Available online at www.sciencedirect.

com

Precambrian Research 162 (2008) 540–558

Isotope chemostratigraphy of marbles in northeastern Mozambique:

Apparent depositional ages and tectonostratigraphic implications
V.A. Melezhik a,∗ , B. Bingen a , A.E. Fallick b , I.M. Gorokhov c , A.B. Kuznetsov c , J.S. Sandstad a ,
A. Solli a , T. Bjerkgård a , I. Henderson a , R. Boyd a , D. Jamal d , A. Moniz e
a Geological Survey of Norway, 7491 Trondheim, Norway
b Scottish Universities Environmental Research Centre, East Kilbride, Glasgow G75 0QF, Scotland, United Kingdom
c Institute of Precambrian Geology and Geochronology, nab. Makarova, 2, 199034 St. Petersburg, Russia
d Eduardo Mondlane University, P.O. Box 257, Maputo, Mozambique
e Provincial Directorate for Geology, Lichinga, Niassa Province, Mozambique

Received 20 February 2007; received in revised form 22 October 2007; accepted 10 November 2007

Abstract
Marbles are minor but characteristic components of metasedimentary units within nappes in the Pan-African Mozambique Belt in NE Mozam-
bique. Metasedimentary units remain largely undated, and carbon and strontium isotope stratigraphy of marbles has been used for indirect dating
of the depositional history in this part of the Mozambique Belt. Sixty-nine samples from nine occurrences of dolomite, calcite and magnesite
marbles in the Montepuez, Xixano, Lalamo, Ocua and Nampula metamorphic complexes were analysed for major and trace elements, and a subset
of 39 samples for C, O and Sr isotopes. The least altered ␦13 C values range from −3.5 to +7.1‰ (V-PDB) and 87 Sr/86 Sr ratios from 0.70504 to
0.70671. These values are considered as the best proxy to seawater composition at the time of deposition. The apparent deposition ages, derived
from available seawater evolution curves, range from c. 1250 to c. 660 Ma. An age of 1250–910 Ma is obtained from a tripartite marble unit
in the Montepuez Complex which is exposed in the Montepuez quarries. Five other age-groups are represented by marble units with apparent
depositional ages of 800–750 Ma (Xixano North), 800–660 Ma (Montepuez West), c. 750 Ma (Nampula), c. 740 Ma (Xixano South and Lalamo),
and 740–670 Ma (Montepuez East). The data suggest that: (i) Pan-African nappes in NE Mozambique include Neoproterozoic and probable
Mesoproterozoic sediments; (ii) Neoproterozoic rocks of the Xixano and Nampula complexes might have different ancestry and were tectonically
juxtaposed during the Neoproterozoic Pan-African orogeny.
© 2007 Elsevier B.V. All rights reserved.

Keywords: Mozambique; Marble; Calcite; Dolomite; Magnesite; Carbon; Oxygen; Strontium; Isotopes

1. Introduction Belt is based on very few radiometric dates that constrain the
ages and provenances of the different tectonostratigraphic com-
The Mozambique Belt in East Africa was first recognised plexes that constitute the belt. These dates (reviewed by Johnson
by Holmes (1951) who demonstrated a structural discontinuity et al., 2005), obtained from igneous and metamorphic rocks by
between the Tanzanian Craton and metamorphic complexes of the U–Pb, Sm–Nd and Rb–Sr methods, are rather few, gener-
an apparent Mezoproterozoic age to the east. Later studies by ally lack high precision, and thus allow various interpretations
Pinna et al. (1993), Muhongo et al. (2001), Kröner et al. (2003) of the geological history (Holmes, 1951; Pinna et al., 1993;
and Johnson et al. (2005) have shown a complex history of the Shackleton, 1996; Muhongo et al., 2001; Meert, 2003). More-
belt, which experienced several tectonometamorphic events that over, there is general lack of dates constraining depositional ages
occurred between 1100 and 500 Ma. In northern Mozambique, of various supracrustal rocks that constitute the belt. To date,
reconstruction of the geological evolution of the Mozambique there has been only one attempt to obtain depositional age from
metasedimentary rocks. The low-grade metalimestones of the
Geci group which was tectonically juxtaposed with granulite-
facies complexes in NW Mozambique, have been constrained to
∗ Corresponding author. Tel.: +47 73 90 42 41; fax: +47 73 92 16 20. 630–585 Ma by means of isotope chemostratigraphy (Melezhik
E-mail address: victor.melezhik@ngu.no (V.A. Melezhik). et al., 2006).

0301-9268/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.precamres.2007.11.002
 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558 541

Marbles associated with metamorphosed complexes are imbricated during the Neoproterozoic Pan-African orogeny, and
minor but consistent components of allochthonous units in the were intruded by several generations of granites; the present-day
Mozambique Belt in NE Mozambique, Tanzania and Madagas- tectonostratigraphy in NE Mozambique is the result of several
car (Pinna et al., 1993; Muhongo, 1994; De Wit et al., 2001; phases of Pan-African orogenic processes.
Fernandez and Schreurs, 2003). The purpose of this work is to Available data provide evidence for a Mesoproterozoic base-
provide further progress in understanding the geological history ment, and an overlying Meso- to Neoproterozoic nappe system.
of the region by constraining the depositional age of high-grade The basement is an integral component of the Nampula, Mar-
marble formations in NE Mozambique with the tool of Sr and rupa and Nairoto complexes, and probably the Meluco Complex.
C isotope chemostratigraphy. Extensive isotopic work in the The basement is mainly made up of felsic orthogneiss with
Caledonian orogenic belt (Melezhik et al., 2001a, b, 2002a, magmatic crystallisation ages ranging from c. 1150 to 940 Ma
b, 2003, 2005; Thomas et al., 2004; Slagstad et al., 2006) has (Bingen et al., 2006a, b, 2007. It is locally unconformably over-
demonstrated that carbonate chemostratigraphy can be in prin- lain by a Neoproterozoic metasedimentary cover (e.g., Mecuburi
cipal applied to high-grade marbles, thus providing the age group, located SW of the area of Fig. 1). The Pan-African
constraints necessary for improving reconstructions of the depo- nappes include the Xixano and Lalamo complexes. They are
sitional history and tectonostratigraphic assembly of areas where mainly Neoproterozoic mafic to intermediate granulites, vari-
it is difficult to use other methods. ous metasediments, metavolcanic rocks and felsic orthogneiss.
Enderbites, metagranites and metarhyolites yield magmatic
2. Geological background crystallisation ages between c. 820 and 740 Ma. The nappes
carry evidence for a pre- or early-Pan-African granulite-facies
Originally the Mozambique Belt, a high-grade north–south metamorphism dated to 735 ± 5 Ma in the Xixano Complex. The
trending orogenic belt in East Africa, was defined to have nappe system probably represents exotic or outboard continen-
formed as the result of a c. 1300 Ma single orogenic event tal fragments accreted to the basement during the Pan-African
(Holmes, 1951). As more radiometric dates were generated, it orogeny. The main Pan-African orogeny took place between
became apparent that the belt is the product of at least two oro- 580 and 530 Ma, and included at least two main phases, par-
genic cycles. Pinna et al. (1993) recognised high-grade gneisses, tially overlapping in time (Bingen et al., 2006a, b, 2007). These
granulites and migmatites comprising a Precambrian basement are (i) NW verging imbrication of the basement and transport of
which was intruded by orogenic plutonic rocks. These rocks the nappes over the basement, and (ii) folding of the basement
were emplaced and deformed during the 1100–850 Ma Mozam- and nappes, especially along a NNE–SSW trend. The Ocua and
bican orogeny, approximately equivalent to the Mezoproterozoic Montepuez complexes are partially interpreted as a melange con-
Kibaran orogeny of central Africa. A younger cycle involved the taining tightly folded lithologies belonging to both the basement
deposition of limestones, phyllites, sandstones and conglom- and the nappes. Available data provide only limited constraints
erates of marine origin, and some other rocks with possible on the deposition of sedimentary sequences (e.g., Melezhik et
glacial affinities. The Geci group metalimestones, belonging to al., 2006). The present study addresses this problem.
this cycle, have an apparent depositional age of 630–585 Ma
(Melezhik et al., 2006). 3. Marble formations studied
Muhongo et al. (2001) and Kröner et al. (2003) constrained
the peak of high-grade metamorphism in northern Mozam- Amphibolite-faces marbles occur sporadically in most
bique to c. 615 Ma. Johnson et al. (2005) reviewed the newly supracrustal complexes (Fig. 1) where they range in thickness
obtained radiometric dates and suggested that previous tectonos- from less than a meter to hundreds of meters with lateral exten-
tratigraphic subdivisions in northern Mozambique should be sion over 100 km. Those involved in our study are from the
considered with caution. Nampula, Montepuez, Xixano, Lalamo, and Ocua complexes
New geological mapping at the 1:250,000 scale was accom- (Fig. 1). Marbles from the Montepuez Complex are the main tar-
plished in northern Mozambique between 2003 and 2006 by get of the research. They were studied in commercially exploited
the Geological Survey of Norway, the British Geological Sur- quarries near Montepuez (Fig. 1). Excellent man-made expo-
vey and the National Directorate of Geology of Mozambique. It sures provide an opportunity for detailed study and systematic
resulted in acquisition of new petrographic, mineralogical, geo- sampling. The level of exposures of marbles from other com-
chemical, structural and geochronological data (Bingen et al., plexes allowed only limited sampling.
2006a, b, 2007). A simplified version of the new geological map The present-day lithology of the Montepuez Complex
over northeasternmost Mozambique is presented in Fig. 1. This includes a variety of amphibolite-facies schists, amphibolites,
map integrates new field, airborne geophysics, structural data quartzites, felsic volcanites and several marble units. Marbles
and a substantial database of U–Pb zircon and monazite ages exposed in four small quarries east of Montepuez comprise three
constraining magmatic and metamorphic events (Bingen et al., distinctive lenses dipping at 46–60◦ southeast. Their visible
2006a, b, 2007). The map features a number of lithotectonic thicknesses range between 20 and 50 m. The lenses are isocli-
units, designated as complexes. New geochronological data nally folded, tectonically dissected, separated by thin slices of
(Bingen et al., 2006a, b, 2007) combined with results reported mica schists and apparently represent a tectonically imbricated
by Kröner et al. (2001, 2003) and Jamal (2005) suggest that unit. Three disparate marble lenses identified in the quarries
the lithological complexes were metamorphosed, deformed and are designated as a tripartite marble unit. A structurally lower
542 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558

Fig. 1. Schematic geological map of northeastern Mozambique with emphasis on marble formations.

member of the unit is termed as Calcite-dolomite marble, a thick bed of bluish, giantly crystalline, massive marble (Fig. 2b)
structurally upper as White dolomite marble, and a middle as occurs in the middle part of the lens. The calcite marbles are
Grey dolomite marble. The Calcite-dolomite marble is mainly structurally overlain by pale grey, medium crystalline, layered,
composed of pale grey, impure, thinly layered and medium dolomite marble. The Grey dolomite marble is dominated by
to coarsely crystalline calcite marbles (Fig. 2a). One 1.5 m- pale grey, medium-crystalline, dolomite marbles with thin lay-
 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558 543

Fig. 2. Lithological varieties of some of the studied marbles. Montepuez Complex marbles: (a) Photograph of an inner corner of sawn surfaces in the quarry №
4 (Calcite-dolomite marble member) showing the typical style of tectonic deformation in the grey, banded, calcite marble; (b) photograph of a banding-parallel
surface in the quarry № 4 (Calcite-dolomite marble member) showing the bluish marble composed of giant calcite crystals; coin = 9 mm; (c) photograph of a sawn
surface in the quarry № 1 (White dolomite marble member) showing banded, pale grey, dolomite marble with numerous inclusions of amphibolite structurally
overlain by massive to weakly banded white dolomite marble with banding-parallel inclusions of amphibolite; hammer head = 12 cm. Marbles of other complexes:
(d) Pale pink, massive, medium-crystalline, dolomitised, magnesite marble from the Lalamo Complex; (e) back-scattered electron image of the micro-brecciated and
retrogressively dolomitised (Dol), calcitised (Cc) and serpentinised (Spt) Lalamo magnesite (Mg); (f) weathered surface of depositional carbonate breccia from the
Lalamo Complex; (g) photomicrograph of the Lalamo carbonate breccia showing dark and pale brown fragments cemented by white calcite.
544 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558

ers of a dolomitised calcite marble. The White dolomite marble ing element extraction (1σ) is ±10% rel. Standard analytical
is distinguished by its white colour and high purity. It con- procedures have been used for measurement of total organic
tains both dolomite and calcite marbles though the latter are carbon (TOC) at NGU. The TOC content has been measured
subordinate. The marbles include roughly equal volumes of mas- from acid-washed material via sealed tube combustion using a
sive and layered varieties and contain sporadic, layer-parallel, Leco SC-444 instrument with a total analytical uncertainty of
strongly foliated amphibolite inclusions, apparently dissected ±15% rel.
and dismembered thin mafic dykes or sills (Fig. 2c). Oxygen and carbon isotope analyses of whole-rock marble
In the tripartite marble unit at Montepuez, both calcite and samples were carried out at the Scottish Universities Environ-
dolomite marbles are composed of crystals arranged in hypid- mental Research Centre, Glasgow, using the phosphoric acid
iomorphic, allotriomorphic and granoblastic textures. In calcite method of McCrea (1950) as modified by Rosenbaum and
marbles, accessory minerals are tremolite, quartz and muscovite, Sheppard (1986) for operation at 100 ◦ C. Carbon and oxy-
whereas dolomite marbles additionally contain sizable amounts gen isotope ratios in carbonate constituents of the whole-rock
of graphite and are commonly devoid of quartz. samples were measured on a VG SIRA 10 mass spectrometer.
Calcite and dolomite marble units occurring in the Mon- Analyses were calibrated against NBS 19, and precision (1σ) for
tepuez Complex outside the quarries (termed in this article as both isotope ratios is better than ±0.2‰. Oxygen isotope data for
East and West marbles, Fig. 1), as well as most marble units dolomites were corrected using the fractionation factor 1.00913,
hosted by other complexes, are very similar to the marbles of and 1.00933 for magnesite recommended by Rosenbaum and
the tripartite unit. They are tectonically dissected bodies asso- Sheppard (1986). The ␦13 C data are reported in per mil (‰)
ciated with paragneisses. They are commonly rather pure and relative to V-PDB and the ␦18 O data in ‰ relative to V-SMOW.
composed of medium-crystalline calcite with variable amount Rb–Sr analyses were carried out at the Institute of Precam-
of dolomite. Impure, dolomitised marbles contain tremolite as brian Geology and Geochronology of the Russian Academy of
the result of quartz + dolomite metamorphic reaction. The mar- Sciences (St. Petersburg) as specified in Gorokhov et al. (1995).
ble lens sampled in the Ocua Complex is composed of pure The Rb and Sr concentrations were determined by isotope dilu-
dolomite. tion. Rb isotopic composition was measured on a MI 1320 mass
One lens from the Lalamo Complex (JS307, Fig. 1) con- spectrometer. Strontium isotope ratios were measured on the
sists of magnesite and calcite marbles. The lens has a thickness Thermo-Finnigan Triton thermal ionisation mass spectrometer
exceeding 200 m and it is traceable along strike over a distance in static collection mode. All 87 Sr/86 Sr ratios were normalised to
of 20 km. It is hosted by amphibolite-facies layered arenites with a 86 Sr/88 Sr of 0.1194, and measurements of the NIST SRM-987
layers of biotite gneiss and quartzite. The magnesite marble bed, run with every batch averaged 0.710255 ± 8 (2␴mean , n = 11).
of unknown thickness, is composed of a pale pink, medium- During the course of the study, the value obtained for the
crystalline, massive lithology (Fig. 2d), which microscopically 87 Sr/86 Sr ratio of the USGS. EN-1 standard was 0.709191 ± 25

appears to be brecciated (Fig. 2e). Extensional micro-fractures (2σ mean , n = 3).

are cemented by retrogressive dolomite and calcite, which are
replaced by later chlorite and serpentinite (Fig. 2e). The same 4.2. Major elements
marble lens, though exposed 5 km north of the magnesite bed,
comprises brown, brecciated, and white, massive varieties of a The major and trace element geochemistry of the tripartite
calcite marble. The former has a visible thickness of over 2 m and marble unit sampled in four quarries at Montepuez is based
a vuggy appearance on weathered surfaces (Fig. 2f). The breccia on analyses of 56 samples (Table 1). Calcite (CM), dolomite
consists of 0.5–5 cm large, red to pale brown, calcite marble frag- (DM) and dolomitised calcite (DCM) marbles have been iden-
ments cemented by coarsely crystalline white calcite (Fig. 2g). tified in all three members of the unit. Both the calcite and
There is no field evidence suggesting a postdepositional tectonic the dolomite marbles are relatively pure rocks. They are almost
brecciation, therefore the breccia is categorised as a collapse devoid of Al2 O, Na2 O and K2 O (Table 1). The DM and DCM
breccia originated either from dissolution of evaporate minerals are relatively enriched in SiO2 (up to 18.7 wt%, Fig. 3a). Sev-
or karstification. eral samples of the DM contain sizable amounts of total organic
carbon (TOC, up to 0.6 wt%). We tentatively suggest a Mg/Ca
4. Geochemistry ratio of 0.1 as the boundary between the CM and the DCM. A
Mg/Ca ratio of 0.5 sets a lower limit for the DM. The CM is
4.1. Analytical techniques characterised by variable Mg/Ca ratios ranging between 0.02
and 0.09 (Fig. 3a), whereas dolomitised varieties show ratios
Major and trace elements were analysed by X-ray fluores- between 0.37 and 0.50 (Table 1). Mg/Ca ratios of the DM range
cence spectrometry at the Geological Survey of Norway (NGU), between 0.51 and 0.63 (average, 0.58 ± 0.03, n = 43), which is
Trondheim, using a Philips PW 1480 X-ray spectrometer. The lower than that for stoichiometric dolomite (0.62), thus indicat-
precision (1σ) is typically around 2% of the major oxide present. ing incomplete dolomitisation. The DCM does not plot along the
Acid-soluble Fe, Ca, Mg and Mn were analysed by ICP-AES at limestone–dolostone mixing line (Fig. 3b) because both Mg and
NGU using a Thermo Jarrell Ash ICP 61 instrument. Detec- Ca concentrations are low due to the high SiO2 content (Fig. 3a).
tion limits for Fe, Mg, Ca, Mn and Sr are 5, 100, 200, 0.2 The East and West marbles in Montepuez Complex (Fig. 1)
and 2 ppm, respectively. The total analytical uncertainty includ- are both pure calcite marbles with Mg/Ca ratios of 0.01 (Table 1).
Table 1
Chemical composition of Montepuez Complex marbles
Field # Lithology SiO2 (wt%) Al2 O3 (wt%) Na2 O (wt%) K2 O (wt%) TOC (wt%) Mg (wt%) Ca (wt%) Fe (ppm) Mn (ppm) Sr (ppm) Mg/Ca
Montepuez quarry 4, UTM 0497505, 8552297
Calcite-dolomite marble member
White to pale grey calcite marbles, partially dolomitised
Mvm040 Coarsely crystalline, weakly banded – – – – – 3.1 33.1 456 28 160 0.09
Mvm041 Coarsely crystalline, weakly banded 0.40 – – 0.04 – 1.9 34.7 312 18 167 0.05
Mvm042 Giant crystalline, massive – – – – – 0.69 37.0 72 4 116 0.02
Mvm043 Coarsely crystalline, banded – – – – – 2.4 34.3 358 14 135 0.07
Mvm044 Giant and medium-crystalline interbedded – – – – – 1.4 35.7 343 20 140 0.04
Mvm045 Coarsely to medium-crystalline, banded 0.04 0.04 – 0.03 – 2.3 34.3 344 12 145 0.07
BB037 Very coarsely-crystalline, massive – – – – – 2.5 33.4 405 18 132 0.07
Calcite marbles, intensively dolomitised
BB039 Pale brown, medium-crystalline 6.0 2.98 0.11 0.70 – 8.7 19.9 4080 150 83 0.44

V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558

Grey dolomite marbles
Mvm046 Medium-crystalline, banded 1.3 0.05 – 0.05 – 12.2 20.6 1300 48 58 0.59
Mvm047 Medium-crystalline, banded 0.48 0.02 – 0.03 – 12.4 21.0 1210 52 93 0.59
BB038 Medium-crystalline, banded – – – 0.03 – 12.2 20.8 1530 57 61 0.59
Montepuez quarry 3, UTM 0497650, 8552251
Grey dolomite marble member
Calcite marbles, intensively dolomitised
Mvm052 Grey, banded, medium-crystalline 2.1 – – – – 10.4 22.7 1690 91 83 0.46

Montepuez quarry 3, UTM 0497650, 8552251

Grey dolomite marble member
Dolomite marbles
Mvm051 Grey, medium-crystalline, banded 6.9 – – – 0.37 10.8 20.2 632 23 99 0.53
Mvm048 Grey, medium-crystalline, banded 0.1 – – – 0.13 12.7 20.6 1540 45 71 0.62
Mvm049 Grey, medium-crystalline, banded 3.0 – – – 0.60 12.0 20.4 1090 35 65 0.59
Mvm050 Grey, medium-crystalline, banded 7.5 – – – 0.33 10.9 19.9 798 27 104 0.55
Mvm053 White, medium-crystalline, banded 2.8 – – – – 11.7 21.3 1280 43 102 0.55
Mvm054 Grey, medium-crystalline, banded 7.1 – – – 0.30 10.7 19.3 666 24 76 0.55
Mvm055 Grey, medium-crystalline, banded 3.0 – – – – 12.0 20.5 2220 81 58 0.59
Mvm056 Grey, medium-crystalline, banded 0.82 – – – 0.32 12.7 21.4 1440 43 86 0.59
BB034 White, coarsely crystalline, massive 0.33 – – – 0.17 12.4 20.8 782 33 86 0.60
BB035 Pale grey, medium-crystalline, banded 7.7 – – – 0.26 10.2 19.2 386 16 99 0.53
Montepuez quarry 2, UTM 0497683, 8552084
White dolomite marble member
White calcite marbles, intensively dolomitised
BB031 Medium-crystalline, massive – – – – – 11.3 22.6 298 12 119 0.50

White dolomite marbles

BB029 Medium-crystalline, massive 1.8 – – – – 12.4 20.9 1530 73 83 0.59
BB030 Medium-crystalline, massive 3.6 – – – – 12.3 19.9 395 20 106 0.62
BB032 Medium-crystalline, massive 0.36 – – – – 12.6 20.4 513 19 71 0.62

Montepuez quarry 1, UTM 0497684, 8552072

White dolomite marble member
White calcite marbles, intensively dolomitised
Mvm065 Finely crystalline, massive 17.6 – – – – 6.7 18.0 328 13 82 0.37
Mvm066 Finely crystalline, massive 8.4 – – – – 9.4 20.6 493 24 99 0.46
Mvm074 Finely crystalline, massive 18.7 – – – – 6.4 17.4 736 52 70 0.37
Mvm075 Finely crystalline, massive 0.04 – – – – 11.2 22.9 758 61 112 0.49

545
 546
Table 1 (Continued )
Field # Lithology SiO2 (wt%) Al2 O3 (wt%) Na2 O (wt%) K2 O (wt%) TOC (wt%) Mg (wt%) Ca (wt%) Fe (ppm) Mn (ppm) Sr (ppm) Mg/Ca
White dolomite marbles
Mvm057 Medium-crystalline, massive; amphibolite inclusions 1.5 – – – – 12.6 20.9 1550 73 79 0.60
Mvm058 Medium-crystalline, massive; amphibolite inclusions 0.72 – – – – 12.4 20.7 1130 47 76 0.60
Mvm059 Medium-crystalline, massive; amphibolite inclusions 11.9 – – – – 9.2 18.1 1540 79 72 0.51
Mvm060 Medium-crystalline, massive; amphibolite inclusions 2.2 – – 0.02 – 12.5 20.8 715 36 78 0.60
Mvm062 Medium-crystalline, massive; amphibolite inclusions 3.8 – – – – 11.0 19.9 2740 98 86 0.55

V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558

Mvm061 Medium-crystalline, massive; amphibolite inclusions – – – – – 12.1 21.8 984 39 108 0.56
Mvm063 Finely crystalline, massive 6.8 – – – – 10.3 19.4 2050 79 88 0.53
Mvm064 Finely crystalline, massive 4.7 – – – – 11.0 20.5 806 40 85 0.54
Mvm067 Finely crystalline, massive 0.73 – – – – 12.1 21.6 1700 49 102 0.56
Mvm068 Finely crystalline, massive 0.08 – – 0.01 – 12.5 20.7 1530 58 80 0.60
Mvm069 Finely crystalline, massive 0.83 – – – – 12.3 21.1 1090 38 89 0.58
Mvm070 Finely crystalline, massive 0.22 – – – – 11.6 21.7 1150 42 108 0.53
Mvm071 Finely crystalline, massive – – – – – 12.4 21.3 447 26 77 0.58
Mvm072 Finely crystalline, massive 0.57 – – – – 12.1 22.1 211 10 98 0.55
Mvm073 Medium-crystalline, massive 0.56 – – – – 12.4 20.5 1570 53 94 0.60
Mvm076 Weakly banded with amphibolite inclusions – – – – – 13.3 21.1 2120 97 65 0.63
Montepuez quarry 1, UTM 0497684, 8552072
White dolomite marble member
White dolomite marbles
Mvm077 Medium-crystalline, massive – – – – 0.10 13.0 21.2 2360 69 89 0.61
Mvm078 Medium-crystalline, massive – – – – – 12.9 20.9 2290 72 86 0.62
Mvm079 Medium-crystalline, massive 0.23 – – – – 12.8 20.7 1100 48 91 0.62
Mvm080 Medium-crystalline, massive – – – – – 13.1 21.0 1180 54 84 0.62
Mvm082 Medium-crystalline, massive 1.8 – – – – 12.5 20.6 863 36 83 0.61
Mvm081 Medium-crystalline, massive 0.52 – – – – 12.2 21.6 507 26 90 0.56
Mvm083 Medium-crystalline, massive 2.1 – – – – 11.3 21.2 2160 70 92 0.53
Mvm084 Medium-crystalline, massive 4.4 – – – – 11.6 20.0 522 20 61 0.58
BB040 Medium-crystalline, massive 4.7 – – – – 11.1 20.1 976 55 121 0.55
BB041 Medium-crystalline, massive – – – – – 12.5 20.5 1210 84 96 0.61
BB042 Medium-crystalline, massive – – – – – 12.5 20.9 829 36 81 0.60
Marbles sampled outside the quarries
Calcite marbles
BB022 White, coarsely-crystalline – – – 0.01 0.17 0.34 37.2 155 17 4120 0.01
TBM-143A Pale grey, medium-crystalline 1.8 0.14 – 0.040 n.d. 0.38 37.2 715 41 1070 0.01
(–) Below detection limits: 0.01 wt% for K2 O; 0.1 wt.% for SiO2 , Al2 O3 , Na2 O and Corg . Mg, Ca, Fe, Mn, and Sr are acid soluble constituents determined by ICP–AES.
 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558 547

Fig. 3. Cross-plots illustrating major geochemical and isotopic features of the tripartite marble unit of the Montepuez Complex.

Marbles of the Xixano and Nampula complexes commonly transitional position between the two end-members (Table 1).
contain a few percent of SiO2 and show variable degrees of The Mn and Fe concentrations in the Montepuez West and
dolomitisation (Table 2). Some varieties contain sizable amounts East calcite marbles are 14–41 ppm and 155–755 ppm, respec-
of Al2 O3 and TOC. The marble of the Ocua Complex is devoid tively. The dolomite and calcite marbles of the Nampula and
of SiO2 and Al2 O3 and is dolomitic in composition. The Lalamo Ocua complexes are enriched in Mn (Table 2) relative to those
Complex contains numerous lenses of a relatively pure cal- in the Montepuez Complex. The Xixano and Lalamo calcite
cite marble containing 0.6 to 6% SiO2 . The magnesite marble marbles are rather similar to the Montepuez Complex marbles.
(Mg/Ca = 1.1) has a low SiO2 content (1.8%) and is devoid of The Lalamo magnesite marble has low Mn (23 ppm) and Fe
Al2 O3 . (250 ppm) concentrations.

4.3. Iron and manganese 4.4. Strontium

In the tripartite marble unit at Montepuez, average Mn In the tripartite marble unit at Montepuez, the Sr concen-
(47 ± 23 ppm, n = 43) and Fe (1208 ± 623 ppm) contents of tration in the CM is low (142 ± 17 ppm, n = 7) and shows no
the DM are much lower than those commonly reported from significant correlation with the Mg/Ca ratio (Fig. 3c). Con-
Proterozoic dolostones (Mn = 180 ppm, Fe = 2020 ppm, e.g., centrations of Sr and Mn are both low, and so is the Mn/Sr
Melezhik et al., 2005). The CM contains less Mn (16 ± 7.3 ppm, ratio (0.11 ± 0.04, n = 7). The DCM and DM contain less Sr
n = 7) and Fe (327 ± 122 ppm), whereas the DCM occupies a (88 ± 15 ppm, n = 6; 86 ± 15 ppm, n = 43, respectively), with
 548
Table 2
Chemical composition of marbles from from Xixano, Lalamo, Ocua and Nampula complexes in NE Mozambique
Sample # Lithology SiO2 (%) Al2 O3 (%) Na2 O (%) K2 O (%) TOC (%) Mg (%) Ca (%) Fe (ppm) Mn (ppm) Sr (ppm) Mg/Ca

Xixano Complex
Calcite marble
IH04.027 Pale grey, medium-crystalline 1.2 0.29 – 0.04 – 0.40 36.2 798 81 1140 0.01

V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558

Calcite, partially dolomitised marbles
JS285 Impure, tremolite-bearing 7.6 – – 0.038 n.d. 2.7 29.4 236 117 58 0.09
JS299 Impure, tremolite-bearing 14.3 – – 0.013 n.d. 4.0 22.5 225 65 70 0.18
Calcite, intensively dolomitised marble
TBM126 Pale grey, medium-crystalline 2.2 – – 0.015 n.d. 10.2 22.9 373 18 88 0.45
Lalamo Complex
Calcite marble
JS200 Marble 0.57 0.05 – 0.03 0.41 0.22 36.8 36.4 15 2160 0.01
Calcite, partially dolomitised marble
JS303 Impure, tremolite-bearing 6.1 – – 0.037 n.d. 1.1 33.7 63.3 14 n.d. 0.03
Magnesite marble
JS307 Pink, brecciated, dolomitised 1.8 – – – n.d. 10.5 9.9 249 23 235 1.06
Ocua Complex
Dolomite marble
JS330 Grey, medium-crystalline – – – 0.003 n.d. 12.1 21.3 331 150 139 0.57
Nampula Complex
Calcite, partially dolomitised marbles
TBM097 Grey, medium-crystalline – – – – 0.36 3.78 31.9 736 271 234 0.12
TBM098 Pale grey, medium-crystalline 1.5 0.21 – 0.17 – 1.53 34.0 1030 187 239 0.05
RB823 Impure, tremolite-bearing 5.3 1.04 – 0.39 n.d. n.d. n.d. n.d. n.d. n.d. n.d.
Calcite, intensively dolomitised marbles
TBM-099 Impure, tremolite-bearing 7.9 1.24 0.19 0.19 – 6.07 22.4 1590 230 347 0.27
TBM-105 Pale grey, medium-crystalline 1.3 – – 0.02 0.34 8.65 24.3 792 186 189 0.36
AS04-26 Pale grey, medium-crystalline 2.1 0.09 – 0.01 0.33 6.62 26.3 321 201 163 0.25

(–) Below detection limits of 0.002 for MnO, 0.003 for K2 O, 0.004 for TiO2 , 0.01 for SiO2 , Al2 O3 , Stot, P2 O5 , 0.1 for Na2 O and TOC.
 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558 549

maximum concentrations of 112 and 121 ppm, thus approaching The Montepuez West marble has a 87 Sr/86 Sr ratio of 0.70566,
only the lowest concentration in the CM (Fig. 3c). The aver- which is more radiogenic than the calcite marbles in the tripartite
age Sr concentration of the DM is higher than that reported for unit. The Montepuez East calcite marble has 87 Sr/86 Sr ratio of
non-metamorphosed Proterozoic dolostones (54 ppm, n = 344; 0.70654, which is the most radiogenic of all in the Montepuez
Melezhik et al., 2005). Mn/Sr ratios range between 0.1 and 1.5 Complex.
averaging 0.58 ± 0.33 (n = 43). In contrast, the Montepuez West Among marbles of other complexes, the Xixano calcite
and East calcite marbles are rich in Sr (1070 and 4120 ppm, marbles show the greatest spread in 87 Sr/86 Sr ratios (0.70574
respectively). The dolomitised calcite and dolomite marbles of to 0.70666), with no dependence on the Sr concentration. In
other complexes are characterised by moderate to low Sr con- contrast, 87 Sr/86 Sr ratios of the Lalamo calcite marbles corre-
centrations. However, pure calcite marbles of the Xixano and late negatively with the Sr concentrations within the range of
Lalamo complexes contain 1140 and 2160 ppm Sr, respectively 0.70627–70699 (Table 4). The magnesite has the most radio-
(Table 2). The Lalamo magnesite has a low Sr content (235 ppm). genic value of 0.70699.

4.5. Oxygen and carbon isotopes

5. Geochemical screening for postdepositional resetting
of C-, O- and Sr-isotope systems
Carbon and oxygen isotope data from the tripartite marble
unit at Montepuez comprise 24 analyses (Table 3). The δ18 O
In general, the carbon isotopes are strongly buffered by the
values of the CM range between 14.6 and 20.3‰ (17.2‰ on
high C concentrations in carbonate minerals relative to fluids
average, n = 7). The δ13 C values are all positive and cluster
(Banner and Hanson, 1990; Jacobsen and Kaufman, 1999) and,
tightly between +1.6 and +2.1‰ (+1.9 ± 0.2‰ on average)
consequently, infiltration of externally derived fluids is likely
(Table 3). There is a very weak positive correlation between
to have a relatively greater effect on O and Sr isotopes. Tra-
δ18 O and δ13 C (r = +0.53, n = 7, >80%). The DM is enriched in
18 O with respect to the CM (Fig. 3d) with δ18 O values tightly ditional screening procedures for postdepositional alteration of
carbonate, originally specified for non-metamorphosed or low-
clustering around 22.1 ± 0.4‰ (n = 12). δ13 C fluctuates between
grade rocks (Brand and Veizer, 1980; Veizer, 1983; Banner
−1.7 and +1.8‰ (Fig. 3e), and low values are associated with
and Hanson, 1990; Denison et al., 1998, 1994; Kaufman and
high TOC content (Fig. 3f); all the samples with negative δ13 C
Knoll, 1995; Azmy et al., 1998; Jacobsen and Kaufman, 1999),
values contain graphite. There is no correlation between δ18 O
are of very limited use in the case studied here. In high-grade
and δ13 C. The DCM shows both δ13 C (0.4–1.7‰) and δ18 O
metamorphic rocks, such screening could only detect the latest
(19.3–22.6‰) transitional from the CM to the DM.
geochemical alteration. For these reasons, the discrimination
The sample obtained from the Montepuez West calcite marble
technique has been based essentially on geochemical criteria,
is characterised by δ13 C (+0.1‰) and δ18 O (16.7‰), which are
namely on relative abundances of Mn, Rb and Sr (e.g. Brand
within the range of the tripartite marble unit. In contrast, the
and Veizer, 1980). Elemental ratios, such as Mn/Sr, Fe/Sr, Ca/Sr
Montepuez East calcite marble shows much higher values for
and Rb/Sr, as well as δ18 O values, are widely used as geo-
both δ13 C (+4.7‰) and δ18 O (24.7‰).
chemical criteria for detecting the least disturbed C-, O- and
The Xixano calcite marbles have δ13 C values fluctuating
Rb-Sr systems. Very variable values and dissimilar combina-
between −1.1 and +3.8‰, whereas δ18 O ranges from 15.1 to
tions of the ratios have been used in the past (Asmerom et al.,
22‰ (Table 4). The Lalamo calcite marbles are characterised by
1991; Derry et al., 1992; Kaufman et al., 1993; Gorokhov et
the highest δ13 C (+3.5 to +7.1‰) and δ18 O (26.1–28.4‰) val-
al., 1995; Semikhatov et al., 2002; Kuznetsov et al., 2003). In
ues. Similarly, the magnesite marble has a high δ18 O (24.3‰)
all cases, the choice is empirical and to some extent arbitrary.
whereas δ13 C is negative (−1.1‰). The Nampula calcite mar-
The published data on non-metamorphosed limestones suggest
bles are uniformly depleted in 13 C (δ13 C = −3.5 to −2.0‰) and
Mn/Sr = 0.065–0.02 and Mg/Ca < 0.02, whereas a significant
have δ18 O in the range of 18.6–20.8‰.
database obtained from high-grade calcite marbles (Melezhik
et al., 2003) suggests Mn/Sr < 0.02, Mg/Ca < 0.02. However,
4.6. Strontium isotope ratios such values have been chosen empirically during intensive stud-
ies in the Scandinavian Caledonides. The values may well be
In the tripartite marble unit at Montepuez, strontium iso- different in cases where the rocks experienced a different geo-
tope ratios have been obtained from 23 samples (Table 3). logical history, and should therefore be used with caution. For
87 Sr/86 Sr ratios measured from the CM range between 0.70508 instance, Pili et al. (1997a, b) performed isotopic studies on mar-
and 0.70521 (n = 7), whereas those obtained from the DCM are bles from a large-scale granulite section in Madagascar, from
more radiogenic and fluctuate between 0.70531 and 0.70560 geological environments which are rather similar to those in
(n = 4). The DM has the most radiogenic 87 Sr/86 Sr ratios ranging the study area. They reported that marbles from major shear
from 0.70540 to 0.70565 (n = 12) (Fig. 3h) with one outstanding zones are depleted in 13 C down to −3‰ due to exchange with
outlier at 0.70488. The 87 Sr/86 Sr ratios of the CM show a sig- mantle-derived carbon. Marbles from minor shear zones may
nificant negative correlation with the Sr concentration (Fig. 3g) show significant isotopic variations at the metre scale reflect-
(r = −0.66, n = 7, 90%) and do not exhibit a correlation with the ing the heterogeneous distribution of fluid flow associated with
Mg/Ca ratio. variable permeabilities. However, even granulite-grade marbles
 550
Table 3
Minerals in insoluble residue, Rb and Sr concentrations and isotopic ratios of marbles from the Montepuez Complex

Sample # Minerals in insoluble residue Mg/Ca Mn/Sr δ13 C (‰) δ18 O (‰) Rb (ppm) Sr (ppm) 87 Rb/86 Sr 87 Sr/86 Sr (measured) 87 Sr/86 Sr (initial)a

Montepuez quarry 4, UTM 0497505, 8552297

Calcite-dolomite marble member
Calcite marbles, partially dolomitised
Mvm040 – 0.09 0.17 +2.0 17.9 0.12 170 0.0021 0.70510 0.70507
Mvm041 Mu, Am (Qu, Tlc) 0.05 0.11 +1.7 18.9 0.14 169 0.0024 0.70516 0.70512
Mvm042 – 0.02 0.04 +2.1 20.3 0.13 123 0.0031 0.70521 0.70516
Mvm043 – 0.07 0.10 +2.0 16.8 0.11 144 0.0022 0.70514 0.70511
Mvm044 – 0.04 0.15 +1.6 14.6 0.12 150 0.0023 0.70508 0.70504
Mvm045 (Mu) 0.07 0.08 +1.9 15.1 0.13 145 0.0026 0.70518 0.70514
BB037 – 0.07 0.13 +1.8 16.6 0.09 128 0.0021 0.70520 0.70517
Montepuez quarry 3, UTM 0497650, 8552251

V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558

Grey dolomite marble member
Calcite marble, intensively dolomitised
Mvm052 Am, Qu 0.46 1.09 +0.4 19.3 0.18 83 0.0063 0.70531 0.70521
Dolomite marbles
Mvm051 Am, Qu 0.53 0.23 −1.5 22.7 0.22 122 0.0053 0.70565 0.70557
Mvm050 Am, Qu 0.55 0.26 −1.7 21.7 0.09 110 0.0024 0.70561 0.70557
Mvm053 Am (Tlc) 0.55 0.42 +1.4 22.4 0.12 106 0.0033 0.70546 0.70541
Mvm054 Am 0.55 0.32 −1.6 22.0 0.12 78 0.0045 0.70553 0.70546
Mvm056 – 0.59 0.50 −0.4 22.7 0.14 85 0.0048 0.70555 0.70547
Montepuez quarry 2, UTM 0497683, 8552084
White dolomite marble member
Calcite marble, intensively dolomitised
BB031 – 0.50 0.097 +1.8 22.0 0.07 124 0.0017 0.70540 0.70537
Montepuez quarry 1, UTM 0497684, 8552072
White dolomite marble member
Calcite marble, intensively dolomitised
Mvm065 Am 0.37 0.16 +1.3 22.2 0.14 80 0.0051 0.70561 0.70553
Mvm066 Am 0.46 0.24 +1.5 22.6 0.1 99 0.0030 0.70560 0.70555
Mvm075 Am 0.55 0.55 +1.7 21.4 0.08 116 0.0020 0.70551 0.70548
Dolomite marbles
Mvm061 – 0.56 0.36 +0.7 19.5 0.11 108 0.0030 0.70488 0.70483
Mvm064 Am 0.54 0.47 +1.6 22.6 0.09 92 0.0029 0.70565 0.70560
Mvm067 Am (Tlc) 0.56 0.48 +1.4 22.0 0.09 104 0.0025 0.70548 0.70544
Mvm070 – 0.53 0.38 +1.7 22.2 0.09 113 0.0023 0.70556 0.70552
Mvm071 Qu 0.58 0.34 +1.8 21.8 0.07 82 0.0025 0.70562 0.70558
Mvm072 Am 0.55 0.11 +1.8 21.2 0.08 102 0.0023 0.70552 0.70548
Mvm081 Tlc, Am, Qu 0.56 0.29 +1.3 21.7 0.06 92 0.0019 0.70556 0.70553
Marbles sampled outside the quarries
Calcite marbles
BB022 – 0.01 0.004 +0.1 16.7 0.23 4125 0.0002 0.70566 0.70566
TBM143A Qu, Fsp 0.01 0.04 +4.7 24.7 1.31 1040 0.0037 0.70654
a For marbles sampled in quarries, the initial strontium ratio was calculated under the assumption that the age of these rocks is equal to 1200 Ma; for other samples—equal to 750 Ma. Abbreviations: Am –

amphibole; Fsp – feldspar; Mu – muscovite; Qu – quartz; Tlc – talc. Rb and Sr concentrations were determined by standard isotope dilution and solid-source mass spectrometry.
Table 4
Minerals in insoluble residue, Rb and Sr concentrations and isotopic ratios of marbles from Xixano, Lalamo, Ocua and Nampula complexes in NE Mozambique

Sample # Minerals in insoluble residue Mg/Ca Mn/Sr δ13 C ‰ δ18 O Rb Sr 87 Rb/86 Sr 87 Sr/86 Sr (measured) 87 Sr/86 Sr (initial)a

Xixano Complex
Calcite marble

V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558

IH04.027 (Mu) 0.01 0.07 −1.1 15.1 0.25 1190 0.0006 0.70616 0.70615
Calcite, partially dolomitised marbles
JS285 Am, Mu 0.09 2.0 +2.4 22.0 0.25 60.6 0.0121 0.70666 0.70653
JS299 Px 0.18 0.9 −0.2 19.0 0.29 73.5 0.0116 0.70574 0.70562
Calcite, intensively dolomitised marble
TBM-126 Am (Tlc, Qu, Mu) 0.45 0.2 +3.8 19.0 0.49 82 0.0175 0.70631 0.70612
Lalamo Complex
Calcite marble
JS200 Grph, Am, Mu, Fsp 0.01 0.007 +7.1 26.1 0.12 2105 0.0002 0.70671 0.70671
Calcite, partially dolomitised marble
JS303 Qu, Am, Px 0.03 n.d. +3.5 28.4 0.73 3930 0.0005 0.70627 0.70626
Ocua Complex
Dolomite marble
JS330 – 0.57 1.1 +0.4 23.1 0.22 130 0.005 0.70684 0.70679
Nampula Complex
Calcite, partially dolomitised marbles
TBM-097 – 0.12 1.16 −3.5 19.6 0.04 229 0.0005 0.70594 0.70593
TBM-098 n.d. 0.05 0.78 −2.0 19.9 n.d. n.d. n.d. n.d. n.d.
Calcite, intensively dolomitised marbles
TBM-099 n.d. 0.27 0.66 −2.8 18.6 n.d. n.d. n.d. n.d. n.d.
TBM-105 n.d. 0.36 0.98 −2.9 20.8 n.d. n.d. n.d. n.d. n.d.
AS04-26 n.d. 0.25 1.23 −2.9 20.1 n.d. n.d. n.d. n.d. n.d.
a The initial strontium ratio was calculated under the assumption that the age of these rocks is equal to 750 Ma. Abbreviations: Am – amphibole; Fsp – feldspar; Grph- graphite; Mu – muscovite; Px – pyroxene;

Qu – quartz; Tlc – talc. Rb and Sr concentrations were determined by standard isotope dilution and solid-source mass spectrometry.

551
552 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558

outside shear zones have carbon and oxygen isotopic composi- Sample Mvm061 of the dolomite marble, the outlier, yielded
tions similar to their protoliths (Pili et al., 1997a, b). an exceptionally low 87 Sr/86 Sr ratio of 0.70488. This is coupled
with depletion in 18 O, and therefore, has no obvious explanation
5.1. Montepuez Complex marbles in terms of preservation potential. The sample was obtained from
the unit containing abundant dismembered amphibolite bod-
The Montepuez CM from the tripartite unit shows significant ies. The amphibolite occurring in the quarries is characterised
fluctuations in δ18 O and a weak positive correlation between by a low 87 Sr/86 Sr ratio (0.70403). The amphibolite contains
δ18 O and δ13 C, which are indicative of postdepositional reset- 320 ppm Sr (Table 5), which is three times higher than in the
ting. Based on the observed alteration trends (Fig. 3e, d and f), dolomite marbles. Thus, the most probable explanation for the
we tentatively suggest that +2‰ and 20‰ represent the least- low 87 Sr/86 Sr in the outlier is an isotopic exchange between the
altered δ13 C and δ18 O values, respectively. In the DM, low δ13 C amphibolite and the dolomite. The low ␦18 O value is consistent
is associated with high TOC content (Fig. 3f), and such an with this model.
effect can be attributed to a local isotopic exchange between The Montepuez West marble shows similarities to those
organic matter/graphite and dolomite. There is no correlation studied in the quarries, whereas the Montepuez East marble is
between δ18 O and δ13 C, and the overall δ18 O values show an distinctly different in that its δ13 C (+4.7‰) and δ18 O (24.7‰)
overlap with those reported from non-metamorphosed, Protero- values are enriched in heavy isotopes. These enriched values plot
zoic dolostones (e.g., Melezhik et al., 2005). This may suggest outside any alteration trends associated with the tripartite marble
a high degree of preservation of original δ18 O values in the DM. unit and therefore cannot be explained as the result of a higher
The observed alteration trend (Fig. 3d and f) suggests +1.8‰ degree of preservation. We tentatively suggest that the calcite
and 22‰ as the best proxies for initial ␦13 C and ␦18 O values, was precipitated from seawater with different C- and O-isotopic
respectively. composition.
In the tripartite unit, the 18 O depletion of the CM relative The Montepuez West and East calcite marbles are distin-
to the DM can be attributed to a sequence of events. The par- guished by higher 87 Sr/86 Sr ratios (Table 3). Because these
tial dolomitisation of limestones commonly creates enhanced rocks have Sr concentrations (1040–4125 ppm) one order
porosity and permeability, whereas intensively and completely of magnitude higher than that of the CM in the tripar-
dolomitised rocks become largely impermeable. As the result, tite unit, their more radiogenic strontium isotope ratios are
the partially dolomitised rocks should be more sensitive towards very unlikely to be attributable to postdepositional alterations.
exchange of their oxygen isotopes with infiltrating postdeposi- Instead, they apparently represent marbles whose carbonate
tional fluids, whereas the δ13 C would be sufficiently buffered by precursor was originally precipitated from seawater with a dif-
the dissolving calcite precursor (cf. Banner and Hanson, 1990; ferent isotopic composition. This is particularly true for the
Land, 1992). Because diagenetic and metamorphic alterations Montepuez East 13 C-rich marble. As far as the Montepuez
affect carbonate material in a similar way (Nabelek, 1991), we West marble is concerned, it was certainly derived from an
are not in a position to distinguish between the two. aragonite precursor as indicated by the very high Sr abun-
In the tripartite marble unit, the strontium isotope ratios of dance. This suggests that the latter may represent not only
the CM are less radiogenic than DCM, and both have less a different age but also that it may signify a different depo-
radiogenic ratios in comparison with the DM (excluding one sitional setting or a different postdepositional geochemical
outlier, Fig. 3h). Such features can be reconciled if the dolomite history.
exchanged Sr with dolomitising fluids. This inference is in
agreement with the relative enrichment of the dolomite in 18 O. 5.2. Marbles of other complexes
However, significant negative correlation between 87 Sr/86 Sr
ratios and the Sr concentration in the CM (Fig. 3g), and the A limited database on marbles (Table 4) sporadically occur-
absence of 87 Sr/86 Sr correlation with the Mg/Ca ratio, sug- ring in several other complexes (Fig. 1) does not allow for robust
gest that the CM were affected by an additional episode of screening against postdepositional resetting of C, O and Rb/Sr
postdepositional alteration, which was not associated with the isotope systems. However, several samples are characterised by
dolomitisation. Based on the reasoning outlined above and con- exceptionally high Sr concentrations, which should provide a
sidering all observed alteration trends (Fig. 3g and h), the least sufficient buffer against postdepostional resetting. Such sam-
radiogenic 87 Sr/86 Sr ratio of 0.70508 is tentatively suggested as ples can be considered as retaining 87 Sr/86 Sr ratios close to
the least altered value for the CM. However, the relatively low depositional values. In the Xixano Complex, this applies to
Sr concentration in the CM was probably not sufficient to pro- the South marble lens (IH04.027, Fig. 1) having 1190 ppm Sr
vide a strong buffer for the Sr isotope system against multiple and 87 Sr/86 Sr of 0.70616. Interestingly, the Xixano North mar-
alterations, although this is currently impossible to quantify. ble (JS299, Fig. 1), has a significantly lower Sr concentration
Table 5
Chemical composition and strontium isotope ratio of amphibolite boudin in marbles from the Montepuez quarry

Sample # SiO2 (%) Al2 O3 (%) Fe2 O3 (%) TiO2 (%) MgO (%) CaO (%) Na2 O (%) K2 O (%) Rb (ppm) Sr (ppm) 87 Rb/86 Sr 87 Sr/86 Sr

BB036 40.7 15.9 15.6 3.63 6.37 11.2 2.41 0.16 1.18 320 0.0118 0.70403
 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558 553

(73 ppm) combined with the lower 87 Sr/86 Sr ratio of 0.70574,

which cannot be attributed to a postdepositional resetting.
Two samples, representing different marble units in the
Lalamo Complex, namely the Centre and South marbles (Fig. 1),
yielded very high Sr concentrations (3939 and 2102 ppm), but
however have different 87 Sr/86 Sr ratios (Table 4). Because of
strong buffering abilities against postdepositional alterations,
both ratios are probably very close to depositional values; thus
calcite in these two marbles was very likely precipitated from
isotopically different seawaters.
The Nampula marbles have negative δ13 C values, which dis-
tinguish them from other marbles in the study area. The δ13 C
values show a very limited spread, implying that they may reflect
the isotopic composition of seawater. Importantly, despite a low
Sr concentration, the 87 Sr/86 Sr ratio (0.70594) is much lower in
comparison to the Sr isotope ratios of the Sr-rich Lalamo and
Xixano marbles. This again suggests that the Nampula carbon-
ates reflect the isotopic composition of a different seawater.

6. Implications for isotope chemostratigraphy

Strontium isotope stratigraphy has the potential to pro-

vide indirect dating of Neoproterozoic sedimentary carbonates
because of the gradual, though irregular, increase in 87 Sr/86 Sr
from c. 850 to c. 550 Ma (Fig. 4); and also because Sr has a
long residence time relative to seawater, resulting in a uniform
87 Sr/86 Sr ratio in seawater at any given time (Veizer et al., 1999).

However, sampling of carbonate sequences through Neopro-

terozoic time remains incomplete and precise radiometric ages
are not always available. Because of this, there have been sev-
eral disagreements among investigators as to how to construct
the seawater reference curve (see discussion in Melezhik et al.,
2001b and Kuznetsov et al., 2003, 2005). The result of the dis-
cussion was the reference curve (Fig. 4) which has been used in
this study.
In the case of high-grade metamorphic rocks, when the Sr
concentration in carbonates is low, the Rb/Sr system might have
been reset while the carbon isotope system may still retain
the original ratio. This is because the sediment/water ratios
of diagenetic-metamorphic systems (cf. Banner and Hanson,
1990) are about 101 , 102 –103 and 104 for oxygen, strontium
and carbon, respectively (cf. Land, 1992). Therefore, although
the strontium isotope ratios of carbonates are a better proxy for
seawater isotopic composition, δ13 C may provide a better result
for low-Sr marbles.
The apparent depositional ages of the studied rocks have
been constrained by projection of the ‘least altered’ isotopic
values obtained from the marble formations onto the δ13 C and
the 87 Sr/86 Sr seawater reference curves. The time when both
the δ13 C and the 87 Sr/86 Sr intercepts are in agreement has been
considered to represent an apparent depositional age (Fig. 4).

Fig. 4. Temporal trends of 87 Sr/86 Sr and ␦13 C in seawater and apparent deposi-
tional ages of the studied marbles from NE Mozambique. The 87 Sr/86 Sr reference and Veizer (2002 and references therein) (green symbols). The least altered
curve (green) is based on Melezhik et al. (2001 and references therein) and 87 Sr/86 Sr δ13 C ratios of the marble units are shown by horizontal lines. Vertical,
Kuznetsov et al. (2003 and references therein) (red symbols) modified by using arrowed-head lines indicate apparent depositional ages (see discussion in the
data from Kuznetsov et al. (2005, 2006) and the database compiled by Shields text).
554 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558

6.1. The Montepuez Complex marbles 660 Ma. The younger intercept should not be considered for the
reason explained above. The least altered 87 Sr/86 Sr of 0.70574 is
The upper age limit for sedimentation within the Montepuez significantly lower than the ratio in the South marble. However,
Complex can be constrained by the penetrative 580–530 Ma the low Sr concentration (73 ppm) suggests that the Rb/Sr sys-
high-grade metamorphism (e.g., Bingen et al., 2006a, b, 2007). tem might have experienced a postdepositional alteration, and
However, the lower age limit cannot be constrained by any, therefore 800, and 750 Ma should be considered as the upper
currently available, published data. Thus, within limitations of limit for an apparent depositional age.
‘isotope stratigraphy’, one should initially consider a wide time In the Lalamo Complex, both the Centre and South marbles
interval from the Archaean up to 580 Ma. Within this time span, (Fig. 1) are exceptionally rich in Sr and their 87 Sr/86 Sr ratios may
the least altered initial ratio 87 Sr/86 Sr of 0.70504 projected onto therefore be a robust proxy for coeval seawater. The 87 Sr/86 Sr
the seawater curve sets the apparent depositional age of the tri- of 0.70627 and δ13 C of +3.5‰ for the Lalamo Centre marble
partite marble as Mesoproterozoic at around c. 1250 Ma (Fig. 4). has a joint intercept at 740 or 670 Ma. Although both 87 Sr/86 Sr
The global carbon isotope records offer additional con- (0.70671) and δ13 C (+7.1‰) ratios of the Lalamo South marble
straints. Kah et al. (1999) reported that δ13 C values of −1.0 differ, they provide a similar apparent depositional age at 740 or
to +4.0‰ are characteristic for the interval between approxi- 670 Ma (Fig. 4). As in the Xixano Complex, the younger inter-
mately 1300 and 800 Ma. This pattern is distinct from that of cept at 670 Ma is in conflict with magmatic ages ranging between
older Mesoproterozoic successions, which record values near 753 ± 10 and 696 ± 13 Ma (Bingen et al., 2006a, b, 2007), and
0‰, and suggests that the 1300–800 Ma moderately positive thus should not be considered as an apparent depositional age.
values may be useful for broad time correlation. In the late All marbles sampled in the Nampula Complex have rel-
Palaeoproterozoic (1800–1500 Ma) the secular δ13 C curve is atively low Sr concentration (up to 350 ppm, Table 3). The
flat, relatively monotonous and fluctuates close to zero much like limited database (Table 4) does not allow reliable discrimination
that seen in the Mesoproterozoic (Lindsay and Brasier, 2002). against postdepositional resetting of the Rb/Sr system. Conse-
Thus, considering all limitations, ages older than 1300 Ma are quently the 87 Sr/86 Sr ratio of 0.70594 obtained from one of the
not credible, as the Montepuez carbonates have slightly elevated Nampula Complex marbles may not represent a robust proxy
δ13 C values fluctuating between +1.6 and +2.0‰, thus distinct for coeval seawater. However, the ratio can set an upper age
from those close to zero. Moreover, Palaeoproterozoic carbon- limit for deposition. Several intercepts on the 87 Sr/86 Sr refer-
ates are characterised by a much lower 87 Sr/86 Sr (Shields and ence curve can be reconciled with δ13 C of −3.5‰ at c. 750 Ma
Veizer, 2002). (Fig. 4). The age cannot be older than 800 Ma, because δ13 C
The currently available Sr-isotope database includes a rather values significantly below −1.0‰ are not characteristic of the
limited number of analyses involved in the constructed reference 1300–800 Ma interval (Kah et al., 1999) and the late Palaeopro-
curve. Therefore, there is chance for more fluctuations than it is terozoic (1800–1500 Ma) (Lindsay and Brasier, 2002).
shown in Fig. 4, however, combined Sr- and C-isotope data sug-
gest that an apparent depositional age cannot be younger than 7. Implications for tectonostratigraphy and basin
910 Ma. The relatively low Sr concentrations in the Montepuez evolution
marbles might not have provided sufficient buffer against post-
depositional resetting of the Sr isotope system. Consequently, Apparent carbonate depositional ages obtained by isotope
the apparent depositional age interval of 1250–910 Ma should chemostratigraphy (Table 6) can provide insight into basin evo-
be considered as a minimum age. lution during the Mezo-Neoproterozoic in NE Mozambique,
The 87 Sr/86 Sr (0.70566) and δ13 C (+0.1‰) values obtained which cannot be done at present by any other means. However,
the Montepuez West marble, when projected onto the seawa- apparent ages should be treated with caution for two reasons: (i)
ter curve, give several interceptions; however, both ratios agree a possible, unresolved, postdepositional resetting; and (ii) a pos-
at 800, 750 and 660 Ma, which can be considered as apparent sible bias towards exposed and selectively sampled marble units.
depositional ages. The Montepuez East marble has 87 Sr/86 Sr of Therefore, initial 87 Sr/86 Sr ratios might have been lower, and so
0.70654. This, combined with δ13 C of +4.7‰, corresponds to a apparent depositional ages older, and the age coverage obtained
740 or 700–660 Ma seawater (Fig. 4). for each complex might be incomplete. Further, we consider only
complexes that contain marbles for which apparent depositional
6.2. Marbles of other complexes ages have been obtained.
The apparent carbonate depositional ages cluster in six
In the Xixano Complex only the South marble (IH04.027) groups (Table 6), namely (1) 1250–910 Ma (tripartite marble unit
could be considered to retain an 87 Sr/86 Sr ratio close to the of the Montepuez Complex); (2) 800 or 750 Ma (Xixano North
depositional value due to the high Sr concentration (1190 ppm). marble); (3) 800, 750 or 660 Ma (Montepuez West marble); (4)
The 87 Sr/86 Sr of 0.70616 and δ13 C of −1.1‰ intersects on the c. 750 Ma (Nampula marble); (5) c. 740 Ma (Xixano South and
seawater reference curve agree at 740, 720 and 670 Ma. Two Lalamo Centre and South marbles); and (6) 740, or 700–670 Ma
younger intercepts should not be considered because the Xix- (Montepuez East marble). All groups, except group 1, have over-
ano Complex underwent metamorphism at 735 ± 5 Ma (Bingen lapping apparent ages. The apparent depositional ages of groups
et al., 2006a, b, 2007). The 87 Sr/86 Sr and ␦13 C of the North 2, 3 and 6 are tentatively considered undistinguishable. Although
Xixano marbles intercepts the reference curve at 800, 750 and the isotopic composition of the Nampula marble (group 4) has
 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558 555

Table 6
Summary of the least altered 87 Sr/86 Sr and δ13 C ratios, and apparent depostional ages of marble units in NE Mozambique

Complex Occurrence 87 Sr/86 Sr ␦13 C (‰) Age (Ma) Age-group

Montepuez Montepuez quarries 0.70504 +2.0 c. 1200 1

Xixano Xixano North 0.70574 −0.2 800 or 750 2
Montepuez Montepuez West 0.70566 +0.1 800, 750 or 660 3
Nampula Nampula 0.70594 −3.5 c. 750. 4
Xixano Xixano South 0.70616 −1.1 c. 740 5
Lalamo Lalamo Centre 0.70627 +3.5 c. 740 5
Lalamo Lalamo South 0.70671 +7.1 c. 740 5
Montepuez Montepuez East 0.70654 +4.7 700–670 or 740 6

a unique resolution at c. 750 Ma, this apparent depositional age view of the very limited sedimentological evidence, the robust
overlaps with those of the Xixano North (group 2) and Mon- reconstruction of depositional settings is not possible.
tepuez West (group 3) marbles. Similarly, the c. 740 Ma Xixano The 800–660 Ma (Montepuez West) and 740–670 Ma (Mon-
South and Lalamo Centre and South marbles (group 5) overlaps tepuez East) marble units are composed of Sr-rich calcite,
with the apparent depositional age of the Montepuez East mar- suggesting an aragonite precursor. The marbles of these two
ble (group 6). Individual marble units within the c. 750 and c. units show considerable isotopic differences (Table 3). The sig-
740 Ma groups are rather inhomogeneous in terms of 87 Sr/86 Sr nificant enrichment of the East marble in 13 C suggests enhanced
ratios, highly variable in Sr abundance; the Nampula marble is burial of organic matter coeval with the carbonate deposition.
significantly depleted in 13 C, whereas the Lalamo South marble Overall, the isotopic dissimilarities of the East and West mar-
is considerably enriched in 13 C (Table 6). All these differences bles, combined with a considerable spatial separation by thick
suggest that their carbonates originally precipitated from differ- piles of gneisses and schists, would argue for their accumu-
ent seawater, and that the depositional ages of individual marble lation in two unrelated basins. However, care must be taken
units might be also different. This is, however, impossible to in comparison of the carbon and strontium isotopic records
resolve within the limitations of the isotope chemostratigraphy due to the greatly different residence times of C (∼105 years)
approach (Fig. 4) and Sr (∼3 × 106 years) in oceanic waters, and consequently
their respective response time to perturbations (e.g., Derry et
7.1. The Montepuez Complex al., 1992; Richter and Turekian, 1993). The greatest temporal
change confidently documented in seawater 87 Sr/86 Sr is at a rate
The Montepuez Complex contains marble units showing of ∼69 × 10−6 Myr−1 that has occurred between 40 and 25 Ma
at least two different depositional ages. The oldest age of in the Cenozoic (e.g., Elderfield, 1986). If such a rate is applied
1250–910 Ma was obtained from the tripartite marble unit at to the two youngest carbonate units, the required difference in
Montepuez quarries. The two younger marble units, one with their time of deposition should be not less than 10 Ma. How-
the apparent depositional ages at c. 800, 750 or 660 Ma (West ever, this can neither prove nor disprove whether or not the two
marble), and the other at c. 740 or between 700–670 Ma (East younger marbles belonged to two separated depositional cycles.
marble), have carbon and Sr isotopic compositions that cor- We tentatively imply isolated carbonate platform settings for the
respond to significantly different seawaters (Table 6). From two younger marble units of the Montepuez Complex because
the tectonostratigraphic implications, it is important to note they are composed of high-Sr, pure, calcite marbles. However,
that the 800–660 Ma marble is closely linked spatially to the limited nature of the geological information available is
the 1250–910 Ma marbles exposed in the Montepuez quarries acknowledged.
(Fig. 1).
The 1250–910 Ma marbles represent the oldest dated carbon- 7.2. The Xixano Complex
ate rocks in the Montepuez Complex. No depositional basement
to the marbles is currently known at the present day. The The Xixano Complex includes two marble units whose iso-
1250–910 Ma estimate suggests that the Pan-African nappe topic composition is respectively consistent with c. 800 and
system in NE Mozambique may carry Mesoproterozoic sed- 750 Ma, and c. 740 Ma seawater. Although their apparent depo-
imentary rocks, and not only Neoproterozoic lithologies, as sitional ages are rather similar, the marble units are separated in
suggested by zircon-monazite geochronology of orthogneisses. space and have a significant difference in 87 Sr/86 Sr (Table 6).
The 1250–910 Ma marbles are low-Sr, intensively dolomi- These two sedimentation episodes are preceded by a volcanic-
tised rocks. Compilation of depositional settings of Mezo- arc type felsic volcanism dated to 818 ± 10 Ma (Bingen et al.,
Neoproterozoic dolostones suggests that dolostones and 2006a, b, 2007). However, detailed geological relationships
dolomitised limestones are common features of intertidal, between the sedimentary and volcanic rocks remain unknown.
supratidal and peritidal settings, whereas limestones tend to There is no sedimentological evidence elucidating depositional
accumulate in shelf and deep basinal environments, very often settings of carbonates in the Xixano Complex. Tentatively, one
below storm waves (Muir et al., 1980; Delaney, 1981; Tucker, volcanic and two sedimentary events are suggested as one
1983a, b; Lindsay, 1987; Preiss, 1987; Southgate, 1989). In continuous depositional cycle, until new data would suggest
556 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558

otherwise. The 818 Ma volcanic rocks and 800–740 Ma carbon- Mozambique within a previously established tectonostrati-
ate units and volcanic rocks underwent only amphibolite-facies graphic and geotectonic framework.
metamorphism, and thus their deposition-metamorphic history 3. Despite amphibolite-facies Neoproterozoic metamorphism,
must be considered separately from that of the mafic and ender- some marbles retain δ18 O (V-SMOW) as high as 28.4‰,
bitic orthogneisses affected by granulite-facies metamorphism suggesting well-preserved depositional signatures. High
at 735 ± 4 Ma in the same complex (Bingen et al., 2006a, b, Sr-abundances in several marble units (1000–4000 ppm)
2007). Such heterogeneity is considered to be the result of tec- provided strong buffering of the Sr-isotope system against
tonic juxtaposition of two plates with different ancestries during postdepositional resetting.
the Neoproterozoic stacking event. 4. The overall spread of the least altered δ13 C (−3.5 to +7.1‰)
and 87 Sr/86 Sr (0.70504–0.70671) values from the Nampula,
7.3. The Lalamo Complex Montepuez, Xixano and Lalamo marbles suggests several
discrete apparent depositional ages within the 1250–660 Ma
Based on the results described above, the geological history time interval. The apparent depositional age of the Ocua
of the Lalamo Complex very likely includes development of marble was not resolved.
one carbonate basin with an apparent age of c. 740 Ma. Two 5. The 300 Ma time lag between felsic volcanism
synorogenic granitic plutonic events dated to 753 ± 10 Ma and (1250–1200 Ma) and carbonate deposition (c. 750 Ma)
696 ± 13 Ma in the Lalamo Complex (Jamal, 2005; Bingen in the Nampula Complex suggests either a non-depositional
et al., 2006a, b, 2007) may imply that the development of unconformity between the two cycles or different ancestry
this carbonate-depositing basin falls in between the two oro- and tectonic juxtaposition during the Neoproterozoic
genic events. This also may imply that these complexes could Pan-African orogeny.
have different ancestries and were tectonically juxtaposed prior 6. Juxtaposition of amphibolite-grade Xixano Complex mar-
to or during the 560–530 Ma Pan-African orogeny. However, bles (originally deposited between 800–750 and at c. 740 Ma)
since the time constraint for the carbonate deposition remains with 735 Ma granulites implies different ancestry of these
imprecise, both above inferences require corroboration by more rock assemblages.
precise age determinations. 7. Apparent carbonate depositional ages suggest that the Nam-
pula, Montepuez, Xixano and Lalamo complexes do not
7.4. The Nampula Complex represent contiguous tectonostratigraphic units and that they
may in fact consist of juxtaposed nappe units.
The Nampula Complex is regarded as one of the basement
complexes in NE Mozambique. It is mainly made up of Meso- Acknowledgements
proterozoic orthogneiss ranging in age between c. 1150 and
1020 Ma (Kröner et al., 1997; Bingen et al., 2006a, b, 2007). This work is the part of the Mineral Resources Management
The Sr- and C-isotopic compositions of marble interbedded Capacity Building Project, Republic of Mozambique; Com-
with graphite schist and paragneiss in the northeastern part ponent 2: Geological Infrastructure Development Programme:
of the complex, south of the Ocua Complex (Fig. 1), sug- Geological Mapping (Lot 1) financed by the Nordic Develop-
gest much younger depositional ages, at around c. 750 Ma. ment Fund. The C and O isotope analyses were performed at the
The difference over 300 Ma between the older igneous rocks Scottish Universities Environmental Research Centre supported
and the younger marbles, both subsequently involved in the by the Consortium of Scottish Universities and the Natural
Pan-African deformation and high-grade metamorphism, can- Environment Research Council. The Rb–Sr isotope study at
not easily be reconciled within a model of the coherent the Institute of Precambrian Geology and Geochronology (St.
tectonostratigraphic unit which experienced a single cycle of Petersburg) was partly supported by the RFBR Project 05-05-
deposition–deformation–metamorphic history. The data suggest 65329. We are grateful to G.V. Konstantinova, N.N. Melnikov
that the marble-bearing sequences represent either an uncon- and E.P. Kutyavin for their assistance in the Rb-Sr analytical
formable Neoproterozoic cover, or tectonic slices, belonging work as well as T.L. Turchenko for the XRD analyses of sili-
to the Neoproterozoic nappe system, folded together with the ciclastic constituents of the carbonate rocks. Two anonymous
basement lithologies. reviewers and the Editor are acknowledged for constructive crit-
icism leading to improvements in the manuscript.
8. Conclusions
References
1. Calcite, dolomite and, to a lesser extent, magnesite marbles Asmerom, Y., Jacobsen, S., Knoll, A.H., Butterfield, N.J., Swett, K., 1991. Stron-
are minor but characteristic components of several metased- tium isotope variations of Neoproterozoic seawater. Implications for crustal
imentary units of unknown age within tectonic nappes in the evolution. Geochim. Cosmochim. Acta. 55, 2883–2894.
Neoproterozoic Mozambique Belt of NE Mozambique. Azmy, K., Veizer, J., Bassett, M.G., Copper, P., 1998. Oxygen and carbon isotope
2. Carbon and strontium isotope stratigraphy of marble units composition of Silurian brachiopods. Implications for coeval seawater and
glaciations. Geol. Soc. Am. Bull. 110, 1499–1512.
has been applied for indirect dating of carbonate sedimenta- Banner, J.L., Hanson, G.N., 1990. Calculation of simultaneous isotopic and
tion and to provide a complementary and independent insight trace element variations during water-rock interaction with applications to
into the depositional history and tectonic development of NE carbonate diagenesis. Geochim. Cosmochim. Acta 54, 3123–3137.
 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558 557

Bingen, B., Boyd, R., Thomas, R.J., Bjerkgård, T., Feito, P., Henderson, I.H.C., Kaufman, A.J., Knoll, A.H., 1995. Neoproterozoic variations in the C-isotopic
Hollick, L.M., Jacobs, J., Jamal, D., Key, R.M., Lutro, O., Melezhik, V.A., composition of seawater. Stratigraphic and biogeochemical implications.
Nordgulen, Ø., Rossi, D., Sandstad, J.S., Skår, Ø., Smethurst, M., Smith, Precambrian Res. 73, 27–49.
R.A., Solli, A., Tveten, E., Viola, G., 2006a. Crustal architecture in North- Kaufman, A.J., Jacobsen, S.B., Knoll, A.H., 1993. The Vendian record of Sr
ern Mozambique. Results from a regional bedrock mapping project. In: and C isotope variations in seawater. Implications for tectonics and palaeo-
Abstract, European Geosciences Union General Assembly 2006, Vienna, climate. Earth Planet. Sci. Lett. 120, 409–430.
Austria, 02–07 April 2006, Geophysical Research Abstracts, 8. A06049. Kröner, A., Sacchi, R., Jaeckel, P., Costa, M., 1997. Kibaran magmatism and Pan-
Bingen, B., Viola, G., Henderson, I.H.C., Boyd, R., Thomas, R.J., Bjerkgård, African granulite metamorphism in northern Mozambique: single zircon
T., Feito, P., Hollick, L.M., Jacobs, J., Key, R.M., Rossi, D., Sandstad, J.S., ages and regional implications. Afr. J. Earth Sci. 25, 467–484.
Skår, Ø., Smith, R.A., Solli, A., Tveten, E., 2006b. Geochronology of Pan- Kröner, A., Willner, A.P., Hegner, E., Jaeckel, P., Nemchin, A.A., 2001. Single
African terrain assembly in NE Mozambique. In: Abstract, 21st Colloquium zircon ages, PT evolution and Nd isotopic systematics of high-grade gneisses
on African Geology, 03–07 July 2006, Maputo. in southern Malawi and their bearing on the evolution of the Mozambique
Bingen, B., Viola, G., Griffin, W.L., Jacobs, J., Boyd, R., Thomas, R.J., Daudi, Belt in southeastern Africa. Precambrian Res. 109, 257–291.
E.X.F., Henderson, I.H.C., Beyer, E., Skar, Ø., Engvik, A., Key, R.M., Solli, Kröner, A., Muhongo, S., Hegner, E., Wingate, M.T.D., 2003. Single-zircon
A., Sandstad, J.S., Smethurst, M., Tveten, E., Bjerkgård, T., Melezhik, V.A., geochronology and Nd isotopic systematics of Proterozoic high-grade rocks
Jamal, D., Smith, R., Hollick, L.M., Feito, P., 2007. Crustal architecture of from the Mozambique Belt of southern Tanzania (Masasi area): implications
the Mozambique Belt in Northeastern Mozambique: perspective from U-Pb for Gondwana assembly. J. Geol. Soc. London 160, 745–757.
geochronology and Lu-Hf isotopes in zircon. In: Abstract, 7th International Kuznetsov, A.B., Semikhatov, M.A., Gorokhov, I.M., Melnikov, N.N., Kon-
Symposium on Applied Isotope Geochemistry, Stellenbosch, South Africa, stantinova, G.V., Kutyavin, E.P., 2003. Sr isotope composition in carbonates
September 2007. of the Karatau Group, Southern Urals, and standard curve of 87 Sr/86 Sr
Brand, U., Veizer, J., 1980. Chemical diagenesis of a multicomponent carbonate variations in the Late Riphean Ocean. Stratigr. Geol. Correl. 11, 415–449.
system. 1. Trace elements. J. Sedim. Petrol. 50, 1219–1236. Kuznetsov, A.B., Krupenin, M.T., Ovchinnikova, G.V., Gorokhov, I.M., Maslov,
Delaney, G.B., 1981. The Mid-Proterozoic Wernecke Supergroup, Wernecke A.V., Kaurova, O.K., Ellmies, R., 2005. Diagenesis of carbonate deposits
Mountains, Yukon Territory. In: Campbell, F.H.A. (Ed.), Proterozoic Basins of the Lower Riphean Bakal Formation, the southern Urals: Sr isotopic
of Canada. Geol. Surv. Can. Paper 81–10, 1–23. characteristics and Pb-Pb age. Lithol. Mineral Resour. 40, 195–227.
Denison, R.E., Koepnick, R.B., Fletcher, A., Howell, M.W., Callaway, W.S., Kuznetsov, A.B., Semikhatov, M.A., Maslov, A.V., Gorokhov, I.M., Prasolov,
1994. Criteria for the retention of original seawater 87 Sr/86 Sr in ancient E.M., Krupenin, M.T., Kislova, I.V., 2006. New data on Sr- and C-isotopic
shelf limestones. Chem. Geol. 112, 131–143. chemostratigraphy of the Upper Riphean type section (Southern Urals).
Denison, R.E., Koepnick, R.B., Burke, W.H., Hetherington, E.A., 1998. Con- Stratigr. Geol. Correl. 14, 602–628.
struction of the Cambrian and Ordovician seawater 87 Sr/86 Sr curve. Chem. Land, L.S., 1992. The dolomite problem: stable and radiogenic isotope clues.
Geol. 152, 325–340. In: Clauer, N., Chaudhuri, S. (Eds.), Isotopic Signatures and Sedimentary
Derry, L.A., Kaufman, A.J., Jacobsen, S.B., 1992. Sedimentary cycling Records. Springer-Verlag, pp. 49–68.
and environmental changes in the Late Proterozoic. Evidence from Lindsay, W.V., 1987. Sequence stratigraphy and depositional controls in late
stable and radiogenic isotopes. Geochim. Cosmochim. Acta 56, Proterozoic-Early Cambrian sediments of Amadeus Basin, central Australia.
1317–1329. Am. Assoc. Petrol. Geol. Bull. 71, 1387–1403.
De Wit, M.J., Bowring, S.A., Ashwal, L.D., Randrianasolo, L.G., Morel, V.P.I., Lindsay, J.F., Brasier, M.D., 2002. Carbon isotope record from 2.6 to 1.9 Ga
Rambeloson, R.A., 2001. Age and tectonic evolution of Neoproterozoic carbonates of Western Australian basins. Precambrian Res. 114, 1–34.
ductile shear zones in southwestern Madagascar, with implications for Gond- McCrea, J.M., 1950. On the isotopic chemistry of carbonates and a paleotem-
wana studies. Tectonics 20, 1–45. perature scale. J. Chem. Phys. 18, 849–857.
Elderfield, H., 1986. Strontium isotope stratigraphy. Palaeogeogr. Palaeoclima- Meert, J.G., 2003. A synopsis of events related to the assembly of eastern
tol. Palaeoecol. 57, 71–90. Gondwana. Tectonophysics 362, 1–40.
Fernandez, A., Schreurs, G., 2003. Tectonic evolution of the Proterozoic Itremo Melezhik, V.A., Gorokhov, I.M., Fallick, A.E., Gjelle, S., 2001a. Strontium
Group metasediments in central Madagascar. In: Yoshida, M., Windley, B.F., and carbon isotope geochemistry applied to dating of carbonate sedimen-
Dasgupta, S. (Eds.), Proterozoic East Gondwana: Supercontinent Assembly tation: an example from high-grade rocks of the Norwegian Caledonides.
and Breakup. Geol. Soc. London, Spec. Publ., pp. 381–399. Precambrian Res. 108, 267–292.
Gorokhov, I.M., Semikhatov, M.A., Baskakov, A.V., Kutyavin, E.P., Mel’nikov, Melezhik, V.A., Gorokhov, I.M., Kuznetsov, A.B., Fallick, A.E., 2001b.
N.N., Sochava, A.V., Turchenko, T.L., 1995. Sr isotopic composition in Riph- Chemostratigraphy of Neoproterozoic carbonates: implications for ‘blind
ean, Vendian, and Lower Cambrian carbonates from Siberia. Stratigraph. dating’. Terra Nova 13, 1–11.
Geol. Correl. 3, 1–28. Melezhik, V.A., Gorokhov, I.M., Fallick, A.E., Roberts, D., Kuznetsov, A.B.,
Halverson, G.P., Hoffman, P.F., Schrag, D.P., Maloof, A.C., Rice, A.H.N., 2005. Zwaan, B.K., Pokrovsky, B.G., 2002a. Isotopic stratigraphy suggests Neo-
Towards a Neoproterozoic composite carbon isotope record. Geol. Soc. Am. proterozoic ages and Laurentian ancestry for high-grade marbles from the
Bull. 117, 1181–1207. north-central Norwegian Caledonides. Geol. Mag. 139, 375–393.
Holmes, A., 1951. The sequence of Pre-Cambrian orogenic belts in south and Melezhik, V.A., Roberts, D., Gorokhov, I.M., Fallick, A.E., Zwaan, K.B.,
central Africa. In: Sandford, K.S., Blondel, F. (Eds.), Proceedings of the 18th Kuznetsov, A.B., Pokrovsky, B.G., 2002b. Isotopic evidence for a complex
International Geological Congress. Assoc. Afr. Geol. Surveys, London, pp. Neoproterozoic to Silurian rock assemblage in the north-central Norwegian
254–269. Caledonides. Precambrian Res. 114, 55–86.
Jacobsen, S.B., Kaufman, A.J., 1999. The Sr, C and O isotopic evolution of Melezhik, V.A., Zwaan, B.K., Motuza, G., Roberts, D., Solli, A., Fallick, A.E.,
Neoproterozoic seawater. Chem. Geol 161, 37–57. Gorokhov, I.M., Kuznetsov, A.B., 2003. New insights into the geology of
Jamal, D., 2005. Crustal evolution in NE Mozambique—implications for Gond- high-grade Caledonian marbles based on isotope chemostratigraphy. Nor-
wana reconstruction. Ph.D. thesis. University of Cape Town, Cape Town. wegian J. Geol. 83, 209–242.
Johnson, S.P., Rivers, T., De Waele, B., 2005. A review of the Mesoproterozoic Melezhik, V.A., Roberts, D., Fallick, A.E., Gorokhov, I.M., Kuznetsov, A.B.,
to early Palaeozoic magmatic and tectonothermal history of south-central 2005. Geochemical preservation potential of high-grade calcite marble ver-
Africa: implications for Rodinia and Gondwana. J. Geol. Soc. London 162, sus dolomite marble. Implication for isotope chemostratigraphy. Chem.
433–450. Geol. 216, 203–224.
Kah, L.C., Sherman, A.G., Narbonne, G.M., Knoll, A.H., Kaufman, A.J., Melezhik, V.A., Kuznetsov, A.B., Fallick, A.F., Smith, R.A., Gorokhov, I.M.,
1999. ␦13 C stratigraphy of the Proterozoic Bylot Supergroup, Baffin Island, Jamal, D., Catuane, F., 2006. Depositional environments and an apparent
Canada: implications for regional lithostratigraphic correlation. Can. J. Earth age for the Geci meta-limestones: constraints on the geological history of
Sci. 36, 313–332. northern Mozambique. Precambrian Res. 148, 19–31.
558 V.A. Melezhik et al. / Precambrian Research 162 (2008) 540–558

Muir, M., Lock, D., von der Borch, C., 1980. The Coorong model for penecon- Semikhatov, M.A., Kuznetsov, A.B., Gorokhov, I.M., Konstantinova, G.V., Mel-
temporaneous dolomite formation in the middle Proterozoic McArthur nikov, N.N., Podkovyrov, V.N., Kutyavin, E.P., 2002. Low 87 Sr/86 Sr ratios
Group, Northern Territory, Australia. In: Zenger, D.H., Dunham, J.B., in seawater of the Grenville and post-Grenville time. Determining factors.
Ethington, R.L. (Eds.), Concepts and Models of Dolomitization, 28. SEPM Stratigr. Geol. Correl. 10, 1–41.
Spec. Publ., pp. 51–67. Shackleton, R.M., 1996. The final collision between East and West Gondwana:
Muhongo, S., 1994. Neoproterozoic collision tectonics in the Mozambique Belt where is it? J. Afr. Earth Sci. 23, 271–287.
of East Africa: evidence from the Uluguru mountains, Tanzania. J. Afr. Earth Shields, G., Veizer, J., 2002. Precambrian marine carbonate isotope database.
Sci. 19, 153–168. Version 1.1. Geochem. Geophys. Geosystems G3 3, 1–12.
Muhongo, S., Kröner, A., Nemchin, A.A., 2001. Single zircon evaporation Slagstad, T., Melezhik, V.A., Kirkland, C.L., Zwaan, K.B., Roberts, D.,
and SHRIMP ages for granulite-facies rocks in the Mozambique Belt of Gorokhov, I.M., Fallick, A.E., 2006. Carbonate isotope chemostratigraphy
Tanzania. J. Geol. 109, 171–189. suggests revisions to the geological history of the West Finnmark Cale-
Nabelek, P.I., 1991. Stable isotope monitors. In: Kerrick, D.M. (Ed.), Contact donides, northern Norway. J. Geol. Soc. London 163, 277–289.
Metamorphism, Reviews in Mineralogy 26. Mineral. Soc. Am, pp. 395–435. Southgate, P.N., 1989. Relationships between cyclicity and stromatolite form in
Pili, E., Ricard, Y., Lardeaux, J.M., Sheppard, S.M.F., 1997a. Lithospheric shear the late Proterozoic Bitter Springs Formation, Australia. Sedimentology 36,
zones and mantle-crust connections. In: Schulmann, K. (Ed.), Thermal and 323–339.
Mechanical Interactions in Deep-Seated Rocks, 280. Tectonophysics, pp. Thomas, C.W., Graham, C.M., Ellam, R.M., Fallick, A.E., 2004. 87 Sr/86 Sr
15–29. chemostratigraphy of Neoproterozoic Dalradian limestones of Scotland and
Pili, E., Sheppard, S.M.F., Lardeaux, J.M., Martelat, J.E., Nicollet, C., 1997b. Ireland. Constraints on depositional ages and time scales. J. Geol. Soc.
Fluid flow vs. scale of shear zones in the lower continental crust and the London 161, 229–242.
granulite paradox. Geology 25, 15–18. Tucker, M.E., 1983a. Sedimentation of organic-rich limestones in the late Pre-
Pinna, P., Jourde, G., Calvez, J.Y., Mroz, J.P., Marques, J.M., 1993. The Mozam- cambrian of southern Norway. Precambrian Res. 22, 295–315.
bique Belt in northern Mozambique; Neoproterozoic (1100–850 Ma) crustal Tucker, M.E., 1983b. Diagenesis, geochemistry, and origin of a Precambrian
growth and tectogenesis, and superimposed Pan-African (800–550 Ma) tec- dolomite; the Beck Spring Dolomite of eastern California. J. Sed. Petrol. 53,
tonism. Precambrian Res. 62, 1–59. 1097–1119.
Preiss, W.V., 1987. The Adelaide Geosyncline; late Proterozoic stratigraphy, Veizer, J., 1983. Trace elements and isotopes in sedimentary carbonates. In:
sedimentation, palaeontology and tectonics. Geol. Surv. S. Austral. Bull. Reeder, R.J. (Ed.), Carbonates. Mineralogy and Chemistry. Reviews in Min-
53, 438 pp. eralogy, 11. Mineral. Soc. Am, pp. 265–299.
Richter, F., Turekian, K., 1993. Simple models for the geochemical response Veizer, J., Ala, D., Azmy, K., Bruckschen, P., Buhl, D., Bruhn, F., Carden,
of the ocean to climatic and tectonic forcing. Earth. Planet. Sci. Lett. 119, G.A.F., Diener, A., Ebneth, S., Godderis, Y., Jasper, T., Korte, C., Pawellek,
121–131. F., Podlaha, O.G., Strauss, H., 1999. 87 Sr/86 Sr, ␦13 C and ␦18 O evolution of
Rosenbaum, J.M., Sheppard, S.M.F., 1986. An isotopic study of siderites, Phanerozoic seawater. Chem. Geol. 161, 59–88.
dolomites and ankerites at high temperatures. Geochim. Cosmochim. Acta
50, 1147–1159.