 520°C, and then Althaus (1967) and Richardson et al. (1969) gave
a temperature around 600°C for it.
According to this new scale of temperature, partial melting would take place
relatively commonly in regional metamorphism. This has shed a new light on the
problem of material migration in metamorphism.
From the beginning of the twentieth century, most of the brilliant petrol-
ogists have had a romantic vision in the application of physics and chemistry to
440/HISTORY OF THE STUDY OF METAMORPHISM

petrology. Reactions in the earth's crust should be governed by the same laws as
for those done in the laboratory. It was expected that petrology should become
the experimental physical chemistry applied to the crust. The determination of
the temperature and pressure of geologic processes was regarded even as the aim
of petrology. This vision appears to have been realized to a certain extent,
though much more remains to be done.
In this respect, Winkler's book, Petrogenesis of Metamorphic Rocks, (1965,
1967) most clearly represents the present-day atmosphere of petrology. It
represent'I the triumphal song of experimental studies.

7. TECTONIC SIGNIFICANCE OF METAMORPHISM AND THE FUTURE

OF METAMORPHIC PETROLOGY IN UNIFIED EARTH SCIENCE

Earth science has begun to change radically in the last decade. Radiometric
dating has greatly modified our views on the structure and historical develop-
ment of orogenic belts and continental crusts. Seismic and gravity studies have
clarified the structure of the crust and upper mantle. Particularly important is
the recent progress in the study of ocean floors. This has led to the hypotheses
of ocean-floor spreading and plate tectonics.
These hypotheses have aroused a new accelerated progress in the study of the
solid earth. All branches of geology and geophysics are now beginning to
collaborate in the establishment of unified earth science.
It is happy that metamorphic petrology has advanced in the last decade so
markedly as to be able to meet the new situation. We can estimate the
temperature and pressure of metamorphism with considerable reliability, and
such estimated values may be used in the construction of geologic models for
orogenic processes and ocean-floor spreading along with geophysical data.
Marked advance in the petrologiC survey in many regions has clarified the nature
and diversity of regional metamorphism and related magmatism. The tectonic
Significance of metamorphism has been well established. Thus, metamorphic
geology is playing an important role in the unified earth science in the
framework of plate tectonics (Miyashiro, 1972a, b). This book is intended to
outline metamorphic geology along this line.
References
References

Ackermann, P. B. and Walker, F. (1960). Vitrification of arkose by Karroo

dolerite near Heilbron, Orange Free State. Q. J. geol. Soc. Lond. 116,
239-54.
Agrell, S. O. (1965). Poly thermal metamorphism of limestones at Kilchoan,
Ardnamurchan. Min. Mag. 34, 1- 15.
- and Langley, J. M. (1958). The dolerite plug at Tievebulliagh near
Cushendall, Co. Antrim. Proc. R. Jr. Acad., B 59,93-127.
Albee, A. L. (1962). Relationships between the mineral association, chemical
composition and physical properties of the chlorite series. Am. Mineral. 47,
851-70.
- (1965a). A petrogenetic grid for the Fe-Mg silicates of pelitic schists. Am. J.
Sci. 263,512-36.
- (1965b). Distribution of Fe, Mg, and Mn between garnet and biotite in
natural mineral assemblages. J. Geol. 73, 155-64.
- (1968). Metamorphic zones in northern Vermont. In Studies of Appalachian
Geology: Northern and Maritime, ed. E-an Zen et al., 329-341. New York,
Interscience.
Alderman, A. R. (1936). Eclogites in the neighbourhood of Glenelg, Inverness-
shire. Q. J. geol. Soc. Lond. 92,488-530.
Althaus, E. (1967). The triple point andalusite-sillimanite-kyanite. Contr.
Mineral. Petrol. 16,29-44.
- (1969). Experimental evidence that the reaction of kyanite to form
sillimanite is at least bivariant. Am. J. Sci. 267,273-7.
Anderson, A. L. (1952). Multiple emplacement of the Idaho batholith. J. Geol.
60,255-65.
Aoki, K. and Oji, Y. (1966). Calc-alkaline volcanic rock series derived from alkali
olivine basalt magma. J. geophys. Res. 71,6127-35.
Armstrong, R. L., Jaeger, E. and Eberhardt, P. (1966). A comparison of K- Ar
and Rb- Sr ages on Alpine biotites. Earth Planet. Sci. Letters 1, 13-19.
Aronson, J. L. (1968). Regional geochronology of New Zealand. Geochim.
Cosmochim. Acta 32,669-97.
Atherton, M. P. (1964). The garnet isograd in pelitic rocks and its relation to
metamorphic facies. Am. Mineral. 49, 1331-49.
- (1965). The chemical significance of isograds. In Controls of Metamorphism,
ed. W. S. Pitcher and G. W. Flinn, 169-202. Edinburgh and London: Oliver
and Boyd.
- (1968). The variation in garnet, biotite and chlorite composition in medium
grade pelitic rocks from the Da1radian, Scotland, with particular reference to
the zonation in garnet. Contr. Mineral. Petrol. 18,347-71.
Atwater, T, (1970). Implications of plate tectonics for the Cenozoic tectonics of
western North America. Bull. geol. Soc. Am. 81,3513-36.
Aubouin, J. (1965). Geosynclines. Amsterdam, Elsevier.
444/ REFERENCES
Aumento, F. (1969). Diorites from the Mid-Atlantic Ridge at 45°N. Science,
N. Y. 165, 1112-13.
- and Loncarevic, B. D. (1969). The Mid-Atlantic Ridge near 45°N, III. Can. J.
Earth Sci. 6, 11-23.
Avias, J. (1967). Overthrust structure of the main ultrabasic New Caledonian
massives. Tectonophysics 4, 531-41.
Bailey, E. H., Blake, M. C. and Jones, D. L. (1970). On-land Mesozoic oceanic
crust in California Coast Ranges. U. S. Geol. Surv., Prof Paper 700-C
70-81.
- , Irwin, W. P. and Jones, D. L. (1964). Franciscan and related rocks, and
their significance in the geology of western California. Calif Div. Mines Geol.
Bull. 183.
Balk, R. (1936). Structural and petrologic studies in Dutchess County, New
York, Part I. Bull. geol. Soc. Am. 47,685-774.
Banno, S. (1958). Glaucophane schists and associated rocks in the Omi district,
Niigata Prefecture, Japan. Jap. J. Geol. Geogr. 29,29-44.
- (1964). Petrologic studies on Sanbagawa crystalline schists in the Bessi-Ino
district, central Sikoku, Japan. Tokyo Univ. Fac. Sci. J. Sec. II, 15203-319.
- (1966). Eclogite and eclogite facies. Jap. J. Geol. Geogr. 37,105-22.
- (1967). Effect of jadeite component on the paragenesis of eclogitic rocks.
Earth Planet. Sci. Letters 2, 249- 54.
- and Hatano, M. (1963). Zonation of metamorphic rocks in the Horokanai
area of the Kamuikotan metamorphic belt in Hokkaido (in Japanese with
English abstract). J. geol. Soc. Japan 69, 388-93.
- and Kanehira, K. (1961). Sulfide and oxide minerals in schists of the
Sanbagawa and central Abukuma metamorphic terranes. Jap. J. Geol. Geogr.
32,331-48.
- and Matsui, Y. (1965). Eclogite types and partition of Mg, Fe and Mn
between clinopyroxene and garnet. Proc. Japan A cad. 41,716-21.
- and Miller, J. A. (1965). Additional data on the age of metamorphism of the
Ryoke- Abukuma and Sanbagawa metamorphic belt, Japan. Jap. J. Geol.
Geogr. 36, 17-22.
- and Yoshino, G. (1965). Eclogite-bearing peridotite mass at Higasiakaisi-
yama in the Bessi area, central Sikoku, Japan. I. U. G. S. Upper Mantle
Symposium, New Delhi, 1964, 150-60. Copenhagen, 1965.
Barazangi, M. and Dorman, J. (1969). World seismicity maps compiled from
ESSA, Coast and Geodetic Survey, epicenter data, 1961-7. Bull. Seismol.
Soc. Am. 59,369-80.
Bard, J. P. (1969). Le Metamorphisme Regional Progressi/ des Sierras d'Aracena
en Andalousie Occidentale (Espagne). University of Montpellier.
- (1970). Composition of hornb1endes formed during the Hercynian
progressive metamorphism of the Aracena metamorphic belt (SW Spain).
Contr. Mineral. Petrol. 28, 117- 34.
Barker, F. (1961). Phase relations in cordierite-garnet-bearing Kinsman quartz
monzonite and the enclosing schist, Lovewell Mountain quadrangle, New
Hampshire. Am. Mineral. 46, 1166-76.
Barrow, G. (1893). On an intrusion of muscovite-biotite gneiss in the south-
eastern Highlands of Scotland, and its accompanying metamorphism. Q. J.
Geol. Soc. Lond. 49,330-58.
- (1912). On the geology of lower Dee-side and the southern Highland Border.
Proc. Geol. Ass. 23,274-90.
 REFERENCES /445

Barth, T. F. W. (1936). Structural and petrologic studies in Dutchess County,

New York. II. Bull. Geol. Soc. Am. 47,775-850.
- (1938). Progressive metamorphism of sparagmite rocks of southern Norway.
NorskGeol. Tidsskr. 18,54-65.
- (1945). Studies in the igneous rock complex of the Oslo region, II. Skrifter
Utgitt Det Norske Videnskaps-Akademi Oslo. I. Mat.-Naturv. Kl. 1944, No.9.
- (1952, 1962). Theoretical Petrology. 1st and 2nd eds. New York: John
Wiley.
- (1961). Abundance of the elements, areal averages and geochemical cycles.
Geochim. Cosmochim. Acta 23, 1-8.
Bateman, P. C., Clark, L. D., Huber, N. K., Moore, J. G. and Rinehart, C. D.
(1963). The Sierra Nevada batholith: A synthesis of recent work across the
central part. U.S. Geol. Surv., Prof. Paper 414-D.
- and Dodge, F. C. W. (1970). Variations of major chemical constituents
across the central Sierra Nevada batholith. Bull. geol. Soc. Am. 81,409-20.
- and Eaton, J. P. (1967). Sierra Nevada batholith. Science, N. Y. 158,
1407-17.
Bearth, P. (1959). Ueber Eklogit, Glaukophanschiefer und metamorphe Pillow-
laven. Schweiz. Mineral. Petrogr. Mitt. 39,267-86.
- (1962). Versuch einer Gliederung alpinmetamorpher Serien der Westalpen.
Schweiz. Mineral. Petrogr. Mitt. 42, 127-37.
Becke, F. (1903). Ueber Mineralbestand und Struktur der kristallinischen
Schiefer. Compte rendu IX. Session du congres geologique internat. (Vienne),
Part 2, pp. 553-70.
Billings, M. P. and White, W. S. (1950). Metamorphosed mafic dikes of the
Woodsville quadrangle, Vermont and New Hampshire. Am. Mineral., 35,
629-43.
Binns, R. A. (1964). Zones of progressive regional metamorphism in the
Willyama complex, Broken Hill district, New South Wales. J. geol. Soc.
Australia 11,283-330.
- (1965a). Hornblendes from some basic hornfelses in the New England
region, New South Wales. Min. Mag. 34, 52-65.
- (1965b). The mineralogy of metamorphosed basic rocks from the Willyama
complex. Broken Hill district, New South Wales, Parts 1 and 2. Min. Mag. 35,
306-26,561-87.
- (1967). Barroisite-bearing eclogite from Naustdal. SOlO og Fjordane,
Norway. J. Petrol. 8,349-71.
- (1969). Ferromagnesian minerals in high-grade metamorphic rocks. Geol.
Soc. Australia Spec. Paper No.2 323-32.
Birch, F. and Le Comte, P. (1960). Temperature-pressure plane for albite
composition. Am. J. Sci. 258, 209-17.
Bird, J. M. and Dewey, J. F. (1970). Lithosphere plate-continental margin
tectonics and the evolution of the Appalachian orogen. Bull. geol. Soc. Am.
81,1031-60.
Blake, M. C., Jr, Irwin, W. P. and Coleman, R. G. (1969). Blueschist-facies
metamorphism related to regional thrust faulting. Tectonophys. 8, 237-46.
Bloxam, T. W. (1956). Jadeite-bearing metagreywackes in California. Am.
Mineral 41, 488-96.
- and Allen, J. B. (1960). Glaucophane-schist, eclogite and associated rocks
from Knockormal in the Girvan-Ballantrae complex, south Ayrshire. Trans.
R. Soc. Edinb. 64,1-27.
446/ REFERENCES
Bocquet, J. (1971). Cartes de repartition de quelques mineraux du meta-
morphisme alpin dans les Alpes franco-italiennes. Eclogae geol. Helvetiae 64,
71-103.
Boettcher, A. L. and Wyllie, P. J. (1967). Revision of the calcite-aragonite
transition, with the location of a triple point between calcite I, calcite II and
aragonite. Nature, Lond. 213,792-3.
- and - (1968). Melting of granite with excess water to 30 kilobars pressure.
J. Geol. 76, 235-44.
Bogdanoff, A. A., Mouratov, M. V. and Schatsky, N. S. (1964). Tectonics of
Europe; Explllnatory Note to the International Tectonic Map of Europe,
Scale 1:2 500 000. Moscow: Nauka.
Boldizsar, T. (1964). Terrestrial heat flow in the Carpathians. J. geophys. Res.
69, 5269-275.
Bosma, W. (1964). The spots in the spotted slates of Stiege (Vosges) and
Vogtland (Saxony). Geol. en Mijnbouw 43,476-89.
Bowen, N. L. (1928). The Evolution of the Igneous Rocks. Princeton: Princeton
University Press.
- (1940). Progressive metamorphism of siliceous limestone and dolomite. J.
Geol. 48,225-74.
Bowes, D. R. (1968). The absolute time scale and subdivision of Precambrian
rocks in Scotland. Geol. Foren. Stockholm Forh. 90, 175-88.
Boyd, F. R. (1959). Hydrothermal investigations of amphiboles. In Researches
in Geochemistry, ed. P. H. Abelson, 377-96. New York: John Wiley.
- and England, J. L. (1959). Pyrope. Geophys. Lab. A. Rep. Director for
1958-195983-7.
Brothers, R. N. (1970). Lawsonite-albite schists from northernmost New
Caledonia. Contr. Mineral. Petrol. 25, 185-202.
Brown, E. H. (1967). The greenschist facies in part of eastern Otago, New
Zealand. Contr. Mineral. Petrol. 14,259-92.
- (1969). Some zoned garnets from the greenschist facies. ~m. Mineral. 54,
1662-77.
- (1971). Phase relations of biotite and stilpnomelane in the greenschist facies.
Contr. Mineral. Petrol. 31,275-99.
Brown, P. E., Miller, J. A. and Grasty, R. L. (1968). Isotope ages of late
Caledonian granitic intrusions in the British Isles. Proc. Yorkshire geol. Soc.
36,2S1-76.
- , - , Soper, N. J. and York, D. (1965). Potassium-argon age pattern of the
British Caledonides Proc. Yorkshire geol. Soc. 35, 103-38.
Brown, W. L. (1960). The crystallographic and petrologic significance of
peristerite unmixing in the acid plagioc1ases. Z. Kristallog. 113, 330-44.
- (1962). Peristerite unmixing in the plagioclases and metamorphic facies
series. Norsk Geol. Tidsskr. 42, 2. Halvbind (Feldspar Volume), 354-82.
Bryhni, I., Green, D. H., Heier, K. S. and Fyfe, W. S. (1970). On the occurrence
of eclogite in western Norway. Contr. Mineral. Petrol. 26, 12-19.
Buddington, A. F. (1939). Adirondack igneous rocks and their metamorphism.
Mem. geol. Soc. Am. 7.
- (1959). Granite emplacement with special reference to North America. Bull.
geol. Soc. Am. 70,671-747.
- , Fahey, J. and Vlisidis, A. (1955). Thermometric and petrogenetic
significance of titaniferous magnetite. Am. J. Sci. 253,497- 532.
 REFERENCES /447

Bullard, E., Everett, J. E. and Smith, A. G. (1965). The fit of the continents
around the Atlantic. In A Symposium on Continental Drift ed. P. M. S.
Blackett et al. Phil. Trans. R. Soc. Land. Ser. A 258,41- 51.
Burnham, C. W. (1959). Contact metamorphism of magnesian limestones at
Crestmore, California. Bull. geol. Soc. Am. 70,879-920.
Burns, R. G. (1966). Origin of optical pleochroism in orthopyroxenes. Min. Mag.
35,715-19.
Burst, J. P., Jr (1959). Postdiagenetic clay mineral environmental relationships
in the Gulf Coast Eocene. Proc. Sixth Nat. Cant. on Clays and Clay Minerals,
327-341. Oxford: Pergamon Press.
Butler, B. C. M. (1967). Chemical study of minerals from the Moine schists of
the Ardnamurchan area, Argyllshire, Scotland. J. Petrol. 8,233-67.
Butler, J. R. and Ragland, P. C. (1969). Petrology and chemistry of meta-igneous
rocks in the Albemarle area, North Carolina slate belt. Am. J. Sci. 267,700-26.
Cady, W. M. (1968). The lateral transition from the miogeosynclinal to the
eugeosynclinal zone in northwestern New England and Adjacent Quebec. In:
Studies of Appalachian Geology: Northern and Maritime, ed E-an Zen et al.,
151- 161. New York: Interscience.
Campbell, A. S. and Pyfe, W. S. (1965). Analcime-albite equilibria. Am. J. Sci.
263, 807-16.
Cann, J. R. (1969). Spilites from the Carlsberg Ridge, Indian Ocean. J. Petrol.
10,1-19.
- and Vine, P. J. (1966). An area on the crest of the Carlsberg Ridge:
petrography and magnetic survey. Phil. Trans. R. Soc. Land. Ser. A 259,
198-217.
Capdevila, R. (1968) Les types de metamorphisme "intermMiaires de basse
pression" dans Ie segment hercynien de Galice nord-orientale. A cad. Sci. C. R.
266, 1924-7.
Carmichael, D. M. (1970). Intersecting isograds in the Whetstone Lake area,
Ontario,J. Petrol. 11,147-81.
Carmichael, I. S. E. (1964). The petrology of Thingmuli, a Tertiary volcano in
eastern Iceland. J. Petrol. 5, 435-60.
Carpenter, A. B. (1967). Mineralogy and petrology of the system Cao-Mgo-
CO 2 - H 2 0 at Crestmore, California. A m. Mineral. 52, 1363- 41.
Chatterjee, N. D. (1961). The Alpine metamorphism in the Simplon area,
Switzerland and Italy. Geol. Rundschau 51,1-72.
- (1970). Synthesis and upper stability of paragonite. Contr. Mineral. Petrol.
27,244-57.
Chinner, G. A. (1960). Pelitic gneisses with varying ferrous/ferric ratios from
Glen Clova, Angus, Scotland. J. Petrol. 1,178-217.
- (1961). The origin of sillimanite in Glen Clova, Angus. J. Petrol. 2,312-23.
- (1962). Almandine in thermal aureoles. J. Petrol. 3,316-40.
- (1965). The kyanite isograd in Glen Clova, Angus, Scotland. Min. Mag. 34
(Tilley vol.), 132-43.
- (1966a). The distribution of pressure and temperature during Dalradian
metamorphism. Q. J. geol. Soc. Land. 122,159-86.
- (1966b). The significance of the aluminum silicates in metamorphism. Earth
Sci. Revs. 12,111-26.
- (1967). Chloritoid, and the isochemical character of Barrow's zones. J.
Petrol. 8,268-82.
448/REFERENCES

Chinner, G. A., Smith, J. V. and Knowles, C. R. (1969). Transition metal

contents of Al2 SiO s polymorphs. Am. J. Sci. 267-A (Schairer Vol.), 96-113.
Christensen, N. I. (1965). Compressional wave velocities in metamorphic rocks
at pressures to ten kilobars. J. geophys. Res. 70,6147-64.
Clark, S. P., Jr, editor (1966). Handbook of physical constants. Revised ed.
Mem. geol. Soc. America 97.
- , Robertson, E. C; and Birch, F. (1957). Experimental determination of
kyanite-sillimanite equilibrium relations at high temperatures and pressures.
Am. J. Sci. 255,628-40. '
Closs, H. (1969). Explosion seismic studies in Western Europe. In The Earth's
Crust and Upper Mantle, ed. P. J. Hart, Geophys. Monogr. 13, 178-88.
American Geophys. Union, Washington, D.C.
Coes, L., Jr, (1953). A new dense crystalline silica. Science, N. Y. 118, 131-2.
Cogne, J., Jeanette, D. and Ruhland, M. (1966). L'Ile de Croix. Servo Carte geol.
Alsace Lorraine Bull., 19, Part 1,41-95.
Coleman, R. G. (1967). Glaucophane schists from California and New Caledonia.
Tectonophys. 4,479-9g.
- (1971). Plate tectonic emplacement of upper mantle peridotites along
continental edges. J. geophys. Res. 76, 1212-22.
- and Clark, J. R. (1968). Pyroxenes in the blueschist facies of California. Am.
J. Sci. 266,43-59.
- and Lanphere, M. A. (1971). Distribution and age of high-grade blueschists,
associated eclogites, and amphibolites from Oregon and California. Bull. geol.
Soc. Am. 82,2397-412.
- and Lee, D. E. (1962). Metamorphic aragonite in the glaucophane schists of
Cazadero, California. Am. J. Sci. 260,577-95.
- and - (1963). Glaucophane-bearing metamorphic rock types of the
Cazadero area, California. J. Petrol. 4, 260- 301.
- ,- , Beatty, L. B. and Brannock, W. W. (1965). Eclogites and eclogites:
Their differences and similarities. Bull. geol. Soc. Am. 76,483-508.
- and Papike, J. J. (1968). Alkali amphiboles from the blueschists of
Cazadero, California. J. Petrol. 9,105-22.
Compston, W. and Arriens, P. A. (1968). The Precambrian geochronology of
Australia. Can. J. Earth Sci. 5,561-83.
Compton, R. R. (1955). Trondhjemite batholith near Bidwell Bar, California.
Bull. geol. Soc. Am. 66, 9-44.
- (1960). Contact metamorphism in Santa Rosa Range, Nevada. Bull. geol.
Soc. Am. 71, 1383-416.
Condie, K. (1967). Composition of the ancient North American crust. Science,
N.Y. ISS, 1013-15.
Coombs, D. S. (1954). The nature and alteration of some Triassic sediments
from Southland, New Zealand. Trans. R. Soc. N.Z. 82,65-109.
- (1960). Lower grade mineral facies in New Zealand. Internat. geol. Congr.
21st Sess. Rep. Part 13,339- 51. Copenhagen.
- (1961). Some recent work on the lower grades of metamorphism. Australian
J. Sci. 24, 203-15.
- , Ellis, A. J., Fyfe, W. S. and Tayler, A. M. (1959). The zeolite facies, with
comments on the interpretation of hydrothermal synthesis. Geochim.
Cosmochim. Acta 17,53-107.
 REFERENCES /449

Coombs, D. S., Horodyski, R. J. and Naylor, R. S. (1970). Occurrence of

prehnite-pumpellyite facies metamorphism in northern Maine. Am. J. Sci.
268,142-56.
- and Whetten, J. T. (1967). Composition of analcime from sedimentary and
burial metamorphic rocks. Bull. geol. Soc. Am. 78,269-82.
Cooper, A. F. and Lovering, J. F. (1970). Greenschist amphiboles from Haast
River, New Zealand. Contr. Mineral. Petrol. 27, 11-24.
Crawford, M. L. (1966). Composition of plagioclase and associated minerals in
some schists from Vermont, U.S.A., and South Westland, New Zealand, with
inferences about the peristerite solvus. Contr. Mineral. Petrol. 13,269-94.
Crawford, W. A. and Fyfe, W. S. (1964). Calcite-aragonite equilibrium at 100°C.
Science, N. Y. 144, 1569-70.
Crook, K. A. W. (1969). Contrasts between Atlantic and Pacific geosynclines.
. Earth Planet. Sci. Letters 5, 429- 38.
Cullen, D. J. (1967). Island arc development in the southwest Pacific.
Tectonophys. 4, 163-72.
Daly, R. A. (1917). Metamorphism and its phases. Bull. geol. Soc. Am. 28,
375-418.
Dalziel, I. W. D. (1966). A structural study of the granitic gneiss of Western
Ardgour, Argyll and Inverness-shire. Scott. J. Geol. 2, 125- 52.
Danielsson, A. (1950). Das Calcit-Wollastonitgleichgewicht. Geochim. Cosmo-
chim. Acta 1,55-69.
Davidson, C. F. (1943). The Archaean rocks of the Rodil district, South Harris,
Outer Hebrides. Trans. R. Soc. Edinb. 61,71-112.
Dawson, J. B. (1967). A review of the geology of kimberlite. In mtramafic and
Related Rocks, ed. P. J. Wyllie, 241-51. New York: John Wiley.
Dearnley, R. (1963). The Lewisian complex of South Harris with some
observations on the metamorphosed basic intrusions of the Outer Hebrides,
Scotland. Q. J. geol. Soc. Lond. 119,243-397.
Dennis, J. G. (1967). International Tectonic Dictionary-English Terminology.
American Association of Petroleum Geologists, Tulsa, Oklahoma.
den Tex, E. (1971). The facies groups and facies series of metamorphism, and
their relation to physical conditions in the earth's crust. Lithos. 4, 23-41.
de Roever, W. P. (1955a). Genesis of jadeite by low-grade metamorphism. Am. J.
Sci. 253,283-98.
- (1955b). Some remarks concerning the origin of glaucophane in the North
Berkeley Hills, California. Am. J. Sci. 253,240-4.
- (1956). Some differences between post-Paleozoic and older regional meta-
morphism. Geol. en Mijnbouw 18, 123-7.
- (1964). On the cause of the preferential distribution of certain metamorphic
minerals in orogenic belts of different age. Geol. Rundschau 54,933-43.
- and Nijhuis, H. J. (1964). Plurifacial alpine metamorphism in the eastern
Betic Cordilleras (SE Spain) with special reference to the genesis of the
glaucophane. Geol. Rundschau 53,324-36.
de Sitter, L. U. (1956). Structural Geology. New York: McGraw-Hill.
de Waard, D. (1965). A proposed subdivision of the granulite facies. Am. J. Sci.
263, 455-6l.
Dewey, J. F. (1969). Evolution of the Appalachian/Caledonian orogen. Nature,
Lond. 222,124-9.
450/ REFERENCES
Dewey, J. F. and Kay, M. (1968). Appalachian and Caledonian evidence for drift
in the north Atlantic. In The History of the Earth's Crust, a Symposium, ed.
R. A. Phinney, 161-7. Princeton: Princeton University Press.
- and Pankhurst, R. J. (1970). The evolution of the Scottish Caledonides in
relation to their isotopic age pattern. Trans. R. Soc. Edinb. 68, 361- 89.
Dickinson, W. R. (1962). Petrogenetic significance of geosynclinal andesite
volcanism along the Pacific margin of North America. Bull. geol. Soc. Am. 73,
1241-56.
- (1966). Structural relationships of San Andreas fault system, Cholame
Valley and Castle Mountain Range, California. Bull. geol. Soc. Am. 77,
707-26.
- (1968). Circum-Pacific andesite types. J. geophys. Res. 73,2261-9.
- (1970). Relations of andesites, granites, and derivative sandstones to
arc-trench tectonics. Rev. Geophys. Space Phys. 8, 813-60.
- (1971). Clastic sedimentary sequences deposited in shelf, slope, and trough
settings between magmatic arcs and associated trenches. In Pacific Geology,
ed. M. Minato, vol. 3, 15-30. Tokyo: Tsukiji Shokan.
- and Hatherton, T. (1967). Andesitic volcanism and seismicity around the
Pacific. Science, N. Y. 157,801-3.
- , Ojakangas, R. W. and Stewart, R. J. (1969). Burial metamorphism of the
late Mesozoic Great Valley sequence, Cache Creek, California. Bull. geol. Soc.
Am. 80,519-26.
Dietz, R. S. (1963). Alpine serpentines as oceanic rind fragments. Bull geol. Soc.
Am. 74, 947- 52.
- and Holden, J. C. (1966). Miogeoclines (miogeosynclines) in space and time.
J. Geol. 74,566-83.
Dobretsov, N. L. (1968). Lawsonite-glaucophane and glaucophane schists of the
U.S.S.R. and some problems of metamorphism in orogenic belts (in Russian
with English abstract). Internat. geol. Congr. 23rd Sess., Dok!. Soviet Geol.,
Problem 3,31-9.
- and Kuroda, Y. (1969). Geologic laws characterizing glaucophane meta-
morphism in northwestern part of the folded frame of Pacific Ocean.
Internat. geol. Rev. 12,1389-407.
Doll, C. G. (1961). Centennial Geologic Map of Vermont. Vermont Geol.
Survey.
Donnelly, T. W., Rogers, J. J. W., Pushkar, P. and Armstrong, R. L. (1971).
Chemical evolution of the igneous rocks of the eastern West Indies: an
investigation of thorium, uranium and potassium distributions, and lead and
strontium isotopic ratios. Mem. geol. Soc. America 130, 181-224.
Eade, K. E. and Fahrig, W. F. (1971). Geochemical evolutionary trends of
continental plates - A preliminary study of the Canadian shield. Bull. geol.
Surv. Can. 179.
- ,- and Maxwell, J. A. (1966). Composition of crystalline shield rocks and
fractionating effects of regional metamorphism. Nature, Lond. 211, 1245-9.
Egeler, C. G. (1956). The Alpine metamorphism in Corsica. Geol. Mijnbouw 18,
115-18.
Engel, A. E. J. (1963). Geologic evolution of North America. Science, N. Y. 140,
143-52.
- and Engel, C. G. (1958, 1960). Progressive metamorphism and granitization
of the major paragneiss, northwest Adirondack Mountains, New York. I, II.
Bull. geol. Soc. Am. 69, 1369-414; 71,1-58.
 REFERENCEs/451

Engel, A. E. J. and Engel, C. G. (1962a). Progressive metamorphism of

amphibolite, Northwest Adirondack Mountains, New York. In Petrologic
Studies (Buddington Vol.), ed. A. E. J. Engel et al., 37-82. Geol. Soc.
America.
- and - (1962b). Hornblendes formed during progressive metamorphism of
amphibolites, northwest Adirondack Mountains, New York. Bull. geol. Soc.
Am. 73,1499-514.
- and - (1964a). Continental accretion and the evolution of North America.
In Advancing Frontiers in Geology and Geophysics, ed. A. P. Subramaniam
and S. Balakrishna, 17- 38. Indian Geophysical Union.
- and - (1964b). Composition of basalts from the Mid-Atlantic Ridge.
Science, N. Y. 144, 1330- 3.
- ,- and Havens, R. G. (1964). Mineralogy of amphibolite interlayers in the
gneiss complex, northwest Adirondack Mountains, New York. J. Geol. 72,
131-56.
- ,- and - (1965). Chemical characteristics of oceanic basalts and the
upper mantle. Bull. geol. Soc. Am. 76,719-34.
Epstein, S. and Taylor, H. P., Jr (1967). Variation of 18 0/ 16 0 in minerals and
rocks. In Researches in Geochemistry, ed. P. H. Abelson, vol. 2,29-62. New
York: John Wiley.
Eric, J. H., Stromquist, A. A. and Swinney, C. M. (1955). Geology and mineral
deposits of the Angels Camp and Sonora quadrangles, Calaveras and
Tuolumne Counties, California. Calif. Div. Mines Spec. Rep 41.
Ernst, W. G. (1960). The stability relations of magnesioriebeckite. Geochim.
Cosmochim. Acta 19, 1-40.
- (1961). Stability relations of glaucophane. Am. J. Sci. 259,735-65.
- (1963a). Petrogenesis of glaucophane schists. J. Petrol. 4, 1-30.
- (l963b). Polymorphism in alkali amphiboles; Am. Mineral. 48,241- 60.
- (1963c). Significance of phengitic micas from low-grade schists. Am.
Mineral. 48, 1357-73.
- (1964). Petrochemical study of coexisting minerals from low-grade schists,
Eastern Sikoku, Japan. Geochim. Cosmochim. Acta 28, 1631-68.
- (1965). Mineral parageneses in Franciscan metamorphic rocks, Panoche Pass,
California. Bull. geol. Soc. Am. 76,879-914.
- (1966). Synthesis and stability relations of ferrotremolite Am. J. Sci. 264,
37-65.
- (1968). Amphiboles, Crystal Chemistry, Phase Relations and Occurrence.
New York: Springer-Verlag.
- (1970). Tectonic contact between the Franciscan melange and the Great
Valley Sequence - crustal expression of a late Mesozoic Benioff zone. J.
geophys. Res. 75, 886-90l.
- (197la). Do mineral parageneses reflect unusually high-pressure conditions
of Franciscan metamorphism? Am. J. Sci. 270,81-108.
- (1971b). Metamorphic zonations on presumably subducted lithospheric
plates from Japan, California and the Alps. Contr. Mineral. Petrol. 34,43-59.
- (1972a). Occurrence and mineralogic evolution of blueschist belts with time.
Am. J. Sci. 272,657-68.
- (1972b). Ca-amphibole paragenesis in the Shirataki district, central Shikoku,
Japan. Mem. Geol. Soc. Am. 135,73-94.
- and Seki, Y. (1967). PetrologiC comparison of the Franciscan and
Sanbagawa metamorphic terranes. Tectonophys. 4,463-78.
452/REFERENCES

Ernst, W. G., Seki, Y., Onuki, H. and Gilbert, M. C. (1970). Comparative study
of low-grade metamorphism in the California Coast Ranges and the outer
metamorphic belt of Japan. Mem. geol. Soc. Am. 124.
Eskola, P. (1914). On the petrology of the Orijiirvi region in southwestern
Finland. Bull. Comm. geol. Finlande. No. 40.
- (1915). On the relations between the chemical and mineralogical com-
position in the metamorphic rocks of the Orijiirvi region. BUll. Comm. geol.
Finlande. No. 44.
- (1920). The mineral facies of rocks. Norsk Geol. Tidsskr. 6, 143-94.
- (1921). On the eclogites of Norway. Videnskaps. Skrifter. I. Mat.-Naturv. Kl.
1921, No.8.
- (1922). On contact phenomena between gneiss and limestone in western
Massachusetts. J. Geol. 30, 265-94.
- (1929). On mineral facies. Geol. Foren. Stockholm Forh. 51, 157-72.
- (1932). On the origin of granitic magmas. Mineral. Petrogr. Mitt. 42,
455-81.
- (1933). On the differential anatexis of rocks. Bull. Comm. geol. Finlande.
No. 103, 12-25.
- (1939). Die metamorphen Gesteine. In Die Entstehung der Gesteine, by
Tom F. W. Barth, C. W. Correns and P. Eskola, 263-407. Berlin: Julius
Springer. Reprinted in 1960 and 1970.
- (1952). On the granulites of Lapland. Am. J. Sci. (Bowen Vol.), 133-71.
- (1954). A proposal for the presentation of rock analyses in ionic percentage.
Acad. Scient. Fennicae Annal. Ser. A, III. No. 38.
- (1957). On the mineral facies of charnockites. Madras Univ. J. B, 27
(Centenary No.), 101-19.
- (1963). The Precambrian of Finland. In The Precambrian, ed. K. Rankama,
vol. 1, 145-263. New York: Interscience.
Eugster, H. P. (1959). Reduction and oxidation in metamorphism. In Re-
searches in Geochemistry, ed. P. H. Abelson, 397-426. New York: John
Wiley.
- and Wones, D. R. (1962). Stability relations of the ferruginous biotite,
annite.J. Petrol. 3,82-125.
- and Yoder, H. S. (1954). Paragonite. Geophys. Lab. A. Rep. Director for
1953-1954,111-14.
- and - (1955). The join muscovite-paragonite. Geophys. Lab. A. Rep.
Director for 1954-1955, 124-6.
Evans, B. W. (1964). Fractionation of elements in the pelitic hornfelses of the
Cashel-Lough Wheelaun intrusion, Connemara, Eire. Geochim. Cosmochim.
Acta 28, 127-56.
- (1965). Application of a reaction-rate method to the breakdown equilibria
of muscovite and muscovite plus quartz. Am. J. Sci. 263,647-67.
- and Guidotti, C. V. (1966). The sillimanite-potash feldspar isograd in
western Maine, U.S.A. Contr. Mineral. Petrol. 12,25-62.
Evernden, J. F. and Kistler, R. W. (1970). Chronology of emplacement of
Mesozoic batholithic complexes in California and western Nevada. U.S. geol.
Surv. Prof. Paper 623.
- and Richards, J. R. (1962). Potassium-argon ages in eastern Australia. J.
geol. Soc. Aust. 9, 1-49.
Fabries, J. (1963). Les formations cristallines et metamorphiques du Nord-Est de
la province de Seville (Espagne). Sciences de la Terre, Mem. 4.
 REFERENCES /453
Fabries, J . (1968). Nature des hornb1endes et types de metamorphisme. Intern at.
Mineral Assoc. 5th General Meeting (1966). Papers and Froc. 204-11.
Mineral. Soc. London.
Fawcett,J. J. and Yoder, H. S., Jr (1966). Phase relationships of ch10rites in the
system MgO-Al 2 0 3-Si0 2 -H 2 O. Am. Mineral. 51,353-80.
Fedotov, S. A. (1968). On deep structure, properties of the upper mantle, and
volcanism of the KurU-Kamchatka Island Arc according to seismic data. In
The Crust and Upper Mantle of the Pacific Area, ed. L. Knopoff et al.,
Geophys. Mongr. 12, 131-9. Washington, D.C.: American Geophys. Union.
Fenner, C. N. (1929). The crystallization of basalts. Am. J. Sci. [5] 18, 225- 5 3.
Fitch, F. J. (1965). The structural unity of the reconstructed North Atlantic
continent. In A Symposium on Continental Drift, ed. P. M. S. Blackett et al.,
Phil. Trans. R. Soc. Land. Ser. A. 258, 191- 3.
Fitton, J. G. and Hughes, D. J. (1970). Volcanism and plate tectonics in the
British Ordovician. Earth Planet. Sci. Letters 8, 223-8.
Fonteille, M. and Guitard, G. (1968). Le comportement de l'eau dans 1es roches
metamorphiques catazonales des Pyrenees et son influence sur 1es assocations
minera1es des paragness d'origine pelitique, en particulier des kinzigites. Acad.
Sci. Paris C. R. 267, 1133- 5.
Foster, H. L. (1965). Geology of Ishigaki-shima, Ryukyu-retto. U.S. Geol. Surv.
Prof Paper 399-A.
Francis, G. H. (1956). Facies boundaries in pelites at the middle grades of
regional metamorphism. Geol. Mag. 93,353-68.
- (1958). Petrological studies in Glen Urquhart, Inverness-shire. British
Museum, Mineral. Bull. 1,123-62.
French, B. M. and Eugster, H. P. (1965). Experimental control of oxygen
fugacities by graphite-gas equilibriums. J. geophys. Res. 70, 1529- 39.
Frey, M. (l969a). Die Metamorphose des Keupers vom Tafeljura bis zum
Lukmanier-Gebiet. Beitr. geol. Karte der Schweiz, neue Folge, 137.
- (1969b). A mixed-layer paragonite/phengite of low-grade metamorphic
origin. Contr. Mineral. Petrol 24, 63-5.
- (1970). The step from diagenesis to metamorphism in pelitic rocks during
Alpine orogenesis. Sedimentol. 15,261-79.
Fry, N. and Fyfe, W. S. (1969). Eclogites and water pressure. Contr. Mineral.
Petrol. 24, 1- 6.
Fyfe, W. S. (1958). A further attempt to determine the vapor pressure of
brucite. Am. J. Sci. 256,729-32.
- , Turner, F. J. and Verhoogen, J. (1958). Metamorphic reactions and
metamorphic facies. Mem. geol. Soc. Am. 73.
- and Valpy, G. W. (1959). The analcime-jadeite phase boundary: some
indirect deductions. Am. J. Sci. 257,316-20.
Fyson, W. K. (1971). Fold attitudes in metamorphic rocks. Am. J. Sci. 270,
373-82.
Ganguly, J. (1968). Analysis of the stabilities of chloritoid and staurolite and
some equilibria in the system FeO-A12 03-Si02-H2 0-0 2 • Am. J. Sci. 266,
277-98.
- and Newton, R. C. (1968). Thermal stability of chloritoid at high pressure
and relatively high oxygen fugacity. J. Petrol, 9,444-66.
Garlick, G. D. and Epstein, S. (1967). Oxygen isotope ratios in coexisting
minerals of regionally metamorphosed rocks. Geochim. Cosmochim. Acta 31,
181-214.
454/ REFERENCES
Gass, I. G. (1968). Is the Troodos Massif of Cyprus a fragment of Mesozoic
ocean floor? Nature, Lond. 220, 39-42.
Geijer, P. (1963). The Precambrian of Sweden. In The Precambrian, ed. K.
Rankama, vol. 1,81-144. New York: Interscience.
Gemuts, I. (1965). Regional metamorphism in the Lamboo complex, East
Kimberley area. W. Aust. geol. Surv. A. Rep. 1964. 36-41.
Gibbs, G. V. (1966). The polymorphism of cordierite I. Am. Mineral. 51,
1068-87.
Gilluly, J. (1965). Volcanism, tectonism, and plutonism in the western United
States. Geol. Soc. Am. Spec. Paper 80.
- (1971). Plate tectonIcs and magmatic evolution. Bull. geol. Soc. Am. 82,
2383-96.
Goldschmidt, V. M. (1911). Die Kontaktmetamorphose im Kristianiagebiet.
Vidensk. Skrifter. L Mat.-Naturv. K. (1911), No. 11.
- (1912a). Die Gesetze der Gesteinsmetamorphose, mit Beispielen aus der
Geologie des siidlichen Norwegens. Vidensk. Skrifter. L Mat.-Naturv. KI.
(1912), No. 22.
- (1912b). Ueber die Anwendung der Phasenregel auf die Gesetze der
Mineralassoziation. Centralblatt Mineral. Geol. Palaeont. (1912),574-6.
- (1915). Geologisch-petrographische Studien im Hochgebirge des siidlichen
Norwegens. III. Die Kalksilikatgneise und Kalksilikatglimmerschiefer im
Trondhjem-Gebiete. Vidensk. Skrifter. L Mat.-Naturv. KI. (1915), No. 10.
- (1921). Geologisch-petrographische Studien im Hochgebirge des siidlichen
Norwegens. V. Die Injektionsmetamorphose im Stavanger-Gebiete. Vidensk.
Skrifter. L Mat.-Naturv. Kl. (1920), No.1 O.
- (1922). On the metasomatic processes in silicate rocks. Econ. Geol. 17,
105-23.
Goldsmith, J. R. and Newton, R. C. (1969). P-T-X relations in the system
CaC0 3 -MgC0 3 at high temperatures and pressures. Am. J. Sci. 267-A
(Schairer vol.), 160-90.
GonzMez-Bonorino, F. (1971). Metamorphism of the crystalline basement of
central Chile, J. Petrol. 12, 149-75.
Gorai, M. (1951). Petrological studies on plagioclase twins. Am. Mineral. 36,
884-901.
Grant, J. A. (1968). Partial melting of common rocks as a possible source of
cordierite-anthophyllite bearing assemblages. Am. J. Sci. 266,908-31.
Green, D. H. (1966). The origin of the 'eclogites' from Salt Lake Crater, Hawaii.
Earth Planet. Sci. Letters 1,414-20.
- and Lambert, I. B. (1965). Experimental crystallization of anhydrous granite
at high pressures and temperatures. J. geophys. Res. 70,5259-68.
- and Ringwood, A. E. (1967). An experimental investigation of the gabbro to
eclogite transformation and its petrological applications. Geochim. Cosmo-
chim. Acta 31,767-833.
Green, T. H. and Ringwood, A. E. (1966). Origin of the calc-alkaline igneous
rock suite. Earth Planet. Sci. Letters 1,307-16.
- and - (1968). Genesis of the calc-alkaline igneous rock suite. Contr.
Mineral. Petrol. 18,105-62.
Green, J. C. (1963). High-level metamorphism of pelitic rocks in northern New
Hampshire. Am. Mineral. 48,991-1023.
Greenwood, H. J. (1961). The system NaA1Siz 0 6 -H z O--argon: total pressure and
water pressure in metamorphism. J. geophys. Res. 66, 3923-46.
 REFERENCES /455

Greenwood, H. J. (1962). Metamorphic reactions involving two volatile

components. Geophys. Lab. A. Rep. Director for 1961-1962,82-5.
- (1963). The synthesis and stability of anthophyllite. J. Petrol. 4,317-51.
- (1967). Wollastonite: stability in H2 0- CO2 mixtures and occurrence in a
contact-metamorphic aureole near Salmo, British Columbia, Canada. Am.
Mineral. 52, 1669-80.
Griffin, J. J., Windom, H. and Goldberg, E. D. (1968). The distribution of clay
minerals in the world ocean. Deep Sea Res. 15,433-59.
Grubenmann, U. (1904-6). Die Kristallinen Schiefer. 1st ed. I (1904), II (1906);
2nd ed. (1910). Berlin: Gebriider Borntrager.
- and Niggli, P. (1924). Die Gesteinsmetamorphose. 1. Berlin: Gebriider
Borntrager.
Guidotti, C. V. (1963). Metamorphism of the pelitic schists in the Bryant Pond
quadrangle, Maine. Am. Mineral. 48,772-91.
- (1968). Prograde muscovite pseudomorphs after staurolite in the Rangeley-
Oquossoc areas, Maine. Am. Mineral. 53, 1368-76.
- (1969). A comment on 'Chemical study of minerals from the Moine schists
of the Ardnamurchan area, Argyllshire, Scotland' by B. C. M. Butler, and its
implications for the phengite problem. J. Petrol. 10, 164-70.
- (1970). The mineralogy and petrology of the transition from the lower to
upper sillimanite zone in the Oquossoc area, Maine. J. Petrol. 11,277-336.
Guitard, G. and Raguin, E. (1958). Sur la presence de gneiss it grenat et
hypersthene dans Ie massif de l'Agly (Pyrenees Orientales). Acad. Sci. (Paris)
C. R. 247,2385-8.
Halferdahl, L. B. (1961). Chloritoid: its composition, X-ray and optical pro-
perties, stability, and occurrence. J. Petrol. 2,49-135.
Hallimond, A. F. (1943). On the graphical representation of the calciferous
amphiboles. Am. Mineral. 28,65-89.
Hamilton, W. (1969a). Mesozoic California and the underflow of Pacific mantle.
Bull. geol. Soc. Am. 80,2409-30.
- (1969b). The volcanic central Andes - A modem model for the Cretaceous
batholiths and tectonics of Western North America. In Proceedings of the
Andesite Conference, ed. A. R, -McBirney, Oregon State Dept. Geol. Mineral
Industr. Bull. 65,175-184.
- and Myers, W. B. (1967). The nature of batholiths. U.S. Geol. Surv. Prof.
Paper 554-C.
Hara, I., Kanisawa, S., Kano, H., Kuroda, Y., Maruyama, T., Mitsukawa, H.,
Nureki, T., Umemura, H. and Uruno, K. (1969). Polymetamorphism in the
Abukuma Plateau with special regards to the discovery of staurolite and
kyanite (in Japanese with English abstract). Mem. geol. Soc. Japan 4, 83-
97.
Harder, H. (1956). Untersuchungen an Paragoniten und an Natrium-haltigen
Muskoviten. Heidelberger Beitr. Mineral. Petrogr. 4,227-69.
Harker, A. (1904). Tertiary Igneous Rocks of Skye. Mem. Geol. Surv.
- (1918). The present position and outlook of the study of metamorphism in
rock masses. Q. J. geol. Soc. Lond. 74, li':"lxxx.
- (1932). Metamorphism. A Study of the Transformations of Rock-Masses.
London: Methuen.
Harker, R. I. and Tuttle, O. F. (1956). Experimental data on the Pco z -T curve
for the reaction: calcite + quartz ~ wollastonite + carbon dioxide. Am. J.
Sci. 254,239-56.
456/REFERENCES

Harper, C. T. and Landis, C. A. (1967). K-Ar ages from regionally meta-

morphosed rocks, South Island, New Zealand, and some tectonic implica-
tions. Earth Planet. Sci. Letters 2, 419-29.·
Harry, W. T. (1950). Aluminium replacing silicon in some silicate lattices. Min.
Mag. 29, 142-9.
- (1958). A re-examination of Barrow's Older Granites in Glen Clova, Angus.
Trans. R. Soc. Edinb. 63,393-412.
Harte, B. and Johnson, M. R. W. (1969). Metamorphic history of Dalradian
rocks in Glens Clova, Esk and Lethnot, Angus, Scotland. Scott. J. Ceol. 5,
54-80.
Hasebe, K., Fujii, N. and Uyeda, S. (1970). Thermal processes under island arcs.
Tectonophys. 10,335-55.
Hashimoto, M. (1968). Glaucophanitic metamorphism of the Katsuyama
district, Okayama Prefecture, Japan. Tokyo Univ. Fac. Sci. Jour. Sec. 2 17,
99-162.
- (1969). A note on stilpnomelane mineralogy. Contr. Mineral. Petrol. 23,
86-8.
- , Igi, S., Seki, Y., Banno, S. and Kojima, G. (1970). Metamorphic Facies Map
o/Japan (scale 1:2000000). Geol. Surv. Japan, Kawasaki.
- and Saito, Y. (1970). Metamorphism of Paleozoic greenstones of the Tamba
Plateau, Kyoto Prefecture. J. geol. Soc. Japan 76, 1-6.
Hashizume, M., Oike, K., Asano, S., Hamaguchi, H., Okada, A., Murauchi, S.,
Shima, E. and Nogoshi, M. (1968). Crustal structure in the profile across the
northeastern part of Honshu, Japan, as derived from explosion seismic
observations, part 2. Bull. Earthquake Res. Inst. 46,607-30.
Hattori, H. (1968). Late Mesozoic to Recent tectogenesis and its bearing on the
metamorphism in New Zealand and in Japan. Ceol. Survey Japan Rep. 229.
Hay, R. L. (1966). Zeolites and zeolitic reactions in sedimentary rocks. Ceol.
Soc. Am. Spec. Paper 85.
Hayase, I. and Ishizaka, K. (1967). Rb-Sr dating on the rocks in Japan, I (in
Japanese with English abstract). J. Jap. Assoc. Mineral. Petrol. econ. Ceol. 58,
201-11.
- and Nohda, S. (1969). Geochronology on the 'oldest rock' of Japan.
Ceochem. J. 3,45- 52.
Hayashi, M. (1968). Chemical characteristics of the serpentinites in Shikoku (in
Japanese with English abstract). J. Jap. Assoc. Mineral. Petrol. econ. Ceol. 59,
60-72.
Heier, K. S. (1956). Thermometric and petrogenetic significance of titaniferous
magnetite - discussion. Am. J. Sci. 254,506-15.
- (1957). Phase relations of potash feldspar in metamorphism. J. Ceol. 65,
468-79.
- (1960). Petrology and geochemistry of high-grade metamorphic and igneous
rocks on Langoy, northern Norway. Norges Ceol. Unders. No. 207.
- (1961). The amphibolite-granulite facies transition reflected in the
mineralogy of potassium feldspars. Cursillos Cont. Fasc. 8, 131-7.
Heirtzler, J. R. (1968). Sea-floor spreading. Scient. Am. 219, No.6. 60-70.
- , Dickson, G. 0., Herron, E. M. Pitman, W. C. and Le Pichon, X. (1968).
Marine magnetic anomalies, geomagnetic field reversals, and motions of the
ocean floor and continents. J. geophys. Res. 73,2119- 36.
 REFERENCES/457

Henley, K. J. (1970). Application of the muscovite-paragonite geothermometer

to a staurolite-grade schist from Sulitjelma, north Norway. Min. Mag. 37,
693-704.
Hensen, B. J. (1971). Theoretical phase relations involving cordierite and garnet
in the system MgQ-FeO-AlzOa-SiOz. Contr. Mineral. Petrol. 33, 191-214.
Herrin, E. (1969). Regional variations of P-wave velocity in the upper mantle
beneath North America. In The Earth's Crust and Upper Mantle, ed. P. J.
Hart, Geophys. Monogr. 13, 242-246. Washington, D.C: American Geophys.
Union.
Hess, P. C. (1969). The metamorphic paragenesis of cordierite in pelitic rocks.
Contr. Mineral. Petrol. 24, 191-207.
Hess, H. H. (1962). History of ocean basins. In Petrologic Studies (Buddington
Vol.), ed. A. E. J. Engel et al., 599-620. Geol. Soc. America.
Hey, M. H. (1954). A new review of the chlorites. Min. Mag. 30,277-92.
Hietanen, A. (1956). Kyanite, andalusite, and sillimanite in the schist in Boehls
Butte quadrangle, Idaho. Am. Mineral. 41, 1-27.
- (1961). Metamorphic facies and style of folding in the Belt Series northwest
of the Idaho batholith. Bull. Comm. geol. Finlande No. 196, 73-103.
- (1962). Staurolite zone near the St. Joe River, Idaho. U.S. geol. Surv. Prof,
Paper 450-C, 69-72.
Higgins, M. W. (1971). Cataclastic rocks. U.S. geol. Surv. Prof, Paper 687.
Hill, M. L. and Dibblee, T. W. (1953). San Andreas, Garlock, and Big Pine faults,
California. Bull. geol. Soc. Am. 64,443-58.
Himmelberg, G. R. and Papike, J. J. (1969). Coexisting amphiboles from
blueschist facies metamorphic rocks. J. Petrol. 10, 102-14.
- and Phinney, W. C. (1967). Granulite-facies metamorphism, Granite Falls-
Montevideo area, Minnesota. J. Petrol. 8, 325-48.
Holdaway, M. J. (1971). Stability of andalusite and the aluminum silicate phase
diagram. Am. J. Sci. 271,97-131.
Holland, Th. H. (1900). The charnockite series, a group of Archaean
hypersthenic rocks in Peninsular India. India geol. Survey, Mem. 28,
119-249.
Hollister, L. S. (1966). Garnet zoning: an interpretation based on the Rayleigh
fractionation model. Science, N. Y. 154, 1647-51.
- (1969a). Contact metamorphism in the Kwoiek area of British Columbia:
An end member of the metamorphic process. Bull. geol. Soc. Am. 80,
2465-94.
- (1969b). Metastable paragenetic sequence of andalusite, kyanite, and
sillimanite, Kwoiek area, British Columbia. Am. J. Sci. 267,352-70.
- (1970). Origin, mechanism, and consequences of compositional sector-
zoning in staurolite. Am. Mineral. 55,742-66.
- and Bence, A. E. (1967). Staurolite: Sectoral compositional variations.
Science, N. Y. 158, 1053- 6.
Holmes, A. and Reynolds, D. L. (1947). A front of metasomatic metamorphism
in the Da1radian of Co. Donegal. Bull. Comm. geol. Finlande. No. 140,25-65.
Hoschek, G. (1967). Untersuchungen zum Stabilitatsbereich von Chloritoid und
Staurolith. Contr. Mineral. Petrol. 14, 123-62.
- (1969). The stability of staurolite and chloritoid and their significance in
metamorphism of pelitic rocks. Contr. Mineral. Petrol. 22, 208- 32.
458/ REFERENCES
Howie, R. A. (1965). The pyroxenes of metamorphic rocks. In Controls of
Metamorphism, ed. W. S. Pitcher and G. W. Flinn, 319-26. Edinburgh and
London: Oliver and Boyd.
Hsii, K. J. (1971). Franciscan melanges as a model for eugeosynclinal sedimenta-
tion and underthrusting tectonics. J. geophys. Res. 76, 1162-70.
Hsu, L. C. (1968). Selected phase relationships in the system AI-Mn-Fe-Si-O; a
model for garnet equilibria. J. Petrol. 9,40-83.
Hunahashi, M. (1957). Alpine orogenic movement in Hokkaido, Japan.
Hokkaido Univ. Fac. Sci. J. Ser. 4, 9,415-69.
Hutton, C. O. (1938). The stilpnomelane group of minerals. Min. Mag. 25,
172-206.
- (1940). Metamorphism in the Lake Wakatipu region, Western Otago, New
Zealand. N. Z. Dep. Sci. Ind. Res. Geol. Mem. No.5.
- and Turner, F. J. (1936). Metamorphic zones in northwest Otago. Trans. R.
Soc. N. Z. 65,405-6.
Hutton, J. (1795). Theory of the Earth with Proofs and Illustrations. 2 vols.
London and Edinburgh. (Reprinted 1959, H. R. Engelmann and Wheldon &
Wesley.)
Hyndman, R. D. and Hyndman, D. W. (1968). Water saturation and high
electrical conductivity in the lower continental crust. Earth Planet. Sci.
Letters 4, 427- 32.
Ichikawa, K., Ishii, K., Nakagawa, C., Suyari, K. and Yamashita, N. (1956). The
Kurosegawa zone (in Japanese with German abstract). J. geol. Soc. Japan 62,
82-103.
Iiyama, T. (1960). Recherches sur Ie role de l'eau dans la structure et Ie
polymorphisme de la cordierite. Bull. Soc. fram;aise mineral crystal. 80,
155-78.
- (1964). Etude des reactions d'echange d'ions Na-K dans la serie muscovite-
paragonite. Bull. Soc. fran(:aise mineral. crystal. 87,532-41.
Isachsen, Y. W., (ed.) (1969). Origin of anorthosite and related rocks. New York
State Museum and Science Service, Mem. 18.
backs, B. and Molnar, P. (1971). Distribution of stresses in the descending
lithosphere from a global survey of focal-mechanism solutions of mantle
earthquakes. Rev. Geophys.space Phys. 9, 103-74.
- , Oliver, J. and Sykes, L. R. (1968). Seismology and the new global
tectonics. J. geophys. Res. 73,5855-99.
Ishizaka, K. (1969). U- Th- Pb ages of zircon from the Ryoke metamorphic
terrain, Kinki district (in Japanese with English abstract). J. Jap. Assoc.
Mineral. Petrol. econ. Geol. 62, 191-7.
Iwasaki, M. (1963). Metamorphic rocks of the Kotu-Bizan area, eastern Sikoku.
Tokyo Univ. Fac. S,ci. J., Sec. 2, IS, 1-90.
- (1969). The basic metamorphic rocks at the boundary between the
Sanbagawa metamorphic belt and the Chichibu unmetamorphosed Paleozoic
sediments (in Japanese with English abstract). Mem. geol. Soc. Japan 4,
41-9.
Izawa, E. (1968). Carbonaceous matter in some metamorphic rocks in Japan. J.
geol. Soc. Japan 74,427-32.
Jaffe, H. W., Robinson, P. and Klein, C., Jr (1968). Exsolution lamellae and
optic orientation of clinoamphiboles. SCience, N. Y. 160,776-8.
 REFERENCES /459
Jakes, P. and White, A. J. R. (1972). Major and trace element abundances in
volcanic rocks of orogenic areas. Bull. geol. Soc. Am. 83,29-40.
James, H. L. (1955). Zones of regional metamorphism in the Precambrian of
northern Michigan. Bull. geol. Soc. Am. 66,1455-88.
Johannes, W. and Puhan, D. (1971). The calcite-aragonite transition, reinvesti-
gated. Contr. Mineral. Petrol. 31,28-38.
Johannsen, A. (1931). A Descriptive Petrography of the Igneous Rocks. Vol. 1.
Chicago: University of Chicago Press.
Johnson, C. G., Alvis, R. J. and Hetzler, R: L. (1960). Military geology of Yap
Islands, Caroline Islands. Intelligence Div., Office of the Engineer, Head-
quarters U.S. Army Pacific.
Johnson, M. R. W. (1962). Relations of movement and metamorphism in the
Dalradians of Banffshire. Trans geol. Soc. Edinb. 19,29-64.
- (1963). Some time relations of movement and metamorphism in the
Scottish Highlands. Geol. en Mijnbouw 42, 121-42.
Joplin, G. A. (1942, 1943). Petrological studies in the Ordovician of New South
Wales. I, II. Proc. Linn. Soc. N.S. W. 67, 156-96; 68,159-83.
- (1964). A Petrography of Australian Igneous Rocks. Sydney: Angus and
Robertson.
- (1968). A Petrography of Australian Metamorphic Rocks. Sydney: Angus
and Robertson.
Kanehira, K. (1967). Sanbagawa crystalline schists in the limori district, Kii
Peninsula. Jap. J. Geol. Geogr. 38,101-15.
- , Banno, S. and Nishida, K. (1964). Sulfide and oxide minerals in some
metamorphic terranes in Japan. Jap. J. Geol. Geogr. 35, 175-91.
Kaneko, S. (1966). Transcurrent displacement along the Median Line, South-
western Japan. New Zealand Jour. Geol. Geophys., 9,45- 59.
Kanisawa, S. (1964). Metamorphic rocks of the southwestern part of the
Kitakami Mountainland, Japan. Tohoku Univ. Sci. Rept., Ser. 3 9,
155-198.
Karakida, Y., Yamamoto, H., Miyachi, S., Oshima, T. and Inoue, T. (1969).
Characteristics and geologic situations of metamorphic rocks in Kyushu (in
Japanese with English abstract). Mem. geol. Soc. Japan 4, 3-21.
Karolus, K., Forgac, J. and Konecny, V. (1968). Neovolcanics of the West
Carpathians. Internat. geol. Congr. 23rd Session, Guide to Excursion 18AC,
Bratislava.
Katada, M. (1965). Petrography of Ryoke metamorphic rocks in northern Kiso
district, central Japan. J. Jap. Assoc. Mineral. Petrol. econ. Geol. 53, 77-90,
155-164,187-204.
- and Sumi, K. (1966). Stilpnomelane co-existing with biotite in a Ryoke
metamorphic rock. J. geol. Soc. Japan 72, 543-4.
Katsui, Y. (1961). Petrochemistry of the Quaternary volcanic rocks of Hokkaido
and surrounding areas. Hokkaido Univ. Fac. Sci. J. Ser. 4, 11,1-58.
Kawachi, Y. (1968). Large-scale overturned structure in the Sanbagawa meta-
morphic zone in central Shikoku, Japan. J. geol. Soc. Japan 74,607-16.
Kawai, N., Ito, H. and Kume, S. (1961). Deformation of the Japanese Islands as
inferred from rock magnetism. Geophys. J. 6, 124- 30.
Kawano, Y. and Ueda, Y. (1967a). Periods of the igneous activities of the
granitic rocks in Japan by K-A dating method. Tectonophys. 4, 523- 30.
460/REFERENCES

Kawano, Y. and Ueda, Y. (l967b). K-Ar datipg on the igneous rocks in Japan
(VI) (in Japanese with English abstract). I. lap. Assoc. Mineral. Petrol. econ.
Geol. 57, 177-87.
- , Yagi, K. and Aoki, K. (1961). Petrography and petrochemistry of the
volcanic rocks of Quaternary volcanoes of Northeastern Japan. Tohoku Univ.
Sci. Rep., Ser. 3, 7,1-46.
Kay, M. (1951). North American geosynclines. Mem. Geol. Soc. Am. 48.
- (1969). Continental drift in North Atlantic Ocean. In North Atlantic -
Geology and Continental Drift, a Symposium, ed. M. Kay, American Assoc.
Petroleum Geol., Mem. 12,965-73.
Keith, T. E. C., Muffler, P. and Cremer, M. (1968). Hydrothermal epidote
formed in the Salton Sea geothermal system, California. Am. Mineral. 53,
1635-44.
Kennedy, W. Q. (1948). On the significance of thermal structure in the Scottish
Highlands. Geol. Mag. 85,229- 34.
Kern, R. and Weisbrod, A. (1967). Thermodynamics for Geologists (translated
by D. McKie). San Francisco: Freeman, Cooper & Co.
Kerrick, D. M. (1968). Experiments on the upper stability limit of pyrophyllite
at 1.8 kilobars and 3.9 kilobars water pressure. Am. I. Sci. 266,204-14.
- (1970). Contact metamorphism in some areas of the Sierra Nevada,
California. Bull. geol. Soc. Am. 81,2913-38.
King, P. B. (1959). The Evolution of North America. Princeton: Princeton
University Press.
- (1969). The tectonics of North America - a discussion to accompany the
tectonic map of North America (scale 1:5000000). U.S. Geol. Surv. Prof
Paper 628.
Kisch, H. J. (1966). Zeolite facies and regional rank of bituminous coals. Geol.
Mag. 103,414-22.
Kitahara, S. and Kennedy, G. C. (1967). The calculated eqUilibrium curves for
some reactions in the system MgO-Si0 2 - H2 0 at pressures up to 30 kilobars.
Am. I. Sci. 265,211-17.
Klein, C., Jr (1968). Coexisting amphiboles. I. Petrol. 9,281-330.
- (1969). Two-amphibole assemblages in the system actinolite-hornblende-
glaucophane. Am. Mineral. 54,212-37.
Kohler, A. (1941). Die Abhiingigkeit der Plagioklasoptik vom vorangegangenen
Wiirmevethalten. Die Existenz einer Hoch- und Tieftemperaturoptik. Mineral.
Petrogr. Mitt. 53,24-49.
- (1949). Recent results of investigations on the feldspars. I. Geol. 57,
592-9.
Koizumi, M. and Roy, R. (1960). Zeolite studies. I. Synthesis and stability of
the calcium zeolites. I. Geol. 68,41-53.
Kojima, G. (Kozima, Z.) (1963). On the fundamental structure of the Sam-
bagawa crystalline schist zone (in Japanese with English abstract). Hiroshima
Univ. geol. Rep., No. 12, 173-82.
- and Suzuki, T. (1958). Rock structure and quartz fabric in a thrusting shear
zone: the Kiyomizu Tectonic Zone in Shikoku, Japan. Hiroshima Univ. I.
Sci., Series C, 2, 173-93.
Konishi, K. (1963). Pre-Miocene basement complex of Okinawa, and the
tectonic belts of the Ryukyu Islands. Kanazawa Univ. Sci. Rep. 8, 569-
602.
 REFERENCES /461

Korzhinskii, D. S. (1936). Mobility and inertness of components in meta-

somatosis. Bull. Akad. Nauk. S.S.S.R. Ser, Geol., 1936, No.1, 35-60.
- (1959). Physicochemical Basis of the Analysis of the Paragenesis of Minerals.
New York: Consultants Bureau.
- (1965). The theory of systems with perfectly mobile components and
processes of mineral formation. Am. J. Sci. 263, 193-205.
Kosminskaya, I. P., Belyaevsky, N. A. and Volvovsky, I. S. (1969). Explosion
seismology in the U.S.S.R. In The Earth's Crust and Upper Mantle, ed. P. J.
Hart, Geophys. Monogr. 13,195-208. Washington, D.C.: American Geophys.
Union.
Kretz, R. (1961). Some applications of thermodynamics to coexisting minerals
of variable composition. Examples: orthopyroxene-clinopyroxene and ortho-
pyroxene-garnet. J. Geol. 69,361-87.
Kuno, H. (1959). Origin of Cenozoic petrographic provinces of Japan and
surrounding areas. Bull. volcanol. Ser. 2, 20,37-76.
-(1960). High-alumina basalt.J. Petrol. 1,121-45.
- (1966). Lateral variation of basalt magma type across continental margins
and island arcs. Bull. volcanol. 29, 195-222.
- (1968). Origin of andesite and its bearing on the island arc structure. Bull
volcanol. 32,141-76.
- (1969). Mafic and ultramafic nodules in basaltic rocks of Hawaii. Geol. Soc.
Am. Mem. 115,189-234.
Kuroda, Y. and Miyagi, H. (1967). Metamorphic rocks of the Ishigakishima,
Ryukyu Islands, Japan (in Japanese with English abstract). Comm. Vol. for
Prof. H. Shibata, 148- 52.
Kushiro, I. (1968). Compositions of magmas formed by partial zone melting of
the Earth's upper mantle. J. geophys. Res. 73, 619- 34.
- and Aoki, K. (1968). Origin of some eclogite inclusions in kimberlite. Am.
Mineral 53, 1347-67.
- and Yoder, H. S., Jr (1966). Anorthite-forsterite and anorthite-enstatite
reactions and their bearing on the basalt-eclogite transformation. J. Petrol. 7,
337-62.
Lambert, I. B. and Heier. K. S. (1968). Geochemical investigations of deep-
seated rocks in the Australian Shield. Lithos 1,30-53.
Lambert, R. St J. (1959). The mineralogy and metamorphism of the Moine schists
of the Morar and Knoydart districts of Inverness-shire. Trans. R. Soc. Edinb.
63,553-88.
Landis, C. A. (1971). Graphitization of dispersed carbonaceous material in
metamorphic rocks. Contr. Mineral. Petrol. 30, 34-45.
- and Coombs, D. S. (1967). Metamorphic belts and orogenesis in southern
New Zealand. Tectonophys. 4, 501-18.
Lapadu-Hargues, P. (1945). Sur l'existence et la nature de l'apport chimique
dans certaines series cristallophylliennes. Bull. Soc. geol. France Ser. 5, 15,
255-310.
Lappin, M. A. (1960). On the occurrence of kyanite in the eclogites of the Selje
and A.heim district, Nordfjord. Norsk Geol. Tidsskr. 40,289-96.
Lapworth, C. (1885). The Highland controversy in British geology: its causes,
course, and consequence. Nature, Lond. 32,558-9.
Larsen, E. S., Jr (1948). Batholith and associated rocks of Corona, Elsinore, and
San Luis Rey quadrangles, southern California. Mem. Geol. Soc. America 29.
462/REFERENCES
Laves, F. (1954). The coexistence of two plagioclases in the oligoclase compo-
sition range. J. Geol. 62, 409-11.
Leake, B. E. (1965). The relationship between composition of calciferous
amphibole and grade of metamorphism. In Controls of Metamorphism, ed.
W. S. Pitcher and G. W. Flinn, 299-318. Edinburgh and London: Oliver &
Boyd.
- (1968). A catalog of analyzed calciferous and sub calciferous amphiboles
together with their nomenclature and associated minerals. Geol. Soc. America
Spec. Paper 98.
- and Skirrow, G. (1960). The pelitic hornfelses of the Cashel-Lough
Wheelaun intrusion, County Galway, Eire. J. Geol. 68,23-40.
Lee, D. E., Coleman, R. G. and Erd, R. C. (1963). Garnet types from the
Cazadero area, California. J. Petrol. 4,460-92.
Lee, W. H. K. and Uyeda, S. (1965). Review of heat flow data. In Terrestrial
Heat Flow, ed. W. H. K. Lee, Geophys. Monogr. 8, 87-190. Washington,
D.C.: American Geophys. Union.
Le Pichon, X. (1968). Sea-floor spreading and continental drift. J. geophys. Res.
73,3661-97.
Liou, J. G. (1970). Synthesis and stability relations of wairakite, CaA12-
Si 4 0 12 .2H 2 0. Contr. Mineral. Petrol. 27,259-82.
- (l97la). Synthesis and stability relations of prehnite, Ca2 A12 Si3 0 10 (OHh.
Am. Mineral. 56,507-31.
- (197lb). Stilbite-laumontite equilibrium. Contr. Mineral. Petrol. 31, 171-7.
Lipman, P. W., Prostka, H. 1. and Christiansen, R. L. (1971). Evolving
subduction zones in the western United States, as interpreted from igneous
rocks. SCience, N.Y. 174,821-5.
Lobjoit, W. M. (1964). Kyanite produced in a granitic aureole. Min. Mag. 33,
804-8.
Loewinson-Lessing, F. Y. (1954). A Historical Survey of Petrology (translated
by S. I. Tomkeieff). Edinburgh and London: Oliver and Boyd.
Loomis, A. A. (1966). Contact metamorphic reactions and processes in the Mt.
Tallac roofremnant, Sierra Nevada, California. J. Petrol. 7,221-45.
Lovering, J. F. and White, A. J. R. (1969). Granulitic and eclogitic inclusions
from basic pipes at Delegate, Australia. Contr. Mineral. Petrol. 21, 9- 52.
Lundgren, L. W., Jr (1966). Muscovite reactions and partial melting in south-
eastern Connecticut. J. Petrol. 7,421- 53.
Luth, W. C., Jahns, R. H. and Tuttle, O. F. (1964). The granite system at
pressures of 4 to 10 kilobars. J. geophys. Res. 69,759-73.
Lyell, C. (1830, 1832, 1833). Principles of Geology. 1st ed., 3 vols. London:
John Murray.
Lyons, J. B. (1955). Geology of the Hanover quadrangle, New Hampshire-
Vermont. Bull. geol. Soc. Am. 66, 105-46.
- and Morse, S.A. (1970). MgjFe partitioning in garnet and biotite from some
granitic, pelitic, and calcic rocks. A m. Mineral. 55, 231-45.
Macdonald, G. A. (1960). Dissimilarity of continental and oceanic rock types. J.
Petrol. 1,172-7.
MacGregor, M. and Wilson, G. (1939). On granitization and associated processes.
Geol. Mag. 76, 193-215.
Mall, A. P. and Singh, C. D. P. (1972). Coexisting plagioclases from basic
charnockites of India. Lithos 5, 105-14.
 REFERENCES/463
Marmo, V. (1962). On granites. Bull. Comm. geol. Finlande 201,1-77.
Mason, B. (1962). Metamorphism in the southern Alps of New Zealand. Bull.
Am. Mus. Nat. Hist. 123,213-47.
Mason, R. (1967). The field relations of the Sulitjelma gabbro, Nordland. Norsk
Geol. Tidsskr. 47,237-47.
Mather, J. D. (1970). The biotite isograd and the lower greenschist facies in the
Dalradian rocks of Scotland. J. Petrol. 11,253-75.
- and Atherton, M. P. (1965). Stilpnomelane from the Dalradian. Nature,
Lond. 207,971-2.
Matsuda, T. (1964). Island-arc features and the Japanese Islands (in Japanese
with English abstract). Chigaku-zasshi 73, 271-80.
- and Kuriyagawa, S. (1965). Lower grade metamorphism in the eastern
Akaishi Mountains, central Japan (in Japanese with English abstract). Tokyo
Univ. Earthquake Res. Inst. Bull. 43,209-35.
- , Nakamura, K. and Sugimura, A. (1967). Late Cenozoic orogeny in Japan.
Tectonophys. 4,349-66.
- and Uyeda, S. (1971). On the Pacific-type orogeny and its model -
Extension of the paired belts concept and possible origin of marginal seas.
Tectonophys. 11,5-27.
Matsumoto, Takashi (1968). A hypothesis on the origin of the late Mesozoic
volcano-plutonic association in East Asia. In Pacific Geology, ed. M. Minato
etal., vol. 1, p. 77-83. Tokyo: Tsukiji Shokan.
Matsumoto, Tatsuro, Yamaguchi, M., Yanagi, T., Matsushita, S., Hayase, I.,
Ishizaka, K., Kawano, Y. and Ueda, Y. (1968). The Precambrian problem in
younger orogenic zones: an example from Japan. Can. J. Earth Sci. 5,643-8.
McKee, B. (1962a). Widespread occurrence of jadeite, lawsonite, and glauco-
phane in central California. Am. J. Sci. 260,596-610.
- (1962b). Aragonite in the Franciscan rocks of the Pacheco Pass area,
California. Am. Mineral. 47,379-87.
McNamara, M. J. (1965). The lower greenschist facies in the Scottish Highlands.
Geol. Foren. Stockholm Forh. 87, 347-89.
Melson, W. G. (1966). Phase equilibria in calc-silicate hornfels, Lewis and Clark
County, Montana. Am. Mineral. 51,402-21.
- , Bowen, V. T., van Andel, T. H. and Siever, R. (1966). Greenstones from
the central valley of the Mid-Atlantic Ridge. Nature, Lond. 209,604- 5.
- , Jarosewich, E., Bowen, V. T. and Thompson, G. (1967). St Peter and St
Paul Rocks: a high-temperature, mantle-derived intrusion. Science, N. Y. 155,
1532-5.
- and van Andel, T. H. (1966). Metamorphism in the Mid-Atlantic Ridge,
22 ON latitude. Marine Geol. 4, 165-86.
Menard, H. W. (1967). Transitional types of crust under small ocean basins. J.
geophys. Res. 72,3061-73.
Metz, P. W. and Puhan, D. (1970). Experimentelle Untersuchung der Metamor-
phose von kieselig dolomitischen Sediment en. Contr. Mineral. Petrol. 26,
302-14.
- and Trommsdorff, V. (1968). On phase equilibria in metamorphosed
siliceous dolomites. Contr. Mineral. Petrol. 18,305-9.
- and Winkler, H. G. F. (1963). Experimentelle Gesteinsmetamorphose - VII.
Die Bildung von Talk aus kieseligem Dolomit. Geochim. Cosmochim. Acta
27,431-57.
464/REFERENCES
Metz, P. W. and Puhan, D. (1964). Experimentelle Untersuchung der Diopsid-
bildung aus Tremolit, Calcit und Quarz. Naturwissenschaften 51,460.
Miller, J. A. (1965). Geochronology and continental drift - the North Atlantic.
In A Symposium on Continental Drift, ed. P. M. S. Blackett et al., Phil. Trans.
R. Soc. Lond. Ser. A, 258, 180-191.
Minear, J. W. and ToksOz, M. N. (1970). Thermal regime of a downgoing slab
and new global tectonics. J. geophys. Res. 75, 1397-419.
Misch, P. H. (1964). Stable association wollastonite-anorthite, and other calc-
silicate assemblages in amphibolite-facies crystalline schists of Nanga Parbat,
northwest Himalayas. Beitr. Mineral. Petrogr. 10,315-56.
- (1966). Tectonic evolution of the northern Cascades of Washington State. In
Tectonic History and Mineral Deposits of the Western Cordillera, etc. Can.
Inst. Mining Metallurgy Spec. Vol. 8, 101-48.
Miyagi, K. (1964). Petrology of alkali rocks, sub-alkali rocks and high-alumina
basalt from the Green Tuff formations of Northeast Japan (in Japanese with
English abstract). J. geo!. Soc. Japan 70,72-87.
Miyashiro, A. (1949). The stability relation of kyanite, sillimanite and andalu-
site, and the physical conditions of the metamorphic processes (in Japanese
with English abstract). J. geol. Soc. Japan 55,218-23.
- (1953a). Progressive metamorphism of the calcium-rich rocks of the
Gosaisyo-Takanuki district, Abukuma Plateau, Japan. Jap. J. Geo!. Geogr. 23,
81-107.
- (1953b). Calcium-poor garnet in relation to metamorphism. Geochim.
Cosmochim. Acta 4, 179-208.
- (1957a). Cordierite-indialite relations. Am. J. Sci. 255,43-62.
- (1957b). The chemistry, optiCS and genesis of the alkali-amphiboles. Tokyo
Univ. Fac. Sci. J., Sec. 2, 11,57-83.
- (1958). Regional metamorphism of the Gosaisyo-Takanuki district in the
central Abukuma Plateau. Tokyo Univ. Fac. Sci. J., Sec. 2, 11,219-72.
- (1959). Abukuma, Ryoke, and Sanbagawa metamorphic belts (in Japanese
with English abstract).J. geol. Soc. Japan 65,624-37.
- (1960). Thermodynamics of reactions of rock-forming minerals with silica.
I-VI. Jap. J. Geol. Geogr. 31,71-8,79-84,107-11,113-20,241-6,247-52.
- (196la). Evolution of metamorphic belts. J. Petrol. 2, 277-311.
- (196lb). Metamorphism of rocks. In Constitution of the Earth, ed. C.
Tsuboi (in Japanese), 243-68. Tokyo: Iwanami-Shoten.
- (1964). Oxidation and reduction in the earth's crust with special reference
to the role of graphite. Geochim. Cosmochim. A eta 28, 717- 29.
- (1965). Metamorphic Rocks and Metamorphic Belts (in Japanese). Tokyo:
I wanami-Shoten.
- (1967a). Aspects of metamorphism in the circum-Pacific region. Tectono-
phys. 4,519-21.
- (1967b). Orogeny, regional metamorphism, and magmatism in the Japanese
Islands. Medd. fra Dansk Geol. Forening 17, 390-446.
- (1967c). Metamorphism of mafic rocks. In Basalts: Treatise on Rocks of
Basaltic Composition, ed. H. H. Hess and A. Poldervaart, vol. 2, 799-834.
New York: Interscience.
- (l972a). Pressure and temperature conditions and tectonic significance of
regional and ocean-floor metamorphism. Tectonophys. 13, 141-59.
 REFERENCEs/465

Miyashiro, A. (1972b). Metamorphism and related magmatism in plate tectonics.

Am. f. Sci. 272, 629-56.
- (1973). Paired and unpaired metamorphic belts. Tectonophysics 17,241-54.
- and Banno, S. (1958). Nature of glaucophanitic metamorphism. Am. f. Sci.
256, 97-11 O.
- and Haramura, H. (1966). Sedimentation and regional metamorphism in the
Paleozoic geosynclinal pile of Japan. Indian Geophys. Union, 1966, No.3
(Proc. Symposium on Tectonics), 45- 55.
- and Seki, Y. (1958). Enlargement of the composition field of epidote and
piemontite with rising temperature. Am. f. Sci. 256, 423- 30.
- and Shido, F. (1970). Progressive metamorphism in zeolite assemblages.
Lithos 3, 251- 60.
- and - (1973). Progressive compositional change of garnet in metapelite.
Lithos 6, 13- 20.
- ,- and Ewing, M. (1969a). Diversity and origin of abyssal tholeiite from
the Mid-Atlantic Ridge near 24 ° and 30 ° N latitude. Contr. Mineral. Petrol.
23,38-52.
- , - and - (1969b). Composition and origin of serpentinites from the
Mid-Atlantic Ridge near 24° and 300 N latitude. Contr. Mineral. Petrol. 23,
117-27.
- ,- and - (1970a). Petrologic models for the Mid-Atlantic Ridge. Deep
Sea Res. 17, 109-23.
- , - and - (1970b). Crystallization and differentiation in abyssal
tholeiites and gabbros from mid-oceanic ridges. Earth Planet. Sci. Letters 7,
361-5.
- , - and - (1971). Metamorphism in the Mid-Atlantic Ridge near 24° and
300 N. Phil. Trans. R. Soc. Lond., Ser. A, 268,589-603.
Monger, 1. W. H. and Hutchison, W. W. (1971). Metamorphic map of the
Canadian Cordillera. Geol. Surv. Can., Paper 70- 33.
Moore, J. G. (1959). The quartz-diorite boundary line in the western United
States. f. Geol. 67, 198-210.
- , Grantz, A. and Blake, M. C., Jr (1961). The quartz diorite line in
northwestern North America. U.S. Geol. Surv. Prof Paper 424-C, 87-90;
also, 1963, Prof. Paper 450-E, 89-93.
Moores, E. M. (1969). Petrology and structure of the Vourinos ophiolitic
complex of northern Greece. Geol. Soc. America, Spec. Paper 118.
- and Vine, F. J. (1971). The Troodos massif, Cyprus and other ophiolites as
oceanic crust; evolution and implications. Phil. Trans. R. Soc. London, Ser. A,
268, 443-66.
Morgan, W. J. (1968). Rises, trenches, great faults, and crustal blocks. f.
geophys. Res. 73,1959-82.
MUller, G. (1970). Kaledonische Intrusivgesteine des Stavanger-Gebietes. Contr.
Mineral. Petrol. 27,52-65.
- and Schneider, A. (1971). Chemistry and genesis of garnets in metamorphic
rocks. Contr. Mineral. Petrol. 31, 178-200.
Mueller, R. F. (1960). Compositional characteristics and equilbrium relations in
mineral assemblages of a metamorphosed iron formation. Am. f. Sci. 258,
449-97.
- (1967). Mobility of the elements in metamorphism. f. Geol. 75,565-82.
466/REFERENCES

Muffler, L. J. P. and White, D. E. (1969). Active metamorphism of Upper

Cenozoic sediments in the Salton Sea geothermal field and the Salton Trough,
southeastern California. Bull. geol. Soc. Am. 80, 157-82.
Nagasaki, H. (1966). A layered ultrabasic complex at Horoman, Hokkaido,
Japan. Tokyo Univ. Fac. Sci. J., Sec. 2, 16,313-46.
Nagata, T. (1961). Rock Magnetism. Revised ed., Tokyo: Maruzen.
Naggar, M. H. and Atherton, M. P. (1970). The composition and metamorphic
history of some aluminium silicate-bearing rocks from the aureoles of the
Donegal granites. J. Petrol. 11,549-89.
Naylor, R. S. (1968). Origin and regional relationships of the core-rocks of the
Oliverian domes. In Studies of Appalachian Geology: Northern and Maritime,
ed. E-an Zen et al. 231-40. New York: Interscience.
Newton, M. S. and Kennedy, G. C. (1968). Jadeite, analcite, nepheline, and
albite at high temperatures and pressures. Am. J. Sci. 266,728-35.
Newton, R. C. (1966a). Some calc-silicate equilibrium relations. Am. J. Sci. 264,
204-22.
- (1966b). Kyanite-anda1usite equilibrium from 700 to 800D C. Science, N.Y.
153,170-2.
- , Goldsmith, J. R. and Smith, J. V. (1969). Aragonite crystallization from
strained calcite at reduced pressures and its bearing on aragonite in low-grade
metamorphism. Contr. Mineral. Petrol. 22,335-48.
- and Kennedy, G. C. (1963). Some equilibrium reactions in the join
CaA1 2 Si 2 0 8 - H2 O. J. geophys. Res. 68,2967-83.
Niggli, E. (1960). Mineral-Zonen der Alpinen Metamorphose in der Schweizer
A1pen. Internat. geol. Congr. 21st Sess. (Norden), Rep. Part 13, 132-8.
Copenhagen.
- (1970). Alpine Metamorphose und alpine Gebirgsbildung. Fortschr. Miner.
47,16-26.
- and Niggli, C. R. (1965). Karten der Verbeitung einiger Mineralien der
alpidischen Metamorphose in den Schweizer Alpen (Stilpnomelan, Alkali-
Amphibo1, Chloritoid, Staurolith, Disthen, Sillimanit). Eclogae Geol.
Helvetiae, 58, 335- 68.
Nockolds, S. R. and Allen, R. (1953, 1954, 1956). The geochemistry of some
igneous rock series, I, II and III. Geochim. Cosmochim. Acta 4, 105-42; 5,
245-85; 9,34-77.
Nozawa, T. (1968). Isotopic ages of Hida metamorphic belt; summary and note
in 1968 (in Japanese with English abstract). J. geol. Soc. Japan 74,447-50.
Nureki, T. (1969). Geological relations of the Sangun metamorphic rocks to the
'non-metamorphic' Paleozoic formations in the Chugoku province (in
Japanese with English abstract). Mem. Geol. Soc. Japan 4, 23-39.
Offler, R. and Fleming, P. D. (1968). A synthesis of folding and metamorphism
in the Mt Lofty Ranges, South Australia. J. Geol. Soc. Aust. 15,245-66.
Oftedahl, C. (1959). Volcanic sequence and magma formation in the Oslo
region. Geol. Rundschau 48, 18-26.
O'Hara, M. J. (1961). Zones u1trabasic and basic gneiss masses in the early
Lewisian metamorphic complex at Scourie, Sutherland. J. Petrol. 2, 248-
76.
- and Mercy, E. L. P. (1963). Petrology and petrogenesis of some garneti-
ferous peridotites. Trans. R. Soc. Edinb. 65, 251- 314.
 REFERENCES/467

Okada, A. (1970). Fault topography and rate of faulting along the Median
Tectonic Line in the drainage basin of the River Yoshino, northeastern
Shikoku, Japan (in Japanese with English abstract). Geogr. Rev. Japan 43,
1-21.
Oki, Y. (1961a). Metamorphism in the northern Kiso Range, Nagano Prefecture,
Japan. Jap. J. Geol. Geogr. 32,479-96.
- (1961b). Biotite in metamorphic rocks. Jap. J. Geol. Geogr. 32,497- 506.
Okrusch, M. (1969). Die Gneishornfelse um Steinach in der Oberpfalz. Eine
phasenpetrologische Analyse. Contr. Mineral. Petrol. 22,32-72.
- (1971). Garnet- cordierite- biotite equilibria in the Steinach aureole, Bavaria.
Contr. Mineral. Petrol. 32, 1-23.
Oliver, R. L. (1969). Some observations on the distribution and nature of
granulite-facies terrains. Geol. Soc. Aust. Spec. Publ. No.2, 259-68.
Ono, A. (l969a). Geology of the Ryoke metamorphic belt in the Takato-Sioziri
area, Nagano Prefecture (in Japanese with English abstract). J. geol. Soc.
Japan 75,491-8.
- (1969b). Zoning of the metamorphic rocks in the Takato-Sioziri area,
Nagano Prefecture (in Japanese with English abstract). J. geol. Soc. Japan 75,
521-36.
- (1970). Origin of pegmatites (in Japanese with English abstract). J. geol.
Soc. Japan 76, 13-21.
Onuki, H. (1968). Almandine hornfelses from the Tono contact aureole,
Kitakami Mountainland. J. Jap. Assoc. Mineral. Petrol. econ. Geol. 59,9-20.
Orville, P. M. (1962). Alkali metasomatism and feldspars. Norsk Geol. Tidsskr.
42,2. Halvbind (Feldspar Vol.), 283-316.
- (1969). A model for metamorphic differentiation origin of thin-layered
amphibolites. Am. J. Sci. 267,64-86.
Osberg, P. H. (1968). Stratigraphy, structural geology, and metamorphism of the
Waterville-Vassalboro area, Maine. Maine geol. Surv. Bull. 20.
- (1971). An equilibrium model for Buchan-type metamorphic rocks, south-
central Maine. Am. Mineral. 56,570-86.
Osborn, E. F. (1962). Reaction series for sub alkaline igneous rocks based on
different oxygen pressure conditions. Am. Mineral. 47,211-26.
Oxburgh, E. R., and Turcotte, D. L. (1971). Origin of paired metamorphic belts
and crustal dilation in island arc regions. J. geophys. Res. 76, 1315-27.
Ozawa, A. (1968). The Green Tu//(in Japanese). Tokyo: Ratisu.
Packham, G. H. and Crook, K. A. W. (1960). The principle of diagenetic facies
and·some of its implications. J. Geol. 68,392-407.
Page, L. R. (1968). Devonian plutonic rocks in New England. In Studies 0/
Appalachian Geology: Northern and Maritime, ed. E-an Zen et al., 371-83. New
York: Interscience.
Pakiser, L. C. and Robinson, R. (1966). Composition of the continental crust as
estimated from seismic observations. In: The Earth beneath the Continents,
ed. J. S. Steinhart and T. 1. Smith, Geophys. Monogr. 10,620-6. Washington,
D. C.: American Geophys. Union.
Palm, Q. A. (1958). Les roches cristallines des Cevennes medianes a hauteur de
Largentiere, Ardeche, France. Geol. Ultraiectina 3.
Pant6, G. (1968). Cenozoic volcanism in Hungary. Internat. geol. Congr. 23rd
Sess., GUide to Excursion 40C, Budapest.
468/REFERENCES

Parras, K. (1958). On the charnockites in the light of a highly metamorphic

complex in southwestern Finland. Bull. Comm. geol. Finlande No. 181.
Peacock, M. A. (1931). Classification of igneous rock series. J. Geol. 39,54-67.
Petrushevsky, B. A. (1971). On the problem of the horizontal heterogeneity of
the earth's crust and uppermost mantle in southern Eurasia. Tectonophys. 11,
29-60.
Phillips, F. C. (1930). Some mineralogical and chemical changes induced by
progressive metamorphism in the Green Bed group of the Scottish Dalradian.
Min. Mag. 22,239-56.
Pidgeon, R. T. and Compston, W. (1965). The age and origin of the Cooma
granite and its associated metamorphic zones, New South Wales. J. Petrol. 6,
193-222.
Pitcher, W. S. (1965). The aluminium silicate polymorphs. In Controls of
Metamorphism, ed. W. S. Pitcher and G. W. Flinn, 327-41. Edinburgh and
London: Oliver and Boyd.
- and Read, H. H. (1960). The aureole of the Maine Donegal Granite. Q. J.
geol. Soc. Lond.1l6, 1-36.
- and - (1963). Contact metamorphism in relation to manner of emplace-
ment of the granites of Donegal, Ireland. J. Geol. 71,261-96.
Poldervaart, A. (1955). Chemistry of the earth's crust. Geol. Soc. America Spec.
Paper 62 (Crust of the Earth), 119-44.
Polkanov, A. A. and Gerling, E. A. (1960). The pre-Cambrian geochronology of
the Baltic shield. In ternatl. Geol. Congr. 21st Sess. (Norden) Rept., part 9,
183-91. Copenhagen.
Powers, M. C. (1959). Adjustment of clays to chemical change and the concept
of the equivalance level. Proc. 6th natn. Con/. on Clays and Clay Minerals,
309-26. Oxford: Pergamon Press.
Pugin, V. A. and Khitarov, N. 1. (1968). The system A1203-Si02 at high
temperatures and pressures (in Russian). Geokhimiya (1968), No.2, 157-65.
Ramberg, H. (1944, 1945). The thermodynamics of the earth's crust. I, II. Norsk
Geol. Tidsskr. 24, 98- 111; 25, 307-26.
- (1948). Radial diffusion and chemical stability in the gravitational field. J.
Geol. 56,448-58.
- (1949). The facies classification of rocks: a clue to the origin of quartzo-
feldspathic massifs and veins. J. Geol. 57, 18- 54.
- (1951). Remarks on the average chemical composition of granulite and
amphibolite-to-epidote amphibolite facies gneisses in west Greenland. Meddr.
Dansk Geol. Foren. 12,27-34.
- (1952). The Origin of Metamorphic and Metasomatic Rocks. Chicago:
University of Chicago Press.
Read, H. H. (1957). The Granite Controversy. London: Thomas Murby.
- (1961). Aspects of Caledonian magmatism in Britain. Liverpool Manchester
geol. J. 2,653-83.
Reed, J. J. (1958). Regional metamorphism in southeast Nelson. N. Z. Geol.
Surv. Bull. N. S. 60.
- (1964). Mylonites, cataclasites, and associated rocks along the Alpine fault,
South Island, New Zealand. N. Z. J. Geol. Geophys. 7,645-84.
Reilly, G. A. and Shaw, D. M. (1967). An estimate of the composition of part of
the Canadian shield in northwestern Ontario. Can. J. Earth Sci. 4,725-39.
 REFERENCES/469
Research Group for Explosion Seismology (1966). Explosion seismological
research in Japan. In The Earth beneath the Continents, ed. J. S. Steinhart
and T. J. Smith, Geophys. Monogr. 10, 334-48. Washington, D.C.: American
Geophys. Union.
Research Group of Peridotite Intrusion (1967). Ultrabasic rocks in Japan. J.
geol. Soc. Japan 73,543-53.
Reverdatto, V. V. (1970). Pyrometamorphism of limestones and the tempera-
ture of basaltic magmas. Lithos 3, 135-43.
- , Sharapov, V. N. and Melamed, V. G. (1970). The controls and selected
peculiarities of the origin of contact metamorphic zonation. Contr. Mineral.
Petrol. 29,310-37.
Richardson, S. W. (1968). Staurolite stability in a part of the system Fe- Al-Si-
O-H. J. Petrol. 9,468-88.
- , Gilbert, M. C. and Bell, P. M. (1969). Experimental determination of
kyanite-andalusite and andalusite-sillimanite equilibria; the aluminum silicate
triple point. Am. J. Sci. 267,259-72.
Rikitake, T. S., Miyamura, S., Tsubokawa, I., Murauchi, S., Uyeda, S., Kuno, H.
and Gorai, M. (1968). Geophysical and geological data in and around the
Japan arc. Can. J. Earth Sci. 5,1101-18.
Ringwood, A. E. (1969). Composition and evolution of the upper mantle. In
The Earth's Crust and Upper Mantle, ed. P. J. Hart, Geophys. Monogr. 13,
1-17. Washington, D. C.: American Geophys. Union.
- - and Green, D. H. (1966a). An experimental investigation of the gabbro-
eclogite transformation and some geophysical implications. Tectonophys. 3,
383-427.
- and - (1966b). Petrological nature of the stable continental crust. In The
Earth beneath the Continents, ed. J. S. Steinhart and T. J. Smith, Geophys.
Monogr. 10,611-19. Washington, D.C.: American Geophys. Union.
- and Lovering, J. F. (1970). Significance of pyroxene-ilmenite intergrowths
among kimberlite xenoliths. Earth Planet. Sci. Letters 7, 371- 5.
Robertson, E. C., Birch, F. and MaCDonald, G. J. F. (1957). Experimental
determination of jadeite stability relations to 25 000 bars. Am. J. Sci. 255,
115-35.
Robie, R. A. and Waldbaum, D. R. (1968). Thermodynamic properties of
minerals and related substances at 298.15 K (25.0 °C) and one atmosphere
(1'013 bars) pressure and at higher temperature. U.S. geol. Surv. Bull.
1259.
Robinson, P., Ross, M. and Jaffe, H. W. (1971). Composition of the antho-
phyllite-gedrite series, comparison of gedrite and hornblende, and the
anthophyllite-gedrite solvus. Am. Mineral. 56, 1005-4l.
Rocci, G. and Juteau, T. (1968). Spilite-keratophyres et ophiolites. Geol.
Mijnbouw 47, 330-9.
Rodgers, J. (1967). Chronology of tectonic movements in the Appalachian
region of eastern North America. Am. J. Sci. 265,408-27.
- (1968). The eastern edge of the North American continent during the
Cambrian and early Ordovician. In Studies of Appalachian Geology: Northern
and Maritime, ed. E-an Zen et al., 141-9. New York: Interscience.
- (1970). The Tectonics of the Appalachians. New York: John Wiley
(Interscience ).
4701 REFERENCES
Ronov, A. B. and Yaroshevsky, A. A. (1969). Chemical composition of the
earth's crust. In The Earth's Crust and Upper Mantle, ed. P. J. Hart, Geophys.
Monogr. 13,37-57. Washington, D.C.: American Geophys. Union.
Rosenbusch, H. (1877). Die Steiger Schiefer und ihre Contactzone an der
Granititen von Barr-Andlau und Hohwald. Abh. zur geol. Specialkarte von
Elsass-Lothringen 1,79-393.
- (1910). Elemente der Gesteinslehre. 3rd ed. Stuttgart: Schweizerbart'sche
VerlagSbuchhandlung.
Runcorn, S. K. (1965). Changes in the convection pattern in the earth's mantle
and continental drift: evidence for a cold origin of the earth. In A
Symposium on Continental Drift, ed. P. M. S. Blackett et al., Phil. Trans. R.
Soc. Lond. Ser. A 258, 228-51.
Saba, P. (1961). The system NaAlSi04 (nepheline)-NaAlSi3 0 s (albite)-H 2 0.
Am. Mineral. 46,859-84. .
Sander, B. (1930). Gefugekunde der Gesteine. Vienna: Springer.
Sato, S. (1968). Precambrian-Variscan polymetamorphism in the Hida massif,
basement of the Japanese Islands. Tokyo Univ. Educ. Sci. Rep. Sec. C, 10,
15-130.
Schreyer, W. (1964-6). Synthetische und natiirliche Cordierite I, II, III. Neues
Jahrb, Mineral. Abh., 102, (1964), 39-67; 103, (1965),35-79; 105, (1966),
211-44.
- (1968). A reconnaissance study of the system MgO-A12 0 3 -Si0 2 -H 2 0 at
pressures between 10 and 20 kbar. Geophys. Lab. A. Rep. Director for
1967-1968,380-92.
Schwarcz, H. P. (1969). Pre-Cretaceous sedimentation and metamorphism in the
Winchester area, northern Peninsular Ranges, California. Geol. Soc. Am. Spec.
Paper 100.
Scott, J. S. and Drever, H. I. (1953). Frictional fusion along a Himalayan thrust.
Proc. R. Soc. Edinb. Ser. B 65,121-40.
Sederholm, J. J. (1907). Om granit och gneis. Bull. Comm. geo/. Finlande No.
23.
- (1923, 1926). On migmatites and associated Pre-Cambrian rocks of south-
western Finland. I, II. Bull. Comm. geol. Finlande Nos. 58 and 77.
- (1925). The average composition of the earth's crust in Finland. Bull.
Comm. geo!. Finlande No. 70.
- (1932). On the geology of Fennoscandia with special reference to the
Pre-Cambrian. Bull. Comm. geol. Finlande No. 98.
- (1967). Selected Works: Granites and Migmatites. Edinburgh and London:
Oliver & Boyd.
Seifert, F. and Schreyer, W. (1970). Low temperature stability limit of Mg
cordierite in the range 1-7 kbar water pressure: a redetermination. Contr.
Mineral. Petrol. 27,225-38.
Seitsaarl, J. (1953). A blue-green hornblende and its genesis from the Tampere
schist belt. Bull. Comm. geol. Finlande No. 159, 83-95.
Seki, Y. (1957). Petrological study of hornfelses in the central part of the
Median Zone of Kitakami Mountainland, Iwate Prefecture. Saitama Univ. Sci.
Rep., Ser, B, 2,307-61.
- (1958). Glaucophanitic regional metamorphism in the Kanto Mountains,
central Japan. Jap. J. Geol. Geogr. 29,233-58.
 REFERENCES /471

Seki, Y. (1959). Petrological studies on the circum-Hida crystalline schists. Saitama

Univ. Sci. Rep., Ser. B, 3,209-20.
- (1960). Jadeite in Sanbagawa crystalline schists of central Japan. Am. J. Sci.
258,705-15.
- (1961a). Pumpellyite in low-grade regional metamorphism. J. Petrol. 2,
407-23.
- (1961b). Calcareous hornfe1ses in the Arisu district of the Kitakami
mountains, northeastern Japan. Jap. J. Geol. Geogr. 32,55-78.
- (1965). Behaviour of carbon dioxide in low-grade regional metamorphism
(in Japanese with English abstract). J. Jap. Assoc. Mineral. Petrol. econ. Geol.
54,1-13.
- (1966). Rock alteration in some geothermal areas of Japan (in Japanese with
English abstract). J. Jap. Assoc. Mineral. Petrol. econ. Geol. 55,212-19.
- (1969). Facies series in low-grade metamorphism. J. geol. Soc. Japan 75,
255-66.
- , Oki, Y., Matsuda, T., Mikami, K. and Okumura, K. (1969a). Metamorphism
in the Tanzawa Mountains, central Japan. J. Jap. Assoc. Mineral. Petrol. econ.
Geol. 61, 1-29,50-75.
- , Onuki, H., Oba, T., and Mori, R. (1971). Sanbagawa metamorphism in the
central Kii Peninsula. Jap. J. Geol. Geogr. 41,65-78.
- , Onuki, H., Okumura, K. and Takashima, I. (1969b). Zeolite distribution in
the Katayama geothermal area, Onikobe, Japan. J. Jap. Geol. Geogr. 40,63-79.
- and Yamasaki, M. (1957). Aluminian ferro anthophyllite from the Kitakami
Mountainland, north-eastern Japan. Am. Mineral. 42,506-20.
Semenenko, N. P. (1970). Geochronological aspects of stabilization of contin-
ental Precambrian platforms. Eclogae geol. Helve'tiae 63, 301- 31 O.
Shaw, D. M. (1965). Geochemistry of pelitic rocks. III. Major elements and
general geochemistry. Bull. geol. Soc. Am. 67,919-34.
- (1968). Radioactive elements in the Canadian Precambrian Shield and the
interior of the earth. In Origin and Distribution of the Elements, ed. L. H.
Ahrens, 855-70. Oxford: Pergamon Press.
- , Reilly, G. A., Muysson, J. R., Pattenden, G. E. and Campbell, F. E. (1967).
An estimate of the chemical composition of the Canadian Precambrian shield.
Can. J. Earth Sci. 4,829- 53.
Shibata, K. (1968). K- Ar age determinations on granitic and metamorphic rocks
in Japan. Geol. Surv. Jap. Rep. No. 227.
- , Adachi, M. and Mizutani, S. (1971). Precambrian rocks in Permian
conglomerate from central Japan. J. geol. Soc. Japan 77,507-14.
- and Igi, S. (1969). K- Ar ages of muscovite from the muscovite-quartz schist
of the Sangun metamorphic terrain in the Tari district, Tottori Prefecture,
Japan. Geol. Surv. Jap. Bull. 20,707-9.
Shido, F. (1958). Plutonic and metamorphic rocks of the Nakoso and Iritono
districts in the central Abukuma Plateau. Tokyo Univ. Fac. Sci. J., Sec. 2, 11,
131-217.
- (1959). Notes on rock-forming minerals (9). Hornblende-bearing eclogite
from Gongen-yama of Higashi- Akaisi in the Bessi district, Sikoku. J. geol.
Soc. Japan 65,701-3.
- and Miyashiro, A. (1959). Hornblendes of basic metamorphic rocks. Tokyo
Univ. Fac. Sci. J., Sec. 2, 12,85-102.
472/ REFERENCES
Shido, F., Miyashiro, A. and Ewing, M. (1971). Crystallization of abyssal
tholeiites. Contr. Mineral. Petrol. 32,251-66.
- and Seki, Y. (1959). Notes on rock-forming minerals (11). Jadeite and
hornblende from the Kamuikotan metamorphic belt. J. geol. Soc. Japan 65,
673-77.
Shieh, Y. N. and Taylor, H. P., Jr (1969). Oxygen and hydrogen isotope studies
of contact metamorphism in the Santa Rosa Range, Nevada and other areas.
Contr. Mineral. Petrol. 20,306-56.
Shimazu, Y. (1961). A geophysical study of regional metamorphism. lap. J.
Geophys. 2,135-76.
Shiraki, K. (1971). Metamorphic basement rocks of Yap Islands, western Pacific:
Possible oceanic crust beneath an island arc. Earth Planet. Sci, Letters 13,
167-74.
Sighinolfi, G. P. (1969). K-Rb ratio in high-grade metamorphism: a confirm-
ation of the hypothesis of a continual crustal evolution. Contr. Mineral.
Petrol. 21,346-56.
- (1971). Investigations into deep crustal levels: fractionating effects and
geochemical trends related to high-grade metamorphism. Geochim. Cosmo-
chim. Acta 35, 1005-21.
Simonen, A. (1953). Stratigraphy and sedimentation of the Svecofennidic, early
Archean supracrustal rocks in southwestern Finland. Bull Comm. geol.
Finlande No. 160, 1-64.
- (1960a). Plutonic rocks of the Svecofennides in Finland. Bull. Comm. geoL
Finlande No. 189.
- (1960b). Pre-Quaternary rocks in Finland. Bull. Comm. geol. Finlande No.
191.
Smith, D. G. W. (1969). Pyrometamorphism of phyllites by a dolerite plug J.
Petrol. 10,20-55.
Smith, J. R. (1958). The optical properties of heated plagioclases. Am. Mineral.
43, 1179-94.
- and Yoder, H. S., Jr (1956). Variations in X-ray powder diffraction patterns
of plagioclase feldspars. Am. Mineral. 41,632-47.
Smith, J. V. (1962). Genetic aspects of twinning in feldspars. Norsk Geol.
Tidsskr. 42,2. Halvbind (Feldspar Vol.), 244-63.
Smith, R. E. (1969). Zones of progressive regional burial metamorphism in part
of the Tasman geosyncline, eastern Australia. J. Petrol. 10,144-63.
Smithson, S. B. and Barth, Tom. F. W. (1967). The Precambrian Holum granite,
south Norway. Norsk Geol. Tidsskr. 47,21-55.
Sobolev, V. S., Dobretsov, N. L., Reverdatto, V. V., Sobolev, N. V., Jr,
Ushrakova, E. N. and Khlestov, V. V. (1967). Metamorphic facies and series
of facies in the U.S.S.R. Meddr. Dansk Geol. Forening, 17, 458-72.
Sobolev, N. V., Jr, Kuznetsova, I. K. and Zyuzin, N. I. (1968). The petrology of
grospydite xenoliths from the Zagadochnaya kimberlite pipe in Yakutia. J.
Petrol. 9,253-80.
Spry, A. (1955). Thermal metamorphism of portions of the Woolomin Group in
the Armidale district, part 2. Proc. R. Soc. N. S. W. 89, 157-70.
Steiner, A. (1953). Hydrothermal rock alteration at Wairakei, New Zealand.
Econ. Geol. 48, 1-13.
- (1955). Wairakite, the calcium analogue of analcime, a new zeolite mineral.
Min. Mag. 30,691-8.
 REFERENCES/473

Steiner, A. (1963). The rocks penetrated by drillholes in the Waiotapu thermal

area, and their hydrothermal alteration.N. Z. Dept. Sci. Indust. Res. Bull. 155,
28-34.
Stewart, 'F. H. (1946). The gabbroic complex of Belhelvie in Aberdeenshire, Q.
J. geol. Soc. Lond. 102,465-98.
Stockwell, C. H. (1964). Fourth report on structural provinces, orogenies and
time-classification of rocks of the Canadian Precambrian Shield. In Age
Determinations and Geological Studies, Part 2. Geol. Surv. Can. Paper 64-17,
1-21,26-29.
Storre, B. (1970). Stabilitlitsbedingungen Grossular-fiirender Paragenesen im
System CaO- Al2 03-Si02-C02- H2 O. Contr. Mineral. Petrol. 29, 145-62.
Stout. J. H. (1971). Four coexisting amphiboles from Telemark, Norway. Am.
Mineral. 56,212-24.
Sturt, B. A. (1962). The composition of garnets from pelitic schists in relation to
the grade of regional metamorphism. J. Petrol. 3, 181-91.
- and Harris, A. L. (1961). The metamorphic history of the Loch Tummel
area, central Perthshire, Scotland. Liverpool Manchester geol. J. 2, 689-711.
Sugi, K. (1930). On the granitic rocks of Tsukuba district and their associated
injection-rocks. Jap. J. Geol. Geogr. 8, 29-112.
Sugimura, A. (1960). Zonal arrangement of some geophysical and petrological
features in Japan and its environs. Tokyo Univ. Fac. Sci. 1., Sec. 2, 12,133-53.
- (1968). Spatial relations of basaltic magmas in island arcs. In Basalt: Treatise
on Rocks of Basaltic Composition, ed. H. H. Hess and A. Poldervaart, Vol. 2,
537-71. New York: Interscience.
- Matsuda, T., Chinzei, K. and Nakamura, K. (1963). Quantitative distribution
of late Cenozoic volcanic materials in Japan. Bull. volcanol. 26, 125-40.
Summerhayes, C. P. (1967). New Zealand region volcanism and structure.
Nature, Lond. 215,610-11.
Suppe, J. (1969). Times of metamorphism in the Franciscan terrain of the
Northern Coast Ranges, California. Bull. geol. Soc. Am. 80, 135-42.
- (1970). Offset of Late Mesozoic basement terrains by the San Andreas fault
system. Bull. geol. Soc. Am. 81,3253-8.
Sutton, J. and Watson, J. (1969). Scourian-Laxfordian relationships in the
Lewisian of northwest Scotland. Geol. Assoc. Can. Spec. Paper 5, 119-28.
Suwa, K. (1961). Petrological and geological studies on the Ryoke metamorphic
belt. Nagoya Univ. Earth Sci. J. 9,224-303.
Suzuki, M. and Kojima, G. (1970). On the association of potassium feldspar and
corundum found in the Hida metamorphic belt. J. Jap. Assoc. Mineral. Petrol.
econ. Geol. 63,266-74.
Suzuki, T. (1964). The Mikabu green rocks in Shikoku, I (in Japanese). Kochi
Univ. Res. Rep. 13 (Natural Science Section), 94-102.
- (1967). The Mikabu green rocks in Shikoku (in Japanese with English
abstract). J. geol. Soc. Japan 73,207-16.
- , Sugisaki, R. and Tanaka, T. (1971). Geosynclinal igneous activity of the
Mikabu green rocks of Ozu City, Ehime Prefecture (in Japanese with English
abstract). Mem. geol. Soc.iapan 6,121-36.
Tagiri, M. (1971). Metamorphic rocks of the Hitachi district in the southern
Abukuma Plateau. J. Jap. Assoc. Mineral. Petrol. econ. Geol. 65,77-103.
Takeuchi, H. and Uyeda, S. (1965). A possibility of present-day regional
metamorphism. Tectonophys 2, 59-68.
474/REFERENCES
Taliaferro, N. L. (1943). Franciscan-Knoxville problem. Bull. American Assoc.
Petroleum Geol. 27, 109-219.
Talwani, M., Le Pichon, X. and Ewing, M. (1965). Crustal structure of the
mid-ocean ridges 2. Computed model from gravity and seismic refraction
data. J.·geophys. Res. 70,341-52.
Taneda, S. (1965). 'Areal rock character' in Japan (in Japanese with English
abstract). Kyushu Univ. Fac. Sci. Rep., Geol. 8, 1-40.
Tayama, R. (1935). Topography, geology and coral reefs of the Yap Islands (in
Japanese). Tohoku Univ. Fac. Sci. Rep., Geol. Paleont. No. 19.
Taylor, H. P., Jr. and Coleman, R. G. (1968). 18 0/ 16 0 ratios of coexisting
minerals in glaucophane-bearing metamorphic rocks. Bull. geol. Soc. Am. 79,
1727-56.
Taylor, S. R. and White, A. J. R. (1965). Geochemistry of andesites and the
growth of continents. Nature, Lond. 208,271-3.
Tazaki, K. (1964). Alkali amphibole-bearing metamorphic rocks in the Kamuiko-
tan belt of the southwestern part of Asahikawa, central Hokkaido (in
Japanese with English abstract). Chikyu-kagaku No. 71,8-17.
Thomas, H. H. (1922). On certain xenolithic Tertiary minor intrusions in the
island of Mull. Q. J. geol. Soc. Lond. 78,229-59.
Thompson, A. B. (1970a). A note on the kaolinite-pyrophyllite equilibrium.
Am. J. Sci. 268,454-8.
- (1970b). Laumontite equilibria and the zeolite facies. Am. J. Sci. 269,
267-75.
- (1971). Analcite-albite equilibria at low temperatures. Am. J. Sci. 271,
79-92.
Thompson, J. B., Jr (1955). The thermodynamic basis for the mineral facies
concept. Am. J. Sci. 253,65-103.
- (1957). The graphical analysis of mineral assemblages in pelitic schists. Am.
Mineral. 42,842-58.
- (1970). Geochemical reaction and open systems. Geochim. Cosmochim.
Acta 34,529-51.
- and Norton, S. A. (1968). Paleozoic regional metamorphism in New England
and adjacent areas. In Studies of Appalachian Geology: Northern and
Maritime, ed. E-an Zen et al. 319-27. New York: Interscience.
- , Robinson, P., Clifford, T. N., and Trask, N. J. (1968). Nappes and gneiss
domes in west-central New England. In Studies of Appalachian Geology:
Northern and Maritime, ed. E-an Ze net al. 203-18. New York: Interscience.
Tilley, C. E. (1924). Contact metamorphism in the Comrie area of the Perthshire
Highlands. Q. J. geol. Soc. Lond. 80, 22-70.
- (1925). A preliminary survey of metamorphic zones in the southern
Highlands of Scotland. Q. J. geol. Soc. Lond. 81, 100-12.
- (1947). The dunite-mylonites of St Paul's Rocks (Atlantic). Am. J. Sci. 245,
483-91.
- (1951). A note on the progressive metamorphism of siliceous limestones
and dolomites. Geol. Mag. 88, 175-8.
Tilling, R. I., Klepper, M. R. and Obradovich, J. D. (1968). K-Ar ages and time
span of emplacement of the Boulder Batholith, Montana. Am. J. Sci. 266,
671-89.
Tobschall, H. J. (1969). Eine Subfaziesfolge der Griinschieferfazies in den
Mittleren cevennen (Dep. Ardeche) mit Pyrophyllit aufweisenden Mineral-
paragenesen. Contr. Mineral. Petrol. 24,76-91.
 REFERENCES /475
Trommsdorff, V. (1966). Progressive Metamorphose Kieseliger Karbonatgesteine
in den Zentralalpen Zwischen Bernina und Simplon. Schweiz. Mineral. Petrogr.
Mitt. 46,431-60.
Tsuboi, C., Jitsukawa, A., Tajima, H., and Okada, A. (1953- 6). Gravity
survey along the lines of precise levels throughout Japan by means of a
Worden gravimeter, Parts 1- 8. Earthquake Res. Inst. Tokyo Univ. Bull. Suppl.
4.
Tsuji, S. (1967). Petrology of the Higo metamorphic complex in the Kosa-
Hamamati area, Kumamoto Prefecture, Kyushu. lap. l. Geol. Geogr. 38,
13-25.
Turner, F. 1. (1948). Mineralogical and structural evolution of the metamorphic
rocks. Geol. Soc. Am., Mem. 30.
- and Verhoogen, J. (1951, 1960). Igneous and Metamorphic Petrology. 1st
and 2nd ed. New York: McGraw-Hill.
Tuttle, O. F. and Bowen, N. L. (1958). Origin of granite in the light of
experimental studies in the system NaAlSi30s-KAlSi30g-Si02-H20. Mem.
Geol. Soc. Am. 74.
- and Harker, R. I. (1957). Synthesis of spurrite and the reaction wollastonite
+ calcite+tspurrite + carbon dioxide. Am. l. Sci. 255,266-34.
Tyrrell, G. W. (1955). Distribution of igneous rocks in space and time. Bull.
Geol. Soc. Am. 66,405-26.
Ueda, Y. and Onuki, H. (1968). K- Ar dating on the metamorphic rocks in
Japan. I (in Japanese with English abstract). J. lap. Assoc. Mineral. Petrol.
econ. Geol. 60, 159-66.
- , Yamaoka, K., Onuki, H. and Tagiri, M. (1969). K-Ar dating on the
metamorphic rocks in Japan. II (in Japanese with English abstract). J. Jap.
Assoc. Mineral. Petrol. econ. Geol. 61, 92-9.
Ueta, S. (1961). Two contrasted regional metamorphic terrains in the Yatusiro
district, Kumamoto Prefecture (in Japanese with English abstract). J. geol.
Soc. Japan 67,526-39.
Uno, T. (1961). Metamorphic rocks of the Tukuba district, Ibaraki Prefecture
(in Japanese with English abstract). J. geol. Soc. Japan 67,228-36.
Ustiyev, Y. K. (1965). Problem of volcanism and plutonism. Internatn. Geol.
Rev. 7, 1994-2016.
Utada, M. (1965). Zonal distribution of authigenic zeolites in the Tertiary
pyroclastic rocks in Mogami district, Yamagata Prefecture. Tokyo Univ. Coll.
Gen. Educ. Sci. Paper 15, 173-216.
Uyeda, S. and Rikitake, T. (1970). Electrical conductivity anomaly and
terrestrial heat flow. J. Geomag. Geoelect. 22,75-90.
- and Vacquier, V. (1968). Geothermal and geomagnetic date in and around the
island arc of Japan. In The Crust and Upper Mantle of the Pacific Area, ed. L.
Knopoff et al., Geophys. Monogr. 12, 349-66. Washington, D.C.: American
Geophys. Union.
Vallance, T. G. (1953). Studies in the metamorphic and plutonic geology of the
Wantabadgery- Adelong-Tumbarumba district, N.S.W. I, II. Proc. Linn. Soc.
N. S. W. 78,90-121,181-96.
- (1960). Notes on metamorphic and plutonic rocks and their biotite from
the Wantabadgery-Adelong-Tumbarumba district, N.S.W. Proc. Linn. Soc.
N.S. W. 85,94-104.
- (1967). Mafic rock alteration and isochemical development of some
cordierite-anthophyllite rocks. J. Petrol. 8,84-96.
476/REFERENCES

van Bemmelen, R. W. (1949). The Geology of Indonesia. The Hague: Govern-

ment Printing Office.
Vance, J. A. (1968). Metamorphic aragonite in the prehnite-pumpellyite facies,
northwest Washington. Am. J. Sci. 266,299-315.
van der Linden, W. J. M. (1967). Structwa1 relationships in the Tasman Sea and
Southwest Pacific Ocean. N. Z. J. Geol. Geophys. 10, 1280-30r.
- (1969). Extinct mid-ocean ridges in the Tasman Sea and in the Western
Pacific. Earth Planet. Sci. Letters 6, 483-90.
van der P1as, L. (1959). Petrology of the northern Adula region, Switzerland.
Leid. Geol. Mededel. 24,411-602.
van Hise, C. R. (1904). A Treatise on Metamorphism. U.S. Geol. Survey Monogr. 47.
Ve1de, B. (1965). Phengite micas: synthesis, stability and natural occurrence.
Am. J. Sci. 263,886-913.
- (1966). Upper stability of muscovite. Am. Mineral. 51,924-9.
Vidale, R. (1969). Metasomatism in a chemical gradient and the formation of
calc-silicate bands. Am. J. Sci. 267,857-74.
Vine, F. J. (1966). Spreading of the ocean floor: new evidence. Science, N. Y.
154,1405-15.
Vinogradov, A. P. and Tugarinov, A. I. (1962). Problems of geochronology of
the pre-Cambrian in eastern Asia. Geochim. Cosmochim. Acta 26, 1283- 300.
Vogt, Th. (1927). Sulitelmafeltets geologi og petrografi, Norges Geol. Unders.
No. 121.
von Platen, H. (1965). Experimental anatexis and genesis of migmatites. In
Controls of Metamorphism, ed. W. S. Pitcher and G. W. Flinn, 203-218.
Edinburgh and London: Oliver & Boyd.
Wager, L. R. and Deer, W. A. (1939). The petrology of the Skaergaard intrusion,
Kangerdlugssuaq, East Greenland. Meddr. Gronland, 105, No.4.
Walker, G. P. L. (1960). Zeolite zones and dike distribution in relation to the
structure of the basalts of Eastern Iceland. J. Geol. 68,515-28.
Walter, L. S. (1963). Experimental studies on Bowen's decarbonation series - I
and II. Am. J. Sci. 261,488-500,773-9.
- (1965). Experimental studies on Bowen's decarbonation series - III. Am. J.
Sci. 263,64-77.
Ward, R. F. (1959). Petrology and metamorphism of the Wilmington complex,
Delaware, Pennsylvania, and Maryland. Bull. geol. Soc. Am. 70, 1425-58.
Warner, J. R., Doyle, R. G. and Hussey, A. M. (1967). Generalized map of
regional metamorphic zones. In Preliminary Geologic Map of Maine, ed. R. G.
Doyle. Maine Geol. Survey.
Washington, H. S. (1901). A chemical study of the glaucophane schists. Am. J.
Sci. 4th ser. 11,35-59.
Waters, A. C. and Campbell, C. D. (1935). Mylonites from the San Andreas fault
zone. Am. J. Sci. [5] 29,473- 503.
Weeks, W. F. (1956). A thermochemical study of equilibrium relations during
metamorphism of siliceous carbonate rocks. J. Geol. 64,245-70.
Weill, D. F. (1966). Stability relations in the Ah03-Si02 system calculated
from solubilities in the Ah03-Si02-Na3AlF6 system. Geochim.
Cosmochim. Acta 30,223-37.
- and Fyfe, W. S. (1964). A discussion of the Korzhinskii and Thompson
treatment of thermodynamic equilibrium in open systems. Geochim. Cosmo-
chim. Acta 28, 565-76.
 REFERENCES/477
Weill, D. F. and Fyfe, W. S. (1967). On equilibrium thermodynamics of open
systems and the phase rule (a reply to D. S. Korzhinskii). Geochim. Cosmochim.
Acta 31, 1167-76.
Wenk, E. (1962). Plagioklas als Indexmineral in den Zentralalpen. Schweiz.
Mineral. petrogr. Mitt. 42, 139-52.
- and Keller, F. (1969). Isograde in Amphibolitserien der Zentralalpen.
Schweiz. Mineral. Petrogr. Mitt. 49, 157-98.
White, A. J. R. (1964). Clinopyroxenes from eclogites and basic granulites. Am.
Mineral. 49, 883-8.
- (1966). Genesis of migmatites from the Palmer region of South Australia,
Chem. Geol. 1, 165-200.
- ,Compston, W. and Kleeman, A. W. (1967). The Palmer granite - a study of
a granite within a regional metamorphic environment. J. Petrol. 8, 29- 50.
- (1971). Comment on granulites. Neues Jahrbuch Mineral. Monatshefte
(1971).116-18.
White, R. W. (1966). Ultramafic inclusions in basaltic rocks from Hawaii. Contr.
Mineral. Petrol. 12,245-314.
Williams, A. F. (1932). The Genesis of the Diamond, Vols. 1,2. London: Ernest
Benn.
Wilson, J. T. (1965). A new class of faults and their bearing on continental drift.
Nature, Land. 207,343-7.
- (1966). Did the Atlantic close and then re-open? Nature, Land. 211,
676-81.
- (1968). Static or mobile earth: the current scientific revolution. In
Gondwanaland Revisited: New Evidence for Continental Drift. Proc. Am.
Phil. Soc. 112, No.5, 309-20.
Windley, B. F. (1970). Anorthosites in the early crust of the earth and on the
moon. Nature, Land. 226, 333- 5.
Winkler, H. G. F. (1965,1967). Petrogenesis of metamorphic rocks. 1st and 2nd
eds. New York: Springer-Verlag.
Wiseman, J. D. H. (1934). The central and south-west Highland epidiorites: a
study in progressive metamorphism. Q. J. geol. Soc. Land. 90, 354-417.
Wones, D. R. and Eugster, H. P. (1965). Stability of biotite: experiment, theory,
and application. Am. Mineral. 50,1228-72.
Woodland, B. G. (1963). A petrographic study of thermally metamorphosed
pelitic rocks in the Burke area, northeastern Vermont. Am. J. Sci. 261,
354-75.
Wyllie, P. J. and Tuttle, O. F. (1961). Hydrothermal melting of shales. Geol.
Mag. 98,56-66.
Wynne-Edwards, H. R. (1969). Tectonic overprinting in the Grenville province,
southwestern Quebec. Geol. Assoc. Canada Spec. Paper 5,163-82.
- and Hasan, Z. (1970). Intersecting orogenic belts across the North Atlantic.
Am. J. Sci. 268,289-308.
Yagi, K., Kawano, Y. and Aoki, K. (1963). Types of Quaternary volcanic
activity in Northeastern Japan. Bull. volcanol, 26,223-35.
Yamada, N. (1966). Nature of the Late Mesozoic igneous activities in and
around southwest Japan (in Japanese with English abstract). Chikyu-kagaku,
Nos. 85-6, 53-8.
- , Kawada, K. and Morohashi, T. (1971). The Nohi rhyolite as pyroclastic
flow deposit (in Japanese with English abstract). Chikyu-kagaku, 25,52-88.
478/REFERENCES
Yamada, N. and Nakai, Y. (1969). Geologic relations between the Nohi rhyolites
and the so-called Ryoke granites in central Japan (in Japanese with English
abstract). Mem. geoL Soc. Japan 4, 51-60.
Yamaguchi, M. and Yanagi, T. (1970). Geochronology of some metamorphic
rocks in Japan. Eciogae geol. Helvetiae 63, 371-SS.
Yamamoto, H. (1962). Plutonic and metamorphic rocks along the Usuki-
Yatsushiro tectonic line in the western part of central Kyushu. Fukuoka
Gakugei Univ. Bull. 12, 93-172.
- (1964). Metamorphic rocks of the Kiyama district, east of Kumamoto City,
Japan (in Japanese). Kyushu Univ. Fac. Sci. Rep. Geol. 7, 33-S.
Yen, T. P. (1959). Soda-amphibole-quartz schist from Taiwan. Proc. geol. Soc.
China 2, 153-6.
- (1962). The grade of metamorphism of the Tananao schist. Proc. geol. Soc.
China 5, 101-S.
- (1963). The metamorphic belts within the Tananao schist terrain of Taiwan.
Proc. geol. Soc. China 6, 72-4.
- and Rosenblum, S. (1964). Potassium-argon ages of micas from the Tananao
schist terrane of Taiwan - a preliminary report. Proc. geol. Soc. China 7,
SO-I.
Yoder, H. S., Jr (1952). The MgQ-Al203-Si02-H20 system and the related
metamorphic facies. Am. J. Sci. (Bowen Vol.), 569-627.
- (1959). Experimental studies on micas: a synthesis. Proc. 6th natn. Cont. on
Clays and Clay Minerals, 42-60. Oxford: Pergamon Press.
- and Eugster, H. P. (1954). Phlogopite synthesis and stability range.
Geochim. Cosmochim. Acta 6. 157-S5.
- and - (1955). Synthetic and natural muscovite. Geochim. Cosmochim.
Acta 8, 225-S0.
- and Tilley, C. E. (1962). Origin of basalt magmas: an experimental study of
natural and synthetic rock systems. J. Petrol. 3,342-532.
Yui, S. (1962). Notes on rock-forming minerals (24). Stilpnomelane from the
Motoyasu mine, Sikoku. J. geol. Soc. Japan 6S, 597-600.
Zemann, J. (1962). Zur Kristallchemie der Granate. Beitr. Mineral. Petrogr. S,
ISo-S.
Zen, E-an (1960). Metamorphism of Lower Paleozoic rocks in the vicinity of the
Taconic Range in west-central Vermont. Am. Mineral. 45, 129-75.
- (196Ia). Mineralogy and petrology of the system Al2 0 3- Si02- H 20 in some
pyrophyllite deposits of North Carolina. Am. Mineral. 46,52-66.
- (196Ib). The zeolite facies: an interpretation. Am. J. Sci. 259, 401-9.
- (1966). Construction of pressure-temperature diagrams for multicomponent
systems after the method of Schreinemakers - a geometric approach. U.S.
Geol. Survey Bull. 1225.
- (196S). Introduction. In Studies of Appalachian Geology: Northern and
Maritime, ed. E-an Zen et al., 1-5. New York: Interscience.
- and Albee, A. L. (1964). Coexisting muscovite and paragonite in pelitic
schists. Am. Mineral. 49,904-25.
Zwart, H. J. (1962). On the determination of polymorphic mineral associations,
and its application to the Bosost area (central Pyrenees). Geol. Rundschau 52,
3S-65.
- (1963). Some examples of the relations between deformation and meta-
morphism from the central Pyrenees. Geol. Mi;nbouw 42, 143- 54.
 REFERENCES /479

Zwart, H. J. (l967a). The duality of orogenic belts. Geol Mijnbouw 46, 283- 309.
- (1967b). Orogenesis and metamorphic facies series in Europe. Meddr. Dansk.
Geol. Forening 17, 504-16.
- (1969). Metamorphic facies series in the European orogenic belts and their
bearing on the causes of orogeny. Geol. Assoc. Can. Spec. Paper S, 7-16.
- , Corvalan, J., James, H. L., Miyashiro, A., Saggerson, E. P., Sobolev, V. S.,
Subramaniam, A. P. and Vallance, T. G. (1967). A scheme of metamorphic
facies for the cartographic representation of regional metamorphic belts.
lUGS Geological Newsletter (1967), No.2, 57-72.
Index

Abukuma Plateau, Japan: Almandine zone (in the Scottish Highlands)

claim of polymetamorphism 174, 368 58,185-7
geology 366-8 Alpine fault, New Zealand:
limestone 268 cataclastic rocks along 424
maps 83, 237, 360 geology 382-4
metabasites 236-8 map 383
metapelites 173-4 Alpine schists, New Zealand 384
progressive changes (diagram) 76 Alpinotype orogeny 94
relation to Ryoke belt 368 Alps, Western Europe:
Abyssal tholeiite 416 geology 343-5
Acadian orogeny (in the Appalachians) limestone 269-70
330,345 maps 190,343,344
ACF diagram (of Eskola) 315 metapelites 189-91
for eclogite facies 315 orogeny 93
for granulite facies 314 paired metamorphic belts 344-5,393
for low-pressure metamorphism 301 Aluminium-silicate minerals (Al, SiDs):
meaning 124-7 occurrence 115,223-5
for medium-pressure metamorphism 306 phase diagrams 40,72
and metamorphic facies 293 stability relations 39,41,86,89
Acidic rocks, definition 29 Al, SiD s (see aluminium-silicate minerals)
Actinolite (amphibole) 33, 114, 117,251-7 Amesite, 206
Actinolite-calcic plagioclase zone (or horn- Amphibole:
fels facies) 117-18,296,300 anthophyllite 258,261
Actinolite-greenschist facies 304 calcic 61,251-7
Adelaide geosyncline, Australia 380 (map) cummingtonite 258
Adirondack Mountains, New York State gedrite 258
243-4,327 glaucophane 262-3
AFM diagram: subcalcic 251,257
meaning 130 Amphibolite:
and the paragenesis of garnet 213 definition 32, 33
for progressive metamorphism 228-30 derived from calcareous sediments, 267
Akaishi (Akaisi) Mountains Japan: derived from igneous rocks 32
auto cannibalism 66 Amphibolite facies:
history and metamorphism 140-2 ACF diagrams 127,128,301,306
maps 141,360,364 A'KFdiagram 130
Akermanite 291 in contact metamorphism 116
AKF diagram (=A' KF diagram) 129 definition 68
A'KFdiagram (of Eskola): in high-pressure metamorphism 309
meaning, 129 in low-pressure metamorphism 300
and metamorphic facies 293 in medium-pressure metamorphism
Alkali rock series 102 305-6
Almandine (garnet): and the name amphibolite 33
in metabasites 259-60 pressure and temperature 90
in metapelites 211-21 Analcime 150, 159-62
482/ INDEX
Andalusite 3940, 72, 115, 199, 223-25, Be1omorides, Baltic shield 338-9
277 (see aluminium-silicate minerals) Benioff zone 350, 401
Andalusite-sillimanite type (= low-pressure Bessi (Besshi)-Ino area, Japan:
type) 74 geology 370
Andesite 101 limestone 270
Anglesey area, Wales 110 map 192,360
Angly area, Pyrenees 176 metabasites 244-5
Ankerite 114,242, 253 metapelites 191-3
Anorthoclase 303 progressive changes (diagram) 78
Anorthosite 107, 329 Betic Cordillera, Spain 3434,344 (map)
Anthophyllite (amphibole) 258,261 Biotite:
Antigorite 206 breakdown 227
Anti-stress minerals (of Harker) 433 equilibrium with garnet 217
Appalachians, North America: in metabasites 258-9
geology 330-2 in metapelites 208
granitic rocks 97 Biotite isograd:
maps 181,328 reactions at 208
metabasites 241-2 in the Scottish Highlands 58
metapelites 92, 180-3, 188-9 Biotite zone (in the Scottish Highlands) 58,
prehnite-pumpellyite facies 140,182 185-6
Aqueous fluid 21,63 blast (suffix) 28
Aracena area, Spain 177-8, 177 (map) Blasto (prefix) 28
Aragonite 89, 153,228-30 Blueschist facies (=glaucophane-schist facies)
Arc-trench gap 403 69
Arisu area, Japan: Bosost area, Pyrenees 176,179-80
actinolite-calcic plagioclase zone 117 Broken Hill area, Australia 243,379
limestone 290 Brucite 266
metabasites 289 Buchan region, Scotland 185, 186 (map)
metapelites 280 341,433
Asiatic Continent 354-5,354 (map) Buchite 27, 287
Atlantic region 345-7,346 (map), 391-2 Buller geosyncline, New Zealand 384
Augite (pyroxene) 260 Burial metamorphism:
Australia 379-81,380 (map) definition 27
and metamorphic facies 70-1,139
Ballantrae area, Scotland 186 (map), 342
Baltic shield 96,33840,339 (map) Calc-alkali rock series 101-2
Bar 38,39 Calcareous sediments 29
Baric types (of metamorphism): Calcic amphibole 61,251-7 (see amphibole)
in contact metamorphism 115 Calcite:
definition 71-3 and aragonite 228-31
and facies series 297 and carbon dioxide pressure 114,253
and geologic age 110,391 in limestone 270
and geothermal gradient 85 Caledonides, Europe 33842, 339 (map),
and igneous rock association 93-6 345
and low-temperature metamorphism 113 Calorie 38,39
and metabasites 235 Canadian shield, North America 327-30,
and metapelites 168 328 (map)
Barr-Andlau area, France 277 Canterbury area, New Zealand 151
Barroisite 257 Carbon dioxide:
Barrovian region, Scotland 57 (map), 59, behaviour 44
185-8,186 (map), 341,433 effect on minerals 113-14,253
Barrovian zones 57-8,67,185-8 mobility 42
Barrow, George (1853-1932) 57.,5-8.,432 mixed with water 44, 270-6
Barth, Tom. F. W. (1899-1971) 188,436 Cashel-Lough Wheelaun area, Ireland 285-6
Basic rocks, definition 29 Cataclasis 27, 421
Becke, F. (1855-1931) 431 Cataclasite 421
 INDEX /483
Cataclastic metamorphism 27-8,421-3 Comrie area, Scotland 279-80
Cataclastic rocks 421-5 Contact metamorphism:
Cazadero area, California 246-7 baric types 115-16
Celebes 386-7,387 (map) defmition 27
Cevennes, France 180 facies names 297
OIabazite (zeolite) 157 geothermal gradient 87
OIalcedony 165 metamorphic facies 69-70, 116-18
OIarnockite 33 migration of materials 118-19
Olemical composition 120-4, 128 progressive metamorphism 277 -92
Olemical potential 43 Continental collision 391
Olichibu geosyncline, Japan 356 Continental drift in the Atlantic region
Ollorite: 345-6
distinction from clay minerals 164 Continental margin (active) 100-1, 390
equilibrium with garnet· 217 Cooma area, Australia 98, 174, 371
in metabasites 248 Cordierite 211-14
in metapelites 206-9 Cordilleran Mountains, North America 327,
in sediments and burial metamorphism 330 (map), 332-7
163-4 CO 2 (see carbon dioxide)
Ollorite zone (in the Scottish Highlands) Crestmore, California 291-2
58,185-6 Cristobalite 303
Olloritoids 205-6, 222 Crust:
Christiania area (see Oslo area, Norway) continental 409-15
Olurchill province, Canada 328 (map) lower 23,410-15
Circum-Hida metamorphics, Japan 360 oceanic 419-20
(map),362 retrogressive metamorphism 23, 413
Circum-Pacific regions 82,391-2 upper 410
Oassification: Cummingtonite 258
of facies series 73,113,293-309
of metamorphic reactions 38
of metamorphic rocks 30 Dalradian Series, Scotland 57 (map), 85,
of metamorphism 22-3 186 (map), 340
of the rocks 19 Decarbonation reactions:
Clausius-Clapeyron equation 41 and behaviour of CO 2 , 62, 267
Oay minerals 163-5,198 in progressive metamorphism 270-6
Clinopyroxene: andP-Testimation 91
calcic in metabasite 260
in eclogites 317-18
jadeite 230-2
°
thermodynamics 38, 50
Deficiency in K2 127
Dehydration reactions:
in limestone 267, 272 and behaviour of H2 0, 43-5, 62
Oinozoisite 248 and geologic age 111-12
Oosed system 120 andP-Testimation 91
Coal seam, metamorphism of 233 thermodynamics 45-8
Coast Ranges, California 193-7, 246-7, in zeolite facies 156-9, 161
335-7,336 (map) Demagnetization of oceanic crust 419
Coast Range thrust, California 335, 336 Depth-zones (of Grubenmann) 431
(map) Diablo Range, California 194-7, 247, 336
Coleman and Lee's nomenclature of Francis- (map)
can rocks 193-4 Diagenesis 19,21
Component: Diopside (pyroxene):
excess 125 in limestone 267, 272
fixed 122 in metabasites 260
independent 120 Dislocation metamorphism, definition 27,
mobile 122 430
perfectly mobile 122-4 Dolomite 114,266,270
Composition-paragenesis diagram 121-3, Donegal granites, Ireland 284-5
295 Double arc (island arc) 404
484/ INDEX
Dutchess County, New York State 181 Fault:
(map),188,436 major 400
Dynamic metamorphism, deimition 27, slab-surface 401
430 transform 400
Dynamometamorphism, deimition 430 Fennoscandia (= Baltic shield) 338, 339
(map)
Ferroactinolite ( =ferrotremolite) 251
East Asiatic volcano-plutonic belt 355-6, Ferrotremolite 251,261
356 (map) Fibrolite (acicular sillimanite) 283
East European platform 338 Fixed component 122
East Japan volcanic belt 350-1, 351 (map) Fluid phase 43-4,63, 125
Eclogites: Foliation 31
definition 33-4,310,315 Forsterite 272
in gneiss areas 316-19 Fossa Magna, Japan 349,360 (map)
and granulite 311-15 Franciscan terrane (or formation), Cali-
in high-pressure metamorphic terranes fornia:
319-20 geology 332,335-7
inclusions 320-3 limestone 270
Eclogite facies: maps 195,196,335
ACF diagram 315 metabasites 246-7
definition 69,310-23 metaclastics 193-7
granitic rocks in 324 ophiolites 94
and granulite facies 311-15 progressive changes (diagram) 80,197
petrographic criteria 313-15 zeolite formation 153
pressure and temperature 312,321 Free energy 39
Edenite (amphibole) 252
Element partition (or distribution) 48-50, Garnet:
217-21,318 disequilibrium growth 61
Epidiorite, Scotland 239-41 in eclogite 318
Epidote 114,248-9 in limestone 274-6
Epidote amphibolite 32 in metabasites 259-60
Epidote-amphibolite facies: in metapelites 211-21
A CF diagram 306 Garnet pyroxenite 315
definition 67 Garnet zone (in the Scottish Highlands) 58,
in high-pressure metamorphism 309 185-7
in low-pressure metamorphism 300 Gedrite (amphibole) 258
in medium-pressure metamorphism 305 Geologic age:
pressure and temperature 90 and dehydration 111-12
and rock nomenclature 33 and metamorphic facies 106
Equilibrium 46,61 and plate motion 391-2
Errol-Bryant Pond area in the Appalachians Geosyncline 24,430
181 (map),182-3 Geothermal curve 72, 86
Eskola, Pentti (1883-1964) 124, 2934, Geothermal field (active) 37, 147-9
435-6 Geothermal gradient 71-3, 85-6, 114,
Europe 338-47,339 (map) 139
Excess component 125 Glaucophane (amphibole):
Excess in K2 0 127 distribution map 107
External conditions 66, 120,295 and geologic age 107
in metabasites 262-3
Facies (see metamorphic facies) origin 436-7
Facies classification 294 Glaucophane schist facies:
Facies group 297 and baric type 77
Facies series: definition 68
and baric type 73 and facies series 113, 308
classification and description 293-309 and geologic age 36, 106, 107
definition 68 and metasomatism 436-7
 INDEX /485
pressure and temperature 90,92 Greenschist 30
Glaucophanitic metamorphism ( = high-pres- Greenschist facies:
sure metamorphism) 77 ACF diagram 301,306
Glen Oova, Scotland 57, 282 definition 67
Gneiss: diversity 112-14
definition 31 in high-pressure metamorphism 309
compositional characteristics 32 in low-pressure metamorphism 299
Gneissosity 31 in medium-pressure metamorphism
Goldschmidt, V. M. (1888-1947) 24,432, 304-5
434-5 pressure and temperature 90,92,439
Gothides, Baltic shield 338-9 Greenville province, Canada 328-30, 328
Grade of metamorphism 60 (map)
Grampian Highlands, Scotland 340 Grossular (garnet):
Granitic rocks: doubt in pyroxene-hornfels facies 279,
in the Alps 344 302
and andesite 85,100 phase diagram 275
in the Appalachians 97-8 stability in limestone 274-6,279
Cooma, Australia 98-9 Grubenmann, U. (1850-1924) 56,431-2
in Cordillera, N. America 332-5
in Finland 96-7
Holum, Norway 99 Haast schist group, New Zealand 384
in Japan 172-3, 365-9 Haematite, 51-2, 264-5
late-kinematic 96 Halloysite 164
and low-pressure metamorphism 85, Harker, A. (1859-1939) 56,433-4
95-6,100 Heat flow 419
Newer, Scotland 97,342 Hercynides, Europe 93, 175, 342-3, 343
Older, Finland 96 (map), 345, 393
Older, Scotland 97, 341 Hercynotype orogeny 94
post-kinematic 96 Heulandite (zeolite) 15 I, 158
and regional metamorphism 25, 95-6, Hida metamorphic complex, Japan:
396,398-9 geologic ages 358
and rhyolite, 368-9 geology 360-1
in the Scottish Caledonides 97 maps 83, 360, 366
synkinematic, 96 Hidaka metamorphic belt, Japan 83 (map),
Younger, Finland 96 359-60,375,376 (map)
Granulite 33 High grade 60
Granulite facies: High-pressure type:
ACF diagram 301,306,314 atypical 77
amphibolite facies, relation to 242-4 definition 72,77
in contact metamorphism 116 facies series 308-9
defmition (discussions) 302,311-13 and geologic age 110
definition (preliminary) 68 summary of features 394-6
distribution map 108 typical 77
eclogite facies, relation to 310-15 Highland boundary fault, Scotland 186
and geologic age 36,107-8 (map),340
in Grenville province, Canada 329 Hokkaido, Japan 359-60,375-6,376 (map)
in low-pressure metamorphism 302-3 Holum granite, Norway 99
in medium-pressure metamorphism Honshu (Honsyu) Arc, Japan 349,358
306-8 Honshu (Honsyu) geosyncline, Japan 356
pressure and temperature 90,311-13 Hornblende (amphibole) 33,61,251-7
and rock nomenclature 33 Hornfels 27, 34
Granulitic texture 33 Hornfels facies (obsolete name) 294
Graphite 53-4, 187,232-3 Hutton, James (1726-97) 429
Great Glen fault, Scotland 186 (map), 340 Hydrothermal alteration 27
Great Valley Sequence, California 335 Hydrothermal metamorphism 27, 147-9
Green Beds, Scotland 239-40 H.O (see water)
486/ INDEX
Iceland 154, 159 Katsuyarna area, Japan:
Icelandite 101 map 360
Igneous facies, deimition 294 progressive metamorphism 153
Igneous rock association 93-6 Keratophyre 95
Igneous rock series 101, 103, 105 Kermadec Arc, Pacific Ocean 380 (map),
Igneous rocks 19 382
Illite 1634,198,201 K-feldspar (see potassium feldspar) 127-30,
llmenite 264-5 132,225
Inner side (= continental side) 357 Kii Peninsula, Japan 151-2,360 (map), 371
Inner arc (volcanic) 406-7 Kilobar 38-9
Intergranular fluid 434,64-5, 125 Kilocalorie 38-9
Interior Lowlands, North America 327 Kimberley (East) area, Australia 175, 379
Inverian orogeny, Scotland 340 Kiyama area, Japan 360 (map), 362
Iritono area, Japan 117,288 Kurile Arc, Pacific Ocean 349
Iron: Kurosegawa zone, Japan 360 (map), 362
formations 54 Kwoiek area, Canada 283
native 51-2 Kyanite (see aluminium-silicate minerals):
oxides 51-2,232,264-5 in contact metamorphism 115,282-5
sulphides 232, 264-5 phase diagrams 40,72
Island arc, 36-7, 1004, 348, 390 in regional metamorphism 72, 223
Isograd,58-60 stability relations 3940,72, 199, 224-5
Itoigawa-Shizuoka (Sizuoka) line, Japan Kyanite-sillimanite type (= medium-pressure
349 (map), 357, 360 (map) type) 75
Izu-Bonin Arc, Pacific Ocean 348-9 Kyanite zone (in the Scottish Highlands)
58,186-7
Jadeite:
occurrence 230-2 Lachlan geosyncline, Australia 381
phase diagrams 40,231 Lamite 291
stability relations 40-1,89,439 Laumontite (zeolite) ISO, 159-60, 162
Japan: Lawsonite 114,228
crust and upper mantle 351-2 Laxfordian orogeny, Scotland 340
heat flow 352-3,353 (map) Lewsian complex, Scotland 340
island arc structure 348-50,349 (map) Limestone:
metamorphic belts 83 (map), 356-60, advanced decarbonation 291-2
360 (map) contact metamorphism 290-2
ophiolites 94, 359 progressive reactions 270-6
Precambrian basement 355-6 regional metamorphism 266-76
Quaternary volcanoes 350-1, 351 (map) Lithospheric plate:
zonation of volcanic rocks 350, 352 motion 104,112,3914
(map) underthrusting 84,398400
Lord Howe Rise, Pacific Ocean 380 (map),
Japan Sea 3524,408
Joule 38-9 382
Lowgrade 60
Low-pressure type:
Kamuikotan metamorphic belt, Japan 83 definition 72, 74
(map), 359-60, 375, 376 (map) facies series 299-303
Kanto Mountains, Japan: and geologic age 110
geology 369 andgranite 85,95-6,100
map 360 summary of features 396-7
metabasites 245-6 Lyell, Charles (1797-1875) 429
metapelites 193
progressive changes (diagram) 79 Macquarie Rise (south of New Zealand)
Kaolinite 164,198 380 (map), 382
Karelides, Baltic shield 33840, 339 (map) Mafic rocks 29
Karroo dolerite, South Africa 287 Magma 19,21
Katarchean, Baltic shield 33,8-9 Magnetite 51-2
 INDEX /487
Maizuru zone, Japan 360 (map), 362 thermodynamics 38-55
Marble 29 Metamorphic recrystallization (see recrystal-
Median (tectonic) line: lization)
in Celebes, 387 (map) Metamorphic rocks:
in Japan 348, 360 (map), 365, 3734, chemical classification 28-30
374 (map),401 (map) definition 20
in New Zealand 383 (map), 386 origin of term 429
Medium-pressure type: Metamorphism:
definition 72,74 baric types 71
facies series 304-8 classification 22-3
and geologic age 110 concept 20
tectonic duality 398 contact type 27
Melting (see partial melting): definition 19,21
in cataclastic metamorphism 422 dislocation 430
in contact metamorphism 285-6 dynamic 430
in regional metamorphism 63-5, 226-7, grade 60
307,412-13 ' high-temperature limit 21
temperature 22 large-scale 20
Merwinite 291 at low temperature 112
Meta (prefix) 28 low-temperature limit 21
Metabasalt 417-18 ocean-floor type 23
Metabasites: origin of word 429
contact metamorphism 288-90 plutonic 430
origin of word 29 regional type 234,430
regional metamorphism 23447 tectonic aspect 23,325425,440
sensitivity 68 transformation-fault type 23
Metaclastics 28, 193 weathering and diagenesis, distinction
Metagabbro 418 from 19-20,21 .
Metagraywacke 28,194-7,296 Metapelites:
Metamorphic belt (regional): A 'KF and AFM diagram for 129-33
facies series variation 82 contact metamorphism 277-85
thermal structure 24,56-8 definition 28
tectonic classification 390-1 progressive textural change 32
tectonics 390408 regional metamorphism 167-93
width 87 Metasediment 28
Metamorphic facies: Metasomatism, definition 62 (see also
and baric type 293-309 migration of materials in metamor-
definition (elaborate) 293-6 phism)
definition (preliminary) 66-7 Metavolcanics 28
and geologic age 106-7 Mica (see biotite, muscovite, and paragon-
and mobile components 295 ite)
names 296-7 Mica schist 28-30
pressure and temperature 88-91 Michigan (northern) 183-5,239
and progressive metamorphism 67-8, Microcline, change to orthoclase 225 (see
294-5 potassium feldspar)
and rock nomenclature 30 Mid-Atlantic Ridge:
systematic description 293-309 cataclastic rocks 425
Metamorphic facies series: metabasalts and metagabbros 417-19
and baric types 73 zeolite-facies rocks 155
definition 68 Mid-oceanic ridge 23, 26, 392-3,417,419
of metamorphic belt 82 Migmatite:
systematic description 293-309 definition and current usage 32
Metamorphic grade 60 and granite 99
Metamorphic reaction: in the Scottish Highlands 341
classification 38 Migration of materials in metamorphism:
history of study 437-9 in contact metamorphism 118-19
488/ INDEX
Migration of materials in metamorphism- Ocean-floor spreading 23,440
continued Oceanic crust 419-20
in ocean-floor metamorphism 418 Older Granite, Finland 96
in regional metamorphism 614 Older Granite, Scotland 341
Omi area, Japan 360 (map), 362
Mikabu tectonic line, Japan 370,401 (map)
Mineral assemblage 48-9,167,434 Onikobe area, Japan 148, 159
Mineral facies, defmition 294 Opal 165
Mineral paragenesis 120 Open system 42, 122
Mineralogical phase rule: Ophiolites:
Goldschmidt's 121 in the Alps 344 (map)
Korzhinskii's 123 concept 94-5
Mixed layer clay minerals 163 in Japan 359,370-1
Mizuho orogeny, Japan 372 non-sequence type 95
Mobile component 122,295-6 sequence type 95
Mogami area, Japan 1424 Organic material 53,232-3
Moine Series, Scotland 185,340 Orijiirvi area, Finland 127-8, 293,339,435
Moine thrust, Scotland 186 (map), 340 Orogenic belt 23
Moldanubian zone in the Hercynides 342 Orogenic metamorphism 24
Monticellite 291 Orogeny 24
Montmorillonite 163 Orthoclase, formation of 225, 243 (see
Montmorillonoids (smectites) 163 potassium feldspar)
Morioka-Shirakawa (Sirakawa) line, JapanOrthopyroxene 260-2
372,402 Oslo (= Christiania) area, Norway 278-9,
Mount Lofty Ranges, Australia 175, 380 290,293,434
Muscovite: Otago schists, New Zealand 384
breakdown 46, 226 Outer arc (non-volcanic) 406
compositional variation 131-2, 208 Outer side (oceanic side) 357
illite and phengite, relation to 2014 Oxidation 514,188
in metabasite 258-9 Oxidation-reduction reaction 38,51-5
paragonite, relation to 203-5 Oxygen:
Mylonite 422 buffer 55
isotope geothermometry 91
Nagasaki metamorphic complex, Japan 360 isotope tracer 62, 118-19
(map), 375 mobility 54
Nanga Parbat, Himalayas 269
Native iron 51-2 Pacheco Pass area, California 153, 195-7,
Neptunists 429 196 (map)
Nernst's partition law 49 Pacific Ocean:
New Caledonia 388 the Atlantic, compared with 391-3
New England geosyncline, Australia 381 circum- 82, 391-2
New Hampshire plutonic series in the Southwest 379-89,380 (map)
Appalachians 97, 332 Paired belts (see paired metamorphic belts)
New Zealand 380 (map), 382-6,383 (map) Paired metamorphic belts:
New Zealand geosyncline 384 in the Alps 344-5
Newer Granite, Scotland 97,342 in California 332,336 (map)
Nohi rhyolites, Japan 368-9, 369 (map) in Celebes 386-7,378 (map)
Norfolk Ridge, Pacific Ocean 380 (map), in Chile 82, 390
382 concept 82
'Normal' regional metamorphism (of in Hokkaido 83 (map), 359-60, 376
Harker) 433 (map)
North America 327,328 (map) in Honshu Arc of Japan, 83 (map), 356-7,
Northeast Japan Arc 348-9, 349 (map), 360 (map)
404-8,405 (map) model for the origin 84, 398400
in New Zealand 382-6,383 (map)
Ocean-floor metamorphism 23, 26, 36, ridge descent, relation to 392
416-20 in Ryukyu Arc 360 (map), 377
 INDEX /489
in Scotland 342 in high-pressure metamorphism 152-3,
summary of features 397 308
in Taiwan 377 (map), 378 in low-pressure metamorphism 140-7,
unpaired belts, relation to 393-4 299
in Washington State 332-3 in medium-pressure metamorphism 149-
Panoche Pass area, California 80, 153, 52, 304
194-5,195 (map), 247 Prehnite-pumpellyite-metagraywacke facies
Paragonite 199-201,203-4 296
Pargasite (amphibole) 252,261 Pressure:
Partial melting: depth, relation to 39, 72
in contact metamorphism 285-6 effect 438-9
in continental crust 412-3 of metamorphic facies 88,90
in granulite facies 307 of metamorphism 437-9
in metapelites 226-7 partial 43
in progressive metamorphism 63 rock- 39,42-3
in pyrometamorphism 286-7 units 38-9
temperature 22 Progressive metamorphism 56, 64, 167,
and water pressure 21-2,65 234,294-5
Partial pressure 43 Progressive mineral changes:
Partition law 49 in metabasites 248-65
Pascal 38-9 in metapelites 198-233,432
Pegmatite, in relation to metamorphism 64, in siliceous limestone 270-6
173 in zeolite facies 156-66
Pelitic metamorphic rocks (see metapelites) Progressive textural changes 32,432
Pennine nappes in the Alps 189-91, 190 Pseudotachylite 422
(map), 343-5 Pumpellyite 70, 114,164
Perfectly mobile component 122, 124-5 Pyrenees, France and Spain 176-80, 268,
Periclase 266,273,291 342-3,343 (map)
Peristerite unmixing 210 Pyrometamorphism 27, 69, 277, 286-7,
Petrogenetic grid 78,81,297 303
Phanerozoic orogenic belt 35 (map), 36, Pyrope (garnet) 217,221,312,314,315,
409-10,414-15 318
Phase rule 120, 125 Pyrophyllite 180,198-201
Phengite (mica) 201-2 Pyroxene (see clinopyroxene and ortho-
Phyllite 32 pyroxene)
Plagioclase: Pyroxene-hornfels facies:
peristerite unmixing 210 ACF diagram 301
progressive change 60,210-11,249-51 definition 69,116
twinning 250-1 in facies series 302
Plate (see lithospheric plate) name change from hornfels facies 294
Plutonists 429
Porphyroblast 28 Quartz:
Porphyrociast 421 and breakdown of micas 226, 228
Potassium feldspar: and reaction temperature 148-9
change of micro cline to orthoclase 225, in zeolite-facies rocks 154, 165-6
243 Quartz vein in relation to metamorphism
effect on mineral assemblages 127-30, 64
132 QuartZ-diorite line, North America, 333
in the sillimanite-K feldspar zone 226, (map), 333-4
241 Quartzo-feldspathic rocks 29
Precambrian shield 34,35 (map), 409-13
Predazzite 266
Prehnite 70,164 Ramberg, H. (1917- ) 44,412,436
Prehnite-pumpellyite facies: Rangitata orogeny, New Zealand 384, 386
definition 70 Rankinite 291
and facies series 113 Reactions (see metamorphic reactions)
4901 INDEX
Recrystallization (metamorphic): Sanidine (potassium feldspar) 303
in cataclastic rocks 422 Sanidinite facies:
defInition 20 ACFdiagram 301
threshold temperatme 113 defInition, difficulty in 303
in zeolite-facies rocks 139 defInition (preliminary) 10
Regional metamorphic belt (gee meta- Schist 30-2
morphic belt) Schistosity 30
Regional metamorphism: Schreinemakers bundle 133-5
definition 23-5 Schreinemakers rule 79,133-5
diversity 71-8,437 Scottish Highlands:
and granite 25-6, 95-6, 398-9 faults 186, 340
'normal' (of Harker) 433 geology 340-2
origin of term 430 granitic rocks 97
Relic 56 history of study 432-3
Riebeckite (amphibole) 262 limestone 268-9
Ripheides, Baltic shield 338-9 maps 57,186
Rock-pressme 39,42-3 metabasites 23941
Rock series (see igneous rock series) metamorphic facies 67-8
Rocks, threefold classification 19-20 metapelites 57-9, 185-8
Rosenbusch, H. (1836-1914) 430,431 progressive changes (diagram) 59
Russian plate 338 Scomian orogeny, Scotland 340
Ryoke metamorphic belt (or terrane), Sedimentary rocks 19
Japan: Septechlorite 206
Abukuma terrane, relation to 368 Sericite 201
geology 363-7 Serpentinite 94,416
and granitic rocks 100,365-7 Sevrne area, Spain 238-9
limestone 267-8 Shimanto (Simanto) geosyncline 371
maps 83,360,364,366,367,374 Shimanto (Simanto) terrane 366 (map),
metabasites 234-6 371-2
metapelites 167-73 Shiojiri (Sioziri)-Takato area, Japan:
progressive changes (diagram) 75 geology 363
and rhyolitic rocks 368-9 granitic rocks 172-3
Ryukyu Arc, Pacific Ocean 349 (map), 377 maps 170, 360
metabasites 243-6
S-smface 31 metapelites 169-73
Saamides, Baltic shield 338-9 progressive changes (diagrams) 75, 171
Salite (pyroxene) 260 Shoshonite 101
Salton Sea area, California 37, 149 Shuksan metamorphic terrane, Washington
San Andreas fault: State 332-3
cataclastic rocks along 4234 Sierra Nevada, California and Nevada:
map 336 geology 332-5
San Juan Islands, Washington State 153 granitic rocks 100,334
Sanbagawa (Sambagawa) metamorphic belt maps 333, 336
(or terrane), Japan: Sillimanite 3940, 72, 115, 199, 224,
faults 401-2 279 (see also aluminium silicate
geology 369-71 minerals)
limestone 270 Sillimanite-K feldspar (or orthoclase) iso-
maps 83,364, 374,401 grad 226
metabasites 244-6 Sillimanite-K feldspar zone (in Appala-
metapelites 191-3 chians) 241
ophiolites 94-5,370-1 Sillimanite zone (in the Scottish Highlands)
progressive changes (diagram) 78, 79 58, 186-8,341
Sangun metamorphic belt, Japan: Sithean Sluaigh area, Scotland 286-7
geology 362-3 Slate 32
maps 83,360 Smectite 1634
progressive metamorphism 153 Solid-solid reactions 38-9,1214,208
 INDEX /491
Solid solution mineral 49,60 Tholeiite series 102
Southern Upland fault, Scotland 186 Tilbuster area, Australia 290
(map), 341 Tilleyite 291
Southern Uplands, Scotland 186 (map), Transform fault metamorphism 23,
341-2 421-425
Southwest Japan Axc 348-9, 349 (map), Tremolite (amphibole) 251,261,273
408 Tridymite 303
Spessartine (garnet) 214-5 Trondheim (Trondhjem) area, Norway
Spilite-keratophyre group 95 24-5,25 (map), 340,434
Spotted slate 277,279 Tschermak substitution 206,251
Spurrite 291 Tschermakite 252
Staurolite 222, 282 Tuhua orogeny, New Zealand 384
Staurolite zone (in the Scottish Highlands) Tukuba (Tsukuba) area, Japan 173, 360
58,187-8 (map),366
Stavanger area, Norway 216,340
Steinach area, Germany 281-2 Ultrabasic rocks 29, 30
Stilbite (zeolite) 157 Units of measurement 38-9
Stilpnomelane (ferro- and ferri-) 263 Unpaired metamorphic belts 393-4
Stress-minerals (of Harker) 433
Subcalcic amphibole 251, 257 (see amphi- Vermiculite 164
bole) Volcanic arc 100
Sulitjelma (Sulitelma) area, Norway 241, Volcanic front 350
340 Volcanic rock series (see igneous rock series)
Superior province, Canada 328-30, 328
(map)
Waiotopu area, New Zealand 149
Svecofennides, Baltic shield:
Wairakei area, New Zealand 147-8,159
geology 339-40
Wairakite (zeolite) 147,159,160
granitic rocks 96-7
Wakatipu metamorphic belt, New Zealand
map 339
383 (map), 384
Wantabadgery-Tumbarumba area, Australia
Taconian (Taconic) orogeny (in the Appala- 174,381
chians) 330, 345 Water:
behaviour 42,64
Taiwan (Formosa) 377-8,377 (map)
dissociation 53
Talc 272
Tananao schists, Taiwan 377 (map), 378 effect 114
mobility 42,62
Tanzawa Mountains, Japan:
autocannibalism 66, 146-7 Waterville-Vassalboro area in the Appala-
geology and metamorphism 144-7, 159 chians 181 (map), 182
map 144 Weathering 19
mineral changes (diagram) 145 Werner, A. G. (1749-1817) 429
Taringatura area, New Zealand: West Japan volcanic belt 350-1, 351 (map)
metamorphism 149-51,159 Wollastonite:
progressive changes (diagram) 150 in contact metamorphism 290-2
Tasman geosyncline, Australia 381 formation 50,273,274
Tasman metamorphic belt, New Zealand in regional metamorphism 268,269
383 (map), 384 Woodsville area in the Appalachians 181
Temperature: (map), 241-2
determined by oxygen isotopes 92 Wiistite 5 1-2
distribution in orogenic belts 398-400
of metamorphic facies 88, 90 Yap Islands, Pacific Ocean 388
of metamorphism 437-40 Younger Granite, Finland 96
units 39
Texture 30 Zeolite 47, 114,156-62
Thermal axis 24 Zeolite facies:
Thermodynamics 38,434,438 definition 70
492/ INDEX
Zeolite facies-continued mineral changes 156-66
and facies series 113 pressure and temperature 90,439
and geologic age 36, 107 Zirkel, F. (1838-1912) 431
high-pressure metamorphism, relation to Zoisite 248
153 Zones:
in low-pressure metamorphism 139-49, depth (of Grubenmann) 431
299 epi-, meso- and kata- 431
in medium-pressure metamorphism progressive mineral 57-60,432
149-52,304 progressive textural 432