Pindell Dewey 1982

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/248815578
Permo-Triassic Reconstruction of Western Pangea and the Evolution of the

Gulf of Mexico/caribbean Region
Article in Tectonics · April 1982

DOI: 10.1029/TC001i002p00179
CITATIONS READS
514 156
2 authors:
James Pindell John F. Dewey

Rice University and Cardiff University University College Oxford, British Museum of Natural History, UC …
79 PUBLICATIONS 2,386 CITATIONS 231 PUBLICATIONS 19,449 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Prelithification and synlithification tectonic foliation development in a clastic sedimentary sequence View project
Paleogeography of the Wilcox View project
All content following this page was uploaded by James Pindell on 03 September 2015.
The user has requested enhancement of the downloaded file.

TECTONICS, VOL. 1, NO. 2, PAGES 179-211, APRIL 1982
PEltO-TRIASSIC RECONSTRUCTION OF WESTERN PANGEA AND THE EVOLUTION

OF THE GULF OF MEXICO/CARIBBEAN REG!ON
JamesPindell and John F. Dewey

1
Department of Geological Sciences, State Universit), of New York at

Albany, Albany, New York 12222
Abstract. A Pe•no-Triassic reconstruction of western Pangea (North

America, South America, Africa) is proposed that is characterized by (1)
definition of the North Atlantic fit by matching of marginal offsets
(fracture zones) along the opposing margins, (2) a South Atlantic fit
that is tighter than the Bullard fit and that is achieved by treating
Africa as two plates astride the Benue Trough and related structures
during the Cretaceous, (3) complete closure of the Proto-Atlantic Ocean
between North and South America, accomplished by placing the Yucatan
block between the Ouachita Mountains and Venezuela, (4) a proposed Her-
cynian suture zone that separates zones of foreland thrusting from zones
of arc-related magmatic activity; to the northwest of this suture lie
the Chortis block and Mexico and most of North America, and to the
southeast lie South America, the Yucatan Block, Florida and Africa, and
(5) satisfaction of paleomagmatic data from North America, South America,
and Africa. Beginning with the proposed reconstruction, the relative
motion history of South America with respect of North America is defined
by using the finite difference method. Within the framework provided by
the proposed relative motion history, an evolutionary model for the
development of the Gulf of Mexico and Caribbean region is outlined in a
series of 13 plate boundary reconstructions at time intervals from the
Jurassic to the present. The model includes (1) formation of the Gulf
of Mexico by 140 Ma, (2) Pacific provenance of the Caribbean plate
through the North America-South America gap during Cretaceous time, (3)
Paleocene-Early Eocene back arc spreading origin for the Yucatan Basin,
whereby Cuba is the frontal arc and the Nicaragua Rise-Jamaica-Southern
Hispaniola is the remnant arc, and (4) 1200 km of post-Eocene cumulative
offset along both the Northern and Southern Caribbean Plate Boundary
Zones, allowing large-scale eastward migration of the Caribbean plate
with respect to the North and South American Plates.
INTRODUCTION
The relative motion of two plates separated by an oceanic ridge can

be derived by assuming that symmetrically-disposed pairs of magnetic
anomalies were once adjacent at the ridge at a time equivalent to the
age of the anomaly pair. Extending this approach to the three-plate
system of plates A, B, and C where the spreading history is known between
the A-B pair and the B-C pair, the relative motion history of the A-C
pair, if it is unknown, may be computed by completing the vector circuit
[McKenzie and Morgan, 1969] or finite difference circuit [Dewey, 1975]
1 Present address: Department of Geological Sciences, Science

Laboratories, Durham DH1 3LE, England.
Copyright 1982 by the American Geophysical Union.

Paper number2T0165. 0278-7407/82/002T-0165510.00
180 Pindell and Dewey: Reconstruction of Western Pangea
for several intervals through time. The spreading histories for the
Central and South Atlantic Oceans are fairly well known. ttowever, much
of the oceanic crust (and its magnetic anomalies) that was produced
during the initial separation between North and South America has been
destroyed by Caribbean evolution and cannot be directly evaluated. The
finite difference method for the North America-Africa-South America
three-plate system is the most valuable first-order technique by which
to examine Caribbean evolution because it provides the relative posi-
tions of the North and South American continents at various times since
the initial breakup of Pangea. The framework provided by this method
indicates whether and when extensional, compressional, or transcurrent
motions occurred between North and South America; thus plate boundary
schemes describing the evolution of the Caribbean Plate may be developed
iteratively by integrating the relative motion data with Caribbean geology.
Of vital importance to establishing the relative motion history is
the early Mesozoic predrift reconstruction of North America, South
America, and Africa. Published initial fits of the major continents are
varied, but all leave a noncontinental gap in the Gulf of Mexico area
that can only be filled with either or both the Yucatan and Chortis
blocks of Central America. The preservation of a remnant hole seems
unlikely because of (1) the intensity and continuity of the
Pennsylvanian-Permian orogeny, (2) the rare occurrence of oceanic holes
in other 'completed' orogenic belts around the world, (5) analyses of
strike-slip lateral motions within present-day collisional zones, which
inevitably fill or replace topographic lows with thick sediment wedges
and/or crustal blocks, and (4) the desert-like climatic conditions indi-
cated by the Early Mesozoic geology of southern North America and
northern South America. We propose an Early Mesozoic continental recon-
struction that (1) provides a tight, continuous fit between the three
major continents and the blocks of Central America and (2) satisfies late
Paleozoic to Early Mesozoic geology in terms of arcs, forelands, accre-
tionary terrains, sedimentary patterns, and paleomagnetism.
Beginning with the proposed continental reconstruction, a relative
motion vector history is reconstructed that describes the motion of
South America with respect to North America since the middle Jurassic.
The relative motion history is produced from an Atlantic opening model
proposed herein that varies significantly in the South Atlantic from
those previously published. It is upon this relative motion history that
the Gulf of Mexico/Caribbean evolutionary model is based.
RECONSTRUCTION OF WESTERN PANGEA
Late Paleozoic Geology of Southern North America

and Evidence for the Hercynian Suture Zone
A fairly continuous belt of Permo-Carboniferous deformation passes

from the northern Appalachians, through the southern Appalachians, the
Ouachitas and the Marathons, into the Huastecan 'belt' of eastern Mexico
(Figure 1). Despite the apparent continuity of this belt, evidence for
locating its associated suture or sutures is very sparse. No clearly
defined suture zone has yet been identified in North America, nor has
one been defined within northwestern South America or western Africa;
it must be assumed that it mostly lies beneath thick coastal plain sequences.
Two probable exceptions exist, however, in the southern United States.
First, the Suwannee Basin of northern Florida and southernmost Georgia
(see Figure 1) contains early Paleozoic faunal assemblages typical of
Africa (Gondwanaland) [King, 1975], and thus Wilson [1966] suggested that
this portion of present-day North America became 'welded on' during the
Late Paleozoic closing of the Proto-Atlantic Ocean. The Hercynian
suture probably passes between the Suwannee Basin and the Southern Appa-
lachians. The latter are definitely composed of material that had be-
longed to North America at least since Devonian times [Hatcher, 1978].
Pindell and Dewey: Reconstruction of Western Pangea 181
Fig. 1. Simplified tectonic mapof Late Paleozoic geology of soutnern

North America. Legend: 1, Appalachian Front; 2, Ouachita Front; 3,
Marathon Front; 4, Huastecan Zones, X = locations of ophiolite bodies;
5, SuwanneeBasin; 6, Sabine Uplift; 7, Llano Uplift; 8, Devil's River
Uplift' 9, Texas Lineament; 10, ChihuahuaTrough' 11, Appalachian
Basin' 12, Black Warrior Basin; 13, ArkomaBasin; 14, Fort Worth Basin;
15, Kerr Basin; 16, Val Verde Basin; 17, Marfa Basin; 18, Pedregosa
Basin; 19, Orogrande Basin; 20, Oquirrh Basin; 21, Paradox Basin; 22,
Central Colorado Trough; 23, Rowe-MoreBasin; 24, Delaware Basin; 25,
Midland Basin; 26, Palo Duro Basin; 27, Anadarko Basin.
The second lies to the south of the Ouachita system in the subsurface
structural high called the Sabine Uplift (Figure 1) where drilling has
recovered Mississippian volcaniclastics that may represent the Permo-
Carboniferous arc associated with the closure of the Proto-Atlantic.
Assuming that the Ouachitas represent allochthonous sheets eraplaced
onto the North American foreland (as described by Graham et al., [1975]
and Kluth and Coney [1981], we suggest that the Hercynian suture
passes between the Sabine Uplift and the Ouachita Mountains.
.Appalachian system. In the Central and Southern Appalachians, defor-
mation began in the Late Mississippian (Chesterian), and large-scale
thrusting continued at least through the late Pennsylvanian, possibly
into the Permian. Total shortening within the southern Appalachians
is at least 500 km as shown by COCORPseismic reflection studies [Cook
et al., 1979]. Deciphering the polarity of subduction before the
continent-continent collision that produced the Appalachian deformation
is difficult, but the westward emplacement of giant thrust sheets onto
the North American craton suggests eastward subduction.
Ouachita system. In the Ouachita segment of the belt, deformation
gradually increased from Late Mississippian into the early Pennsylvanian
[Tomlinson and McBee, 1959; Flawn, 1961; Goldstein, 1961; Frezon and
Dixon, 1975], with the climax of folding and thrusting occurring in the
Late Atokan [Goldstein, 1961] or early Desmoinesian (Frezon and Dixon,
1975]. The Ouachita Mountains are characterized by northward-verging
recumbent folds and southward-dipping thrust faults, the motion upon

which has emplacedpre-Pennsylvanian Paleozoic deep-sea sediments and
Appalachian-Ouachita flysch [Graham et al., 1975] onto the shallow water
Paleozoic shelf of Arkansas, Oklahoma, and northern Mississippi. The
Ouachitas have been interpreted as having been caused by continent-
continent collision about various subduction geometries (see Wickham
etal. [1976] for a review), but the likeliest is that of continent-
continent collision across a southward-dipping subduction zone with
emplacement of allochthonous thrust sheets composedof continental rise,
abyssal and flysch type sediments onto the southward-facing Paleozoic
shelf of southern North America [Briggs and Roeder, ]975; Grahamet al.,
•S].
Marathon system. Stratigraphy and deformation in the Marathon
region, West Texas, is similar to that in the Ouachita system. The two
systems are continuous beneath the Cretaceous and Cenozoic cover that
onlaps the Llano Uplift (Figure 1). Right-lateral Late Paleozoic motion
on the Devil's River Uplift or fault system (Figure 1) produced the
apparent offset between the Ouachita and Marathon belts [Muehlberger,
1965]. Marathon folding and thrusting occurred during Late Pennsylvanian
to early Permian time, culminating in the Wolfcampian [King, 1977], and
thus the final Marathon deformation postdates the final Ouachita defor-
mation by perhaps 15-20 million years.
Huastecan system. South of the Marathons it is difficult to trace
the continuation of the Hercynian deformation belt with confidence.
There is, however, a somewhat disrupted trend of Paleozoic rocks that
continues to the south in eastern Mexico (Figure 1) known as the
Huastecan Structural Belt [de Cserna, 1976], which enjoyed Late Paleo-
zoic folding and westward migrating thrusting similar to that in the
Marathon-Ouachita areas. These rocks range from early Paleozoic
[de Cserna, 1971] to Late Paleozoic [Denison et al., 1970] age and con--
tain ophiolite bodies (Figure 1) at Catorce, west of Ciudad Victoria,
in the Tecomatlan area of Puebla, and north of Oaxaca City [de Cserna,
1976]. The Huastecan belt has been described as 'consisting of a sedi-
mentary nonmetamorphosed external part or outer zone, and a metamor-
phosed internal part or inner zone...where in several localities the
two parts are interposed, due to the allochthonous nature of the
sequence or sequences' [de Cserna, 1976], probably by the Mid-Permian.
The present-day disrupted nature of the Huastecan belt probably
resulted from Jurassic left-lateral motions upon several WNW trending
strike-slip faults through Mexico. This motion was at least partially
responsible for the eraplacementof Mexico into its 'overlap position'
with South America, a problem that characterizes most circum-Atlantic
continental reconstructions. Major left-lateral offset of Jurassic age
has also occurred along the Texas Lineament (Figure 1) which sinistrally
offsets the Huastecan Belt from the Marathon Belt by about 700 kin.
Motion along the Texas Lineament was probably responsible for most of
Mexico's southeastward migration into the overlap position with South
America, after North America had rifted from South America-Africa.
Silver and Anderson [1974] proposed that a similar offset occurred
across the Mohave-Sonora Megashear. Exact offsets on the various
faults are still uncertain, but it is clear that motions of this kind
must have occurred if we are to accept most circum-Atlantic
reconstructions of Pangea.
Sedimentary basins of the foreland. In south-central and southwest-
ern North America, in the foreland of the main zones of Hercynian
thrusting, basin development was extensive. The basins are of two
general types; some formed as a result of transtensional strike-slip
(pull-apart basins), while others are clearly foreland trough or
foredeep-type basins, which were partially overriden by, and included
in, Hercynian thrusting. Pull-apart basins include the Anadarko, Palo
Duro, Central Colorado Trough, Rowe-Mora, Oquirrh, Paradox, Orogrande,
Pedregosa, Delaware, and Midland basins (although many of the basins

have been modified by thrusting); and fore!and troughs include the
Appalachian, Black Warrior, Arkoma, Fort Worth, Kerr, Val Verde, and
Marfa basins (Figure 1). The 'sedimentary nonmetamorphosedexternal
zone' of the Huastecan Belt [de Cserna, 1976] includes Permian sediments
that were probably deposited in foreland troughs. Pull-apart basins •n
Mexico are poorly known, although Permian deposits in the Chihuaua
Trough (Figure 1) [Gries, 1979] may indicate an early episode of right-
lateral transtensional motion of the Texas Lineament. Kluth and Coney
[1981] equate the development of the pull-apart basins and associated
strike-slip activity to foreland deformation as southern North America
underwent collision with South America.-Africa. Several west to north-
west trending strike-slip faults formed in response to the stress field
created by the collision, and some of these reactivated elder crustal
structures (for example, the South Oklahoma and Tabosa aulacogens),
which were related to the Late Precambrian-Early Paleozoic rifting
event of eastern North America [Hoffman et al., 1974; Stewart, 1976].
Time-stratigraphic analysis of the foredeep basins and the pull-
apart basins indicates an east to west migration of Hercynian thrusting
and foreland deformation from Late Mississippian to Early Permian time,
indicating oblique collision and progression ef suturing toward the
southwest [Graham et al., 1975; Kluth and Coney, 1981]. Most of the
rift-type basins showminor early deposition beginning in the Late
Mississippian coeval with the beginning of the Alleghanian Orogeny
(Southern Appalachimns). Many basins also show two periods of very
rapid subsidence coeval with the climax of deformation in the Ouachita
(Atokan-Desmoinesian) and Marathon (Wolfcampi•) belts, separated and
followed by periods of relatively slow subsidence. The Southern
Appalachians, the Ouachita and the Marathon systemsprobably represent
localized climaxes during the suturing process, when promontories in
the opposingmargins producedepisodes of greater deformation, and thus
episodic periods of intense foreland reactivation occurred synchronously
with the individual collisional episodes. The alternating periods of
fast and relatively slow subsidence rates can be related respectively to
crustal stretching [McKenzie, 1978] and thermal subsidence.
Summaryof Hcrcynian deformation. The Hercynian collision between
North America and South .Aanerica-Africa produced deformations, all of
which suggest that subdaction before collision was down to the east
and/or south. North America served as the foreland during oblique
continent-continent collision, and as a result suffered much foreland
deformation. The major zones of Hercynian thrusting and the timing of
the thrusting as evidenced by the youngest overthrust flysch in the
foredeeps are outlined in Table 1.
TABLE 1. Summary of Itercwnian Deformation
Zonc Time of Thrusting/Collision
Southern Appalachians began in Late Mississippian (Chesterian), con-

tinued into the Late Pennsylvanian
Ouachitas Atokan to Desmoinesian
Marathons Virgilian to l•olfcampian
Huastecans Early to [,•iddle Permian

Late Paleozoic Geology of Northern South America
As noted earlier, direct evidence to indicate the location of the

suture zone(s) associated with the Hercynian collision between southern
North America and South America-Africa is minimal. In the South Ameri-
can Andes from Colombia to at least as far south as northern Chile,
pre-Upper Devonian igneous and sedimentary rocks show late Devonian up-
lift and deformation and are unconformably overlain by thick sequences
of Carboniferous to Permian granodiorites, andesitic volcanics, and
volcano-clastic sediments [Irving, 1975; Almeida, 1978]. In the
Colombian Andes, volcanics related to this active arc form significant
portions of the Cordillera Central, the Sierra de Santa Marta, and the
Guajira Peninsula (see Figure 2).' Deep water sediments of equivalent
age flank and are interbedded with the andesires on the western shoul-
ders of the three complexes. The three complexes appear as separate
masses, but Tertiary strike-slip faulting has segmented the once
continuous arc complex [Case and MacDonald, 1975]. To the east of the
arc complex, Carboniferous to Permian black shales and limestones accumu-
lated, probably in an environment similar to that of the Java Sea, a
shallow basin overlying continental crust. During Permian time, the
arc and related sediments underwent deformation with the most intense
deformation occurring in the Central Cordillera, the Guajira Peninsula,
and the Sierra Nevada de Santa Marta of Colombia JAimeida, 1978].
Paleomagnetic studies [Valencio and Vilas, 1976; Smith and Briden,
1977] and paleoclimatic studies [Ziegler et al., 1979] indicate that
Pertoo-Carboniferous plate motion of South America (Gondwana) possessed
a strong northward component. The presence of the Upper Paleozoic
arc indicates subduction of oceanic crust beneath northwestern and wes-
tern South America; the andesitic volcanics and granodiorites are inter-
preted as the arc complex that must have formed during the convergence
between North and South America. Continental collision between the
Colombian portion of the Andes and the Marathon-Huastecan segment of the
North American foreland is indicated by synchronous deformation (Early
Fig. 2. Late Paleozoic terrains of northern South America. 1, Cordil-

lera Central (arc); 2, Sierra Nevada de Santa Marta (arc); 3, Guajira
Peninsula (arc); 4, Deformed granites and gneisses of Venezuela; 5,
shallow water seaway behind Late Paleozoic Arc, similar to the Java Sea.
to Middle Permian) and the ceasing of arc-related volcanism in Colombia

during Middle Permian time.
To the east of Guajira, Colombia, evidence for a well-developed Upper
Paleozoic arc is largely lacking. In northern Venezuela, the most sig-
nificant aspect of Upper Paleozoic geology is the presence of Pennsyl-
vanian-Permian granitic intrusions and metamorphics. Whether the
granitic intrusions are related to the arc in northwest Colombia or to
simple heating of the crust due to thickening cannot be specified.
Extent of Pre-Mesozoic Continental Crust:

The Margins to be Fitted
For the purposes of the Penno-Triassic reconstruction, the islands of

the Greater and Lesser Antilles, the Panama-Costa Rica isthmus, and the
Bahamas Platform may be disregarded because they are all Jurassic or
younger, developed during the course of Caribbean evolution. Pre-
Mesozoic continental crust to be considered is limited to the continents
of North and South America and Africa, and the blocks of Yucatan and
Chortis. The basement of Mexico is probably best considered as a set of
several blocks because of the deformation it underwent during emplace-
ment into the 'overlap position' with South America in most circum-
Atlantic reconstructions.
Northern limit of continental crust of South America. Complex post-
Jurassic tectonic events have obscured and greatly modified the rifted
margin of northern South America. Late Cretaceous collision with the
Aruba-Blanquilla island arc [Gealy, 1980; Maresch, 1974] emplaced giant
thrust sheets onto the Venezuelan margin. The arc is preserved off-
shore, and the suture apparently lies within the Bonaire Basin and con-
tinues into the Trinidad-Tobago area [[4axwell, 1948]. In western Colom-
bia, Neogene collision with the Panama arc has led to accretion of signi-
ficant amounts of deep water and oceanic material [Irving, 1975].
Development of the South Caribbean Plate Boundary Zone [Burke et al.,
1978] during later Cenozoic time has greatly modified the morphology of
the margin. Fortunately, no single onshore fault has been shown to have
accumulated a large offset, and thus it is likely that much of the rela-
tive displacement between the Caribbean and South American plates has
occurred along transform faults to the north of the island-arc chain.
Despite these modifications, Fiooure 3 approximately defines the limit
of the Jurassic rifted margin. Precambrian and Paleozoic basement out-
crops continuously along northern Colombia [Tschanz et al., 1974;
MacDonald, 196S], and it is inferred that basement extends offshore to
the continental shelf break. In central and southern Colombia, the edge
of the Early Mesozoic basement was approximately defined by the western
margin of the Cordillera Central arc [Irving, 197S]. In the north, from
Peninsula de Paraguana to the Cariaco Basin, the Jurassic basement limit
probably closely approximates the suture zone of the Late Cretaceous arc-
continent collision. To the east of the Cariaco Basin, basement is
difficult to define. b•ch of the sediment of the Paria and Trinidad
area has been accreted to the continent in the Late Cretaceous, and high
rates of sediment influx from the Orinoco River have produced a massive
delta, beneath which lies oceanic and/or attenuated continental crust.
It is speculated that the Jurassic limit of basement lies slightly
inland of these areas (Figure 3). Continuing to the southeast, the limit
again returns offshore to the continental margin.
Southern limit of continental crust of North America. Whether the
basement of the Gulf Coastal Plains south of the Ouachita-Marathon belts
is comprised of continental or oceanic basement has been heavily debated
(see Cebull and Shutbet [1980] for a review). The presence of more than
1S km of post-Triassic sediments within the Gulf Coast 'Geosyncline'
[Antoine et al., 1974] dictates that if continental crust is present, it
is extremely attenuated. It has also been suggested that oceanic crust
Fig. 3. Northern limit of continental crust of South America in Juras-

sic timez. 1, Cordillera Central; 2, Santa Marta; 3, Guajira; 4, Base-
ment of Panama arc (Late Cretaceous-Tertiary; 5, Paraguana; 6, Bonaire
Basin (Tertiary); 7, Cariaco Basin (Tertiary); 8, Trinidad-Paria area;
of Paleozoic age is present beneath the coastal plains ICebull and

Shurbet, 1980], a position, however, that fails to explain adequately
the intensely thrusted Ouachita-Marathon belt to the north. Late Paleo-
zoic continental suturing followed by Jurassic rifting is the most satis-
factory explanation for the observedgeology. Whetherthe coastal plain
basement is oceanic (Jurassic) or highly attenuated continental crust is
a trivial matter; either means that the original limit of continental
crust lay well inland of the present-day coastline. The position
adopted here is that the pre-rift limit of continental crust of southern
North America falls between the northern limit of the Louann Salt and
the Cretaceous reef trend to the south (Figure 4). Structural highs
such as the Sabine Uplift and the Wiggins arch in the rifted margin are
consideredto be stranded horsts separated from the mainlandby marginal
grabens.
trend
'" -
•'.'i'.'
- -'-
Sun•'iland
F.Z.
Hercyman front
Limit of cont. crust
Fig. 4. Limit of continental crust along the Gulf margin of the

United States.
Pindell and Dewey: Reconstruction of Western Pangee 187
400 km
Fig. 5. Generalized map of the Yucatan 01ock showing (1) the area
occupied by the Central Guatemalan Arc (cross-pattern) and (2) the
limit of continental crust (CL) to be considered in the fit of Western
Pangea.
In southernmost Florida (Figure 4), Klitgord et al. [1982] show that

pre-Hesozoic crust is absent and that Lower Cretaceous isopachs indicate
a sharply rifted margin, probably attributable to transform faulting
rather than to extension rifting. The transforms (Sunniland and Bahama)
were probably very long, connecting the ridge system of the Atlantic
Ocean to that in the Gulf of Mexico.
In eastern Mexico, continental crust seems to exist very near the
present coastline. However, considerable left-lateral syn-or post-
rifting offset may have occurred upon the Texas Lineament so that, in
northeastern Mexico, there exists an extensive embayment of salts, clas-
tics, and marine deposits (Figure 4).
The Yucatan Block
Although the Yucatan Block (Figure 5) is of lñmited geographic extent,

it contains critical information for Gulf and Caribbean evolution, com-
prising six stages of development as follows: (1) Devonian uplift,
deformation, and metamorphism along the southern portion of the block
(present coordinates); (2) Late Paleozoic Andean-type arc vulcanism
along the southern border of the block (Central Guatemalan arc); (5)
Middle Jurassic rifting that produced a highly stretched rifted margin
along the northen• and western margins; (4) Late Cretaceous collision
with Nicaragua Rise-Jamaica, with Yucatan acting as the foreland; (5)
Late Cretaceous-Paleocene uplift, erosion, and subsequent block-faulting
along the entire eastern margin; and (6) considerable left-lateral off-
set between the Yucatan and Chortis Blocks since Oligocene time that
has produced a complex plate boundary zone between the two. The Devonian
deformational episode [McBirney and Bass, 1969] is based on isotopic
ages of Late Devonian age from the Chuacus Group, which is unconformably
188 Pindell and Dewey- Reconstruction of Western Pangea
Jurass.
Todos Santos Red Beds Rifting, subsidence
195 .....
Triass.
225 ....
Ochoan
240
255 Guadal.
_ TuilonSS--
Leona. ChocalLS
•
•
268 -- • •• .
Wolf. Esperanza Fm. shale, minor LS •

280 dølø' •
Virgil • >800
meters
brown,
black •
o shales, silts, sands o
Missour •
300 UPLIFT
(COLLISION)
.
Desmoin
305 I >1ooo
meters
conglomerati•
•
itokan • sandy,
tuffaceous • •
310 •o volcanics, lavas, o•
Morrow • volcaniclastics, • •
317 o ••
ChesterI leucocratic
dikes,
ANDESITIC
breccia • •
330
Meramec
335
Osage
Kinder
346
UPLIFT, DEFORMATION,
UPPER METAMORPHISM
Chuacus Group
DEVONIAN
=UNCONFORMITY
Fig. 6. Stratigraphic column and tectonic interpretation of the Late

Paleozoic section of central Guatemala, Yucatan Block. The proposed
arc is represented by tile Chicol Formation. After Anderson (1969),
Anderson eta]. (1973), and McBirney and Bass (1969).
overlain by Carboniferous strata. A late Paleozoic Andean arc is

inferred from stratigraphic descriptions from Central Guatemala(see
Figure 6 for a summary). The Late Paleozoic evolution of southern
Yucatan is strikingly similar to that of the Cordillera of western South
America, where Devonianuplift was followed also by the arc-related mag-
matic activity. It is suggested that the two may be correlated and that
the Central Guatemalanarc was a northern, westwardfacing extension of
the Andeanarc during Mid to Late Paleozoic time. The only significant
difference betweenthe two terrains is that vulcanism apparently ceased
during Pennsylvanian time in Yucatan, whereas it continued into the Per-
mian in the Andean Cordillera. ø
Seismic and drilling data [Bufflet et al., 1980, 1981] indicate that
the Gulf of Mexico was formed by the separation of the Yucatan Block
from the southern margin of the United States in Middle to Late Jurassic
time. However, the opposing rifted margins are so highly attenuated

that an accurate location of the ocean-continent transition cannot
readily be made. Both margins contain a thick evaporitic section, in-
dicating the existence of a restricted basin along the site of the de-
veloping rift. While geological data from the Gulf of Mexico are suf-
ficienct to substantiate a Jurassic origin-by rifting between Yucatan and
North America, flow lines and the pre-rift alignment of the margins can-
not yet be proven. First, much of the opening of the Gulf probably
occurred during the Jurassic quiet zone, and thus no magnetic anomalies
may exist. Second, considerable down-slope migration of the salt sec-
tion [Antoine et al., 1974] has distorted its original geographic
extent. Third, the tnick post-Triassic carbonate and evaporite section
(1-4 km) across the entire Yucatan Platform indicates that the Yucatan
basement was significantly stretched during the rifting episode; the
original extent of the block may have been 10-20% smaller.
In the Ouachita system, deformation and thrusting culminated in early
Desmoinesian time. To obtain a closed ocean fit between North and South
America, continental mass must have been present between Venezuela and
the Ouachitas. It is suggested that the Yucatan Block occupied this
position on the basis of (1) the correlation of Jurassic aged salt se-
quences known from the northwestern margin of Yucatan and the southern
margin of the United States, (2) the nearly identical pre-rift geome-
tries of the Yucatan Block and the unclosed 'hole' between the margins of
North America and Venezuela, (3) the presumption that the Central
Guatemalan arc formed a northern, westward-facing extension of the Upper
Paleozoic Andean Cordillera, and (4) the equivalent ages of thrusting in
the Ouachitas and the termination of vulcanism in the Central Guatema-
lan arc of Yucatan.
The Chortis Block
Little is known of the pre-Hesozoic basement of the Chort•s Block

but its extent is roughly defined by the Motagua-Swan transform fault
on the north and the Santa Elena Peridotite on the south (Figure 7).
A thick accretionary pile has accumulated along the western margin as
a result of slightly extensional Tertiary subduction at the Middle Amer-
ø'..
ND
CL
L400km
Fig. 7. Limit oF pre-Mesozoic continental crust of the Chortis block.
Offshore position of boundaryis approximately the 100-fm depth contour.
MZ = Motagua Fault Zone; ND = Nicaragua Depression.
ica Trench. The Nicaragua Depression [Muehlberger, 1976] (Figure 7) is

the site of normal faulting and active arc-related vulcanism. The wes-
tern margin of the Paleozoic basement is roughly coincident with the
eastern boundary of the Nicaraguan Depression. To the east of the block
lies the Nicaragua Rise, a submarine western extension of the southern
Greater Antilles. Delineation of a precise boundary separating the
Chortis continental basement from the Antillean arc-related basement is
probably not possible, but block-faulting in the basement beneath the
100-fm contour east of Nicaragua (Figure 7) suggests a possible eastern
limit of basement; the 100-fro contour is assumed to mark the extent of
Paleozoic basement along the Nicaragua Rise.
The African Connection
The most valid criterion for obtaining a fit between opposing rifted
margins is the realignment of marginal offsets because they are a direct
result of intracontinental rifting [LePichon and Fox, 1971; Klitgord
and Schouten, 1980]. If we assume (1) tortional rigidity for the
circum-Atlantic continents and (2) that the Bullard fit for the South
Atlantic is valid, proper correlation of the Late Paleozoic features be-
tween North and South America cannot be achieved. Two solutions are
plausible: (1) that right-lateral strike-slip motion occurred between
Africa and eastern North America after the Late Paleozoic oroœeny to
approximate the LePichon and Fox configuration before middle Jurassic
rifting, so that formation of much of the Gulf of Mexico predates the
Atlantic Ocean (following Van der Voe and French [1974]), or (2) that
Africa was not a torsionally rigid single plate during Cretaceous times
[Burke and Dewey, 1974] so that the Bullard fit in the South Atlantic
may be improved by closing an apparently noncontinental gap along the
Amazon section of the South Americmn shelf. The second solution allows
excellent correlation of Late Paleozoic features between North and South
America without any pro-rift differential motion between Africa and
North America; this alternative is favored here and is discussed in de-
tail below.
Two Plates in Africa and the Revised .•puth Atlantic Fit
The most commonly accepted South Atlar•tic fit is that of Bullard et

al., [196S], which is further supported by alignment of marginal frac-
ture zones [LePichon and Hayes, 1971]. However, this fit assumes that
continental material underlies the entire Amazonportion of the South
American shelf. The enormousthickness of post-rifting sediments (Fi-
gure 8a) beneath most oft•at portion of the shelf (8-10 km, Kumar et
al., [1979]) indicates that only oceanic or extremely thinned conti-
nental crust can be present there. Thus the Bullard fit apparently
leaves a substantial gap {Figure 8b) between the northern Brazilian and
Guinea margins. In light of the fact that Gondwana was one of t?•e most
stable cratonic masses of the Phanerozoic until the Cretaceous, the
existence of this gap during Late Paleozoic to Cretaceous time is highly
unlikely, and thus the Bullard fit probably cam be improved. It is
apparent that a satisfactory fit for the South Atlantic cannot be made
if torsional rigidity is assm•ed; rotation of Africa with respect to
South America to close the Amazongap produces a gap in the southern
South Atlantic.
Rabinowitz and LaBrecque [1979] noted thane the rifting history of the
South Atlantic was such that rifting occurred earlier in the southern
South Atlantic (130 Ma) than in t?•e equatorial Atlantic (anomaly MO,
which is 110 Ma, after Van Hinte [1979a]). Starting from an initial fit
similar to that of Bullard et al. [1965], they attributed this to a
counterclockwise rotation of the whole African continent with respect to
South America from 130 Ma to MO (110 •.•a) about a pole just off the
S.A. km
0 5(•0
,
Fig. 8. (a) Sediment isopachs and 2000-m contour along the Amazon
shelf, assumed by Bullard et al. (1965) to be underlain by continental
crust. (b) Revised interpretation of continental limit following the
principle that 8-10 km of sediment cannot lie upon normal co.ntinental
crust. Stipi•ed pattern is the gap in the Bullard fit.
northeast coastline of Brazil. This rotation requires that 100 km of

shortening and compression occurred between the Ivory Coast section of
Africa and northernmost Brazil for the period 130-110 Ma. Although
minor compression probably associated with imperfect transform motion
during the opening phase of the equatorial Atlantic is known from be-
neath the Demerara Plateau [Hayes et al., 1972], evidence for major
shortening (100 km) has not been documented. Of great importance,
though, is that the postrotational configuration (110 Ma) of the equa-
torial portions of South America and Africa fills the Amazongap of
Figure 8b in the Bullard fit. Further, 110 Ma has been suggested as the
time of initial rifting in the equatorial Atlantic [Kumar et a1.,1976;
Asmus and Ponte, 1973; Rabinowitz and LaBrecque, 1979]. These consi-
derations lead to the conclusion that the MO configuration of Rabinowitz
and LaBrecque [1979] closely approximates the Early Hesozoic configura-
tion as well, but that their (150-110 Ma) rotation of northern Africa
with respect to South America did not occur. Thus we can dispose of the
Amazongap of Bullard et al. [1965]. however, satisfactory closure
of the entire South Atlantic Ocean cannot be achieved by assuming per-
fectly rigid plates; post-Jurassic internal deformation must have oc-
curred within Africa and/or South America.
192 Pindell and Dewey' Reconstruction of Western Pangea
65
MAE
CAM 70
76 t.,_....,•j•
'.del
SAN deform.
CON '82 m•rmeVVV
TUR9•.
CEN88
100
AI,B
106
shol?•
ß.'.
APT
112
BER 118 bos•l

sst.
HALl 12•, '
VAL 130
..
BER 136 vvvvv
Fig. 9. Generalizedmapof Africa showing Early Cretaceous

rift and
strike-slip fault systems;J. C. Cooper,personal communication,
1981.
Inset: Schematicstratigraphic columnof BenueTrough area, after
Burke (1976, p.100), Burke and Dewey(1974), and Hurat (1972).
Although Grabert [1977] reported Late Jurassic faulting and minor

strike-slip motion in the upper AmazonValley of South America, large-
scale post-Jurassic deformation of this continent is not supported by
geological or geophysical evidence. By contrast, intense internal de-
formation in central Africa of Cretaceous age is abundant. Burke and
Dewey [1974] called attention to widespread basaltic magmatismand
block-faulting across central and northern Africa (summarized in Figure
9) that they consider to have been a poorly defined plate boundary zone
of net differential motion between northwest and southeast Africa. The
only narrow plate boundary within this system may have developed in the
Benue Trough, whose stratigraphic column is shown in Figure 9. The a-
mount of early Cretaceous extension and late Cretaceous shortening that
occurred cannot be defined. Any relative motions between northwest and
southeast Africa that may have led to differences in spreading rates
along the early South Atlantic ridge cannot be observed because •ost of
the Benue episode occurred during the Cretaceous quiet period. All that
can be said of the Benue episode (and its extensions into central and
northern Africa) is that the amount of extension exceeded the amount of
closure; this is indicated by a pronounced positive gravity anomaly and
the subdued, trough-like character of the underlying basement. There-
fore, it is inferred that 'remnant extension' exists today which did not
exist before Early Cretaceous time, and that the present-day shape of
Africa should not be used in pre-Cretaceous paleoreconstructions.
The net effect of the Cretaceous deformation cannot be determined
from African geology, but it is suggested here that the net deformation
is equivalent to the difference between Africa's present shape and the
shape that is required to produce a fit with South America closed in
both the Amazon and southern South Atlantic regions. Figure 10 shows
this closed fit and compares the predeformational and postdeformational
shapes of the African continent. Marginal fracture zones commonlyused
for matching the opposing margins in the southern South Atlantic are
also shownin Figure 10, showing clearly that the proposed predrift re-
construction satisfies the fracture zone alignment criterion.
A pole of rotation that defines the early motion of southeastern
Africa with respect to northwestern Africa-South America must satisfy
fracture zone trends and magnetic anomalies associated with the early
(•po
LE ~19N,2E
$A.
AF.
FP AFZ
Fig. 10. Difference between pre-Cretaceous and present shapes of

Africa. Solid lines are present shapes of the continents and continent-
al saelves, dashed line is proposed pre-deformational position of south-
ern Africa with respect to South America. FP = Falkland Plateau;
AFZ = Agulhas Fracture Zone; 1 and 2 are other fracture zones proposed
by Rabinowitz and LaBrecque (1979). Pole position is that which defines
early motion of souti•ern Africa with respect to North Africa-South
America.
opening phase of the southern South Atlantic Ocean (Mll-MO, or 125-110

Ma). Thepole usedby RabinowitzandLaBrecque
O [1979]
- lies at 2.5øS•
45.0øW
with an angularrotationof +11.0 , all with respectto South
America. This rotation is based upon alignment of the ocean/continent
boundaries, the seaward edge of salt boundaries and the trend of margi-
nal fracture zones, particularly the Agulhas-Falkland fracture zone.
We propose an additional constraint that requires minor adjustment of
the pole position and angular rotation of Rabinowitz and LaBrecque
[1979]; the early South Atlantic pole must be located so that the left-
lateral motion across the Benue Trough be accompanied by an extensional
component to account for the rift-like nature of the trough. This
requires that the pole be located at a more northerly position and,
according to J. LaBrecque (personal communication, 1981), the play in
the data allows northward adjustment. The pole suggested here that
seems to satisfy all constraining data lies at 19øN, 2øE with respect to
present-day Africa (Figure 10), with an angular rotation of about 8•.
It is assumed here that Africa had attained approximately its present
shape by 110 Ma. Any further opening in the Benue area followed by
significant closure cannot be recognized from ocean data because of the
Cretaceous quiet period from 110 •a to 80 Ma. Thus Africa is regarded
as a rigid plate after 110 Ma in the proposed South Atlantic opening
data (Table 3).
Fig. 11. Sketch map showing revised Pertoo-Triassic continental recon-

struction of western Pangea. Note that Proto-Atlantic Ocean is closed,
and that zones of foreland thrusting (H, Huastecan; M, Marathon;
O, Ouachita; A, Appalachian) are separated from Andean Arc terrains by
the proposed Hercynian Suture (MS), S, Suwannee Basin with Gondwana
faunal affinity; Y, Yucatan Block; C, Chortis Block.
Reconstruction of Western Pan•ea
Figure 11 depicts the pre-rift relationship of the circum-Atlantic

continents and Central American blocks foliowing the considerations and
constaints presented in this paper. It is suggested that this configu-
ration existed from Middie Permian to Middie Jurassic time with little
tectonic activity or relative motions between the individual constitu-
ents. The excellent fit was achieved b7 a gradual oblique closing of an
ocean with subduction to the south and east, whereby deformation exter-
nal to the plate boundary zones (particularly the North American fore-
land) accommodated salients with complex strike-slip motions of crustal
segments. The southwest peninsular extension of North America (Mexico)
probably migrated an unknown but considerable distance toward the north-
west during the collision, possibly allowed by 'free face' overriding
of plates in the Pacific.
Paleomagnetic data for Permian rocks of the circum-Atlantic conti-
nents [Van der Voo and French, 1974] are satisfied by the proposed
reconstruction. South America and southern Africa are in approximately
the same position as that suggested by Van der Voo and French [1974],
which is based on paleomagnetic data.
EVOLUTION OF THE GULF OF MEXICO/CARIBBEAN REGION
Relative Motion History of the Major Continents
The following data sources are used to define the poles and rotations
for the opening histories of the Central and South Atlantic oceans, and
these in turn have been used to compute the relative motion history of
South America with respect to North America.
Africa with respect to North America. In the Central Atlantic the
poles and rotations of Sclater et al. [1977] were used and are summa-
rized in Table 2. These poles are based on LePichon and Fox [1971] for
165-80 Ma, and on Francheteau [1973] for 80 Ma to the present. Fol-
Pindell and Dewey' Reconstruction of Western Pangea 195
TABLE 2. Relative Motion Poles,

Africa With Respect to North America
Time Magnetic Finite Difference Stage Poles

Ma Anomaly Latitude, Longitude, Angle, Latitude, Longitude, Angle,
deg. deg. deg. deg. deg. deg.
165 CM 66.0 -12.0 -74.8

55.9 -17.3 11.3
150 interpolated 67.8 -13.8 -63.7
57.9 -21.7 5.4
140 •!-22 68.8 -14.2 -58.4
60.9 -29.5 7.2
125 M-6 70.2 -12.9 -51.3
58.1 -22.0 7.2
110 interpolated 72.3 -13.3 -44.3
58.4 -21.9 7.3
95 interpolated 75.1 -14.0 -37.3
58.5 -21.7 7.3
80 34 79.1 -15.7 -30.4
82.9 153.7 9.9
65 29 70.9 -10.9 -21.2
70.3 -3.8 4.8
53 22 70.9 -13.1 -16.4
70.7 -3.2 6.0
36 13 70.5 -18.7 -10.4
70.5 -18.7 4.7
21 6 70.5 -18.7 -5.7
70.5 -18.7 3.00
10 5 70.5 -18.7 -2.7
70.5 -18.7 1.35
5 interpolated 70.5 -18.7 -1.35
70.5 -18.7 1.35
0 ......... 0.0
lowing Sclater et al. [1977], the time scales of Sclater et al. [1974]
and Van Hinte [1976a,b] are used for the Cenozoic and Mesozoic, respec-
tively.
South America with respect to Africa. In the South Atlantic, because
the Bullard et al. [1965] initial fit is not employed, several data
sources are used. The initial Jurassic fit is that of Rabinowitz and
LaBrecque [1979] for the time of anomaly MO (110 Ma). The additional
rotation of the 'southern plate' of Africa which is required to close
the southern South Atlantic Ocean is defined in this study and occurred
from the Valanginian (125 Ma) to 110 Ma. The motion of South America
with respect to Africa during the interval 110-80 Ma (Cretaceous Quiet
Zone) is that of Rabinowitz and LaBrecque [1979], whose 80 Ha position
is taken from Ladd [1974, 1976]. From 80 Ma to present, the poles and
rotations of Sibuet and Mascle [1978] are used. They too have taken
Ladd's [1974, 1976] 80 Ma position of South America with respect to
Africa and have recognized one minor change in spreading history at 36
Ma (anomaly 13). This provides additional control over the analysis of
Sclater et al. [1977] for the South Atlantic, who assume a single pole
from 80 Ma to present, after Francheteau [1973]. Table 3 summarizes
poles and rotations used in this study for the opening of the South
Atlantic.
South America with respect to North America. Using the finite
difference method, relative motion poles of South America with respect
196 Pindell and Dewey: Reconstructionof WesternPangea

South America with Respect to Africa
Time Magnetic Finite Difference

Ma Anomaly Latitude, deg. Longitude, deg Angle, deg.
165-110 MO 55. ] -35.7 50.9
95 interpolated 58.3 -35.8 42.3
80 34 63.0 -36.0 33.8
65 29 62.0 -36.3 26.8
53 22 60.8 -36.6 21.3
36 13 57.4 -37.5 13.4
21 6 57.4 -37.5 7.8
10 5 57.4 -37.5 3.7
5 interpolated 57.4 -37.5 I. 9
0 ......... 0.0
to North America have been computed. Table 4 summarizes these poles,

both the total finite difference and the stage poles; from these data,
relative motion vectors (Figure 12) have been plotted (equal angular
rotations). The relative motion vectors are plotted with respect to
present-day North American coordinates and provide the framework for the
evolutionary model of the Gulf of Mexico/Caribbean region.
Plate Boundary Reconstructions, 16_5.Ma to .the Present
Figures 13-25 are a set of 13 plate tectonic reconstructions that

outline the proposed model of Gulf and Caribbean evolution since the
middle Jurassic breakup of Pangea. Each reconstruction depicts a plate
boundary system that appears to satisfy Gulf/Caribbean geology for the
period ranging from the age of that to the subsequent reconstruction.
Major geologic features incorporated into the model are shown throughout
the sequence. Where appropriate, instantaneous vector triangles are
shown, as well as arrows defining the motions of Africa and South
America with respect to North America for the time period following a
particular figure. The orientations of extensional and transform plate
boundaries are defined in this way. Note that the relative motions
between the 36 Ma, 21 Ma, 10 Ma, and 0 Ma reconstructions are so small,
although generally compressional, that a well-defined plate boundary
between North and South America may not have existed during these
periods.
165 Ma. Throughout Late Triassic to Middle Jurassic time, a wide-
spread belt of doming and rifting formed within Pangea that closely
followed the line of Hercynian suturing. By 165 Ma, a discrete plate
boundary system had formed and plate accretion had begun. Figure 13
shows a plate boundary system that may have been responsible for the
breakup of Pangea with some of the major failed rifts. The geometry of
the spreading system in the Central Atlantic is that defined by Klitgord

South America With Respect to North America
Time Magnetic Finite Difference Stage Poles

Ma Anomaly Latitude, Longitude, Angle Latitude, Longitude, Angle
deg. deg. deg. deg. deg. deg.
165 ... 52.1 37.6 -28.8

55.9 -17.3 11.3
150 ... 45.2 61.6 -20.1
57.9 -21.7 5.4
140 ... 35.9 75.5 -17.1
60.9 -29.5 7.2
125 ß. ß 15.4 91.8 -15.6
58.1 -22.0 7.2
110 ßßß 5.0 -73.5 17.0
5.2 -92.7 -3.6
95 ß. . 4.1 -68.5 13.6
2.4 -85.3 -3.8
80 ... 3.9 -62.1 10.0
41.9 138.7 4.8
65 . ß. 30.6 -70.3 7.8
28.2 -98.9 -1.4
53 .. ß 30.0 -64.6 6.6
43.5 -81.1 -2.5
36 ... 21.1 -57.7 4.3
11.1 -63.3 -1.5
21 ... 26.4 -54.1 2.8
26.5 -55.1 -1.5
10 ßß. 26.2 -52.9 1.3
23.8 -54.2 -0.6
5 ... 28.2 -51.7 0.7
28.2 -51.7 -0.7
0 ......... 0.0
et al. [1982]; that surrounding the Yucatan block (Y) is poorly con-
strained, although a system similar to that shown is required to move
Yucatan to its present position. The Chortis block (C) is viewed as a
southern extension of Mexico. Extensional left-lateral motion along the
Chihuahua Trough (CT) and Texas Lineament allowed southeastward migra-
tion of Mexico into its 'overlap position,' which was occupied by South
America during the early Mesozoic.
150 Ma. Spreading between North America and Gondwana continued; by
150 Ma, a ridge jump had occurred in the Central Atlantic that is respon-
sible for the apparent asymmetry of the Atlantic Ocean with respect to
its ridge. Rotation of Yucatan with respect to South America is con-
sidered analogous to the early Cretaceous motion of Iberia with respect
to Africa; the formation of the Bay of Biscay, then, can be compared to
that of the Gulf of Mexico. The rifting of Yucatan from North America
left stranded horsts such as the Sabine Uplift (S) and Wiggins Arch (W)
along southern North America. Louann salts (small cross pattern) were
deposited around these structures and along the north-facing rifted mar-
gin of Yucatan in the restricted seaway between Yucatan and North
America. Salts of Jurassic and Early Cretaceous age are found in the
Chihuahua Trough [Gries, 1979] indicating extension and subsidence of
that age along this feature. Left-lateral motion on other faults in
Mexico [de Czerna, 1976] was responsible for the completion of Mexico's
emplacement into the 'overlap position' probably by 140 Ma.
165
NORTH
AHERICA
-,... •..•s \1 "...
• _.-%....'• • o
SOUTH
AHERICA
Fig. 12. Relative motion histories of three points along northern South
Americawith respect to North America. Early relative motion of Africa
is also shown. These provide the relative positions of the three con-
tinents at various times and thus provide a framework in which to base
Caribbean/Gulf evolution. Vectors are the result of data in Tables 2,
3, and4.
In the Bahamasarea (B), the probable existence of a long transform

provides a situation in which compression and extension are likely
during subtle changes in relative motion. The parallel relationship of
the Bahamaswith Atlantic fracture zones nearby to the north supports
the hypothesis that compressional transform motion caused uplift of
oceanic basement to the photic zone so that carbonate deposition could
be initiated.
140 Ma. By about 140 Ma, formation of the Gulf of Mexico is con-
sidered complete. The rotation of Yucatan and the eastward migration of
the Mexican blocks led to juxtaposition of Yucatan and southern Mexico
in the Tehuantepec area. The juxtaposition was probably achieved mainly
by strike-slip, although some amount of compression may have occurred.
Present knowledge of the geology of the Tehuantepec area is too obscure
to obtain a clear understanding of this event.
Klitgord et al. [1982] have mapped fracture zones just north of the
Bahamasthat have a consistent kink such that left-lateral motion upon a
transform fault of the ridge system would enter compression around 140
Ma. We suggest that the minor pole change associated with this kink was
substantial enough to raise oceanic crust adjacent to the long transform
in the Bahamasarea to the photic zone. Since then, carbonate deposi-
tion has kept pace with subsidence, which occurred at rates several
times greater than thermal rates during the remainder of the early
Cretaceous.
125 Ma. Continued spreading between North America and Gondwanapro-
duced an oceanic seaway between North and South America into which crust
of Pacific provenancewas able to enter. This was accomplishedby ini-
tiation of southward dipping subduction at a previously formed Atlantic
fracture zone (zone of weakened crust). Proto-Caribbean crust was sub-
ducted beneath Pacific crust as the Pacific crust entered the Caribbean
area. Northern and southern portions of Pacific crust were subducted
beneath the Andean margins of North and South America, respectively.
Volcanics that were formed along the leading edge of the entering
Pacific crust would become the Greater Antilles Island Arc.
At 125 Ma, the southern South Atlantic began to open by eastward
migration of southeast Africa with respect to South America/northwest
Fig. 13. Paleoreconstruction at 165 Ma.
Africa; this, however, had no effect on the Caribbean area because we

are assuming no significant motion in the Equatorial Atlantic until 110
Ma, althouKh dominK and initial phases of rifting had begun.
110 Ma. At 110 Ma, separation between South America and northwest
Africa began,thus producinga three-plate system. The independent
motionof SouthAmericarequires a neworientation for the ridge between
North and SouthAmerica. It is suggestedthat the ridge jumpedfrom its
Bahamian position to that shownin Figure 17; at 80 Ma, this newridge
position wouldbecomethe site of a northward dipping subductionzone
(VenezuelanAntilles). The Greater Antilles Arc (GA) continued its
migration toward the Yucatan and Bahamas platforms.
95 Ma. At 95 Ma, opening of the Equatorial Atlantic continued and
the Greater Antilles Arc migrated farther into the North-South American
gap consumingthe Proto-Caribbean ridge system continously.
i! , W
,_.._.
Fñg. 14. ?aleoreconstruction at 150

80 Ma. The Late Cretaceous was a time of great complexity. The

western portion of the Greater Antilles Arc, which would become the
Nicaragua Rise and Jamaica, collided with the Yucatan block sending
thrust sheets northward onto the shelf bank of southern Yucatan
[Donnelly, 1977]. Remnants of rocks involved in this collision are seen
today in the Motagua Suture Zone (msz). The eastern remainder of the
arc continued to migrate northwards, with much internal deformation
(Wagwater and Montpellier--Newmarket Troughs), toward the BahamasPlat-
form. Volcaniclastic, flysch-like sequences of Late Cretaceous age in
southern Haiti [F. Maurrasse, personal communication, 1982] may have
originated at the Jamaica-Yucatan orogenic zone.
Because of a major change in the relative motion of South America
with respect to North America, the plate boundary between the two became
compressional and thus plate convergence was initiated. The former
ridge became the site of a north dipping subduction zone whose associ-
ated andesitic volcanism formed the Netherlands-Venezuelan Antilles Arc
(nva). Subduction of oceanic crust of the South America Plate led to
arc-continent collision at the end of the Cretaceous [Gealey, 1980].
The dashed lines of Figure 19 indicate that these features, appar-
ently, did not become active until later in the 80 Ma to 65 Ma interval.
Initiation of subduction of Pacific crust beneath the Panama-Costa Rica
125 mo
-•,•.

Arc (p-cr) [Galli-Oliver, 1979] and motion on the Hess Escarpment (h.e.)
occurred during this interval. These features apparently formed in
oceanic crust that was affected by the widespread B'' volcanic event
during early Late Cretaceous time [Burke et al., 1978]. Apparent B"
occurrences at Nicoya [Schmidt-Effing, 1979] and within the Caribbean
Sea indicate that the Caribbean has a deep, plateau-type crust.
As the leading edge of the Caribbean Plate migrated north of the
North America-South America plate boundary, a new plate boundary was
formed that would become the site of subduction whose volcanism would
95 mo
SA / :
AF/SA

80 me
•"< p.-c.r.
B• • •
form the calc-alkaline rocks of the Aves Ridge (at) in latest Cretaceous-
early Tertiary time.
6S Ma. At about 6S Ma, arc-continent collision occurred between
South America and the Venezuelan Antilles. The subsequent relative
motion between North and South America from 6S to S5 Ma had a slightly
extensional component and it is suggested that early Eocene rifting
(Bonaire, Maracaibo-Falcon) in northern South America are related to
this minor extensional phase.
In the Greater Antilles, continued northward migration was complex.
65 me
Fig. 20. Paleoreconstruction at 65 Ha.

The Yucatan Basin (yb) was formed by back-arc spreading between Cuba
(frontal arc) and the southern Hispaniola block (remnant arc) from
latest Cretaceous to middle Eocene times [Gealey, 1980]. Age-depth
relationships of the Yucatan Basin floor indicate an age within the
early Eocene-Paleocene interval, and grabens of southern Cuba which are
probably associated with the extensional event are filled with Paleocene-
Lower Eocene basalts and clastics [Iturralde-Vinent, 1978]. Uplift and
erosion along the eastern Yucatan Peninsula in Paleocene time [Weidie
et al., 1980] are interpreted as a result of strike-slip passage of the
frontal arc (Cuba). Thus Cuba migrated toward eventual collision with
the Bahamas Platform by the formation of a back-arc basin behind it.
To the east, the remainder of the Greater Antilles migrated toward
the Bahamas in a different fashion. Motion along the Hess Escarpment
(h.e.) severed the southern portion of the arc and allowed northward
migration at a rate similar to the rate of spreading in the Yucatan

Basin. Thus the Hess Escarpment was connected to the Yucatan Basin
ridge system, in the Cauto Basin of Cuba (½b), and to the trench asso-
ciated with the Panama-Costa Rica Arc. The motion on the []ess Escarp-
ment also allowed juxtaposition of the Panama-Costa Rica Arc with the
Chortis block; the suture between the two is defined by the Santa-Elena
Peridotite [aleBoer, 1979].
53 Ma. In about middle Eocene time, the Greater Antilles collided
with the BahamasPlatform [Gealey, 1980]. Eocene migration of rodents
from North America to South America [C. Woods, personal communication,
1982] may have been achieved via the Bahamas/Greater Antilles-Aves Ridge
connection. In the Greater Antilles, carbonate deposition accompanied
general subsidence and tectonic quiescence throughout the remainder of
the Eocene. Similarly, subduction at the Aves Ridge effectively ceased
as the Caribbean Plate moved very little with respect to the North
American Plate. Motion between the North and South American plates,
although minor, was approximately strike-slip and probably occurred
along the northern South American borderland.

Pind•11 and Dewey: Reconstruction of Western Pangea 205
Fig. 25. Plate boundaries at present.
36 Ma. At 36 Ma, eastward migration of the Caribbean Plate with

respect to North and SouthAmericabeganand wasaccompanied by develop-
ment of the North and SouthCaribbeanPlate BoundaryZones[Burke et al.,
1978, 1980]. In a mantle reference frame, westwardmigration of North
and SouthAmericaprobably has exceededeastwardmigration of the
Caribbean Plate. For simplicity and because the relative motion between
North and South America since 36 Ma is very minor, the motions of the
Caribbean Plate are described in terms of North and South America
reference frames.
Post-Eocenetectonics in the Greater Antilles have been complex,
involving large scale strike-slip and associated basin and thrust
development
(transtensionand transpression). Spreadingbeganat the
Caymanspreading center and transform motion has occurred predominantly
along the Motagua-Swan-Orientefault system. Estimatesof the length of
the Cayman Troughindicate that about 1200 kmof strike-slip has
occurredbetweenthe North Americanand Caribbeanplates. This figure
is a minimum, owingto the fact that manysmaller scale strike-slip
faults in Jamaicaand southernHispaniola have undergonemotion of the
same sense. The Oriente fault apparently developed within the remnant
arc of the Greater Antilles, thereby separating the Cayman Ridge and
Oriente Province of Cubafrom southernHispaniola and central Hispaniola.
One result of the Oligocene to present strike-slip phase has been the
juxtaposition of at least three blocks that comprisethe island of
Hispaniola. Figure 22 showsonemajor splay of the fault systempassing
betweenthe central and northern blocks of Hispaniola, which today are
separated by the Cibao Valley. The Tabera Basin in the southern Cibao
has recently been interpreted as an Oligocene aged pull-apart basin asso-
ciated with early eastward migration of the Caribbean Plate
(J. C. Cooper, M.S. thesis in preparation). Another feature in this
area probably associated with the transform motion is the uplift of ser-
pentinite slivers along the northern flank of the Cordillera Central.
The timing of relative motion betweencentral and southernHispaniola
remains unclear, although basalts interbedded with Oligocene (?) lime-
stones in the Sierra Neiba [F. Maurrasse, personal communication,1982]
may be related to strike-slip motions in this area at that time as well.
In the east, renewed subduction of Atlantic crust beneath the
Caribbean Plate led to developmentof the Lesser Antilles Arc (la). It
is suggestedthat a temporary lull in subduction or perhaps back-arc
spreading from 53 to 36 Ma is responsible for the eastward migration of
206 Pindell and Dewey- Reconstruction of Western Pangea
•/early
Triassic
NORTH
AMERICA
/ •••65
, 'P"'
6'.y
140
ß
' _ • 140
SOUTH
AMERICA
Fig. 26. Relative motion vectors of South America and Africa with
respect to North America based on data presented in Sclater et al.
[1977].
the volcanic axis along the eastern Caribbean Plate boundary. In the
south, a large portion of the Caribbean Plate was subducted beneath
Colombia/Ecuador throughout the latter part of the Tertiary. The fact
that the Netherlands-Venezuelan Antilles arc has not been greatly
separated from the Villa de Cura basement thrust which was obducted onto
northern South America [Gealey, 1980] indicates that most of the
Caribbean-South American plate offset has occurred to the north of the
island chain. The complex South Caribbean Plate Boundary Zone has
developed due to the shear couple created at the primary strike-slip
interface. Thus, plate boundary zone related deformation swept eastward
across Venezuela, having begun earlier in western Venezuela than in
eastern Venezuela.
21 Ma. Hastward migration of the Caribbean Plate continued. The
Lesser Antilles overrode Atlantic oceanic crust so that magnetic data
pertaining to relative motions between North and South America were des-
troyed. Motion between central and southern Hispaniola had definitely
begun by this time [J. C. Cooper, M.S. thesis, in preparation].
10 Ma. Continued eastward migration consumed more Atlantic crust,
juxtaposed the three masses of Hispaniola via compressional strike-slip
motion along the flanks of the Cibao and San Juan/Enriquillo Valleys,
and led to arc-continent collision between Panama and northwest Colombia
in late Miocene or early Pliocene time, which led to the uplift of the
three cordillera of Colombia [Irving, 1975]. Numerouspull-apart basins
within the North and South Caribbean Plate Boundary Zones formed in Late
Miocene time due to imperfect strike-slip motion on many minor faults.
Three causes for this seem likely. First, the Panama-Colombia arc-
continent collision may have complicated continued eastward motion of
the Caribbean Plate. Second, convergent relative motion between North
and South America during the past 21 million years (Figure 12) may have
reached a critical point by Late Miocene time that was sufficient to
create internal deformation within the Caribbean Plate, thereby affecting
Plate boundary flow lines. The position and the Neogene uplift of the
Beata Ridge support this viewpoint. Third, the pole describing the
motion of the Caribbean Plate with respect to North and South America
may have shifted position in Late Miocene times so that strike-slip
faults and flow lines about the new pole became non-parallel.
0 Ma: the present. The present plate boundary configuration differs
little from the 10 Ma reconstruction except that an exceedingly complex
zone of deformation has developed in the Panama-Colombiaregion in
response to continued eastward migration of the Caribbean Plate today.
Strike-slip faulting with associated basin and thrust development is
continuing throughout the Northern and Southern Plate Boundary Zones.
In the western portion of the Chortis block, a back-arc basin may be
developing along the Nicaragua Depression. If this is so, the Caribbean
Plate may be moving eastward at a slow rate with respect to the mantle.
D I SCUSS I ON
To be sure, the preceding evolutionary model is not intended as a

definitive statement on the evolution of the Caribbean Plate. It is
merely presented as a model to be chewed over and improved. The authors
are aware that considerable error may exist in the finite difference
data of Table 4.
Both the initial Pangean reconstruction and the relative motion vec-
tors are highly dependent upon the proposed South Atlantic fit which
requires 1) considerable Cretaceous African deformation and 2) a pre-
rift relationship between Brazil and the Guinea margin that is tighter
than that described by Bullard et al. [1965]. It is felt that sub-
stantial evidence for this proposal exists, in so far as a qualitative
analysis may be made, but more quantitative work is needed to gain a
more confident assessment of the plate kinematics involved. Perhaps a
simple re-examination of existing raw data would suffice.
If the proposal turns out to be unacceptable after examination of the
data, the authors suggest that the most plausible alternative would
involve a continental reconstruction similar to that proposed by
Van der Voo and French [1974]. The Yucatan block could remain in its
Gulf of Mexico position described in this study so that North and South
American Late Paleozoic geology is satisfied. However, to establish the
pre-rift relationship between North America and Africa of LePichon and
Fox [1971] and Klitgord and Schouten [1980], major right-lateral shear
must have occurred along the Atlantic borderlands about a pole in the
Sahara region of Africa [as proposed by Van der Voo and French, 1974].
Following this approach and using data presented in Sclater et al.
[1977], the relative motion vectors of South America with respect to
North America are as shown in Figure 26. Another similar model of
Gulf/Caribbean evolution could be constructed but significant differences
would result accordingly. For example, the model would predict 1) an
earlier age of origin for the Gulf of Mexico than for the Atlantic
Ocean; 2) subduction at the Venezuelan Antilles from 12S Ma to 6S Ma;
and 3) a period of oceanic spreading between the North and South
American Plates during the 6S to 36 Ma interval.
Acknowledgments. The first author offers his gratitude to Florentin

Maurrasse and the Miami Geological Society for providing the special
opportunity to tour the geology of Haiti in early 1982, and to Florentin
Maurrasse, Fred Nagle, Gren Draper, Geoff Wadge, Fred Taylor, Vernon
Hunter and Tony Eva for their commentsand criticism of the evolutionary
modelpresented herein. During 1979 and 1980, the authors were provided
with the opportunity to study in detail the Late Paleozoic and Mesozoic
geology of the circum-Gulf/Caribbean area with Phillips Petroleum
Cmnpany. We extend our thanks to Phillips for allowing us access to
proprietary information that helped with the conception of the model
presented herein. We are indebted to Kevin Burke, John LaBrecque, Cal
Cooper, Paul Mann, David Rowley, Eric Rosencrantz, Kim Klitgord, and
Walter Pitman for many helpful discussions during the past two years.
REFERENCES
Almeida, F. F. N., Expanatory note to tectonic map of South America,

scale 1:5,000,000. 21 pp., Departmento Nacional da Producgo Miner-
al.-Commission for the Geological Map of the World.-United Nations
Educational, Scientific and Cultural Organization, 1978.
Anderson, T. H., Geology of the San Sabastian Huehuetenango quadrangle,
Guatemala, Central America, Ph.D. dissertation, Univ. of Tex. at
Austin, Austin, 1969.
Anderson, T. H., B. Burkart, R. E. Clemons, O. H. Bohnenberger, and
D. N. Blount, Geology of the Western Altos Cuchumantanes, Northwestern
Guatemala, Geol. Soc. Am. Bull., 84, 805-826, 1973.
Antoine, J. W., R. Martin, T. Pyle and W. R. Bryant, Continental Margins
of the Gulf of Mexico, in The Geology of Continental Margins, edited
by C. A. Burk and C. L. Drake, pp. 683-693, Springer-Verlag, New
York, 1974.
Asmus, H. E., and F. C. Ponte, The Brazilian marginal basins, in The
Ocean Basins and Margins, vol. 1, The South Atlantic, edited by A.
E. M. Nairn and F. G. Stehli, pp.87-132, Plenum, New York, 1973.
Briggs, G., and D. }t. Roeder, Sedimentation and plate tectonics, Ouachita
biountains and Arkoma Basin, in Sedimentology of Paleozoic Flysch and
Associated Deposits, Ouachita Mountains-Arkoma Basin, Oklahoma, edited
by G. Briggs, pp. 1-22, Dallas Geological Society, Dallas, Tex., 1975.
Buffler, R. T., J. S. Watkins, F. J. Shaub, andJ. L. Worzel, Structure
and early geologic history of the deep central Gulf of Mexico basin
in The Origin of the Gulf of Mexico and the Early Opening of the Cen-
tral North Atlantic Ocean, edited by R. Pilger, pp. 3-16, Louisiana
State University, Baton Rouge, 1980.
Buffler, R. T., F. J. Shaub, R. Huerta, A. B. Ibrahim, and J. S. Watkins,
A model for the early evolution of the Gulf of Mexico basin, Oceanol.
Acta, 129-136, 1981.
Bullard, E. C., J. E. Everett, and A. G. Smith, The fit of the continents
around the Atlantic, Philos. Trans. R. Soc. London, 258A, 41-51, 1965.
Burke, K., Development of graben associated with the initial ruptures of
of the Atlantic Ocean, Tectonophysics, 36(1-3), 93-112, 1976.
Burke, K., and J. F. Dewey, Two plates in Africa during the Cretaceous,
Nature, 249, 313-316, 1974.
Burke, K., P. J. Fox, and S. Fl. C. SengOr, Bouyant ocean floor and the
evolution of the Caribbean, J. Geophy. Res., 83(B8), 3949-3654, 1978.
Burke, K., J. Grippi, and A.M. C. SengOr Neogene structures in Jamaica
and the tectonic style of the northern Caribbean plate boundary zone,
J. Geol., 88, 375-386, 1980.
Case, J. E., and W. R. MacDonald, Regional gravity anomalies and crustal
structure in northern Colombia, Bull. Geol. Soc. Am., 84(9), 2905-2916,
1•73.
Cebull, S. E., and D. H. Shurbet, The Ouachita Belt in the Evolution of
the Gulf of Mexico in The Origin of the Gulf of Mexico and the Early
Opening of the Central North Atlantic, edited by R. H. Pilger, pp. 17-
26, Louisiana State University, Baton Rouge, 1980.
Cook, F. A., D. S. Albough, L. D. Brown, S. Kaufman, J. E. Oliver, and
R. D. Ilatcher, Jr., Thin-skinned tectonics in the crystalline Southern
Appalachians, COCORP seismic-reflection profiling of the Blue Ridge
and Piedmont, Geology, 7, 563-567, 1979.
de Boer, J., The outer ar• of the Costa Rican orogen (oceanic basement
complexesof the Nicoya and Santa Elena Peninsulas), Tectonophysics,
56, 221-259, 1979.
de Cserna, Z., Taconian (early Caledonian) deformation of the Huastecan
structural belt of eastern Mexico, Am. J. Scio, 271, 544-550, 1971.
de Cserna, Z., Mexico--Geotectonics and mineral deposits, N.M. Geol.
Soc. Spec. Publ., 6, 18-25, 1976.
Denison, R. E., W. H. Burke, Jr., E. A. Hetherington, and J. B. Otto,

Basement rock framework of parts of Texas, southern New Mexico, and
northern Mexico, in The Geologic Framework of the Chihuahua Tectonic
Belt, edited by K. Seewald and D. Sundeen, pp. 3-14, West Texas Geolo-
gical Society, Midland, Tex., 1970.
Dewey, J. F., Finite plate rotations' Some_implications for the evolution
of rock masses at plate margins, Am. J. Sci., 275A, 260-284, 1975.
Donnelly, T., Metamorphic rocks and structural history of the Motagua
Suture Zone, eastern Guatemala, paper presented at Eighth Caribbean
Geological Conference, Curacao, Willemstad, 1977.
Flawn, P. T., Tectonics, The Ouachita System, Publ. 6120, pp. 163-172,
Bur. of Econ. Geol., Univ. of Tex. at Austin, Austin, 1961.
Francheteau, J., Plate tectonics model of the opening of the Atlantic
Ocean south of the Azores, in Implications of Continental Drift to
the Earth Sciences, edited by D. H. Tarling, and S. K. Runcorn, pp.
197-202, Academic, New York, 1973.
Frezon, S. E., and G. H. Dixon, Texas Panhandle and Oklahoma, U.S.
Geol. Surv. Prof. Pap. 853, 177-196, 1975.
Galli-Olivier, C., Ophiolitic and island-arc volcanisn in Costa Rica,
Geol. Soc. Am. Bull., 90, 444-452, 1979.
Gealey, W. K., Ophiolite obduction mechanism, in Ophiolites' Proceedings
of the International Ophiolite Symposium, edited by A. Panayiotou,
pp. 228-243, Geological Survey Department, Nicosia, Cyprus, 1980.
Goldstein, A., Jr., The Ouachita Mountains of Oklahoma and Arkansas,
The Ouachita System, Publ. 6120, pp. 21-48, Bur. of Econ. Geol., Univ.
of Tex. at Austin, Austin, 1961.
Grabert, H., The tectonogemsis of the Amazon basin, Mem. Segundo Congr.
Latinoam. Geol., 3, 2201-2217, 1977.
Graham, S. A., W. R. Dickinson , and R. ¾. Ingersoll, Himalayan-Bengel
model for flysch dispersal in Appalachian-Ouachita system, Geol. Soc.
Am. Bull., 86, 273-286, 1975.
Gries, J. C., P---roblems
of delineation of the Rio Grande Rift into the
Chihuahua Tectonic Belt of Northern Mexico, in Rio Grande Rift' Tec-
tonics and Magmatism, edited by R. E. Riecker, pp. 107-114, AGU,
Washington, D. C., 1979.
Hatcher, R. K., Jr., Tectonics of the Western Piedment of Blue Ridge,
Southern Appalachian' Review and speculation, Am. J. Sci., 278,
276-304, 1978.
Hayes, D. E., A. C. Pimm, J. B. Beckman, W. E. Benson, W. H. Berger,
P. H. Roth, P. R. Supko, and U. von Rad, Shipboard site reports, site
137, Initial Rep. Deep Sea Drill. Proj., 14(975), 85-134, 1972.
Hoffman, P. J. F. Dewey, and K. C. A. Burke, Aula½ogensand their
genetic relation to geosynclines,with a Proterozoi½examplefrom
Great Slave Lake, Canada, Soc. Econ. Paleontol. Mineral., Spec. Publ.
19, 38-55, 1974.
Irvi-•ng,E. M., Structural evolution of the northernmostAndes,Colombia,
U.S. Geol.Surv. Prof. Pap. 846, 1-47, 1975.
Iturralde-Vinent, M. A., Los movimientos tectonicos de la etapa de
desarrollo plat aformico en Cuba, Geol. Mijnbouw, 57(2), 205-212, 1978.
King, P. G., The Ouachita and Appalachian orogenic b•lts, in The Ocean
Basins and Margins, vol. 3, Gulf of Mexico and Caribbean, edited by
F. G. Stehli, and A. Nairn, pp. 201-241, Plenum, New York, 1975.
King, P. B., Marathon revisited, in Symposium on the Geologyof the
Ouachita Mountains, vol. 1, pp. 41-69, Arkansas Geological Commission,
Little Rock, 1977.
Klitgord, K. D., and It. Schouten,Mesozoicevolution of the Atlantic,
Caribbean and Gulf of Mexico, in The Origin of the Gulf of •iexico and
the Early Openingof the Central North Atlantic, edited by R. H.
Pilger, Jr., pp. 100-101, Louisiana State University, Baton Rouge,
1980.
Klitgord, K. D., P. Popehoe, and J. Schouten, Florida' A Jurassic trans-

form plate boundary, submitted to J. Geophys. Res., 1982.
Kluth, C. R., and P. J. Coney, Plate tectonics of the ancestral Rocky
Mountains, Geology, 9, 10-15, 1981.
Kumar, N., G. Bryan, M. Gorini, and J. Carva!ho, Evolution of the conti-
nental margin off northern Brazil- Sediment distribution and carbon
potential, An. Acad. Bras. Cienc., 48, 151-145, 1976.
Kumar, N., R. Leydon, J. Carvalho, and O. Francisconi, Sediment isopach
map, Brazilian continental margin, Number 2752, Lamont-Doherty Geolo-
gical Observatory, Palisades, N.Y., 1979.
Ladd, J. W., South Atlantic sea-floor spreading and Caribbean tectonics,
Ph.D. thesis, 251 pp., Columbia Univ., New York, 1974.
Ladd, J. W., Relative motiQn of South America with respect to North
America and Caribbean tectonics, Geol. Soc. Am. Bull., 87(60702),
969-976, 1976.
LePichon, X., and P. J. Fox, Margin offsets, fracture zones and the early
opening of the North Atlantic, J. Geophys. Res., 76, 6294-6508, 1971.
LePichon, X. and D.C. Hayes, Marginal offsets, fracture zones and the
early opening of the South Atlantic, J. Geophys. Res., 76, 6285-6295,
1971.
MacDonald, W. D., Geology of the Serrania De Macuria Area Guajira Penin-
sula, Northeast Colombia, Caribb. Geol. Conf. Trans. 4th, 267-275,
1965.
Maresch, W. V., Plate tectonics origin of the Caribbean Mountain System
of northern South America' Discussion and proposal, Geol. Soc. Am.
Bull., 85, 669-682, 1974.
Maxwell, J. C., Geology of Tobago, British West Indies, Geol. Soc. Am.
Bull., 59, 801-854, 1948.
McBirney, A. R., and M. N. Bass, Structural relations of pre-Mesozoic
rocks of northern Central America, Am. Assoc. Pet. Geol., Mem. 11,
269-280, 1969.
McKenzie, D. P., Some remarks on the development of sedimentary basins,
Earth Planet. Sci. Lett., 40, 25-52, 1978.
McKenzie, D, P., and W. J. Morgan, Evolution of triple junctions, Nature,
224, 125-255, 1969.
Muehlberger, W. R., Late Paleozoic movement along the Texas Lineament,
Trans. N.Y. Acad. Sci., 2(27), 585-592, 1965.
Muehlberger, W. R., The Honduras depression, Publ. Geol. ICAITI, 5, 45-
51., 1976.
Murat, R. C., Stratigraphy and palaeography of the Cretaceous and Lower
Tertiary in Southern Nigeria, in African Geolo.gM, edited by T. F. J.
Dessauvagie, and A. J. Wiseman, pp. 251-268, University of Ibidan,
Nigeria, 1972.
Rabinowitz, P. D. and J. LaBrecque, The Mesozoic South Atlantic Ocean
and evolution of its continental margins, J. Geophy. Res., 84(Bll),
5975-6002, 1979.
Schmidt-Erring, R., Geodynamic history of oceanic crust in Southern Cen-
tral America, Fourth Latin American Geological Congress, Geol.
Rundsch., 68, 457-492, 1979.
Sclater, J. B., R. Jarrat, B. McGowran, and J. Gartner, Comparison
of the magnetic and biostratigraphic time scales since the Late
Cretaceous, Initial Rep. Deep Sea Drill. Proj., 22, 581-586, 1974.
Sclater, J. B., S. [tellinger, and C. Tapscott, The paleobathymetry of the
Atlantic Ocean from the Jurassic to the Present, J. Geol., 85, 509-552,
1977.
Sibuet, J. C., and J. Mascle, Plate kinematic implications of Atlantic
equatorial fracture zone trends, J. Geophys. Res., 85(B7), 5401-5421,
1978.
Silver, E. A., and L. T. Anderson, Possible left-lateral early to middle
Mesozoic disruption of the southwestern North American craton margin,
Geol. Soc. Am. Abstr. Prog., 6, 955, 1974.
Smith, A. G., and J. C. Briden, Mesozoic and Cenozoic Paleo-contñnental

Maps, p. 63, Cambridge University Press, New York, 1977.
Stewart, J. H., Late Precambrian evolution of North America--Plate tecto-
nics implication, Geology, 4, 11-15, 1976.
Tomlinson, C. W., and W. D. McBee, Jr., Pennsylvanian sediments and
orogenies of the Ardmore district, Oklahoma, paper presented at Petro-
leum Geology of Southern Oklahoma Symposium, Ardmore Geol. Soc.,
Ardmore, Okla., 1959.
Tschanz, C. M., R. F., Marvin, B. J. Cruz, H. Mehnert, and G. T. Cebula,
Geologic evolution of the Sierra Nevada de Santa Marta, northeastern
Colombia, Geol. Soc. Am. Bull., 85, 273-284, 1974.
Valencio, D. A., and J. F. A. Vilas, Sequence of the continental move-
ments occurred prior to and after the formation of the South Atlantic,
An. Acad. Bras. Cienc., 48, 377-386, 1976.
Van der Voo, R., and R. B. French, Apparent polar wandering for the
Atlantic bordering continents' Late Carboniferous to Eocene, Earth ,
Sci. Rev. , 10, 99-119, 1974.

Van Hinte, J. E., A Cretaceous time scale, Am. Assoc. Pet. Geol. Bull.,
60(4), 498-516, 1976a.
Van Hinte, J. E., A Jurassic time scale. Am. Assoc. Pet. Geol. Bull.,
60(4), 489-497, 1976b.
Weidie, A.D., W. C. Ward, and R. H. Marshail, Geology of the Yucatan
Platform, in Geology of Cancun, Quintana Roo, Mexico, edited by W. G.
Thorsen, pp. 50-68, West Texas Geological Society, Midiand, Tex., 1980.
Wickham, J., D. Roeder, and G. Briggs, Plate tectonics modeis for the
Ouachita foldbelt, Geology, 4(3), 1973-176, 1976.
Wilson • J ß T .• Did the Atlantic close and the re-open ? Nature ., ,
211(5050), 676-681, 1966.

Ziegler, A., C. Scotese, W. McKerrow, M. Johnson, and R. Bambach, Pal-
eozoic paleogeography, Annu. Rev. Earth Planet. Sci., 7, 473-502, 1979.
(Received January 20, 1982;

accepted January 25, 1982.)
View publication stats

Pindell Dewey 1982

Uploaded by

Copyright:

Available Formats

Pindell Dewey 1982

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pindell Dewey 1982

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author proﬁles for this publication at: https://www.researchgate.

Permo-Triassic Reconstruction of Western Pangea and the Evolution of the

Article in Tectonics · April 1982

James Pindell John F. Dewey

SEE PROFILE SEE PROFILE

Paleogeography of the Wilcox View project

The user has requested enhancement of the downloaded ﬁle.

PEltO-TRIASSIC RECONSTRUCTION OF WESTERN PANGEA AND THE EVOLUTION

JamesPindell and John F. Dewey

Department of Geological Sciences, State Universit), of New York at

Abstract. A Pe•no-Triassic reconstruction of western Pangea (North

The relative motion of two plates separated by an oceanic ridge can

1 Present address: Department of Geological Sciences, Science

Copyright 1982 by the American Geophysical Union.

RECONSTRUCTION OF WESTERN PANGEA

Late Paleozoic Geology of Southern North America

A fairly continuous belt of Permo-Carboniferous deformation passes

Fig. 1. Simplified tectonic mapof Late Paleozoic geology of soutnern

recumbent folds and southward-dipping thrust faults, the motion upon

Pedregosa, Delaware, and Midland basins (although many of the basins

TABLE 1. Summary of Itercwnian Deformation

Zonc Time of Thrusting/Collision

Southern Appalachians began in Late Mississippian (Chesterian), con-

Ouachitas Atokan to Desmoinesian

Marathons Virgilian to l•olfcampian

Huastecans Early to [,•iddle Permian

Late Paleozoic Geology of Northern South America

As noted earlier, direct evidence to indicate the location of the

Fig. 2. Late Paleozoic terrains of northern South America. 1, Cordil-

to Middle Permian) and the ceasing of arc-related volcanism in Colombia

Extent of Pre-Mesozoic Continental Crust:

For the purposes of the Penno-Triassic reconstruction, the islands of

Fig. 3. Northern limit of continental crust of South America in Juras-

of Paleozoic age is present beneath the coastal plains ICebull and

Fig. 4. Limit of continental crust along the Gulf margin of the

In southernmost Florida (Figure 4), Klitgord et al. [1982] show that

The Yucatan Block

Although the Yucatan Block (Figure 5) is of lñmited geographic extent,

Wolf. Esperanza Fm. shale, minor LS •

Fig. 6. Stratigraphic column and tectonic interpretation of the Late

overlain by Carboniferous strata. A late Paleozoic Andean arc is

time. However, the opposing rifted margins are so highly attenuated

The Chortis Block

Little is known of the pre-Hesozoic basement of the Chort•s Block

ica Trench. The Nicaragua Depression [Muehlberger, 1976] (Figure 7) is

The African Connection

Two Plates in Africa and the Revised .•puth Atlantic Fit

The most commonly accepted South Atlar•tic fit is that of Bullard et

northeast coastline of Brazil. This rotation requires that 100 km of

BER 118 bos•l

Fig. 9. Generalizedmapof Africa showing Early Cretaceous

Although Grabert [1977] reported Late Jurassic faulting and minor

Fig. 10. Difference between pre-Cretaceous and present shapes of

opening phase of the southern South Atlantic Ocean (Mll-MO, or 125-110

Fig. 11. Sketch map showing revised Pertoo-Triassic continental recon-

Reconstruction of Western Pan•ea

Figure 11 depicts the pre-rift relationship of the circum-Atlantic

EVOLUTION OF THE GULF OF MEXICO/CARIBBEAN REGION

Relative Motion History of the Major Continents

TABLE 2. Relative Motion Poles,