The Proto-Andean Margin of Gondwana (Geological Society Special Publication No. 142)

The Proto-Andean Margin of Gondwana

Geological Society Special Publications Series Editors A. J. FLEET A. C. MORTON A. M. ROBERTS

It is recommended that reference to all or part of this book should be made in one of the following ways. KELLER, M., BUGGISCH, W. & LEHNERT, O. 1998. The stratigraphical record of the Argentine Precordillera and its plate-tectonic background. In: PANKHURST, R. J. & RAPELA, C. W. (eds) The Proto-Andean Margin of Gondwana. Geological Society, London, Special Publications, 142, 35-56. PANKHURST, R. J. & RAPELA, C. W. (eds) 1998. The Proto-Andean Margin of Gondwana. Geological Society, London, Special Publications, 142.

GEOLOGICAL SOCIETY SPECIAL PUBLICATION NO. 142

The Proto-Andean Margin of Gondwana

EDITED BY

R O B E R T J. P A N K H U R S T British Antarctic Survey and NERC Isotope Geosciences Laboratory, UK AND

CARLOS W. R A P E L A Universidad Nacional de La Plata, Argentina

1998 Published by The Geological Society London

THE GEOLOGICAL SOCIETY The Society was founded in 1807 as The Geological Society of London and is the oldest geological society in the world. It received its Royal Charter in 1825 for the purpose of 'investigating the mineral structure of the Earth'. The Society is Britain's national society for geology with a membership of around 8500. It has countrywide coverage and approximately 1500 members reside overseas. The Society is responsible for all aspects of the geological sciences including professional matters. The Society has its own publishing house, which produces the Society's international journals, books and maps, and which acts as the European distributor for publications of the American Association of Petroleum Geologists, SEPM and the Geological Society of America. Fellowship is open to those holding a recognized honours degree in geology or cognate subject and who have at least two years' relevant postgraduate experience, or who have not less than six years' relevant experience in geology or a cognate subject. A Fellow who has not less than five years' relevant postgraduate experience in the practice of geology may apply for validation and, subject to approval, may be able to use the designatory letters C Geol (Chartered Geologist). Further information about the Society is available from the Membership Manager, The Geological Society, Burlington House, Piccadilly, London WlV 0JU, UK. The Society is a Registered Charity, No. 210161. Published by The Geological Society from: The Geological Society Publishing House Unit 7, Brassmill Enterprise Centre Brassmill Lane Bath BA1 3JN UK (Orders: Tel. 01225 445046 Fax 01225 442836)

Distributors

USA AAPG Bookstore PO Box 979 Tulsa OK 74101-0979 USA (Orders: Tel. (918) 584-2555 Fax (918) 560-2652)

First published 1998

Australia The publishers make no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility for any errors or omissions that may be made. 9 The Geological Society 1998. All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with the provisions of the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 9HE. Users registered with the Copyright Clearance Center, 27 Congress Street, Salem, MA 01970, USA: the item-fee code for this publication is 0305-8719/98/$10.00.

Australian Mineral Foundation 63 Conyngham Street Glenside South Australia 5065 Australia (Orders: Tel. (08) 379-0444 Fax (08) 379-4634)

India Affiliated East-West Press PVT Ltd G-l/16 Ansari Road New Delhi 110 002 India (Orders: Tel. (11) 327-9113 Fax (11) 326-0538)

Japan British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. ISBN 1-86239-021-5

Typeset by Aarontype Ltd, Unit 47, Easton Business Centre, Felix Road, Bristol BS5 0HE, UK Printed by The Alden Press, Osney Mead, Oxford, UK

Kanda Book Trading Co. Cityhouse Tama 204 Tsurumaki 1-3-10 Tama-Shi Tokyo 206-0034 Japan (Orders: Tel. (0423) 57-7650 Fax (0423) 57-7651)

Contents

PANKHURST, R. J. & RAPELA, C. W. The proto-Andean margin of Gondwana: an introduction

1

ASTINI, R. A. Stratigraphical evidence supporting the rifting, drifting and collision of the Laurentian Precordillera terrane of western Argentina

11

KELLER, M., BUGGISCH, W. & LEHNERT, O. The stratigraphical record of the Argentine Precordillera and its plate-tectonic background

35

BENEDETTO, J. L. Early Palaeozoic brachiopods and associated shelly faunas from western Gondwana: their bearing on the geodynamic history of the pre-Andean margin

57

DICKERSON, P. W. & KELLER, M. The Argentine Precordillera: its odyssey from the Laurentian Ouachita margin towards the Sierras Pampeanas of Gondwana

85

HUFF, W. D., BERGSTRC)M, S. M., KOLATA, D. R., CINGOLANI, C. A. ~; ASTINI, R . A . Ordovician K-bentonites in the Argentine Precordillera: relations to Gondwana margin evolution

107

BAHLBURG, H. The geochemistry and provenance of Ordovician turbidites in the Argentine Puna

127

RAMOS, V. A., DALLMEYER, R. D. & VUJOVICH, G. I. Time constraints on the Early Palaeozoic docking of the Precordillera, central Argentina

143

VUJOVICH, G. I. & KAY, S. M. A Laurentian? Grenville-age oceanic arc/back-arc terrane in the Sierra de Pie de Palo, Western Sierras Pampeanas, Argentina

159

RAPELA, C. W., PANKHURST, R. J., CASQUET,C., BALDO, E., SAAVEDRA,J., GALINDO, C. & FANNING, C. M. The Pampean Orogeny of the southern proto-Andes: Cambrian continental collision in the Sierras de C6rdoba

181

DALLA SALDA,L. H., LOPEZDE LUCHI, M. G., CINGOLANI, C. A. & VARELA, R. LaurentiaGondwana collision: the origin of the Famatinian-Appalachian orogenic belt (a review)

219

VON GOSEN, W. & PROZZI, C. Structural evolution of the Sierra de San Luis (Eastern Sierras Pampeanas, Argentina): implications for the proto-Andean margin of Gondwana

235

SIMS, J. P., IRELAND, T. R., CAMACHO, A., LYONS, P., PIETERS, P. E., SKIRROW, R.G., STUART-SMITH, P. G. • MIRO R. U-Pb, Th-Pb and Ar-Ar geochronology from the southern Sierras Pampeanas, Argentina: implications for the Palaeozoic tectonic evolution of the western Gondwana margin

259

SAAVEDRA, J., TOSELLI, A., ROSSI, J., PELLITERO, E. & DURAND, F. The Early Palaeozoic magmatic record of the Famatina System: a review

283

GRISSOM, G. C., DEBARI, S. M. & SNEE, L. W. Geology of the Sierra de Fiambalfi, northwest Argentina: implications for Early Palaeozoic Andean tectonics LLAMBiAS, E. J., SATO, A. M., ORTIZ SUAREZ, A. & PROZZI, C. The granitoids of the Sierra de San Luis

297

PANKHURST, R. J., RAPELA, C. W., SAAVEDRA,J., BALDO, E., DAHLQUIST,J., PASCUA, I. & FANNING, C. M. The Famatinian magmatic arc in the central Sierras Pampeanas: an Early to Mid-Ordovician continental arc on the Gondwana margin

343

Index

369

325

The proto-Andean margin of Gondwana: an introduction R. J. P A N K H U R S T 1 & C. W. R A P E L A 2

1 British Antarctic Survey, c/o N E R C Isotope Geosciences Laboratory, Kingsley Dunham Centre, Keyworth, Nottingham NG12 5GG, UK 2 Centro de Investigaciones Geol6gicas, Universidad Naeional de la Plata, 644 Calle No. 1, 1900 La Plata, Argentina Abstract: A basic background is presented for the discussion of the Early Palaeozoic

geology of western Argentina covered by this book. This includes the definition and terminology of orogenic cycles on this part of the Gondwana margin, represented by the Eastern Sierras Pampeanas. The Pampean orogeny (Early Cambrian) relates to an intense but short-lived period of terrane collision predating the rifting of the Precordillera terrane from Laurentia. The Famatinian cycle is predominantly represented by intense subductionrelated magmatism of Early-Middle Ordovician age, developed on the continental margin of Gondwana during the rifting and drifting of the Precordillera terrane. The Grenvillian basement of the latter is further exemplified by a new Rb-Sr whole-rock isochron age of 1021 + 12Ma for orthogneisses from the Western Sierras Pampeanas. A mid-Ordovician granite in this area (dated at 481 4-6 Ma by U-Pb ion microprobe data) may be related to rifting while the Precordillera terrane was still attached to Laurentia. A divergence of opinion is pointed out between some authors in this book who favour mid-Ordovician collision of the Precordillera with Gondwana, and others who place it much latter, in Silurian or Devonian times. The Palaeozoic geological history of western South America has been the focus of intense international interest during the last five years. The development of this subject, and the growing state of our knowledge, can be traced from the idea of Palaeozoic continent-continent collision between North and South America along this margin (Dalla Salda et al. 1992a, b), through the proceedings of international conferences in M6rida, Spain ('El Paleozoico Inferior de IberoAmerica', Gutierrez-Marco & Saavedra 1992), in San Juan, Argentina (Penrose Conference 'The Argentine Precordillera: a Laurentian Terrane', Dalziel et al. 1996), and the international symposium 'The Proto-Andean Margin of Gondwana', co-organized by IGCP projects 345 and 376 (XIII Congreso Geol6gico Argentino, Actas V, 257-561, Buenos Aires, 1996). A major landmark was the overall consensus reached by participants in the 1995 Penrose Conference that the Precordillera of western Argentina is an allochthonous terrane, originally detached from Laurentia (Dalziel 1997 and references therein; see also Dickerson & Keller this volume) and accreted to the southwestern margin of Gondwana. The timing of this transfer, and the events and processes involved in the arrival and collision of the Precordillera allochthon, have far-reaching consequences for geological, biological and palaeogeographical interpretations and the distribution of mineral resources. Many of the implications are only just beginning to be understood.

The papers presented in this book come mainly from participants in the Buenos Aires meeting in October 1996, where the latest results from various Palaeozoic provinces were discussed, including recent information on the history of the Gondwana foreland before and during the accretion of the Precordillera terrane. For a general audience, appreciation of the arguments presented requires a basic knowledge of the common geographical, geological and tectonic terminology used by South American geologists. It is the purpose of this introduction to provide an adequate background for the non-specialist. The region concerned and its relevant geographical features are illustrated in Figs 1 and 2. The Sierras Pampeanas of western Argentina are elevated ranges of pre-Cenozoic basement rock, thought to result from uplift on reverse faults during upper Tertiary Andean deformation. This structural domain roughly coincides with a near-horizontal segment of the presently subducting Nazca p l a t e - the 'flat-slab zone' between 27~ t and 33~ (Isacks 1988; Jordan & Allmendinger 1986) (Fig. 1), also characterized by Miocene arc magmatism more than 700km inboard of the Chile trench (Kay & Gordillo 1994). The pre-Mesozoic geological record of the Sierras Pampeanas and the immediately adjacent regions is characterized by evidence of several periods of convergence along the proto-Pacific margin of Gondwana. This is the basis of the recognition of the major orogenic cycles, such as

PANKHURST, R. J. & RAPELA, C. W. 1998. The proto-Andean margin of Gondwana: an introduction. In: PANKHURST,R. J. & RAPELA, C. W. (eds) The Proto-Andean Margin of Gondwana. Geological Society, London, St3eciai Publications, 142, 1-9.

2

R. J. PANKHURST & C. W. RAPELA

Fig. 1. Sketch map of southern South America, showing the location of the main areas of Lower Palaeozoic outcrop in western Argentina and their general nomenclature.

the Pampean and Famatinian (Acefiolaza & Toselli 1973), Gondwanian (Llambias et al. 1984) and Andean cycles; a brief description of the meaning that we attach to each of these terms is given in Table 1. Based on local and distinct geological characteristics, they have effectively displaced imported European and North American terms such as 'Caledonian', 'Hercynian', 'Acadian' and 'Taconic', although ironically, with a new and better understanding of continent and supercontinent configurations, the real equivalence of some of these events is now ripe for re-evaluation. Recognition, definition and ordering of the discrete phases of deformation, metamorphism and magmatism that make up these cycles requires intensive research and interpretation, and is thus sometimes a contentious subject. The chronostratigraphical limits of these large-scale geological cycles in southern South America can be ascribed to major plate re-arrangements in the evolutionary stages of the Gondwana supercontinent. The changes between these stages, and therefore in the geodynamic environment along the margin, are mostly marked by initiation, or sudden acceleration, of subduction (see Table 1). It must be stressed that the word cycle

Fig. 2. Sketch map of the region of the Sierras Pampeanas and the Argentine Precordillera, the area encompassing the subject of most papers in this volume.

INTRODUCTION

3

Table 1. Phanerozoic geological cycles along the Pacific margin of South America

Pampean (Neoproterozoic-Late Cambrian) First recognized in the basement of the Eastern Cordillera, the Puna and the northwestern Sierras Pampeanas (Acefiolaza et al. 1990 and references therein), this Neoproterozoic-Cambrian cycle is now regarded as a major orogenic event in southwestern Gondwana. Along the southern proto-Andean margin, the Pampean cycle includes formation of a passive margin sedimentary basin, emplacement of metaluminous granitoid sequences, penetrative folding and low-high-grade metamorphism during mid-Cambrian time (see Rapela et al. this volume), and extensional collapse in the late Cambrian. The Pampean orogeny is broadly coeval with the collisional Rio Doce Orogeny, the last event of the Brasiliano-Pan African cycle along the southern coastal region of Brazil (Campos Neto & Figueiredo 1995). Famatinian (Early Ordovician-Early Carboniferous) A major accretional and orogenic episode that started with subduction along the Cambrian proto-Pacific margin (at c. 490 Ma). The Famatinian orogeny in the Gondwana foreland and the collision of the Precordilleran Terrane along the continental margin are usually both ascribed to this cycle, although they may not be causally related (see Pankhurst et al. this volume). Most of the contributions in this book are devoted to different aspects of this complex process, a summary of which is presented in Fig. 3. It should be noted, however, that mid-Palaeozoic tectonic phases registered in the clastic sequences of the Precordillera ('Precordilleran' and 'Chanic' tectonic phases, Fig. 3) have been related to the subsequent collision of another terrane ('Chilenia') to the west of the Precordillera (Ramos et al. 1986; Astini 1996). These tectonic events, as well as the broadly coeval intrusion of within-plate granites in the Gondwana foreland, are here tentatively included as the latest events of the Famatinian Cycle (see Fig. 3). Gondwanian (Early Carboniferous-Early Cretaceous) This was a period of the maximum extent and relative stability of the super-continent, from Lower Carboniferous (c. 330-340 Ma, see insert, Fig. 3) to final break-up. Following the Early and mid-Palaeozoic orogenies and accretionary events, intrusion of Carboniferous Cordilleran-type batholiths associated with a new subduction regime along the palaeo-Pacific margin heralded a major plate re-assembly. Development of extensive Permian to Jurassic rhyolitic provinces and inner cordilleran plutonic belts are characteristic of the Gondwanian cycle (Rapela et al. 1996). Andean (Early Cretaceous-Present) From the final break-up of Gondwana to the present. The Andean Cycle started with the opening of the South Atlantic and separation of South America from the African-Indian plate. This event was closely associated with extrusion of the Paranh-Etendeka flood basalt province (c. 130 Ma Renne et al. 1992; Turner et al. 1994). The generation of plutonic and volcanic rocks typical of the modern Andes became predominant as the convergence velocity at the palaeo-Pacific margin of the newly formed South American plate increased. The term 'Andean Orogeny' is linked to several compressive phases during this period, but is most commonly used for those of Tertiary and modern age. Formation of intra-continental extensional basins and a passive Atlantic margin are also typical of the Andean period.

is used here to describe stages during the accretion and break-up of a supercontinent, and not necessarily a regular or repeated succession of events. Not even the magmatic products of these subduction regimes are identical; the Famatinian, Gondwanian and Andean calcalkaline rocks each have their own distinctive characteristics. A summary of the major geological features and inferred tectonic events for the Precambrian and Early to mid-Palaeozoic provinces from 22 to 33~ is shown in Fig. 3, based on established literature but in some cases modified by the new results presented in this volume. The basic distinction of a Precordillera terrane and a Gondwana foreland seems useful to describe and contrast the main features of the adjacent continental masses that are inferred to have been amalgamated during the Lower Palaeozoic. The term Precordillera terrane is used to encompass

the Lower Palaeozoic sedimentary sequences of the Precordillera, and its Grenvillian-age basement represented by the adjacent parts of the Western Sierras Pampeanas, as in Astini et al. (1995); see also Ramos et al. (this volume) who use the name 'Cuyania' for this terrane. Early Palaeozoic magmatism and metamorphism is clearly recorded in central and N W Argentina (Sierras Pampeanas, Sierra de Famatina and the Puna), extending into southern Bolivia, northeastern Chile, and the coastal areas of southwestern Peru. The main components are the Ordovician plutonic and volcanic rocks of the Puna, the Famatina System, and the Central Batholithic Belt, which together constitute the Famatinian magmatic arc, and the Cambrian metamorphic and plutonic igneous sequences of the Eastern Sierras Pampeanas and the Eastern Cordillera (Fig. 1). Sedimentation in the Precordillera and granitoid magmatism in

4

R.J.

PANKHURST

& C. W .

RAPELA

O R O G E N I C E V E N T S IN T H E P R O T O - P A C I F I C M A R G I N OF G O N D W A N A System and

Gondwana

dillera T e r r a n e

foreland

Series I Pampean foreland :

[PC (~)] and

The Famatinian arc :

ts Pampeanas [WSP (~)1

Puna [P (~)], Famatina System (FS (~)] & Central batholithic belt [CBB (~)]

Eastern Cordillera [EC (~)1 & Eastern Sierras Pampeanas [ESP (~)]

of cordilleran plutons: CARBONIFEROU;

anicextensive tectonic phase mic basic and ultrabasic rocks layered with Silurian?)nian sediments

DEVONIAN

A

,cordilleran tectonic phase

SILURIAN

Ashgill

ne

Gondwanian

Llandeilo ORDOVICIAN Llanvirn

Gondwania~ faunas ] dominant

ii Angular Ocloyictectonic phase: unconformity

basic and ultrabasic rocks red with early Caradocian andacol tectonic phase

gional unconformity; psephitic deposits / ~ Mixed faunas

Arenig Tremadoc

i i

nites

31atform

t

~

S

Laurentian faunas dominant

- ?Continued magmatism, including emplacement of post-compressional monzogranite plutons - Syntectonic metaluminous batholiths. Calc-alkaline volcanics interlayered with shallow marine volcaniclastics. Opening of backarc basin.

- Transpressive structures associated with low-to-high grade metamorphism. Inner cordilleran plutons. High-A1 and Na trondhjemitic bodies. A

Famatinian Cycle i i i

M CAMBRIAN facies: evaporites bed sediments

] PRECAMBRIAN

between Lower Ordovician & Ashgillian deposits in EC and P. Regional mylonites in the ESP.

- Regional mylonites

i i

PC

t

Intrusion of large post-orogenic batholiths dominated by high-K porphyritic monzogranitic facies in the ESP (e.g. Achaia batholith) and the CBB (e.g. Capillitas and Velasco batholiths). Cordierite -bearing facies are common in the CBB

the PC Caradoc

Cycle

Intrusion of within- plate REE-enriched plutons

oic "Grenville" metamorphic the WSP and basement of the lites, marbles, schists and ing immature arc or oceanic ~s.

- Extension associated with collapse of the Pampean orogen. Late Cambrian platform sediments unconformably overlie the folded Puncoviscana Fm (EC, P). Alkalic volcanism in the EC & P. - Obduction of MORB-type ophiolites in the ESP. - Early-to-Mid-Cambrian compressive events. Folding and low-grade (EC, FS) to high- grade (ESP) metamorphism of the Puncoviscana Fm. and equivalents. Widespread anatexis and S-type plutonism in the ESE - Intrusion of metaluminous orthogneisses. - Passive margin sequence composed of widespread thick turbiditic deposits carrying Neoproterozoic to Lower Cambrian trace fossils in the EC and eastern Puna; southern equivalents in the FS, CBB and ESP, the Puncoviscana Formation.

Pampean Cycle Arequipa-Antofalla craton IAA (~)] - Early Proterozoic igneous and sedimentary rocks with an overprinted high-grade Grenville age event in the late Middle Proterozoic. Ages ranging from 1900 to 980 My.

Fig. 3. S c h e m a t i c r e p r e s e n t a t i o n o f the m a i n g e o l o g i c a l events r e c o g n i z e d in the P a l a e o z o i c h i s t o r y o f the region, with a c o m p a r i s o n o f s u c h event in the a l l o c h t h o n o u s P r e c o r d i l l e r a t e r r a n e a n d the G o n d w a n a f o r e l a n d ( E a s t e r n a n d W e s t e r n Sierras P a m p e a n a s ) .

INTRODUCTION the Eastern Sierras Pampeanas continued into Devonian-Carboniferous times, after the final assembly of this accreted margin. Carboniferous calc-alkaline magmatism to the west of the Precordillera, in Argentina and Chile, marks a yet later stage in the development of the Andean active margin that is not considered here. The uplifted Palaeozoic rocks of the foreland reach altitudes of over 6000 m in the Famatina System. The level of crustal exposure seems to be higher in the north, with shallow marine sediments of Lower Ordovician age and coeval volcanic and volcaniclastic sequences in the Famatina System and the Puna (Saavedra et al. this volume), and Neoproterozoic-Lower Cambrian turbiditic sediments with trace fossils in the Eastern Cordillera (see review by Bahlburg & Herv~ 1997; Bahlburg this volume). In contrast, such supracrustal rocks are relatively rare in the southern areas, where both low- and medium-tohigh grade metamorphic rocks predominate as the host rocks to Ordovician magmatism. South of 33~ the typical uplifted foreland block of the 'fiat-slab' segment abruptly disappears. However, rocks carrying fossils typical of the carbonate platform of the Precordillera have been described at 35~ 68~ in the Ponon Trehue area (Bordonaro et al. 1996), and poorly preserved carbonate rocks of La Pampa at 37~ 67~ are also considered to be a southeasterly extension of the Precordilleran sequences (Criado Roqu6 1972; Astini et al. 1995). Scattered outcrops of igneous and metamorphic rocks in this region have given late Proterozoic K-Ar ages (Linares et al. 1980), suggesting that the crystalline basement of the Western Sierras Pampeanas might also extend to these southerly latitudes (see Fig. 1 and Ramos et al. this volume). Equivalents of the CambrianLower Ordovician carbonate platform of the Precordillera, the key sequences that demonstrate a Laurentian biogeographical affinity (Benedetto this volume and references therein), do not crop out either in Patagonia or to the north of the Precordillera; whether this is a real geological discontinuity or a coincidence of tectonic or erosion processes is still not well understood but is clearly important to the terrane accretion concept (see below).

The Precordillera Terrane Although the basement of the Precordillera is not exposed, indirect evidence of its 'Grenvillian' characteristics and age come from U-Pb studies of xenoliths from Miocene lava domes (Kay et al. 1996). This has been taken as further support for

5

the hypothesis that the Precordillera was derived from part of Laurentia corresponding to the southern Appalachians (e.g., Dalla Salda et al. 1992a; Astini et al. 1995). Other 'Grenvillian ages' (c. 1000 i 100 Ma) have been obtained in the adjacent Sierra de Pie de Palo to the east, including U-Pb ages on orthogneiss, diorite, rhyolite and amphibolite (938 to 1091Ma, McDonough et al. 1993, analytical data and errors not reported). This has lead to a growing consensus that the whole of the Western Sierras Pampeanas might represent basement remnants of an allochthonous Grenvillian terrane of which the Precordillera is only a part. Rb-Sr analysis of equivalent orthogneiss suites in the Sierra de Umango, 250km to the north of Pie de Palo (Fig. 2), has produced a similar age of 1030 + 30Ma (Varela et al. 1996). Finally, we have ourselves obtained a new Rb-Sr whole-rock isochron of 1021 + 12 Ma (2a; MSWD = 2.9) for seven samples of granitic orthogneisses from the highest part of the Sierra de Pie de Palo (Fig. 4). Together with the initial 87Sr/86Sr ratio of 0.7045 +0.0003, this shows that the parent magmas of these gneisses were generated in Grenvillian times from an immature source. Significantly, in view of the intense Palaeozoic metamorphic and magmatic history prevalent to the east, their Rb-Sr systems show no evidence of subsequent events. The evidence for possible Laurentian derivation of the Precambrian metamorphic sequences to the north, west, and south of the Sierras Pampeanas, is more difficult to evaluate. Small outcrops of pre-Carboniferous metamorphic rocks west of the Precordillera in both Argentina and Chile have been taken as evidence for a later exotic terrane ('Chilenia', Ramos et al. 1986), and some of these have now yielded Grenvilleage zircon (Ramos & Basei 1997). Their reality as a separate terrane, or as part of the basement of the Precordillera terrane, is not clear, and the extent to which more distant metamorphic terfanes along the proto-Andean margin can be assigned 'Laurentian' derivation is largely model dependent. For example, the Laurentian 'Occidentalia Terrane' of Dalla Salda et al. (1992a, b, and this volume), includes all known Precambrian outcrops along the Andean margin: the Arequipa-Antofalla craton, the Western Sierras Pampeanas and the western edge of the Patagonian massifs. A continent-continent collision between eastern Laurentia and southwestern Gondwana along the whole length of this margin is postulated in this model. However, Dalla Salda et aL (this volume) base many correlations on the original, very broad, definitions of the Pampean and Famatinian cycles (Acefiolaza &

6

R. J. PANKHURST & C. W. RAPELA

0.76 0.75 0.74 0.73

~

.

.

,

Sjerra de Pie de Palo orthogneiss ~ )~-~ J L~ ~, .j~ I~ k ~

0.72

ige~/i2:ii20.0M~03

I1

i

,

,

,

i

.

.

,

i

,

,

.

i

.

,

,

@ El Indio granite $PP-723 ',

~-~ ~c~

i

.

'~,

Age

0.15

OlO ',

0.71

0.05 b

0.70

~ J

~ I~l

0.5

I

I

I

J

1.0

I

J

I

I

i

1.5

II

~ I

J

2.0

I

I

~ I

tl

2.5

[

ti

[

[,

3.0

L

~

4

i

i

I

6

!

i

i

I

8

i

I

a

I

10

,

,

,

I

,

12

Rb-Sr analytical data for orthogneisses and migmatites, Sierra de Pie de Palo: Sample Rb ppm Sr ppm 87Rb/86Sr % s . d . 87Sr/86Sr SPP407 Gneiss 53.2 74.7 2.0639 0.5 0.734410 SPP411 Gneiss 61.8 116.5 1.5384 0.5 0.727311 SPP412 Gneiss 44.8 131.3 0.9891 0.6 0.719007 SPP413 Granite 96.5 90.8 3.0853 0.5 0.749937 SPP414A Gneiss 41.1 121.9 0.9755 0.7 0.718620 SPP414B Granite 88.3 104.5 2.4525 0.5 0.740044 SPP416 Gneiss 27.9 100.4 0.8036 1.0 0.716315 Sample localities: SPP 407, Quebrada La Petaquita (31~ 68~ remainder around antenna (31~ 67~ 8ZSr/SSSrprecise to :1:0.01%, relative to NBS987= 0.710235

,

,

I

,

14

television

Fig. 4. New geochronological data relating to the Sierra de Pie de Palo, Western Sierras Pampeanas. (a) Rb-Sr isochron for orthogneisses from the central (highest) part of the sierra, indicating a Grenvillian age of formation, (b) U-Pb SHRIMP data for a granite emplaced into metamorphic basement in the SE part of the sierra, corresponding to an Early to Mid-Ordovician age (data available from R.J.P.).

Toselli 1973), which are at variance with the more restricted definitions shown in Fig. 3 and adopted by other authors in this volume (Pankburst et aL; Rapela et aL; Sims et aL). Moreover, Pb isotope studies in the Arequipa-Antofalla craton have been taken to suggest that it represents a crustal reservoir distinct from that of the Western Sierras Pampeanas and the Precordillera basement (Tosdal 1996). Another approach is that of Astini et al. (1995), Thomas & Astini (1996), and Astini (this volume), who consider the platform of the Precordillera and its Grenvillian basement as a more restricted microcontinent (the Precordillera Terrane or Cuyania), detached from Laurentia during the Lower Cambrian and accreted to Gondwana in mid-Ordovician times after travelling independently across the Iapetus Ocean. The size of this terrane has been estimated as about 1000km long and 400kin wide when Andean shortening is restored (Keller & Dickerson 1996), or as a block c. 800km square (Thomas & Astini 1996). Dalziel (1997) proposed an alternative model in which the micro-continent that collided with Gondwana

was a stretched passive margin plateau that was still connected to Laurentia during a MidOrdovician collision. Doubt is cast on this model by current considerations of stratigraphical and palaeontological evidence (Asfini, this volume; Benedetto this volume). New ages and petrology relating the Grenville basement of the Western Sierras Pampeanas and its deformation are presented by Ramos et al. (this volume). The geochemistry and tectonic interpretation of mafic sequences in these western areas as representing a variety of Precambrian oceanic environments are considered by Vujovich & Kay (this volume) who identifiy some of the most primitive Precambrian magma types known. Also shown in Fig. 4 are new U - P b zircon data obtained by SHRIMP (ANU, Canberra) for a granite cutting the metamorphic sequence, Zircon cores show a range of inherited ages up to at least 1400Ma, whereas data from the rims indicate an emplacement age of 481 i 6 Ma. If this were considered part of the Famatinian cycle, it would imply that the Grenvillian basement of the Sierra de Pie de Palo had already collided with the Gondwana foreland by Early

INTRODUCTION Ordovician times. This would be consistent with the models of Astini and Ramos et al. (this volume), who argue for mid-Ordovician collision. However, models in which the collision of the Precordillera terrane occurred much later, in Silurian-Devonian times, are supported by some of the new evidence and interpretations presented in this volume (see Keller et al.; Benedetto; Pankhurst e t ai.). The palaeontological content of the PrecordiUera is revisited by Benedetto, stressing the palaeo-biogeographical links with eastern Laurentia. He suggests that the Precordillera and Gondwana faunas show differences throughout the Late Ordovician and that the first unequivocal evidence for geographical continuity between them seems to be in the Wenlockian. Keller et al. (this volume) identify major unconformities in the Precordillera succession as relating to the rifting and drifting stages of separation from Laurentia, and refute the possibility of an Ordovician collision with Gondwana. Von Gosen & Prozzi (this volume) consider that post-Famatinian compression in the Sierra de San Luis was probably synchronous with collision. The La Escalerilla granite, affected by this deformation, is of Devonian age according to Stuart-Smith et al. (unpublished, cited by Sims et al. this volume). Finally, the subduction episode that closed the Southern Iapetus Ocean, resulting in the collision of the Precordillera terrane, did not start until c.490Ma (Pankhurst e t aL this volume; Sims et al. this volume) and it is therefore likely that the collision itself was much later. Our suggestion is that the mid-Ordovician granite in the Sierra de Pie de Palo is related to magmatism during the initial rifting of the terrane from Laurentia rather than to Gondwana events. Mid-Ordovician K-bentonites intercalated in the Precordillera carbonate sequences might be explained in the same way, although some authors in the present volume (Huff et al.; Dalla Salda et al.) prefer their previous interpretation of derivation from the Famatinian arc itself during a mid-Ordovician collisional episode.

The Gondwana foreland Most of the geological constraints for the collision models outlined above have come from recent sedimentological and biostratigraphical studies on the Precordilleran sequences and petrological-geochronological studies of its Grenvillian basement. It is nonetheless obvious from Fig. 3 that some of these models will be severely constrained, or even discarded, when

7

the timing, geochemical signatures and P - T - t paths of the foreland sequences and the Famatinian arc can be compared in detail with those of the inferred exotic terranes. One purpose of this book is to initiate progress towards that goal, as well as to encourage further research on the many large igneous-metamorphic areas that still lack detailed structural, petrological and isotopic studies. Rapela e t al. and Sims et al. discuss the Cambrian to Lower Ordovician geological characteristics of the Gondwana continental margin before the collision of the Precordillera terrane, represented by the Eastern Sierras Pampeanas and the Eastern Cordillera. Based on a new structural, metamorphic, geochemical, and geochronological study, Rapela et al. conclude that the Pampean orogeny itself resulted from a major earlier collisional event in Early to MidCambrian times.

The Famatinian arc Finally, the development of the Famatinian arc and subsequent transformation of the margin into the Gondwana foreland is treated in number of papers in this volume; in the Famatina System (Saavedra et al.), the Sierra de Fi~tmbala (Grissom et al.), the central southern Sierras Pampeanas (Pankhurst et al.), and the Sierras de San Luis (Llambias et al.; yon Gosen et aL). Further evidence for the geochronological constraints on the timing of both Pampean and Famatinian orogenic events is provided by Sims et al. These papers provide an overall picture of a calcalkaline subduction-related magmatic arc which was active from earliest Ordovician (Tremadocian) times. In the south there is only a plutonic record and the arc was clearly developed on the margin of the Gondwana foreland that had been established during the Pampean orogeny (Pankhurst e t ai. this volume). Now, even in the north, the sedimentological evidence discussed by Bahlburg (this volume) appears to favour a continental arc rather than the earlier model of an island arc for the Famatinian environment at this Early Ordovician stage. Subsequently, more evolved and less deformed granitoids, some with a marked peraluminous tendency, were emplaced into the batholithic belt until mid-Ordovician times. Activity then ceased until peraluminous intra-plate magmatism restarted to the east of the old arc, in Devonian times. At this point, the Palaeozoic geological record of the Sierras Pampeanas ceased, indicating the stability of the new continental margin, at least until the start of the Andean cycle in Cretaceous times.

8

R.J.

P A N K H U R S T & C. W. R A P E L A

Interpretations presented in the contributions in this volume have in m a n y cases developed since the 1996 congress in Buenos Aires, and we have not attempted to ensure total consistency between the various contributions. We have simply tried to point out here that radically different views are expressed concerning the d u r a t i o n and extent o f the F a m a t i n i a n arc, the question o f w h e t h e r this arc was an island arc or a continental one, and the timing o f the final collision of the Precordillera terrane with G o n d w a n a . The reader is invited to c o m p a r e these views critically, as the resolution is necessary to reach a fully coherent u n d e r s t a n d i n g of the processes involved in this, one of the best-constrained allochthonous terrane histories k n o w n . The new data referred to in this introduction were obtained as part of contract CII-CT92-0088" from the European Commission and was also supported by grant PIP No 4148 from the Consejo Nacional de Investigaciones Cientificas y Trcnicas de la Repfiblica Argentina (CONICET); this paper is listed as NIGL Publication No. 238. The volume as a whole is a contribution to IGCP Projects 345 (Andean Lithospheric Evolution) and 376 (Laurentia-Gondwana Connections before Pangea).

References ACE/qOLAZA, F. G. & TOSELLI, A. 1973. Consideraciones estratigr~ificas y tectbnicas sobre el Paleozoico inferior del noroeste argentino. 2 ~ Congreso Latinoamericano de Geologia, Actas, 2, 755-763. , MILLER, H. & TOSELLI, A. J. (eds) 1990. El Ciclo Pampeano en el Noroeste Argentino. Universidad Nacional de Tucumfin, Serie Correlaci6n Geol6gica, 4. ASTINI, R. A. 1996. Las fases diastr6ficas del paleozoico medio en la Precordillera del oeste Argent i n a - evidnecias estratigrfificas. XIII Congreso Geoldgico Argentino y III Congreso de Exploracidn e Hidrocarburos, Buenos Aires, Actas, V, 509-526. 1998. Stratigraphical evidence supporting the rifting, drifting and collision of the Laurentian Precordillera Terrane of western Argentina. This volume. --, BENEDETTO, J. L. & VACCARI, N. E. 1995. The early Paleozoic evolution of the Argentine Precordillera as a Laurentian rifted, drifted and collided terrane. Geological Society of America Bulletin, 107, 253-273. BAHLBURG, H. 1998. The geochemistry and provenance of Ordovician turbidites in the Argentine Puna. This volume. & HERVE, F. 1997. Geodynamic evolution and tectonostratigraphic terranes of northwestern Argentina and northern Chile. Geological Society of America Bulletin, 109, 869-884. -

-

BENEDETTO, J. L. 1998. Early Palaeozoic brachiopods and associated shelly faunas from western Gondwana: their bearing on the geodynamic history of the pre-Andean margin. This volume. BORDONARO, O., KELLER, M. & LEHNERT, O. 1996. E1 Ordovicico de Pon6n Trehue en la provincia de Mendoza (Argentina): redefiniciones estratigrfificas. XIII Congreso Geolrgico Argentino y III Congreso de Exploracirn e Hidrocarburos, Buenos Aires, Actas, I, 541-550. CAMPOSNETO, M. C. & FIGUEIREDO,M. C. H. 1995 The Rio-Doce orogeny, southeastern Brazil. Journal of South American Earth Sciences, 8, 143-162. CRIADO ROQUI~, P. 1972. Cinturbn mbvil mendocinopampeano. In: LEANZA, A. F. (ed.) Geologia Regional Argentina. Academia Nacional de Ciencias de la Repfiblica Argentina, Cbrdoba, 297-303. DALLA SALDA, L. H., CINGOLANI,C. A. 8,~VARELA,R. 1992a. Early Paleozoic orogenic belt of the Andes in southwestern South America: result of Laurentia-Gondwana collision? Geology, 20, 617-620. - - , DALZIEL,I. W. D., CINGOLANI,C. A. & VARELA, R. 1992b. Did the Taconic Appalachians continue into southern South America? Geology, 20, 1059-1062. --, LOPEZ DE LUCHI, M. G., CINGOLANI, C. A. & VARELA, R. 1998. Laurentia-Gondwana collision: the origin of the Famatinian-Appalachian orogenic belt (a review). This volume. DALZIEL, I. W. D. 1997. Neoproterozoic-Paleozoic geography and tectonics: Review, hypothesis, environmental speculation. Geological Society of America Bulletin, 109, 16-42. , DALLA SALDA, L. n., CINGOLANI, C. A. 8,~ PALMER, P. 1996. The Argentine Precordillera: a Laurentian Terrane? Penrose Conference Report. GSA Today, February 1996, 16-18. DICKERSON, P. W. & KELLER, M. 1998. The Argentine Precordillera: its odyssey from the Laurentian Ouachita margin towards the Sierras Pampeanas of Gondwana. This volume. GRISSOM, G. C., DEBARI, S. M. & SNEE, L. W. 1998. Geology of the Sierra de Fiambalfi, northwest Argentina: implications for Early Palaeozoic Andean tectonics. This volume. GUTII~RREZ-MARCO,J. C., SAAVEDRA,J. & RABANO,I. (eds) 1992. Paleozoico Inferior de IberoamHica. Universidad de Extremadura, Spain. HUFF, W. D., BERGSTROM, S. M., KOLATA, D. R., CINGOLANI, C. & ASTINI, R. A. 1998. Ordovician K-bentonites in the Argentine Precordillera: relations to Gondwana margin evolution. This volume. ISACKS, B. 1988. Uplift of the central Andean plateau and bending of the Bolivian Orocline. Journal Geophysical Research, B93, 3211-3231. JORDAN, T. E. & ALLMENDINGER, R. W. 1986. The Sierras Pampeanas of Argentina: a modern analogue of Rocky Mountain foreland deformation. American Journal of Science, 286, 737-764. KAY, S. M. & GORDILLO, C. E. 1994. Pocho volcanic rocks and the melting of depleted continental lithosphere above a shallowly dipping subduction zone in the central Andes. Contributions to Mineralogy and Petrology, 117, 25-44.

INTRODUCTION

--,

ORELL, S. • ABRUZZI, J. M. 1996. Zircon and whole rock Nd-Pb isotopic evidence for a Grenville age and a Laurentian origin for the basement of the Precordillera in Argentina. Journal of Geology, 104, 637-648. KELLER, M. & DICKERSON, P. W. 1996. The missing continent of Llanoria- was it the Argentine Precordillera? XIII Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, V, 355-367. - - , BUGGISCH, W. & LEHNERT, O. 1998. The stratigraphical record of the Argentine Precordillera and its plate-tectonic background. This volume. LINARES, E., LLAMBiAS,E. J. & LATORRE, C. O. 1980. Geologia de la provincia de La Pampa, Repfiblica Argentina y geocronologia de sus rocas metam6rficas y eruptivas. Revista de la Asociaci6n Geol6gica Argentina, 35, 87-146. LLAMBiAS, E. J., CAMINOS, R. & RAPELA, C. W. 1984. Las plutonitas y vulcanitas del Ciclo Eruptivo Gondwfinico. In: RAMOS, V. A. (ed.) Geologia y Recursos Minerales de la Provincia de Rio Negro. Relatorio del 9 ~ Congreso Geol6gico Argentino, 83-117. - - - , SATO, A. M., ORTIZ SUAREZ, A. & PROZZI, C. 1998. The granitoids of the Sierra de San Luis. This volume. MCDONOUGH, M., RAMOS, V. A., ISACHSEN, C. & BOWRING, S. 1993. Edades preliminares de circones del basamento de la Sierra de Pie de PaiD, Sierras Pampeanas Occidentales de San Juan: sus implic/mcias para el supercontinente proterozoico de Rodinia. XII Congreso Geol6gico Argentino y II Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, III, 340-343. PANKHURST, R. J., RAPELA, C. W., SAAVEDRA, J., BALDO, E., DAHLQUIST, J., PASCUA, I. & FANNING, C. M. 1998. The Famatinian magmatic arc in the central Sierras Pampeanas. This volume. RAMOS, V. A. &BASEI, M. 1997. The basement of Chilenia: an exotic continental terrane to Gondwana during the early Paleozoic. In: BRADSHAW,J. D. & WEAVER, S. D. (eds) Terrane Dynamics-97. International Conference on Terrane Geology, Christchurch, New Zealand, Abstracts, 140-143. , DALLMEYER, R. D. & VUJOVICH, G. I. 1998. Time constraints in the Early Palaeozoic docking of the Precordillera, central Argentina. This volume. , JORDAN, T. E., ALLMENDINGER, R. W., MPODOZIS, C., KAY, S. M., CORTI~S, J. M. & PALMA, M. A. 1986. Paleozoic terranes of the central Argentine-Chilean Andes. Tectonics, 5, 855-880. RAPELA, C. W., PANKHURST, R. J., LLAMBiAS,E. J., LABUDIA, C. & ARTABE, A. 1996. "Gondwana"

9

magrnatism of Patagonia: Inner Cordilleran calcalkaline batholiths and bimodal volcanic province. Third International Symposium on Andean Geodynamics, St.-Malo, France, R~sumbs 6tendus, ORSTOM, Paris, 791-794. - - , CASQUET, C., BALDO, E., SAAVEDRA,J., GALINDO, C. & FANNING, C. M. 1998. The Pampean Orogeny of the southern proto-Andes: Cambrian continental collision in the Sierras de Cbrdoba. This volume. RENNE, P. R., ERNESTO, M., PACCA, I. G., GLEN, J. M., PREVOT, M. & PERRIN, M. 1992. The age of Paranfi flood volcanism, rifting of Gondwanaland, and the Jurassic-Cretaceous boundary. Science, 258, 975-979. SAAVEDRA,J., TOSELLI, A., ROSSI, J., PELLITERO, E. & DURAND, F. 1998. The Early Palaeozoic magmatic record of the Famatina System: a review. This volume. SIMS, J. P., IRELAND,T. R., CAMACHO,A., LYONS, P., PIETERS, P. E., SKIRROW, R. G., STUART-SMITH, P. G. & MIR0, R. 1998. U-Pb, Th-Pb and Ar-Ar geochronology from the southern Sierras Pampeanas, Argentina: implications for the Palaeozoic tectonic evolution of the western Gondwana margin. This volume. THOMAS, W. A. & ASTINI, R. A. 1996. The Argentine Precordillera: A traveler from the Ouachita embayment of North America Laurentia. Science, 273, 752-757. TOSDAL, R. M. 1996. The Amazon-Laurentian connection as viewed from the Middle Proterozoic rocks in the central Andes, western Bolivia and northern Chile. Tectonics, 15, 827-842. TURNER, S., REGELOUS, M., KELLEY, S., HAWKESWORTH,C. J. & MANTOVANI,M. 1994. Magmatism and continental break-up in the South Atlantic: high-precision 4~ geochronology. Earth and Planetary Science Letters, 121, 333-348. VARELA, R., LOPEZ DE LUCHI, M., CINGOLANI, C. & DALLA SALDA, L. 1996. Geocronologia de gneises y granitoides de la Sierra de Umango, La Rioja. Implicancias tect6nicas. XIII Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, III, 519-527. YON GOSEN, W. & PROZZI, C. 1998. Structural evolution of the Sierra de San Luis (Eastern Sierras Pampeanas, Argentina): implications for the proto-Andean margin of Gondwana. This volume. VUJOVlCH, G. I. & KAY, S. M. 1998. A Laurentian? Grenville-age oceanic arc/back-arc terrane in the Sierra de Pie de Palo, Western Sierras Pampeanas, Argentina. This volume.

Stratigraphical evidence supporting the rifting, drifting and collision of the Laurentian Precordillera terrane of western Argentina RICARDO

A. A S T I N I

Estratigrafia y Geologia Hist6rica, Departamento de Geologia, Facultad de Ciencias Exactas Fisicas y Naturales, Universidad Nacional de C6rdoba, Vdlez Sarsfield 299, CC 395, 5000 C6rdoba, Argentina (e-mail: [email protected]) Abstract: Major stratigraphical evidence points toward initial rifting of the Precordillera

terrane of western Argentina during the Early Cambrian from the Ouachita embayment in southeastern Laurentia. A first fully developed rifting episode is suggested from subsidence analysis and the sedimentary record. Restricted graben-fill red-beds and evaporites in the eastern tectofacies, together with coarse arkose and quartz conglomerate preserved as blocks in the western tectofacies, are interpreted as deposits of the initial rift-related brittle extension. Stratigraphical similarities with localities surrounding the Ouachita embayment leads to a model in which the Precordillera was the southern part of the pre-Appalachian Laurentia margin and the conjugate margin of the Texas promontory, separated along the Alabama-Oklahoma transform and the Ouachita rift, respectively. Overstepping of syn-rift facies by a thick succession of Lower Cambrian to Lower Ordovician carbonates indicates completion of rift-to-drift transition and initiation of thermo-tectonic subsidence characteristic of continental terrace passive-margin development. Slope facies, marginal faunal belts and recently dated mafic rocks indicate the development of an ocean basin separating the Precordillera from Laurentia by the Mid-Cambrian. A 20-30 Ma period of drifting and isolation is indicated from faunal and sedimentological evidence. Subsidenceinduced incipient drowning and abundant K-bentonites suggest the influence of subduction and the progressive approach of the terrane to western Gondwana. Continuous Mid- and Late Ordovician post-collisional extension brackets the collision, although no clear evidence of shortening has been detected in the preserved former sedimentary bank. Post-collisional extension is separated by >70 Ma from the initial rifting. The Argentine Precordillera has basement rocks and lowermost Palaeozoic cover (Cambrian and most of the Ordovician) which strongly contrast with those of the surrounding geological provinces in southern South America (Fig. 1). For this reason, it has been considered a suspect terrane and has recently been the target of much interest due to the current controversy on early Palaeozoic palaeogeography and plate tectonics. Although it is accepted that the Precordillera should be considered an allochthonous or exotic terrane (Benedetto 1993; Astini et al. 1995a, 1996; Dalziel et al. 1996), there is much debate on the source and route it took to Gondwana. Two major end-member hypotheses have been suggested to explain the transfer of the Precordillera from Laurentia to Gondwana, each implying different kinematics and timing: (1) the terrane hypothesis largely considers the Precordillera as a rifted-drifted microcontinent (Astini et al. 1995a); and (2) the tectonic tracer model (Dalziel, 1993), where after a continentcontinent collision and later rifting, a sliver of Laurentia was left attached to Gondwana (Dalziel et al. 1994; Dalziel 1997). The contrasts between the two models are much greater than is apparent, with different geological implications

and consequences apart from the origin of the Ocloyic orogeny and the Taconic orogeny in the Mid-Ordovician, respectively in western Gondwana and in eastern Laurentia. Stratigraphical history, including the record of subsidence and tectonic history, and sensitivity to climate change, should be regarded as the critical tool to discriminate between these two models. The original departure of the Precordillera block from Laurentia is undisputed, because its Cambrian faunas (Borrello 1971; Vaccari 1994; Lehnert et al. 1997) and sedimentary successions are nearly identical to those of the continentalshelf successions of Laurentia (Bond et al. 1984; Ramos et al. 1986; Benedetto, 1993; Astini et al. 1995a, 1996). Based on close stratigraphical comparisons and kinematic constraints, Thomas & Astini (1996a) concluded that the Precordillera continental fragment was most probably rifted from the Ouachita embayment of the southern margin of Laurentia. New palaeomagnetic data (Rapalini & Astini 1998) confirm this conclusion. Docking with Gondwana apparently occurred between the Early and Late Ordovician, when the Precordillera stratigraphy (Astini et al. 1995a, 1996), faunal affinities (see Benedetto this volume), and

ASTINI, R. A. 1998. Stratigraphical evidence supporting the rifting, drifting and collision of the Laurentian Precordillera terrane of western Argentina. In: PANKHURST, R. J. & RAPELA, C. W. (eds) The Proto-Andean Margin of Gondwana. Geological Society, London, Special Publications, 142, 11-33.

12

R . A . ASTINI

Fig. 1. Map of the main Ordovician outcrops of southern South America showing the lithological contrast between the Argentine Precordillera carbonates and the surrounding siliciclastic and volcaniclastic basins (after Astini 1995a).

sedimentary overlap first show a definitive connection with Gondwana. At present, timing for the collision appears to be in the Middle Ordovician, based mainly on palaeo-biogeographical arguments. Benedetto & Astini (1993), Benedetto (1993), Astini et al. (1995a) and Thomas & Astini (1996a) have suggested that between the initial rifting from Laurentia and the accretion to western Gondwana (the present western side of South America) the Precordillera drifted free and isolated in the Iapetus Ocean, where it had some faunal exchange with other intra-Iapetan island arcs and terranes. The purpose of this paper is to summarize the stratigraphical evidence for the allochthonous nature of the Precordillera and the data that have particularly helped in establishing the model of rifting, drifting, and collision of this exotic-to-Gondwana terrane.

Geological setting and previous work The Argentine Precordillera is a Grenville-age lithospheric block (Kay et al. 1996; Astini et al. 1996) with a thick Palaeozoic cover of folded marine to non-marine sedimentary rocks and Tertiary strata (Allmendinger et al. 1990). Post-

Triassic Mesozoic sediments are absent. Structurally, the Precordillera is considered as part of the extensive fold-and-thrust belt located east of the main Andes (Fig. 2). To the east it is bounded by the broken foreland of the Sierras Pampeanas, a metamorphic basement presently considered as late Precambrian-Early Cambrian in age (Kraemer et al. 1995; Rapela et al. this volume; Sims et al. this volume). The Precordillera thrust-fold belt extends for 400 km between Mendoza and La Rioja provinces in western Argentina, but its present-day features are strongly controlled by Andean segmentation (Jordan et al. 1983). Thus, the real north-south length of the exotic block might exceed 800 km if its extension into the San Rafael block, south of Mendoza, is considered (Astini et al. 1995a; Bordonaro et al. 1996; Ramos et al. this volume). The Precordillera terrane is over 200 km wide (Fig. 2) and may include the Grenvillian basement that crops out immediately to the east (see Pankhurst & Rapela this volume) in the Western Sierras Pampeanas. Identification of terranes can be made by considering the regional extent and overlapping relationships of the sedimentary units. The Lower Cambrian to lower Upper Ordovician

THE LAURENTIAN PRECORDILLERA TERRANE

13

Fig. 2. Location map of the Argentine Precordillera in its present location to the east of the High Andes and detail of the eastern and western tectofacies recognised by Astini (1992) for the Ordovician. Note that part of the basement included in the western Pampeanas as well as the outcrops to the south of San Rafael are included in the Precordillera terrane. Localities are numbered. I, 2, 3, 7, 8, 10 and 11 stand for horst-block nucleating carbonates in the Mid- and Late Ordovician.

section, as well as the Grenvillian basement, seem to be restricted to the Precordillera domain, thus constraining the limits of the exotic terrane. Starting in the uppermost Ordovician, sedimentary units and depositional environments (Hirnantian glacial horizon and post-glacial transgressive deposits) can be traced throughout the rest of the Andean and Sub-

andean belt, which gives a minimum age for the proposed accretion of the Precordillera terrane. Different Cambrian-Ordovician tectono-stratigraphical assemblages reflecting major basin evolutionary stages were identified by Astini & Benedetto (1996). A basal, dominantly carbonate, continental platform cycle (including its rifting and drifting history) can be separated

14

R . A . ASTINI

from a complex, mostly siliciclastic, wedge representing the syn- to post-collisional cycle. A two-fold subdivision of the Precordillera into eastern and western tectofacies can be made on the basis of substrate nature, dominant sedimentary composition, and structural behaviour and response (Fig. 2). According to Astini (1992), the eastern tectofacies comprises most of the shallow-marine carbonate stratigraphy and overlying environmentally restricted clastic rocks, whereas the western tectofacies includes a deep-marine siliciclastic assemblage. Important features in the latter are the association of mafic and ultramafic igneous rocks related to remnants of spreading-ridge centres (Kay et al. 1984) and the widespread low-grade metamorphism. The eastern domain substrate may thus be interpreted as normal continental lithosphere (35-40 kin), whereas the western domain is characteristic of thinner crust. The outstanding stratigraphical feature of the Precordillera is the thick Cambrian and Lower Ordovician carbonate succession (Baldis & Bordonaro 1981; Keller et al. 1994; Keller et al. this volume), which contrasts markedly with coeval clastic and volcaniclastic successions in the rest of the South American margin. This, together with the Laurentian trilobite faunas, led Poulsen (1958) and Borrello (1965, 1971) to consider its possible exotic origin (Ramos et al. 1986). More recently, palaeo-biogeographical analysis based on brachiopods led Benedetto (1993; see also Astini et al. 1995a, fig. 7) to suggest that the Precordillera had remained close to Laurentia until the earliest Ordovician. Further detailed taxonomic studies of trilobites allowed Vaccari (1994; see also Palmer et al. 1996) to establish a close comparison with the south-eastern margin of Laurentia (see Astini et al. 1995a, fig. 5). Finally, Astini et al. (1995b), on stratigraphical grounds, pointed out the very close match between the northern Precordillera and the Alabama Appalachians. Astini (1995a) elaborated a wander path based on lithologically sensitive climate indicators, showing that the Precordillera remained in low latitudes ('Pacific realm') during the Cambrian and earliest Ordovician and then migrated toward temperate environments, with clear indications of the Gondwana glaciation by Late Ordovician (Hirnantian) time. Thus, while the parent continent Laurentia remained within the tropics, the Precordillera terrane drifted southward, and by the early Late Ordovician it occupied a relatively high southern latitude. This suggests that the Precordillera terrane was most probably isolated from Laurentia long before the onset of the Ocloyic Orogeny.

In the past there has been little palaeomagnetic work in the Lower Palaeozoic rocks of the Argentine Precordillera. Now, for the first time, primary magnetisation has been isolated by Rapalini & Astini (1998), at several sites in northern Precordillera. This study confirms the suggestion that the Precordillera constitutes an exotic-to-Gondwana terrane, and its position during the Early Cambrian is remarkably coincident with that suggested by Astini et al. (1995a) and Thomas & Astini (1996a).

Hypotheses and background Because the original rifted margins of both the Appalachian-Ouachita region and the Precordillera have been deformed by later orogenies and at present are partly covered by younger rocks, details of the rift site and geometry, as well as the paired rifted-margin palaeogeography, are difficult to determine. For this reason, alternative interpretations have been suggested for the relative timing of separation from Laurentia and of accretion to Gondwana, as well as of the exact positioning and kinematics of rifting (e.g. Dalla Salda et al. 1992a; Dalziel et al. 1994; Dalziel, 1997; Thomas & Astini 1996a; Astini et al. 1996; Keller & Dickerson 1996; Buggisch 1996). Current discussion focuses on whether or not the Precordillera represented a 'bridge' or a 'raft' between the two principal Early Palaeozoic land masses. This, in turn, is important to deciding between the conventional (archetypal) and alternative plate palaeo-geographical schemes (see Torsvik et al. 1995), in view of the longitude uncertainty in palaeomagnetically based reconstructions. Much evidence has been put forward to support isolation of the Precordillera in an intra-Iapetus position before it collided with Gondwana (Astini et al. 1995a, 1996; Thomas & Astini 1996a; Dalziel et al. 1996). However, in seeking a longitude constraint for the wandering of Laurentia in the 'end-run' hypothesis (Dalziel 1991), an alternative solution to the traditional northern hemisphere continental puzzle was suggested by Dalziel (1997) as a modification of his earlier work (Dalziel et al. 1994). A keystone in this most recent reconstruction is the Argentine Precordillera, which is proposed to have had limited separation from Laurentia through initial crustal extension in the Early Cambrian. In this model the Precordillera block would have had to move outboard of the Ouachita embayment to occupy an 'intraIapetus' position (but still intra-shelf), analogous to the present-day Maurice Ewing bank

THE LAURENTIAN PRECORDILLERA TERRANE southwest of the Falkland/Malvinas plateau in the South Atlantic. According to Dalziel (1997) this model explains the increasing isolation of the Precordillera, while still united to Laurentia through a continental bridge (extended Grenvillian crust); the Precordillera 'peninsula' would have collided with Gondwana. Only after collision did Dalziel suggest development of oceanic crust within the western Precordillera clastic wedge, interpreted as the result of rifting between Laurentia and Gondwana. The main difference from the microcontinent model of Astini et al. (1995a) and Thomas & Astini (1996a) is that (as previously suggested in Dalla Salda et al. 1992a, b) the Precordillera would have drifted as an attached part of Laurentia. Hence, both would have shared the same polar wander path during the Early Ordovician, approaching that of Gondwana. In contrast, Astini et al. (1995a) and Thomas & Astini (1996a) considered the Precordillera block as an independent microcontinent, so that the Laurentian and Gondwanan paths would have never converged to indicate a common Ocloyic-Taconic orogeny. During supercontinent break-up, smaller continental fragments surrounded by various rift branches and transfer faults (later transforms) are left between the larger continents (e.g., Rosendahl 1987; Etheridge et al. 1989). A rifted microcontinent may remain near the parent continent and, during a subsequent collision, be reattached to the parent continent (e.g., the Appalachian internal basement massifs of Thomas 1977 and Hatcher 1983). Alternatively, the rifted lithosphere fragment may separate fully from the parent continent and drift independently to collide with a foreign continent (e.g., Howell 1989). In this case it would be interpreted as a far-travelled terrane. Rift-related intracratonic fault systems indicate brittle crustal extension of the parent continent (Thomas 1991). Failed rifts or aulacogens partially or completely outline basement blocks (Braile et al. 1986), indicating interrupted microcontinental evolution due to incomplete break-up of the parent continent (e.g., SE Laurentia). In the context of rifted margins, a fragment of Grenvillian continental crust, now recognised as the Precordillera, appears to have rifted from the Ouachita embayment of the Laurentian margin during the Early Cambrian (Thomas 1991), as a result of major diachronous rifting associated with the opening of the Iapetus. Three major lines of stratigraphical evidence support the microcontinental model for the Precordillera: (a) the stratigraphical record of rifting and passive margin history, (b) the

15

stratigraphical record of drowning and progressive flexure recording foredeep and forebulge formation in a peripheral foreland during accretion, and (c) the stratigraphical record of major palaeogeographical rearrangement related to widespread extension after collision. Faunal evidence provides precise constraints on the timing and degree of isolation (Benedetto this volume).

Stratigraphical record of rifting and drifting E a r l y C a m b r i a n - E a r l y Ordovician carbonate bank

As previously mentioned the striking feature of the Argentine Precordillera is the Early Cambrian-Early Ordovician marine carbonate succession, which is nearly 3000m thick. This was originally recognized as a typical passive margin peri-cratonic succession by Gonzfilez Bonorino (1975) and Baldis & Bordonaro (1981, 1985), and consists of several units, mainly of shallow to very shallow marine origin (Fig. 3) (Keller et al. 1994). Lower to Middle Cambrian thick, shoaling-upward, carbonate units, alternating with siliciclastic rocks (cf. Aitken 1966; grand cycles according to Baldis & Bordonaro 1982) are succeeded by thick dolostones with sparse stromatolite-bearing horizons. These, in turn, grade into a rhythmically arranged thick- to medium-bedded, mostly dolomitic Upper Cambrian unit composed of recurrent small-scale, shallowing-upward peri-tidal cycles (Keller et al. 1989; Cafias 1990; Armella 1994) and sub-tidal cycles (Cafias 1995a). An exposure-and-drowning surface separates this from the lowermost Ordovician unit, which, according to Cafias (1995b) corresponds to a typical Bahamian restricted platform, bordered by grainstone shoals toward the open sea. The uppermost unit in the carbonate bank is a calcareous, dominantly sub-tidal, muddy succession with a rich open-marine calcareous fauna, punctuated by sea-level fluctuations responsible for the generation of reef-mounds and biostromes (Cafias & Carrera 1993; Cafias & Keller 1993; Keller & Bordonaro 1993). Similar facies types of the Cambro-Ordovician carbonate bank and vertical arrangement compared with the coeval Laurentian carbonate rim, led Astini et al. (1995a) to favour correlation of the Precordillera with the southern Appalachian platform. By the Late Cambrian, the Precordillera had developed into an agraded, flat-topped rimmed-platform dominated by peritidal facies (Cafias 1995a; Armella 1994; Astini et al. 1995a), similar to that interpreted for the

16

R . A . ASTINI

Fig. 3. Generalized time-stratigraphic chart of the Cambrian and Ordovician across the Argentine Precordillera. Note contrasting stratigraphy of the eastern and western tectofacies (see distribution in Fig. 2). CTF, Cerro Totora Formation; LLF, La Laja Formation; ZF, Zonda Formation; LFF, La Flecha Formation; LSiF, La Silla Formation; SJF, San Juan Formation; GF, Gualcamayo Formation; ChF, Las Chacritas Formation; LAF, Las Aguaditas Foramtion; LPF, Las Plantas Formation; LVF, Las Vacas Formation (La Cantera Formation); SF, Sassito Formation; TF, Trapiche Formation; DBF, Don Braulio Formation; LChF, La Chilca Formation; LSF, Los Sombreros Formation; YLF, Yerba Loca Formation; AF, Alcaparrosa Formation. Time-scales according to Tucker & McKerrow (1995). Olistoliths and basement boulders (granites and metamorphic rocks) present in the Los Sombreros Formation (western tectofacies) are represented according to their individual ages (Benedetto & Vaccari 1992; Bordonaro & Banchig 1996; Astini unpublished). (Sources for construction: Albanesi et al. 1990; Astini 1991; Astini & Brussa 1997; Astini & Cafias 1995; Astini & Vaccari 1996; Benedetto & Herrera 1987; Brussa 1994; Keller et al. 1994; Lehnert 1995; Ortega & Brussa 1990; Ortega et al. 1991; Vaccari 1994). central-southern Appalachian passive margin (Demicco 1985; Koerschner & Read 1989; Osleger 1993). The remarkable similarities indicate analogous substrates and latitudes to those of the southern Appalachians. Studies of internal architecture and lateral correlation between Precordilleran units are still limited, although neither very proximal nor really distal environments are well represented in the Argentine Precordillera. It is reasonable then to suggest that the Precordillera is somewhat incomplete, having been a platform at least 400-600km wide, when compared with the similar Appalachian setting (Cafias 1995a; Astini e t al. 1996; Keller 1996). Accordingly,

the assumption that the Precordillera terrane may have been c. 800 km wide (Thomas & Astini 1996a) seems to be reasonable. E - W shortening may have greatly affected the Precordillera terrane in post-Early Ordovician time, stripping the carbonate cover from the Grenvillian basement that now crops out to the east of the Precordillera thrust belt. Tectonic reworking also has affected and removed clear evidence of the western talus (present co-ordinates), which has been greatly shortened and overprinted by both Andean and Early Devonian tectonism (Astini 1996). Nevertheless, an E - W passive margin polarity is the consensus among the Precordilleran workers, based on most of

THE LAURENTIAN PRECORDILLERA TERRANE

17

the olistoliths and slumped blocks of the western tectofacies (Fig. 3) being interpreted as former slope and platform edge facies, gravitationally reworked into more distal environments. Using data from Baldis & Bordonaro (1981), Bond et al. (1984) compared the Precordillera subsidence history with that of the various Laurentian margin domains and concluded that it had most similarity with that of the Appalachian platforms. Therefore, Bond et al. (1984) considered the Precordillera as a probable Appalachian conjugate margin. According to their continental break-up hypothesis, the Precordillera was first positioned as a centralnorthern Appalachian counterpart, a scheme later adopted by Herrera & Benedetto (1991). Nevertheless, Bond et al. (1984) also suggested the alternative that the Precordillera be considered a displaced terrane that was part of the Appalachian miogeocline in the earliest Palaeozoic, a scheme favoured by Ramos et al. (1986) and Benedetto (1993). E v i d e n c e f o r rifting

One premise of similar thick carbonate platforms, such as that of the Precordillera, is that they have always been recognized as typifying low-latitude stable passive-margin continental terrace deposits (e.g., Bond et al. 1983, 1995). These carbonate edifices reflect the thermal subsidence of a continental margin after rifting and hence indirectly indicate the age of break-up of the margin (Bond et al. 1984). The presence of such a typical passive margin carbonate bank implies a minimum thickness of continental crust, sufficient to maintain shallow bathymetry. The first-order control on subsidence and growth of the Precordilleran carbonate platform, as well as for the Laurentian analogues, was thermal contraction of heated lithosphere following rifting (Bond et al. 1989). This argument can be taken as a powerful indicator of fully developed Early Cambrian rifting between the Precordillera and the Ouachita margin. Recent work in the northern Precordillera (Astini & Vaccari 1996) identified a new Lower Cambrian unit, the Cerro Totora Formation (Fig. 3), composed of a mixed succession of evaporites, red siliciclastic rocks and carbonates (Fig. 4), underlying the thick Cambrian to Lower Ordovician carbonates (>2500m). The Cerro Totora Formation records the transition from syn-rift evaporites and marginal-marine red or variegated arkosic sedimentary rocks to quartz arenites and intercalated olive green/ glauconitic shales that reflect generalised transgression associated with the cessation of rifting.

Fig. 4. Stratigraphic section of the syn-rift clastic and evaporitic graben fill succeeded by limestones and dolostones after the main transgressive surface that represents the rift-drift transition in the northern Precordillera (modified from Astini & Vaccari 1996). The basal anhydrite-gypsum succession (>250m) is a common depositional facies in the early stages of restricted-circulation divergent margins in low-latitude arid-climate settings (Kinsman 1975; Rona 1982; Warren 1989). The evaporites are strongly recrystallized and interbedded with dolomitized crypto-microbial and oolitic tabular limestones, and are succeeded by a predominantly siliciclastic interval (c. 50m). This red-bed interval is composed of

18

R . A . ASTINI

red to purplish silty shales and minor quartzose to sub-feldspathic sandstones and intra-clastic conglomerates. Thin beds of reworked detrital evaporites and carbonate layers also occur. Halite hopper-cube casts and intra-sedimentary gypsum crystals indicate extensive evaporation in arid supra-tidal environments, as shown by modern analogues (Shin 1983; Handford 1991). Mud-cracked horizons, tepee structures and brecciated horizons are also common features, suggesting subaerially exposed mud-flat and salt-flat environments (Kendall & Warren 1987). Facies analysis, together with the general stratigraphical development, allows interpretation of this unit as a typical syn-rift succession (Astini & Vaccari 1996). Starting in the late Early Cambrian, the onset of the passive-margin stage is recorded in the upward transition to the thick carbonate bank. Astini et al. (1995b) pointed out the time parallelism and the nearly identical lithology of

the Cerro Totora Formation with the contemporaneous stratigraphy of the Rome (Waynesboro) Formation and related units in the southern Appalachians and Black Warrior Basin, where the rifting history developed similar graben-filled successions (e.g., Mississippi Valley trough and Birmingham graben system, Mellen 1977; Thomas 1991; Raymond 1991). On this basis, Thomas & Astini (1996a) concluded that the Precordillera block was rifted from the Ouachita embayment, initially bounded by the Blue Ridge rift to the east, the Alabama-Oklahoma transform to the north and the Ouachita rift to the west (Fig. 5). NWpropagation of the Alabama-Oklahoma transform fault apparently transferred extension from the mid-Iapetus ridge, along the Blue Ridge rift to the northern part of the Ouachita rift, initiating effective separation of the Precordillera block (Fig. 5). The Alabama-Oklahoma transform fault and the Ouachita rift, outlining the

/

r

,,4'

1

Rifled margin of continental crust Transform fault Intracratonic basement fault ,, ,, Cratonward limit of AppalachianOuachita detachment ,, Thrust fault . . . . . . . Margin of Gulf and Atlantic Coastal Plains

Atlantic Ocean

!\ Gulf of Mexico

•N I

Scale 200 km

~-~ . . a

Fig. 5. Plan view of the southeastern USA with locations of the Ouachita embayment of Laurentia, the Argentine Precordillera block and the general outline of the Late Precambrian-Cambrian rifted margin and related structures (after Thomas & Astini 1996a).

THE LAURENTIAN PRECORDILLERA TERRANE Ouachita embayment and the Texas promontory on the Laurentian margin, are defined by deep borehole and geophysical data (summarized in Thomas 1991). Syn-rift igneous rocks range between 552-4-7 Ma and 525 + 25 Ma. Two intracratonic graben systems northeast of the Alabama-Oklahoma transform (the Mississippi Valley and the Birmingham graben systems) indicate extension of continental crust parallel to the transform fault. The Cerro Totora graben (Astini et al. 1995b; Astini & Vaccari 1996) in the northern Precordillera could be considered an analogue to those developed in the Laurentian plate. In the Precordillera, as well as in the Laurentian examples (Thomas 1991), post-rift carbonates seem to overlap the graben-filled successions and the graben-boundary faults, suggesting the end of the brittle stretching phase of rifting and the onset of drifting dominated by thermo-tectonic subsidence. A fundamental line of evidence in continental margins is the recognition of the riftdrift transition (Scrutton 1982). This corresponds to a time of major change in basin tectonics, from block-faulting and graben formation during extension, to regional subsidence controlled by lithospheric cooling and sediment loading at the onset of sea-floor spreading (Bond et al. 1995). Although identification of the break-up unconformity is not always straightforward, in the Precordillera terrane as well as in several sites of the southeastern United States, the transgressive contact (Fig. 4) may represent the rift-drift transition. This transition is late Early Cambrian in the Precordillera, the age of the first clear transgressive limestones that yields open-marine inner-shelf Olenellid trilobites (Astini & Vaccari 1996). Furthermore, rift-drift completion can be deduced from the overstep of continental and transitional syn-rift facies by Mid- and Late Cambrian passive-margin facies (Thomas 1991; Astini & Vaccari 1996). This is a very strong argument in favour of rifting and sea-floor spreading associated with the Ouachita rift, which supports the hypothesis of complete rifting and progressive separation of the Precordillera during the Cambrian. New support for Early Cambrian rifting comes from recent observations on the mafic and ultramafic rocks in the western tectofacies, formerly considered to be 'Famatinian ophiolites' (Haller& Ramos 1984) of Late Ordovician age. The preliminary crystallization age of a mafic ophiolite section in the south-west Precordillera is 565 + 45 Ma (Davis et al. 1997) and indicates, according to these authors, early stages of ocean spreading in the latest Proterozoic, as the Precordillera began to rift from Laurentia.

19

Subsidence analysis, when used as the basis of palaeogeographical reconstructions (e.g., Bond et al. 1984), does not usually consider different rifting models, wherein the thermal history varies greatly. According to Thomas & Astini (1996b, in press), the stratigraphical and kinematic constraints of rifting favour an asymmetric rifting model for the separation of the Precordillera and the Texas promontory. The asymmetric simple-shear rifting model predicts that lower-plate margins will develop thicker carbonate banks, whereas the carbonate fabric on the upper plate would be delayed due to thermal doming. For this reason opposite margins of a conjugate rift pair may not necessarily represent a mirror image as was suggested by Bond et al. (1984). Moreover, asymmetry is one of the most convincing features to support simple-shear detachment along rifted margins (e.g., Etheridge et al. 1989). The detachment model (Lister et al. 1991) predicts that when ongoing extension leads to continental break-up and the formation of an ocean basin, the resulting opposing conjugate passive margins display contrasting structural and uplift-subsidence histories. A passive margin was established around the Ouachita embayment by latest Mid- to Late Cambrian time (Thomas 1991). Thermal doming beneath the Ouachita margin prevented development of syn-rift rocks and notably retarded passive-margin sedimentation. Uppermost Middle Cambrian strata unconformably overlie basement rocks, indicating broad uplift of the margin during rifting (passive-margin mountains), probably related to thermal buoyancy. Local palaeo-topographical relief on the sub-Cambrian unconformity is >200m (Barnes et al. 1972). The Upper Cambrian-Lower Ordovician passive margin succession is c. 1000m thick across the Texas promontory, indicating minor subsidence of the crustal column in post-rift stages. In contrast, a >2500m thick carbonate bank overlying locally important graben-fills developed from late Early to early Mid-Cambrian time. The combination of a narrow zone of transitional crust, a steep slope at the margin of continental crust, the lack of synrift rocks, passive-margin strata directly overlapping the basement, and lower rates of post-rift subsidence, suggests an upper-plate configuration for the Ouachita rift (Thomas 1993). In contrast, the combination of a thick passive-margin succession indicating higher rates of post-rift subsidence and a longer subsidence history, overstepping of graben fill towards the continental block interior, and apparent early stage syn-rift rocks of an extended crust in the

20

R . A . ASTINI

western tectofacies (quartz-rich conglomerate blocks) suggests that the Precordillera behaved as a lower-plate margin (Thomas & Astini 1996b, in press). Moreover, a lower-plate configuration for the Precordillera terrane would likely explain the diachronous stratigraphical development on the two sides of the rift and the strong stratigraphical similarities between the Precordillera and the Southern Appalachians. The absence of igneous activity on both could reflect their inherited continuity. A major carbonate rim surrounded much of North America during Cambrian and Early Ordovician times, due to the general stability of the craton and its average equatorial position (Sloss 1963; Witzke 1990), although the subsidence history and thermo-tectonic evolution along the margin differ depending on the inherited margin geometry and the kinematics of the different extended crustal blocks during and after rifting (e.g., Thomas 1993). The Precordillera block had its own evolution in terms of subsidence associated with a lowerplate configuration and recorded the rift-drift transition in the late-Early Cambrian. After the Late Cambrian, the Precordillera block apparently moved beyond the Alabama promontory, breaking all connections with Laurentia. Subsequent drift served to isolate the microcontinent. Because of its low-latitude location, a warm climate (Pacific realm) persisted through the earliest Ordovician, producing reef communities and allowing growth of warm water inhabitants.

Stratigraphical signals o f isolation during the Early Ordovician

Continuous exponential-decay thermal subsidence inferred from subsidence analysis (Bond et al. 1984) indicates motion away from the heated rift zone, leading to progressive separation of the Precordillera terrane. This explains the progressive faunal isolation noted by Benedetto (1993) and is strongly supported by palaeo-biogeographical analysis (see Benedetto this volume). The Early Ordovician limestones, in contrast to those of the Late Cambrian, contain no detrital components (Astini et al. 1995a; Cafias 1995a), although neither is there any evidence to suggest an extra-basin source for siliciclastic rocks of the Cambrian units. This may be interpreted as indicating major sea-level fluctuations or the existence of a barrier between the Precordillera bank and continental Laurentia. Scarce detrital content in similar Appalachian

settings is explained by extremely flat topography and wide platform formation with intrabasin depocentres (cf. Markello & Read 1982; Demicco 1985). Although possible, the simplest explanation is to consider the Early Ordovician Precordillera carbonate platform as completely isolated from the continent and lacking relief, similar to the present-day Bahamas (Cafias 1995a, b; Astini et al. 1995a, 1996). Even with sea-level fluctuations as large as those apparently affecting the Precordilleran bank (Keller et al. 1994; Cafias 1995a), continental clastic material would have never reached the Precordillera because of its isolation. Palaeontological criteria (see Benedetto this volume), in addition to those suggested by Astini et al. (1995a, 1996), Benedetto et al. (1995), Vaccari (1995) and Thomas & Astini (1996a), are provided by Lehnert et al. (1997) and Albanesi & Barnes (1997), who remark that the evolution of conodont faunas suggests that the Precordillera was isolated for some time in the Early Ordovician so as to generate endemism, allopatric speciation, and changeover of faunas.

Stratigraphicai record of drowning and progressive flexuring Drowning of the Precordilleran carbonate bank during early Mid-Ordovician time is recorded by progressive westward and southward onlapping of black shales onto the carbonates (Fig. 3) (Astini 1995b; Astini et al. 1995a). Diachronous drowning indicates an increase in water depth across the platform, caused by lithospheric flexure as the Precordillera approached the east-dipping subduction zone beneath the western margin of Gondwana. Variations in thickness of the black shales along strike and across the platform might be related to brittle flexuring, and consequent block-faulting (e.g., Bradley & Kidd 1991). Forebulging towards the western border of the eastern tectofacies (Fig. 2) hindered black shale deposition. There (e.g., Cord6n de Las Chacritas, Mogotes Azules) a small break in sedimentation and keep-up carbonate successions with dramatic shallowing (grainstones) indicate the swelling topography. Progressive intercalation of ash beds, now K-bentonites, in the mid Arenig-early Llanvirn interval (see Huff et al. this volume) serves as independent evidence of arc-related volcanism, most probably related to progressive proximity of the Precordillera to the overriding hanging wall of the orogen, which could have been either an island-arc complex or an active volcanic-arc

THE LAURENTIAN PRECORDILLERA TERRANE seated on western Gondwana margin (the Famatina System?). Although Keller & Dickerson (1996) have suggested their relation with probable bentonites in the Marathon area, considering them as of intraplate extensional origin, there is clear evidence now that there is no counterpart of the Precordilleran K-bentonites in North America, which in nature and timing (mostly Caradoc, Mohawkian of the American subdivision) match more closely those of northwest Europe. The provenance of coarse siliciclastic conglomerates and sandstones (e.g., the Las Vacas and La Cantera formations) (Fig. 3) suggests a source in easterly uplifted areas, most probably related to the Ocloyic orogeny and closure of the marginal ocean basin between the Precordillera and Gondwana. Little is known about direct evidence of shortening linked to the collision of Precordillera. Recent work by Vujovich (1995), and Ramos et al. (1996) recognizes important mylonitic belts affecting the Grenville basement on the eastern border of the Precordillera terrane, close to the boundary with Ramos' (1995) Pampia terrane. Subsurface analyses by Cominguez & Ramos (1991) and Zapata (1996) have also suggested typical collisional zone mylonite fabrics affecting the Grenvillian basement to the east of the Precordillera. Ramos et al. (1996) provided radiometric dates (c. 464Ma) on an associated peak metamorphism during the MidOrdovician (Llanvirn). Associated magmatism in the Sierras Pampeanas yields Early to Late Ordovician ages (c. 490-455 Ma) (see Pankhurst et al. this volume). Finally, west-directed thrusting affecting the passive margin limestones is known in the eastern Precordillera, and has been ascribed to pre-Carboniferous deformation (von Gosen 1992), most probably Ocloyic.

Stratigraphical record of post-collisional rearrangement Astini (1991, 1995c) noticed an important palaeogeographical rearrangement of depocentres and source areas throughout the Mid- to Late Ordovician (Fig. 3). Most of the units in this interval were included in the Las Plantas and the Trapiche alloformations that represent unconformity-bounded heterogeneous mosaics of varied-calibre siliciclastic and carbonate units. Limited lateral continuity, abrupt facies changes (strong lateral grain-size and compositional variations), development of local hiatuses, and restricted sedimentation are some of the common characteristics of this stratigraphical interval. Extensional tectonics have long been

21

used to explain the complex architecture of the Mid- to Late Ordovician Precordilleran stratigraphy (see Borrello 1969). Three different kinds of end-member deposits can be differentiated in the Mid- to Late Ordovician rocks of the Precordillera: (a) coarse- and medium-grained siliciclastic deposits, mainly of extra-basin origin, (b) fine-grained, mostly distal and/or pelagic strata, signifying restricted circulation and slow settling in isolated depocentres, and (c) carbonate intra-basin sedimentation. Although the clastic deposits of both eastern and western tectofacies are mostly extrabasin, and have probably been funnelled to the basin through transfer zones or tectonic troughs, there is a minor basement component derived from intra-basin exposure and erosion of local basement highs. Such is the case for granite boulders in the basal debris flows of the Los Sombreros Formation, the migmatite and schist blocks and pebbles mixed with abundant coarse limestone debris in the La Invernada Formation in the western Precordillera, and the granite and gneiss cobbles and pebbles mixed in the basal conglomerate in the Pon6n Trehue high. The Pon6n Trehue region in southern Mendoza, >200km south of the typical outcrops of the Precordilleran thrust-fold belt (Fig. 2), was formally incorporated into the Precordillera terrane by Astini et al. (1995a), although Baldis & Blasco (1973), Heredia (1982) and Baldis et al. (1985) had previously noted that this region had strong affinities with the Precordillera. Recently, Bordonaro et al. (1996) reinterpreted the local stratigraphy, defining different units in the area that unconformably overlie Grenvillian basement (see Astini et al. 1996). Although the lower unit defined by these authors (Tremadoc-Arenig in age, Lehnert et al. in press) is treated as inplace, following Heredia (1996) I suggest that these limestone outcrops are fallen blocks (olistoliths), which are very common in the middle Upper Ordovician interval (e.g., San Isidro, Los Tflneles, Fig. 2). These limestones are embedded in conglomeratic matrix with abundant basement source components, and the basal conglomerates in nearby outcrops of demonstrated upper Llanvirn-lower Caradoc age have similar composition. Unfortunately, due to later tectonism, it is difficult to trace either outcrop laterally, although this might be ascribed to abrupt depositional topography, with horsts and graben. Also, where exposed, the contacts of the olistoliths with the basement rocks usually display strong shearing. In considering these blocks as in-place, Keller (1996) assigned them to a cratonic section due to their reduced thickness compared to the

22

R . A . ASTINI

expanded sections to the north (the typical Precordilleran outcrops). Nevertheless, no clastic content has been detected in the blocks: hence, it is more logical to consider them as samples of a 'condensed' section in a less subsiding part of the basin. They indicate that even during the Ordovician there was some N-S (present co-ordinates) variation in the Precordilleran carbonate bank architecture. The Pon6n Trehue limestones (Baldis & Blasco 1973) have been correlated on the basis of faunal content and similar lithologies with the Las Aguaditas Formation, south-west of Jfichal (Fig. 2), one of the areas considered as isolated carbonate remnants nucleated on horst blocks. The sedimentary succession of the Pon6n Trehue block reflects a deepening-upwards trend with greywackes and silty shales overlying mixed composition siliciclastic-carbonate conglomerates, bioclastic grainstones and deep-water dark carbonates. Several other carbonate remnants are detected along the central Precordillera (Fig. 2). These include the San Isidro block, the Sassito block, the Los Blanquitos (Las Aguaditas), the Potrerillo Block, the Las Plantas block;

Middle Silurian stata Sassito Formation [late Caradocearly Ashgill] San Juan Formation E (Arenig] o

Jnsgressive conglomerate erosive unconformity

.4,,,,, forced regression [sudden shallowing] ransgressive conglomerate ~""" [sudden deepening]

I

~. ~

skeletal wackestones ~

I[~l with stromatoporoids intrabasinal ~

carbonate

they can be seen as separated blocks or as more or less aligned horsts. Most reveal some kind of active carbonate fabric, nucleated on bottom swells that reached the photic zone under relatively low detrital inflow. This is characteristic of isolated horsts surrounded by relatively deep-water troughs (grabens) that functioned as traps for clastic material. The Pon6n Trehue block is perhaps one of the most remarkable examples, where both basement and overlying strata have been exposed. Two main sedimentary features, indirectly related to block-faulting, are drowning unconformities and forced regressions (Astini 1997a). Drowning unconformities (Schlager 1989) are relatively abrupt terminations of carbonate fabric induced by relative sea-level rise. Conversely, forced regressions (Posamentier et al. 1992) typically indicate rapid sea-level drops. Both features are developed in several localities during the Mid- to Late Ordovician interval. Similar tectonic drowning unconformities are very commonly observed in the Mesozoic Tethyan platform (Bernoulli & Jenkyns 1974; Santantonio 1994) and are also related to widespread

conglomerate

//

erosive unconformity

calcarenites with ~ minor clastic content black shales

~

green siltstones and sandstones chert & phosphate conglomerate

Fig. 6. Sketch showing the stratigraphic relationships and trends observed in the Sassito Formation type section, cropping out in the Sassito creek along the San Juan river (modified from Astini & Cafias 1995). 1, Exposure and erosion (locally developed conglomerate); 2, rapid submergence and restriction (relative sea-level rise); 3, sudden regression and development of shoreface facies (relative sea-level drop); 4, exposure and erosion; 5, renewed transgression and submergence.

THE LAURENTIAN PRECORDILLERA TERRANE extension. Abrupt downward movement of faulted blocks usually generates incipient or full drowning (e.g., the La Chilca block), whereas upward movement of a relatively deep substrate generates abrupt shallowing and, occasionally, produces exposure and erosion (e.g., the Pon6n Trehue block). Some blocks in the Precordillera even reveal a 'yo-yo' tectonic record (cf. BoseUini 1989), having been alternately exposed to erosion and submerged several times. Such is the case of the Sassito block (Fig. 6), which records deep erosion into Arenig limestones, sudden drowning and smothering by black-shale deposition and abruptly overlying shallow shoreface ramp facies (Astini & Cafias 1995), indicating a forced regression. The development of isolated carbonate remnants is perhaps the most straightforward evidence suggesting extension (Fig. 7). Elevated and relatively isolated blocks have served as ideal places for the nucleation of carbonate

23

sediments that might keep up, catch up or drop out according to relative sea level. Fallen blocks, slumped features, intra-formational scars and debris flows are convincing evidence of escarpments and high-gradient slopes surrounding these highs. Hemipelagic carbonates (e.g., the Las Aguaditas Formation) are also common in this environment. Conglomerate and sandstone fills are widely developed in the easternmost grabens in the Precordillera. In south-west Guandacol (northern Precordillera, Fig. 2), angular unconformities and wedging geometry allow reconstruction of half-graben geometry (Fig. 8). Fine-grained black shales in turn indicate restricted sediment inflow from surrounding highs or extra-basin funnelling and stratification of the water column. These are common attributes of isolated grabens (Fig. 7). Maximum stretching, and consequent block faulting, were apparently achieved in the early

Fig. 7. Environmental reconstruction for the main carbonate and restricted facies associations related to horsts and grabens that characterize the Mid- and Late Ordovician extension in the Precordillera (modified from Santantonio 1994). (a) Half-graben development. (b) Horsts and graben development with isolated restricted depocentres in between.

24

R . A . ASTINI

local source escarpment boulders (intrabasinal)

longitudinal or transversal source through canyons or transfer zones (extrabasinal) J stratal wedging ,.

/~ ,k, master normal fault active during sedimenation

hanging-wall e l

1-5 km I

local source (intrabasinal)

rotated block (foot-wall)

I

Fig. 8. Wedged coarse siliciclastic rocks and fallen olistoliths filling an Upper Ordovician hemi-graben in the region to the south-west of Guandacol (northern Precordillera). Note internal angular unconformities and different sources of the infill (after Astini in press). Late Ordovician (Caradoc) when thick massive wedged conglomerates of extra-basin origin were deposited in the eastern tectofacies, and deep-seated extension led to the generation of important volumes of basaltic pillow-lavas in the western Precordillera (Fig. 9). The origin of these mafic submarine volcanic rocks, interlayered with mid-Caradocian graptolitic shales (Blasco & Ramos 1976; Ortega et al. 1991), has been a matter of debate (Kay et al. 1984; Hailer & Ramos 1984, 1993; Ramos et al. 1986; von Gosen 1992; Dalla Salda et al. 1992a; Astini et al. 1996). Based on stratigraphical evidence it seems most reasonable to consider the MORBlike mafic rocks in the Alcaparrosa Formation and associated units of the western tectofacies as products of submarine volcanism related to some kind of incipient spreading-ridge axis,

distinct from the older and real oceanic lithosphere exposed to the south-west of Precordillera (Davis et al. 1997). Recently, Astini et al. (1995a) and von Gosen et al. (1995) have presented explanations for widespread extension, although they argue different contexts for its occurrence. Whereas Astini et al. consider it to be post-collisional, von Gosen et al. visualize it as pre-collisional, related to passive margin extension. Due to its separation from the Cambrian rifting stage (more than 70 Ma apart) and because a thick stable carbonate platform succession separates both stages, it is difficult to interpret this extension episode as related to episodic stretching of the passive margin. Instead, it is proposed that this extension has a different origin related to post-collisional relaxation (Astini 1997b).

Fig. 9. Schematic east-west cross section showing the block-faulted setting for the Precordillera during the Midand Late Ordovician, herein suggested as the main control on sedimentation and architecture for that time interval.

THE LAURENTIAN PRECORDILLERA TERRANE Because syn-collisional extension seems to be rather exceptional, is better interpreted in terms of post-docking disruption (cf. Howell 1989). A major change in plate motion after docking could have been responsible for triggering extension, particularly near the edge of continental lithosphere, hence producing masswasting deposits such as those observed in the western Precordillera (Astini 1991; Keller 1995) and pillow lavas associated with later deepseated extension. In contrast, in the central Precordillera (eastern tectofacies), there seems to be some overlap of both syn-collisional brittle flexural subsidence and post-collisional relaxation. Elastic flexural subsidence would better relate to the relative proximity of Precordillera and Gondwana during the peripheral foreland stage as previously suggested. This would have happened between the mid-Arenig and the early Llanvirn, whereas post-collisional relaxation would correspond to the more severe blockfaulting that occurred during the late LlanvirnCaradoc. Fault overstepping cannot be observed in outcrop so as to infer the end of this extensional episode, but the Late Ordovician units (e.g., the Trapiche-Empozada formations, Fig. 3) still show very restricted areal distribution and limestone conglomerate mega-beds (Astini 1994). These indicate important block faulting with local exposure of limestone cores that provided the intra-basin source for the megabreccias. Different uppermost Ordovician (postHirnantian) and Silurian successions (Fig. 3) overlap, in most of the central Precordillera, different Ordovician substrates (Astini & Maretto 1996). This would constitute the first clear overlap of Late Ordovician extension although, as suggested by Astini & Maretto (1996), seismicity related to extension faulting would have continued during most of the Silurian.

Discussion The selection of an evolutionary model of the Precordillera must rely entirely on evaluation of the available stratigraphical and associated geophysical and geochemical data. Although the microcontinental model does not allow precise relative longitudinal positioning of Laurentia and Gondwana, it is well supported by the stratigraphical data discussed in this and previous papers. Furthermore, indirect positioning can be obtained on reasonable bases as discussed by Thomas & Astini (1996a). As stressed throughout this paper, the geological evidence favours Early Cambrian rifting and progressive isolation of the Precordilleran microcontinent

25

during the drifting stage. This is more consistent with the traditional plate palaeogeography, where the Iapetus Ocean was the epicentre of intense 'traffic' of terranes and island-arcs (one of which was the Precordillera) in a complex mosaic that resembles the present-day SW Pacific (e.g., MacNiocaill et al. 1997). It is important to note that considering the Precordillera terrane as an independent microcontinent does not preclude the 'end-run' hypothesis proposed by Dalziel (1991) which seems to correspond to the slightly younger history of Laurentia (see Meert et al. 1994; Torsvik et al. 1996). Denying the tectonic tracer hypothesis for the Precordillera unfortunately implies that longitudinal positioning of Laurentia during its relative drifting toward the north remains unconstrained. According to present stratigraphical knowledge, while Laurentia remained within the tropics, the Precordillera terrane drifted southward and by early Late Ordovician (a time when the Iapetus Ocean was fully developed) it occupied a relatively high southern latitude (Astini 1995a). Convergence between Laurentia and Gondwana to form the Taconic and Ocloyic (Famatinian) orogenies (Dalla Salda et al. 1992b), respectively, in South and North America, is presently unsupported. The Precordillera was most probably isolated from Laurentia by a western portion of the Iapetus Ocean during the Taconic Orogeny, the latter having its origin in different terranes that accreted against Laurentia as part of the complex mosaic of terranes and island arcs that characterized the Iapetus Ocean (Fig. 10). The Ocloyic orogeny can, in turn, be considered as the product of collision of island-arc complexes, such as the Famatina System and the Puna (Conti et al. 1996), and the Precordillera terrane against the western margin of Gondwana (Fig. 10). Early Cambrian rifting of the Precordillera block, at variance with the 'bridge' model, explains the development of graben and grabenfilled successions and the evolution of the passive margin stage. Mafic rocks from an ophiolite section proved older than Caradoc have recently been found in the western Precordillera (Davis et al. in press). Pods of serpentinite and metagabbro within the Ouachita thrust belt/ accretionary prism in the Laurentian margin suggest oceanic crust beneath the Late Cambrian and younger slope-and-rise deposits of the Ouachita embayment (Nielsen et al. 1989; Viele & Thomas 1989). Additionally, faunal assemblages diagnostic of shelf margin-slope biofacies (Palmer et al. 1996) were found in the Middle Cambrian olistoliths of the western Precordillera

26

R . A . ASTINI

(Bathyurids)

. - . ........ ,(~

.

....... "--.

..

~

,

_

30.~

shales and fine-grained siliciclastic rocks . . . . . . . .

.

,

carbonates low-latitude

.

~

volcaniclastic and

volcanic rocks

.-'"

(B

'~

shallow-marine siliciclastic rocks BA

/

_~-. . . . . . . . . . . . . . ....0-Oo ~

"".4Y

/'

\

. ~//(Northern)

/

Lt:~#' "~ g .~, / ~"

9 . . . ~ ~ut~s*

\

"" i,

w-'"" lll'"..~,

co

~

| SOUTH POLE

d

i

i

i

! /'

~,J-

..

.J

/

,/

-'//

/' //'

....,

/

,/ /

~RECORDI TERRANE LLERA,~, ~"'"", ! (mixed Toquima-Table

(Bath~iurids) + enderiti~s

/

/

/

(Southern),i

Head and'Celtic genera)

/

-'0 ~

/

y/

FAMATINA ' ISLAND ARC

(mixed Celtic and ".Mediterranean genera)

c. 470 Ma

,/

/ ,+'/

~N l

.."

Fig. 10. Early Ordovician palaeogeographic reconstruction showing distribution of brachiopod faunas, trilobites and facies belts. L, Laurentia; S, Siberia; B, Baltica; G, Gondwana; NCH, North China block; SCH, South China block; AM, Armorica; BM, Bohemia; AN, Antarctica; AUST, Australia; IN, India. Northern and Southern Iapetus are indicated. (modified from Astini et al. 1995a; Benedetto & S{mchez 1996; Torsvik et al. 1996; Mac Niocaill et al. 1997).

(Vaccari & Bordonaro 1993). In the absence of an ophiolite suite in structurally complicated areas, Fortey & Cocks (1986) have suggested that marginal faunal belts or platform edge biofacies can be used, in principle, to indicate the proximity of former ocean basins. These findings reinforce the suggested Early Cambrian rifting phase and expansion of an oceanic basin between the Precordillera and the Texas promontory, which would further develop into a true barrier between the Precordillera and Laurentia. A possible modern analogue of the Precordillera-microcontinental model, where the jump of

a spreading ridge occurs inboard into continental crust to separate a smaller crustal block, is the Jan Mayen Ridge. This ridge is a microcontinent rifted from east Greenland, with transform motion on the Jan Mayen fracture zone (Eldholm e t al. 1990; Thomas in press). Dalziel (1997) pointed out that offset lengths on transform segments usually shorten rather than lengthen in present-day examples and, hence, this would constitute a serious objection to the microcontinent model. Nevertheless, Vink e t al. (1984) have described several cases of rifting of small continental fragments or microplates

THE LAURENTIAN PRECORDILLERA TERRANE where inboard ridge-jumps and lengthening offsets on transform segments have preferentially occurred. Major elements that collectively support the rifting, drifting, and collisional model are: (1) rifted margin stratigraphy, (2) development of a typical thermo-tectonically controlled passive-margin carbonate bank, (3) progressive differentiation in faunas and lithology, (4) apparent isolation of the carbonate platform through most of the Tremadoc and Arenig (basically much of the North American Ibexian), (5) beginning of widespread drowning together with inflow of new faunas and general change toward cooler water environments, (6) ash beds reflecting progressive the proximity of volcanic arcs, (7) pervasive instability throughout the basin (suggested from stratigraphical architecture and irregular distribution of sediments and hiatuses), and (8) direct glacial evidence related to Hirnantia fauna. This evidence means that the Precordillera could be considered one of the best constrained allochthonous terranes in Earth history.

Conclusions (1) Based on a reappraisal of the available stratigraphical evidence it is suggested that the Precordillera terrane should not be considered a tectonic tracer, but should instead be categorized as an independent lithospheric fragment exotic to Gondwana. (2) Passive margin stratigraphy, subsidence analysis and marginal faunal belts, associated with a straightforward west-polarity, indicate a fully developed rift stage during the Early Cambrian. (3) A mixed succession of evaporites, red siliciclastic rocks and dolomites underlying the thick Cambrian to Lower Ordovician carbonate platform record a syn-rift, restricted circulation, low latitude, divergent margin setting. Rift to drift completion can be deduced from the overlap of the syn-rift facies by transgressive glauconitic sandstones and limestones with the first open-marine faunas, of late Early Cambrian age. (4) The Precordillera block has its source in the Ouachita embayment in southeastern Laurentia. Stratigraphical and kinematic constraints favour an asymmetric rifting model for the separation of Precordillera and the Texas promontory. (5) After the Late Cambrian, the Precordillera block apparently moved beyond the Alabama promontory, breaking all connections with

27

Laurentia. Progressive separation of the Precordillera block from Laurentian during the drifting stage is mainly supported by the changeover in faunas, although progressive isolation is also indicated by sedimentary features and exponential decay of thermal history. (6) Gradual proximity to Gondwana and bordering island- or volcanic-arcs is marked by prominent ash layers included in the upper part of the Lower Ordovician limestones and a forebulge flexural stage inferred by progressive drowning of the carbonate platform. (7) Mid- to Late Ordovician widespread extension is deduced from the complex architecture and the development of local hiatuses, and restricted depocentres with dominantly siliciclastic sedimentation and scarce carbonate remnants in the Precordillera. This second extension episode is separated by more than 70 Ma from the initial rifting of the Precordillera from Laurentia, and is related to post-collisional disruption and major change in plate motions following collision. (8) Late Ordovician glacial deposits typical of Gondwana overlap the Precordillera region, yielding a minimum age for the proposed accretion of the Precordillera terrane, although pervasive indirect sedimentary evidence of extensional tectonics by Mid-Ordovician allows bracketing the collision near the early MidOrdovician. (9) Stratigraphical evidence points toward different timing and nature of the Ocloyic and Taconic orogenies in South and North America, respectively, although both are related to the collision of terranes and island arcs on opposite margins of the evolving Iapetus. Various Argentine institutions supported my research in the Argentine Precordillera: Consejo Nacional de Investigaciones Cientificas y T+cnicas (CONICET), Consejo de Investigaciones Cientificas y T~cnicas de la Provincia de C6rdoba (CONICOR), Fundaci6n Antorchas, and Secretaria de Ciencia y T~cnica de la Universidad Nacional de C6rdoba (SECyT). Support from the University of Western Michigan and the University of Texas at Austin allowed me to visit type Laurentian localities. I am grateful to Bob Hatcher Jr., for patient advice on a draft manuscript and suggestions that greatly improved this work, as did constructive reviews by John Cooper and David Macdonald. I very much appreciate continuous discussion of Precordilleran topics with colleagues in C6rdoba, especially Luis Benedetto, Marcelo Carrera, Edsel Brussa, Emilio Vaccari and Fernando Cafias, and I also wish to thank Augusto Rapalini, Victor Ramos, Bill Thomas, Steven Davis, Sarah Roeske, Warren Huff, Stig Bergstr6m, Chuck Mitchell, Carlos Cingolani, Christopher Schmidt and Ian Dalziel.

28

R

e

f

R.A.

e

r

e

n

c

e

s

ALBANESI, G. L. & BARNES, C. R. 1997. The Middle Ordovician conodont Paroistodus horridus in the Argentine Precordillera. 6th International Conodont Symposium, Warsaw, Abstracts, 1. , HONICKEN, M. & ORTEGA, G. 1990. Amorphognathus supervus (Conodonta) from Trapiche Formation (Upper Ordovician), Cerro Potrerillos, J/tchal Department, San Juan Province, Argentina. First Latin American Conodont Symposium, Academia de Ciencias, C6rdoba, Abstracts, 109-110. AITKEN, J. D. 1966. Middle Cambrian to Middle Ordovician cyclic sedimentation, southern Rocky Mountains of Alberta. Bulletin of Canadian Petroleum Geology, 14, 405-441. ALLMENDINGER, R. W., FIGUEROA, D., SYDNER, D., BEER, J., MPODOZIS, C. & ISACKS, B. L. 1990. Foreland shortening and crustal balancing in the Andes at 30~ latitude. Tectonics, 9, 789-809. ARMELLA, C. 1994. Thrombolitic-stromatolitic cycles of the Cambro-Ordovician boundary sequence, Precordillera oriental basin, western Argentina. In: BERTRAND SARFATI, J. & MONTY, C. (eds) Phanerozoic Stromatolites, 2, Kluwer Academic Publishers, Netherlands, 421-441. ASTINI, R. A. 1991. Paleoambientes sedimentarios y secuencias depositacionales del Ordovicico cldstico de la Precordillera Argentina. PhD thesis, Universidad Nacional de C6rdoba. - - 1 9 9 2 . Tectofacies ordovicicas y evoluci6n de la cuenca eopaleozoica de la Precordillera Argentina. Estudios Geol6gicos, 48, 315-327. - - 1 9 9 4 . Las megaturbiditas de la Formaci6n Trapiche (Ordovicico superior de la PrecordiUera): procesos sedimentarios y marco geol6gico. Quinta Reunidn Argentina de Sedimentologia, San Miguel de Tucumfin, Argentina, 1, 107-112. - - 1 9 9 5 a . Paleoclimates and paleogeographic paths of the Argentine Precordillera during the Ordovician: evidence from climatically sensitive lithofacies. In: COOPER, J. D., DROSER, M. L. & FINNEY, S. C. (eds) Ordovician Odyssey. Society of Economic Paleontologists and Mineralogists, Book 77, 177-180. --1995b. Geologic meaning of Arenig-Llanvirn diachronous black shales (Gualcamayo Alloformation) in the Argentine Precordillera, tectonic or eustatic? In: COOPER, J. D., DROSER, M. L. & FINNEY, S. C. (eds) Ordovician Odyssey. Society of Economic Paleontologists and Mineralogists, Book 77, 217-220. - - 1 9 9 5 c . Sedimentologia de la Formaci6n Las Aguaditas (talud carbon/~tico) e implicancias estratigr/lficas en la cuenca precordillerana oriental durante el Ordovicico. Revista de la Asociacidn Geoldgica Argentina, 50, 143-164. --1996. Las fases diastr6ficas del Paleozoico medio en la Precordillera del oeste argentinoevidencias estratigraficas-. XIII Congreso Geoldgico Argentino y III Congreso de Exploracidn de Hidrocarburos, Buenos Aires, Actas, V, 509-526.

ASTINI - - 1 9 9 7 a . Las unidades calc~treas del Ordovicico Medio y Superior de la Precordillera Argentina como indicadores de una etapa extensional. Segundas Jornadas de Geologia de Precordillera, San Juan, Argentina, 1, 8-14. 1997b. Stratigraphic evidence of two-stage rifting and collison in the Laurentian derived Precordillera terrane, south-central Andes. Geological Society of America, Annual Meeting, Abstracts with Programs, 29, A-116. in press. El "Conglomerado de Las Vacas" y el Grupo Trapiche de la Precordillera: Tectbnica distensiva en el Ordovicico Superior. Revista de la Asociacidn Geoldgica Argentina. & BENEDETTO, J. L. 1996. Tectonostratigraphic development and history of an allochthonous terrane in the Pre-Andean Gondwana margin: the Argentine Precordillera. 3rd International Symposium on Andean Geodynamics, St Malo, France, Extended Abstracts, 759-762. & BRUSSA, E. D. 1997. Dos nuevas localidades fosiliferas en el Conglomerado de Las Vacas (Caradociano) en la Precordillera Argentina: Importancia cronoestratigrfifica. Reuni6n de Comunicaciones Paleontol6gicas. Ameghiniana, 43, 231. & CA~AS, F. L. 1995. La Formaci6n Sassito, una nueva unidad calc/~rea en la Precordillera de San Juan: sedimentologia y significado estratigrfifico. Asociacidn Argentina de Sedimentologla, Revista, 2, 19-37. & MARETTO, H. M. 1996. Anfilisis estratigr/lfico del Silflrico de la Precordillera Central de San Juan y consideraciones sobre la evoluci6n de la cuenca. XIII Congreso Geolbgico Argentino y III Congreso de Exploracibn de Hidrocarburos, Buenos Aires, Actas, I, 351-368. • VACCARI,N. E. 1996. Sucesi6n evaporitica del Cfimbrico inferior de la Precordillera: significado geol6gico. Revista de la Asoeiacidn Geoldgica Argentina, 51, 9%106. - - , BENEDETTO,J. L., & VACCARI, N. E. 1995a. The Early Paleozoic evolution of the Argentine Precordillera as a Laurentian rifted, drifted and collided terrane: a geodynamic model. Geological Society of America Bulletin, 107, 253-273. , RAMOS, V. A., BENEDETTO, J. L., VACCARI, N. E. & CAlqAS, F. L. 1996. La Precordillera: un terreno exotico a Gondwana. XIII Congreso Geoldgico Argentino y III Congreso de Exploracidn de Hidrocarburos, Buenos Aires, Actas, V, 293-324. - - - , VACCARI, N. E., THOMAS, W. A., RAYMOND, D. E. & OSBORNE, W. E. 1995b. Shared evolution of the Argentine Precordillerra and the Southern Appalachians during the Lower Cambrian. Geological Society of America, Abstracts with Programs, 27, 458. BALDIS, B. A. & BLASCO, G. 1973. Trilobites Ordovicicos de Pon6n Trehue, Sierra Pintada de San Rafael, provincia de Mendoza. Ameghiniana, 10, 72-88. & BORDONARO, O. 1981. Evoluci6n de las facies carbon/lticas en la cuenca C~tmbrica de la Precordillerra de San Juan. VIII Congreso Geoldgico Argentino, San Luis, Actas, II, 385-397. -

-

-

-

-

-

-

-

-

-

-

-

THE LAURENTIAN PRECORDILLERA TERRANE --

& 1982. Comparaci6n entre el Cfimbrico de la "Great Basin" norteamericana y la Precordillera de San Juan, Argentina, su implicancia intercontinental. V Congreso Latinoamericano de Geologia, Buenos Aires, 1, 97-108. & 1985. Variaciones de facies en la cuenca cfimbrica de la Precordillera Argentina, y su relaci6n con la generacidn del borde continental. VI Congreso Latinoamerieano de Geologia, Bogotfi, Actas, 1, 149-161. , ARMELLA, C. & CABALERI, N. 1985. Desarrollo de la plataforma carbonfitica Ordovicica argentina. VI Congreso Latinoamericano de Geologia, Bogota, Actas, 1, 165-174. BARNES, V. E., BELL, W. C., CLABAUGH,S. E., CLOUD, P. E., JR., MCGEHEE, R. V., RODDA, P. U. & YOUNG, K. 1972. Geology of the Llano region and Austin area, field excursion. Guidebook No. 13. Texas Bureau of Economic Geology, University of Texas, Austin. BENEDETTO, J. L. 1993. La hip6tesis de la aloctonia de la Precordillera Argentina: un test estratigr/tfico y biogeogrfifico. XII Congreso Geoldgico Argentino y H Congreso de Exploracion de Hidrocarburos, Mendoza, Actas, III, 375-384. - - 1 9 9 8 . Early Palaeozoic brachiopods and associated shelly faunas from western Gondwana: their bearing on the geodynamic history of the pre-Andean margin. This volume. -& ASTINI, R. A. 1993. A collisional model for the Early Paleozoic stratigraphic evolution of the Argentine Precordillera. 2nd International Symposium on Andean Geodynamics, Oxford, Extended Abstracts, 501-504. -& HERRERA, Z. A. 1987. Primer hallazgo de braqui6podos y trilobites de la Formaci6n Trapiche (Ordovicico Tardio), Precordillera Argentina. X Congreso Geol6gico Argentino, Tucumfin, Actas, III, 73-76. & SANCHEZ, T. M. 1996. Paleobiogeography of brachiopod and molluscan faunas along the South American margin of Gondwana during the Ordovician. In: BALDIS, B. A. & ACElqOLAZA, F. G. (eds) Early Paleozoic Evolution in NW Gondwana. Serie Correlaci6n Geol6gica, Universidad Nacional de Tucum/m, 12, 23-38. & VACCARI,N. E. 1992. Significado estratigrfifico y tect6nico de los complejos de bloques cambroordovicicos resedimentados de la Precordillera Occidental, Argentina. Estudios Geol6gicos, 48, 305-313. - - , CARRERA, M. G. & S,h,NCHEZ, T. M. 1995. ~Fhe evolution of faunal provincialism in the Argentine Precordillera during the Ordovician: new evidence and paleontological implications. In: COOPER, J. D., DROSER, M. L. & FINNEY, S. C. (eds) Ordovician Odyssey. Society of Economic Paleontologists and Mineralogists, Pacific Section, Book 77, 181-184. BERNOULLI, D. & JENKYNS, H. 1974. Alpine, Mediterranean, and central Atlantic Mesozoic facies in relation to the early evolution of the Tethys. In: DOTT, R. H. & SHAVER, R. H. (eds) Modern and ancient geosyndinal evolution. Society of -

-

-

-

29

Economic Paleontologists and Mineralogists, Special Publications, 19, 129-160. BLASCO, G. & RAMOS, 1976. Graptolitos caradocianos de la Formaci6n Yerba Loca y del Cerro La Chilca, Departamento J/lchal, provincia de San Juan. Ameghiniana, 13, 312-329. BOND, G. C., KOMINZ, M. A & DEVL1N, W. J. 1983. Thermal subsidence and eustasy in the Lower Paleozoic miogeocline of western North America. Nature, 306, 775-779. - - , KOMINZ, M., STECKLER, M. & GROTZINGER, J. 1989. Role of thermal subsidence, flexure and eustasy in evolution of Early Paleozoic passivemargin carbonate platforms. In: CREVELLO,P. D., WILSON, J. L., SARG, J. F. & READ, J. F. (eds) Controls on carbonate platform and basin development. Society of Economic Paleontologists and Mineralogists, Special Publications, 44, 39-61. SHERIDAM, R. E. 1995. Continental terraces and rises. In: BusBY, C. J. & INGERSOLL, R. V. (eds) Tectonics of Sedimentary' Basins. Blackwell Science, Oxford, 149-178. --, NICKESON, P. A. & KOMINZ, M. A. 1984. Breakup of a supercontinent between 625 Ma and 555Ma: New evidence and implications for continental histories. Earth and Planetary Science Letters, 70, 325-345. BORDONARO,O. L & BANCHIG,A. L. 1996. Estratigrafia de los olistolitos cfimbricos de la Precordillera Argentina: XIII Congreso GeoldgicoArgentino y Ill Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, V, 471-479. - - , KELLER, M. & LEHNERT, O. 1996. E10rdovicico de Pon6n Trehue en la Provincia de Mendoza (Argentina): redefiniciones estratigrfificas..gill Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, I, 541-550. BORRELLO, A. V. 1965. Sobre la presencia del Cb,mbrico Inferior olenellidiano en la Sierra de Zonda, Precordillera de San Juan. Ameghiniana, 3, 313-318. 1969. Embriotect6nica y tect6nica tensional. Su importancia en la evoluci6n estructural de la Precordillera. Revista de la Asociacidn Geol6gica Argentina, 24, 5-13. 1971. The Cambrian of the South America. In: HOLLAND, C. H. (ed.) Cambrian of the New World. Lower Paleozoic Rocks of the World, Wiley Interscience, New York, 385-438. BOSELLINI, A. 1989. Dynamics of Tethyan carbonate platforms. In: CREVELLO, P. D., WILSON, J. L., SARG, J. F. & READ, J. F. (eds) Controls on carbonate platform and basin de velopment. Society of Economic Paleontologists and Mineralogists, Special Publications, 44, 3-13. BRADLEY, D. C. & KIDD, W. S. F. 1991. Flexural extension of the upper continental crust in collisional foredeeps. Geological Society America Bulletin, 103, 1416-1438. BRAILE,L. W., HINZE, W. J., KELLER,G. R., LIDIAK,E. G. & SEXTON,J. L. 1986. Tectonic development of the New Madrid rift complex, Mississippi Embayment, North America. Tectonophysics, 131, 1-21.

30

R.A.

BRUSSA, E. D. 1994. Las graptofaunas ordovicicas del sector central de la Precordillera Occidental sanjuanina. PhD thesis, Universidad Nacional de C6rdoba. BUGGISCH, W. 1996. The Argentine Precordillera Terrane -critical remarks and remaining questions-. XIII Congreso Geol6gico Argentino y III Congreso de Exploraciones de Hidrocarburos, Buenos Aires, Actas, V, 469. CAt'AS, F. L. 1990. Biolititos microbianos (microbialitos) del C~imbrico tardio, Formaci6n Los Sapitos, Precordillera de La Rioja. III Reuni6n Argentina de Sedimentologia, San Juan, Actas, 81-86. - - 1 9 9 5 a . Estratigrafia y evolucidn paleoambiental de las sucesiones carbondticas del Cdmbrico tardlo y Ordovicico temprano de la Precordillera Septentrional, Repfiblica Argentina. PhD thesis, Universidad Nacional de C6rdoba. - - 1 9 9 5 b . Early Ordovician carbonate platform facies of the Argentine Precordillera: Restricted shelf to open platform evolution. In: COOPER, J. D., DROSER, M. L. & FINNEY, S. C. (eds) Ordovician Odyssey. Pacific Section, Society of Economic Paleontologists and Mineralogists, Book 77, 221-224. ~; CARRERA, M. 1993. Early Ordovician spongesalgal reef mound of the Precordillera Basin, Western Argentina. Facies, Erlangen, 29, 169-178. & KELLER, M. 1993. Arrecifes y monticulos arrecifales en la Formaci6n San Juan (Precordillera Sanjuanina, Argentina): Los arrecifes mils antiguos de Sudambrica. Boletin de la Real Sociedad Espa~qola de Historia Natural (Secci6n Geologla), 88, 12%136. COMIN~UEZ, A. H. & Ramos, V. A. 1991. La estructura profunda entre Precordillera y Sierras Pampeanas de la Argentina. Evidencias de la sismica de reftexi6n profunda. Revista Geol6gica de Chile, 18, 3-14. CONTI, C. M., RAPALINI, A. E., COIRA, C. & KOUKHARSKY, M. 1996. Paleomagnetic evidence of an early Paleozoic rotated terrane in northwest Argentina: A clue for Gondwana-Laurentia interaction? Geology, 24, 953-956. DALLA SALDA, L., CINGOLANI, C. A. & VARELA, R. 1992a. Early Paleozoic belt of the Andes and Southwestern South America: Result of Laurentia-Gondwana collision? Geology, 20, 617-620. , DALZIEL,I. W. D., C1NGOLANI,C. A. & VARELA, R. 1992b. Did the Taconic Appalachians continue into Southern South America? Geology, 20, 1059-1062. DALZIEL, I. W. D. 1991. Pacific margins of Laurentia and East Antarctica as a conjugate rift pair: Evidence and implication for an Eocambrian supercontinent. Geology, 19, 598-601. - - 1 9 9 3 . Tectonic tracers and the origin of the protoAndean margin. XII Congreso Geol6gico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, III, 367-374. - - 1 9 9 7 . Neoproterozoic-Paleozoic geography and tectonics: review, hypothesis, environmental speculation. Geological Society of America Bulletin, 109, 16-42. -

-

-

-

ASTINI --,

DALLA SALDA, L. H. & GAHAGAN, L. M. 1994. Paleozoic Laurentia-Gondwana interaction and the origin of the Appalachian-Andean mountain system. Geological Society of America Bulletin, 106, 243-252. - - , - - , CINGOLANI,C. • PALMER,A. R. 1996. The Argentine Precordillera: A Laurentian terrane: Penrose Conference Report. GSA Today (February), 16-18. DAVIS, J. S., MCCLELAND, W. C., ROESKE, S. M., GERBI, C. & MOORES, E. M. 1997. The early to middle Paleozoic tectonic history of the SW Precordillera terrane and Laurentia-Gondwana interactions. Geological Society of America, Annual Meeting, Abstracts with Programs, 29, A-380. DEMICCO, R. V. 1985. Platform and off-platform carbonates of the Upper Cambrian of western MARYLAND, U. S. A. Sedimentology, 32, 1-22. ELDHOLM, O., SKOGSEID, J., SUNDVOR, E. & MYHRE, A. M. 1990. The Norwegian-Greenland Sea. In: GRANTZ, A., JOHNSON, L. & SWEENEY,J. F. (eds) The Arctic Ocean region. Geological Society of America, The Geology of North America, L, 351-364. ETHERIDGE, M. A., SYMONS, P. A. & LISTER, G. S. 1989. Application of the detachment model to reconstruction of conjugate passive margins. In: TANKARD, A. J. & BALKWILL, H. R. (eds) Extensional tectonics and stratigraphy of the North Atlantic margins. American Association of Petroleum Geologists, Memoirs, 46, 23-40. FORTEY, R. A. & COCKS, L. R. M. 1996. Marginal faunal belts and their structural implications, with examples from the Lower Palaeotoic. Journal of the Geological Society, London, 143, 151-160. GONZALEZ BONORINO, G. 1975. Sedimentologia de la Formacidn Punta Negra y algunas consideraciones sobre la geologia regional de la Precordillera de San Juan y Mendoza. Revista de la Asociaci6n Geol6gica Argentina, 30, 223-246. HALLER, M. J. & RAMOS, V. A. 1984. Las ofiolitas famatinianas (Eopaleozoico) de las Provincias de San Juan y Mendoza. IX Congreso Geoldgico Argentino, San Carlos de Bariloche, Actas, II, 66-83. & 1993. Las ofiolitas y otras rocas afines. In: RAMOS, V. A. (ed.) Geologia y recursos naturales de Mendoza. XII Congreso Geol6gico Argentino y II Congreso de Exploraci6n de Hidrocarburos, Mendoza, Relatorio, 31-39. HANDFORD, C. R. 1991. Marginal marine halite: sabhkas and salinas. In: MELVIN, J. L. (ed.) Evaporites, petroleum and mineral resources. Elsevier, Amsterdam, Developments in Sedimentology, 50, 1-66. HATCHER, R. D. JR. 1983. Basement massifs in the Appalachians: their role in deformation during the Appalachian orogenesis. Geological Journal, 18, 255-265. HEREDIA, S. 1982. Pygodus anserinus Lamont & Lindstr6m (Conodonto) en el Llandeilliano de la Formacidn Pondn Trehue. Ameghiniana, 19, 101-104. -

-

THE LAURENTIAN PRECORDILLERA TERRANE - - 1 9 9 6 . El Ordovicico del Arroyo Pon6n Trehue, sur de la provincia de Mendoza. XIII Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, I, 601-605. HERRERA,Z. A. & BENEDETTO,J. L. 1991. Early Ordovician brachiopod faunas of the Precordillera basin, Western Argentina: Biostratigraphy and paleobiogeographical affinities. In: McKINNON, D. I., LEE, D. E. & CAMPBELL,J. D. (eds) Brachiopods through time. Balkema, Rotterdam, 283-301. HOWELL, D. G. 1989. Tectonics of suspect terranes. mountain building and continental growth. Chapman and Hall, London. HUFF, W. D., BERGSTROM, S. M., KOLATA, D. R., CtNGOLANI, C. & ASTINI, R. A. 1998. Ordovician K-bentonites in the Argentine Precordillera: relations to Gondwana margin evolution. This volume. JORDAN, T. E., ISACKS,B. L., ALLMENDINGER,R. W., BREWER, J. A., RAMOS,V. A. & ANDO, C. J. 1983. Andean segmentation related to geometry of subducted Nazca plate. Geological Society of America Bulletin, 4, 341-361. KAY, S. M., RAMOS, V. A. & KAY, R. W. 1984. Elementos mayoritarios y trazas de las vulcanitas ordovicicas de la Precordillera Occidental: basaltos de rift ocefinico temprano(?) pr6ximos al margen continental. IX Congreso Geol6gico Argentino, San Carlos de Bariloche, Actas, II, 48-65. , ORRELL, S. & ABRUZZl, J. M. 1996. Zircon and whole rock Nd-Pb isotopic evidence for a Grenville age and Laurentian origin for the basement of the Precordilleran terrane in Argentina. Journal of Geology, 104, 637-648. KELLER, M. 1995. Continental slope deposits in the Argentina Precordillera: sediments and geotectonic significance. In: COOPER, J. D., DROSER, M. L. & FINNEY, S. C. (eds) Ordovician Odyssey. Society of Economic Paleontologists and Mineralogists, Pacific Section, Book 77, 211-215. - - 1 9 9 6 . Anatomy of the Precordillera (Argentina) during Cambro-Ordovician times: implications for the Laurentia-Gondwana transfer of the Cuyania Terrane. 3rd International Symposium in Andean Geodynamics, St Malo, France, Extended Abstracts, 775-778. & BORDONARO, O. 1993. Arrecifes de estromatoporoideos en el Ordovicico inferior del oeste argentino y sus implicancias paleogeogr~ficas. Revista Espa~ola de Paleontologia, 8, 165-169. & DICKERSON, P. W. 1996. The Missing continent of Llanoria- was it the Argentine Precordillera?. XIII Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, V, 355-367. , BUGGISCH, W. & BERCOWSKI, F. 1989. Facies and sedimentology of Upper Cambrian shallowing-upward cycles in the La Flecha Formation (Argentine Precordillera). Zentralblatt fiir Geologic und Palaeontologie, 1, 999-1011. -& LEHNERT, O. 1998. The stratigraphical record of the Argentine Precordillera and its platetectonic background. This volume.

--,

31

CANAS, F. L., LEHNERT, O. • VACCARI, N. E. 1994. The Upper Cambrian and Lower Ordoviclan of the Precordillera (Western Argentina): some stratigraphic reconsiderations. Newsletter on Stratigraphy, 31, 115-132. KENDALL, C. G. ST. C. 8~ WARREN, J. 1987. A review of the origin and setting of tepees and their associated fabrics. Sedimentology, 34, 1007-1027. KINSMAN, D. J. J. 1975. Rift valley basins and sedimentary history of trailing continental margins. In: FISCHER, A. G. 8~ SHELDON, J. (eds) Petroleum and global tectonics. Princeton University Press, Princeton, 83-126. KOERSCHNER,W. F. III. & READ, J. F. 1989. Field and modelling studies of Cambrian carbonate cycles, Virginia Appalachians. Journal of Sedimentary Petrology, 59, 654-687. KRAEMER, P. E., ESCAYOLA,M. P. & MARTINO, R. D. 1995. Hip6tesis sobre la evoluci6n tect6nica neoproterozoica de las Sierras Pampeanas de C6rdoba (30~176 Argentina. Revista de la Asociaci6n Geol6gica Argentina, 50, 47-59. LEHNERT, O. 1995. Ordovizische Conodonten aus der Prdkordillere Westargentiniens: Ihre Bedeutung ffir Stratigraphie und Palaogeographie. PhD thesis, Erlager Geologische Abhandlungen, Erlangen, Germany. --, MILLER, J. L. & REPETSKI, J. E. 1997. Paleogeographic significance of Clavohamulus hintzei Miller (Conodonta) and other Ibexian conodonts in an early Paleozoic carbonate platform facies of the Argentine Precordillera. Geological Society of America Bulletin, 109, 429-443. --, KELLER, M. 8r BORDONARO, O. in press. Early Ordovician conodonts from the southern Cuyania terrane (Mendoza Province, Argentina). Proceedings of ECOS VI, Paleontologica Polonica. LISTER, G. S., ETHRIDGE,M. A. & SYMONDS,P. A. 1991. Detachment models for the formation of passive continental margins. Tectonics, 10, 1038-1064. MAC NIOCAILL, C., VAN DER PLUIJM, B. & VAN DER VOO, R. 1997. Ordovician paleogeography and the evolution of the Iapetus ocean. Geology, 25, 159-162. MARKELLO, J. R. & READ, J. F. 1982. Upper Cambrian intrashelf basin, Nolichucky Formation, southwest Virginia Appalachians. American Association of Petroleum Geologists Bulletin, 66, 860-878. MEERT, J. G., VAN DEL VOO, R. & PAYNE, T. W. 1994. Paleomagnetism of the Catoctin volcanic province: A new Vendian-Cambrian apparent polar wander path for North America. Journal of Geophysical Research, 99, 4625-4641. MELLEN, F. F. 1977. Cambrian System in Black Warrior basin. American Association of Petroleum Geologists, Bulletin, 59 193-216. NIELSEN, K. C., VIELE, G. W. 8z ZIMMERMAN,J. 1989. Structural setting of the Benton-Broken Bow uplifts. In: HATCHER, R. D. JR., THOMAS, W. A. 8z VIELE, G. W. (eds) The Applachian-Ouachita orogen in the United States. Geological Society of America, The Geology of North America, F-2, 635-660.

32

R.A.

ORTEGA, G. A. & BRUSA, E. D. 1990. La subzona de Climacograptus bicornis (Caradociano temprano) en la Formacidn Las Plantas en su localidad tipo, Precordillera de San Juan, Argentina: Ameghiniana, 27, 281-288. ---, -& ASTINI, R. A. 1991. Nuevos hallazgos de graptolitos en la Formacibn Yerba Loca y su implicancia estratigr/tfica (Precordillera de San Juan, Argentina). Ameghiniana, 28, 163-178. OSLEGER, D. A. 1993. Cyclostratigraphy of Late Cambrian carbonate sequences: an interbasinal comparison of the Cordilleran and Appalachian passive margins. In: COOPER, J. D. & STEVENS, C. H. (eds) Paleozoic Paleogeography of the Western United States. Society of Economic Paleontologists and Mineralogists, Pacific Section Pacific Section, Book 67, 801-828. PALMER, A. R., VACCARI, N. E., ASTINI, R. A., CAiqAS, F. L. & BENEDETTO,J. L. 1996. Cambrian history of the Argentine Precordillera: Adios Laurentia! Geological Society of America South Central Meeting, Abstracts with Programs, 28, 57. PANKHURST, R. J. & RAPELA, C. W. 1998. Introduction. This volume. , SAAVEDRA,J., BALDO, E., DAHLQU1ST,J., PASCUA, I. & FANNING, C. M. 1998. The Famatinian magmatic arc in the central Sierras PampeaHas. This volume. POSAMENTIER, H. W., ALLEN, G. P., JAMES, D. P. & TESSON, M. 1992. Forced regressions in a sequence stratigraphic framework: concepts, examples and exploration significance. American Association of Petroleum Geologists Bulletin, 76, 1687-1709. POULSEN, V. 1958. Contribution to the Middle Cambrian paleontology and stratigraphy of Argentina. Copenhague Matematisc fysiske Meddelelser, Der kongelige Danske Videnskabernes Selsbak, 31, 1-22.

RAMOS, V. A. 1995. Sudambrica: un mosaico de continentes y oc6anos. Ciencia Hoy, Buenos Aires, 6, 24-29. - - , DALLMEYER,R. D. & VUJOVICH,G. I. 1998. Time constraints in the Early Palaeozoic docking of the Precordillera, central Argentina. This volume. RAMOS, V. A., JORDAN, T. E., ALLMENDINGER,R. W., MPODOZIS, C., KAY, S. M., CORTES, J. M. & PALMA, M. A. 1986. Paleozoic terranes of the Central Argentine Chilean Andes. Tectonics, 5, 855-880. , VUJOVICH,G. I. & DALMEYER,D. 1996. Klippes y ventanas tect6nicas de la Sierra de Pie de Palo: sus relaciones con la colisi6n de Precordillera. XIII Congreso Geol@ico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, V, 377-391. RAPALINI, A. E. & ASTINI, R. A. 1998. Paleomagnetic confirmation of the Laurentian origin of the Argentine Precordillera. Earth Planetary and Science Letters, 155, 1-14. RAPELA, C. W., PANKHURST, R. J., CASQUET, C., BALDO, E., SAAVEDRA, J., GALINDO, C. & FANNING, C. M. 1998. The Pampean Orogeny of the southern proto-Andes: Cambrian continental collision in the Sierras de C6rdoba. This volume.

ASTINI RAYMOND, D. E. 1991. New subsurface information on Paleozoic stratigraphy of the Alabama fold and thrust belt and the Black Warrior basin. Alabama Geological Survey Bulletin, 143, 185. RONA, P. A. 1982. Evaporites at passive margins. In: SCUTTON, R. A. (ed.) Dynamics of passive margins. American Geophysical Union and Geological Society of America, Geodynamics Series, 6, 16-132. ROSENDAHL, B. R. 1987. Architecture of continental rifts with special reference to East Africa. Annual Review Earth and Planetary Sciences, 15, 445 503. SANTANTONIO, M. 1994. Pelagic carbonate platforms in the geologic record: their classification, and sedimentary and paleotectonic evolution. American Association of Petroleum Geologists Bulletin, 78, 122-141. SCHLAGER, W. 1989. Drowning unconformities on carbonate platforms. In: CREVELLO, P. D., WILSON, J. L., SARG, J. F. & READ, J. F. (eds) Controls on carbonate platform and basin development. Society of Economic Paleontologists and Mineralogists Special Publications, 44, 15-25. SCRUTTON, R. A. 1982. Passive continental margins: a review of observations and mechanisms. In: SCRUTTON, R. A. (ed.) Dynamics of passive margins. American Geophysical Union, Geodynamics Series, 6, 5-11. SHIN, E. A. 1983. Tidal flat environments. In: SCHOLLE, P. A., BEBOUT, D. G. & MOORE, C. H. (eds) Carbonate depositional environments. American Association of Petroleum Geologists, Memoirs, 33, 173-210. SIMS, J. P., IRELAND,Z. R., CAMACHO,A., LYONS, P., PIETERS, P. E., SKIRROW, R. G., STUART-SMITH, P. G. & MIRO, R. 1998. U-Pb, Th-Pb and Ar-Ar geochronology from the southern Sierras Pampeanas, Argentina: implications for the Palaeozoic tectonic evolution of the western Gondwana margin. This volume. SLOSS, L. L. 1963. Sequences in the intracratonic interior of North America. Geological Society of America Bulletin, 74, 93-114. THOMAS, W. A. 1977. Evolution of AppalachianOuachita salients and recesses from reentrants and promontories in the continental margin. American Journal of Science, 277, 1233-1278. 1991. The Appalachian-Ouachita rifted margin of southeastern North America. Geological Society of America Bulletin, 103, 415-431. 1993. Low-angle detachment geometry of the Late Precambrian-Cambrian Appalachian-Ouachita rifted margin of southeastern North America. Geology, 21, 921-924. --in press. Neoproterozoic-Paleozoic geography and tectonics: Review, hypothesis, environmental speculation: Discussion. Geological Society of America, Bulletin. & ASTINI, R. A. 1996a. The Argentine Precordillera: a traveler from the Ouachita embayment of North American Laurentia. Science, 273, 752-757.

THE LAURENTIAN PRECORDILLERA TERRANE & 1996b. Asymmetric conjugate pair rift margins of the Ouachita rift and the Argentine Precordillera. Geological Society of America, Abstracts with Programs, 28, 66. & in press. Simple-shear conjugate rift margins of the Argentine Precordillera and the Ouachita embayment of Laurentia. Geological Society of America Bulletin. TORSVIK, T. H., TAIT, J., MORALEV, V. M., MCKERROW, W. S., STURT, B. A. & ROBERTS, D. 1995. Ordovician paleogeography of Siberia and adjacent continents. Journal of the Geological Society, London, 152, 279-287. --, SMETHURST, M. A., MEERT, J. G., VAN DER VOO, R., MCK~RROW, W. S., BRASIER, M. D., STURT, B. A. & WALDERHAUG, H. J. 1996. Continental break-up and collision in the Neoproterozoic and Palaeozoic - a tale of Baltica and Laurentia. Earth-Science Reviews, 40, 229-258. TUCKER, R. D. & MCKERROW, W. S. 1995. Early Paleozoic chronology: a review in light of new U-Pb zircon ages from Newfoundland and Britain. Canadian Journal of Earth Sciences, 32, 368-379. VACCARI, N. E. 1994. Las faunas de trilobites de las sucesiones carbondticas del Cdmbrico y Ordov&ico temprano de la Precordillera Septentrional, Rep~blica Argentina. PhD thesis, Universidad Nacional de C6rdoba, C6rdoba. - - 1 9 9 5 . Early Ordovician trilobite biogeography of Precordillera and Famatina, western Argentina: preliminary results. In: COOPER, J. D., DROSER, M. L. & FINNEY, S. C. (eds) Ordovician Odyssey. Society of Economic Paleontologists and Mineralogists Pacific Section, Book 77, 193-196.

33

& BORDONARO, O. 1993. Trilobites en los olistolitos cfimbricos de la Formaci6n Los Sombreros (Ordovicico), Precordillera de San Juan, Argentina. Ameghiniana, 30, 383-393. VIELE, G. W. & THOMAS, W. A. 1989. Tectonic synthesis of the Ouachita orogenic belt. In: HATCHER, R. D. JR., THOMAS, W. A. & VIELE, G. W. (eds) The Applachian-Ouachita orogen in the United States. Geological Society of America, The Geology of North America, F-2, 695-728. VINK, G. E., MORGAN, W. J. & ZHAO, Wu-LING, 1984. Preferential rifting of continents: A source of displaced terranes. Journal of Geophysical Research, 89, 10 072-10 076. voN GOSEN, W. 1992. Structural evolution of the Argentine Precordillera: the Rio San Juan section. Journal of Structural Geology, 14, 643-667. - - , BUGGISCH, W. & LENHERT, O. 1995. Evolution of the Early Paleozoic m61ange at the eastern margin of the Argentine Precordillera. Journal of South American Earth Science, 8, 405-424. VUJOVICH, G. I. 1995. Geologia del basamento igneometam6rfico de la Loma de Las Chacras, Sierra de la Huerta, San Juan. Revista de la Asociaci6n Geol6gica Argentina, 49, 321-336 WARREN, J. K. 1989. Evaporite sedimentolgy. Prentice Hall, Englewood Cliffs. WITZKE, B. J. 1990. Palaeoclimatic constraints for Palaeozoic palaeolatitudes of Laurentia and Euramerica. In: MCKERROW, W. S. & SCOTESE, C. R. (eds) Palaeozoic Palaeogeography and Biogeography. Geological Society of London, Memoirs, 12, 57-73. ZAPATA, T. R. 1996. Crustal evolution of the Precordillera thrust belt- Bermejo basin, Argentina. PhD thesis, Cornell University, Ithaca.

The stratigraphical reeord of the Argentine Preeordillera and its plate-tectonic background MARTIN

KELLER,

WERNER

BUGGISCH

& OLIVER

LEHNERT

Institut ffir Geologie und Mineralogie, Universitdit Erlangen, Schlossgarten 5, D-91054 Erlangen, G e r m a n y Abstract: The stratigraphical record of the Argentine Precordillera from Early Cambrian to

Late Devonian times reveals its plate-tectonic history from incipient rifting and the evolution of a marginal platform to the separation from Laurentia, its drift in higher latitudes and amalgamation with Gondwana. This pre-Carboniferous succession can be subdivided into four supersequences bounded by major unconformities and their main features are discussed with respect to plate-tectonic implications. The basal two supersequences which include the carbonate platform deposits are subdivided into 13 third-order sequences, each with a duration of 2-10 Ma. Supersequence A reflects intracratonic rifting, creating a graben system and forming a marginal plateau. In its in higher part a progradational carbonate complex covered the entire platform. Supersequence B shows the development of an aggradational carbonate succession and the evolution of reef ecosystems comparable to those that developed around the margins of the Ouachita embayment. It also shows the demise of the carbonate platform by drowning. Deposits of Supersequence C reflect crustal extension and rifting, which led to the final separation of the Precordillera from mainland Laurentia. Supersequence D reveals the approach, and probably the accretion, of the Argentine Precordillera to Gondwana.

The Argentine Precordillera (AP) is a morphostructural entity between the main Andean Cordillera to the west and the Sierras Pampeanas to the east (Fig. 1). It differs from the surrounding geological provinces in exposing a thick Cambro-Ordovician carbonate platform succession. Within this succession, there is a Cambrian trilobite fauna typical of the Early Palaeozoic margins of Laurentia (Borrello 1971; Vaccari 1995). Recent investigations have shown that not only the trilobite fauna is similar to that of Laurentia, but also the Late Cambrian and Early Ordovician conodont faunas (Lehnert 1995). In addition, there is a close similarity between reef ecosystems in the AP and those around the Ouachita embayment along the southern margin of Laurentia (Keller & Flfigel 1996; Keller 1997). The uniqueness of the carbonate succession, together with the palaeontological data and isotope data from the presumed basement of the AP, led to speculation that the AP was a terrane exotic to South America, derived from Laurentia (Bond et al. 1984; Keppie 1991; Dalla Salda et al. 1992a, b, 1993; Astini et al. 1995; 1996, Dalziel et al. 1994; Dalziel 1997). Although there are models which try to interpret the origin of the AP as being autochthonous (Gonzfilez Bonorino & Gonzf, lez Bonorino 1991) or parautochthonous (Baldis et al. 1989; Loske 1992), there has been increasing agreement that it is

indeed a Laurentia-derived terrane (see Dalziel 1997 and Dalziel et al. 1996 for discussion). However, models explaining the provenance, the timing of the possible transfer, and the accretion of the AP to Gondwana are highly controversial (Dalla Salda et al. 1992a; Dalziel et al. 1994; Astini et al. 1995, 1996; Keller & Dickerson 1996; Keller 1997). Although many authors believe in Mid-Ordovician accretion, we take the Carboniferous deposits of the Paganzo basin, which unconformably rest on the AP and adjacent terranes, as the first unequivocal sign that the AP had finally been accreted to Gondwana. Based on this minimum age of accretion, we will discuss the pre-Carboniferous sediments of the AP and their depositional environment in order to interpret the geotectonic background in which they formed. Many of our conclusions are drawn under the assumption that the AP is indeed a Laurentia-derived terrane. We will discuss the most important lithostratigraphical units and make some reference to less important sediments, but will not attempt to give a complete stratigraphy of the AP. Similarly, we will only briefly mention the most important sequence-stratigraphical features of the thirdorder cycles in the carbonate platform deposits, i.e. those that help in elucidating the platetectonic history. For a more detailed sequencestratigraphical discussion of the carbonates the reader is referred to Keller (1997).

KELLER, M., BUGGISCH,W. & LEHNERT,O. 1998. The stratigraphical record of the Argentine Precordillera and its plate-tectonic background. In: PANK~tJRST, R. J. & RAPELA, C. W. (eds) The Proto-Andean Margin of Gondwana. Geological Society, London, Special Publications, 142, 35-56.

36

M. KELLER E T A L .

Fig. 1. Morpho-structural units of NW Argentina and adjacent areas and components of the Cuyania terrane (1, Eastern Sierras Pampeanas; 2, Sierra de Famatina; 3, Western Sierras Pampeanas; 4, valley of Iglesias-Calingasta-Uspallata).

Regional overview Geologically, the AP constitutes a high-level fold and thrust belt of mainly Miocene age. The sedimentary succession starts with Lower Cambrian sediments and, with several gaps, continues into the Triassic. Younger Mesozoic deposits are absent but, with the onset of crustal shortening and the corresponding uplift of the Andes, a thick succession of Tertiary and, locally, Quaternary sediments was deposited in terrestrial environments. The Cambrian to lower Middle Ordovician sediments are predominantly carbonates, whereas in younger deposits terrigenous clastic rocks prevail. The contact with the underlying basement is nowhere exposed, but there is indirect evidence for a metamorphic basement from xenoliths within Tertiary volcanic rocks (Abbruzzi et al. 1993) and from basement clasts in the Ordovician continental margin facies (Keller 1995). Isotope data from the xenoliths show values typical of the Grenville-age belts along the Appalachian margin of Laurentia. Similar

Fig. 2. Distribution of Cambro-Ordovician sediments in the Argentine Precordillera (modified from Keller & Bordonaro 1993) and localities mentioned in the text. values have been obtained recently from basement rocks in the western Sierras Pampeanas (McDonough et al. 1993; Kay et al. 1996). The recognition of Grenville-type rocks both beneath the AP and in the western Sierras Pampeanas (Sierra Pie de Palo, Fig. 1) led Ramos (1995) to propose the existence of a 'Cuyania' terrane composed of Grenvillian crust and a carbonate platform cover with Laurentian faunas. An additional fragment of the Cuyania terrane has been identified near San Rafael (Bordonaro et al. 1996; Fig. 1), where Lower Ordovician carbonates with a Laurentian fauna are exposed.

PRECORDILLERA STRATIGRAPHY Within the AP proper (Fig. 2), two fundamentally different tectono-sedimentary environments are recognized (e.g., Baldis et al. 1982; Astini 1992; Keller 1997): a carbonate platform of Cambrian to early Middle Ordovician age overlain by predominantly siliciclastic platform deposits (hereafter referred to as the platform or former platform area), and a western siliciclastic basin which developed mainly during the

37

Mid- and Late Ordovician as a response to rifting (here referred to as the western basin).

Stratigraphy The pre-Carboniferous sedimentary succession of the AP (Fig. 3) can be subdivided into several major, unconformity-bounded supersequences:

Fig. 3. Lithostratigraphy of pre-Carboniferous rocks in the AP. This chart is a compilation of published information from many different sources.

M. KELLER ET AL.

38

Supersequence A (Late Early Cambrian-Late Cambrian), Supersequence B (Early Ordovicianearliest Mid-Ordovician), Supersequence C (MidOrdovician-Late Ordovician), and Supersequence D (Silurian-Late Devonian). Supersequences A and B are composed of a total of 13 sequences (Fig. 4) spanning the Mid-Cambrian to earliest Mid-Ordovician. The duration of each sequence varies between 2 M a and 10Ma. This is within the average duration of third-order cycles (1-10Ma) as originally described by Vail et al. (1977). Some of these sequences are composed of several smaller-scale sequences, which from their size and average duration are intermediate between

I

2000m

Ill -[

~ ~ ~, -g

" ffl

O, (DI

|

,7

|

lO00m


525m)

P,g N

Strata of Supersequence A are exposed in the Guandacol area, in the Sierra Chica de Zonda, and in the Sierra de Villicum. Four formations have been recognized in the carbonate platform deposits of the AP. In addition, deep-water limestones of Middle and Late Cambrian age (Bordonaro & Banchig 1996) are preserved as olistoliths in Ordovician continental margin sediments.

Lithostratigraphy.

1500m

S

Supersequence A (Late Early Cambrian-Late Cambrian)

Cerro Totora Formation (c. 340m)

..~--._~ = ~

S

the third-order sequences and the typical fifthorder small-scale cycles with thicknesses varying between 1 m and 5 m. Consequently, these intermediate-scale cycles are regarded as fourthorder cycles.

I

] ] 0m Fig. 4. Third-order sequences of the Argentine Precordillera carbonate platform succession and their thicknesses. Note increasing cycle thickness towards the Cambrian-Ordovician boundary (sequence boundary 9) and decreasing thicknesses thereafter.

Lithostratigraphy. The La Laja Formation is the oldest carbonate platform unit preserved in the AP. It displays a varied spectrum of carbonate rocks with intercalated intervals of siliciclastic deposits. The sediments have been grouped into depositional assemblages (Keller & Bordonaro in press) or facies associations (Keller 1997). The lime mudstone-wackestone-intraclast packstone association is dominated by muddy textures containing well-preserved fossils, among which are trilobites, hyolithids, rare phosphatic brachiopods, and sponge spicules. Some of the mudstones are strongly burrowed. Intraclast packstones are graded and in places show an erosional contact with the underlying

PRECORDILLERA STRATIGRAPHY beds. The sediments were deposited below fairweather wave base in a well-aerated environment. The intraclast packstones represent tempestites, their clast content was derived from shallow shelf areas. The oncolite-wackestone association is composed of thick, in places cross-bedded, oncolite beds intercalated in wackestones with trilobites, hyolithids, and brachiopods. Erosional features at the base of the oncolites are present locally. The depositional environment of this association is similar to that of the lime mudstone-wackestoneintraclast packstone association. The oncolites are also interpreted as tempestites, although their source area was different. The original site of oncoid formation has not been found. The oolite-oncolite association is composed of oolitic and/or oncolitic packstones and grainstones. Many of the horizons show crossbedding, and locally well-developed herringbone cross-stratification has been observed. Similar sedimentary features were found in the packstone-grainstone association, but these rocks are predominantly composed of fossil hash (trilobites, hyolithids, eocrinoids, brachiopods). Detrital quartz is present is some beds. Sediments of both associations represent winnowed sand bars, barriers, and subtidal shelf sands. The depositional environment was shallow subtidal and lower intertidal where wave and current activities constantly removed the lime mud and led to the formation of herringbone cross-stratification. The siltstone association is easily recognized in the field by its yellow and brown colours which contrast with the dark grey and black colours of the limestones. Besides siltstones, silty marlstones, marlstones, eocrinoidal or oolitic grainstones are part of this association. Wave ripples and linsen and flaser bedding are the most prominent bedforms. The sandstonegrainstone association is composed of bioclastic, in places oolitic, grainstones and thin beds of quartz arenites or calcareous sandstones. Coquinas of olenellids are common. In both associations, lithology in combination with sedimentary structures indicates a shallow subtidal shelf with a strong tidal influence. However, in three sections thick quartz arenites alternating with black shales are present. Tectonic deformation has obliterated all sedimentary structures; consequently, an interpretation of this succession is difficult. Sediments of the La Laja Formation represent three depositional systems: a carbonate shoal complex with dominantly grainstones and packstones, an open marine environment with mudstones and wackestones, and a shallow shelf association influenced by terrigenous detritus.

39

Mudstones and wackestones are interpreted to represent both back-barrier, lagoon deposits and sediments which accumulated seaward of the main high energy zone.

Sequence stratigraphy. Two types of shallowing-upward successions have been observed. There are calcareous successions with a thickness between some tens of metres and more than 100m. The successions start with sediments of the lime mudstone-wackestone-intraclast packstone association. As this association represents the deepest environment present in the La Laja Formation, the basal bounding surface is always a flooding surface. Up section, bioclastic packstones and grainstones are developed. Oolites or oncolites form the top of the carbonate succession. Within the individual successions no smallscale cycles have been observed. Several of these cycles are stacked to form larger-scale sequences. Four of the latter within the La Laja Formation (Fig. 4) span approximately 10 Ma, hence they have to be regarded as third-order cycles. The most characteristic feature of these cycles are basal units composed of the siltstone association. No dramatic environmental change is indicated by the switch from carbonate production at the top of a carbonate cycle to siliciclastic deposition of the siltstone associations. In contrast, the upper boundary of the terrigenous units is always sharp and records important flooding. Consequently, the surface separating the siltstone units from the overlying carbonate intervals is interpreted as the transgressive surface (Keller 1997) which marks the beginning of the transgressive systems tract. The transgressive systems tract is relatively thin and is composed mainly of rocks with muddy textures. There is no distinct surface which might be interpreted as a maximum flooding surface. Nevertheless, the change from mud-dominated textures in the lower part of the succession to grain-supported textures in the upper part seems to reflect the transition from the transgressive systems tract to the highstand systems tract. The relatively thick highstand systems tract together with the internal architecture of the entire succession mark a catch-up system (Kendall & Schlager 1981). In such systems, shoal deposition (oolite-oncolite association in the La Laja Formation) is restricted to the late highstand. In most sequences within the La Laja Formation, sediments of the oolite-oncolite association were deposited during this interval related to sealevel stillstand or relative sea-level fall. The position of the siltstone units between late highstand deposits and the transgressive surface at the base of the subsequent carbonate interval

40

M. KELLER ET AL.

indicates that the siltstone units are related to a sea-level lowstand and that their lower boundary is the sequence boundary. No signs of subaerial exposure have been observed along these contacts, nor is there a basinward shift in facies. Consequently, the sequence boundaries are type-2 boundaries, and the siltstone intervals represent the corresponding shelfmargin systems tracts.

The Zonda and La Flecha Formations Lithostratigraphy. Both formations are composed predominantly of peritidal dolomites. Secondary dolomitization has obliterated most of the primary structures in the Zonda Formation (200-300 m). Where preserved, sedimentary structures are comparable to those of the La Flecha Formation. The sediments of the latter (400-700m) have been described in detail by Keller et al. (1989), Cafias (1995a), and Armella (1989a, b). Rocks of the mudstone-wackestone association contain a sparse fauna (trilobites, hyolithids) or are bioturbated. The sediments were deposited in shallow subtidal areas, probably under restricted conditions. A few intercalated trilobite packstones represent episodic storms. In the intraclast grainstone-oolite-thrombolite association, the intraclasts reflect erosional processes that affected the entire spectrum of rocks present in both formations. Oolite grainstones are present as continuous beds, often with well developed herringbone cross-stratification. However, they also fill erosional channels which in places cut down almost 2m into thrombolite mounds. Locally, the oolites onlap individual mounds demonstrating a close relation between mound growth and oolite deposition. The individual thrombolites either grew as isolated mounds or they form laterally linked structures several metres high and hundreds of metres wide. This association records complex sediment build-up interactions in shallow subtidal and lower intertidal areas. The stromatolite association is composed of different types of stromatolites and accompanying mudstones and grainstones. LLH-stromatolites are most abundant, but SH morphotypes have also been observed. The association formed in environments extending from the shallow subtidal to the lowermost supratidal. In the microbial laminite-breccia association, thick mudstone intervals, in places with evaporite pseudomorphs, represent storm events during which lime mud was transported from its original site of formation, mainly the shallow

subtidal, onto the supratidal flats. Mudcracks, tepee structures, and pseudomorphs after evaporites testify to prolonged episodes of subaerial exposure, desiccation, and evaporation. The association described here was mainly formed in higher intertidal and supratidal environments. Finally, there are some mudstones with large spheroidal mudlayers and abundant microscopic cracks. These have been interpreted as calcrete horizons (Keller et al. 1989) and were attributed to a terrestrial setting. Rocks of both formations were deposited in a peritidal environment, many of them under restricted, hypersaline conditions. Small-scale (fifth-order) cycles are a prominent feature in all sections, their internal architecture, however, does not permit a distinction of more proximal from more distal cycles. The only evidence of somewhat 'deeper' environments is present in the Guandacol area, where there is a higher percentage of shallow-subtidal mudstones in the cycles (Cafias 1995a).

Sequence stratigraphy. Sedimentologically, the lower part of the Zonda Formation is a continuation of the uppermost sequence of the La Laja Formation (sequence 5; Fig. 4). This is indicated by the transition from oolites with herringbone cross-stratification (uppermost part of the La Laja Formation) to inter- and supratidal dolostones of the basal Zonda Formation. These rocks are abruptly overlain by dark subtidal mudstones at the base of another major shallowing-upward succession. The upper boundary of this cycle coincides with the boundary between the Zonda and La Flecha Formations. The La Flecha Formation is composed of two similar shallowing-upward sequences (sequences 7 and 8; Fig. 4), but there is no well-defined boundary between them. Within the lower cycle the abundance of calcrete horizons increases towards the top but calcretes are absent above the presumed cycle boundary. A reversed pattern is visible in the distribution of thrombolites: they become less abundant towards the top of the cycle, are absent in the uppermost interval, but regain importance at the base of the next cycle. All three sequence boundaries within the Upper Cambrian strata either show signs of subaerial erosion, coarse detrital quartz, abundant evaporites, or concentrations of calcrete horizons just beneath the main surface. Consequently, each of these sequence boundaries has to be regarded as a type-1 sequence boundary (Keller 1997). These sequences were deposited during approximately 10Ma, which qualifies them as third-order sequences.

PRECORDILLERA STRATIGRAPHY

Middle and Upper Cambrian deep-water limestones and marlstones These rocks, which recently have been combined under the informal name of the 'La Cruz limestones' (Keller 1997), are mainly present as olistoliths within the Ordovician continental margin facies. Only the outcrop at Cerro Pelado shows these deep-water limestones in a presumably autochthonous position above peritidal dolomites of the La Flecha Formation. The limestones and marlstones are dark grey and fine grained, and thin- to medium-bedded. Shales have been observed only locally. The fauna consists of agnostid trilobites and sponge spicules. Graded beds, in places with erosional structures at the base, and slumped beds are also present. In general, the components are less than 1 mm in diameter and composed of platformderived material. The horizons described here are interpreted as distal tempestites, deposited in a background environment of deep-water limestones and marlstones. It is noteworthy that wherever these rocks are exposed as olistoliths within the continental slope facies, there are no indications of mass-flow deposits which might indicate the existence of a Mid- or Late Cambrian platform margin or slope. In contrast, we believe that the platform gradually passed into deep-water environments without a pronounced slope. Platform configuration might have been a homoclinal ramp (Read 1982, 1985; Burchette & Wright 1992). Within the most important olistolith, at the Los Tuneles section (Fig. 2), an unconformity is present separating Middle Cambrian deep-water limestones from latest Cambrian or even lowermost Ordovician slope deposits (Keller 1995). These are the oldest slope sediments hitherto known from the AP.

Evolution of Supersequence A The basal succession of supersequence A consists of redbeds alternating with evaporite layers (Cerro Totora Formation). Astini et al. (1995) interpreted this succession as a rift-related sequence indicating the separation of the AP from Laurentia. In our view, its localized presence and relatively reduced thickness, together with the absence of any indications of riftrelated volcanism or ocean floor, all point to intracontinental rifting similar to that of the coeval Birmingham graben of Laurentia (Thomas 1991). The uppermost horizons of the Cerro Totora Formation are coeval to the basal beds of

41

the La Laja Formation (El Estero Member). In this part of the succession, both formations show an alternation of sandstones, shales, and grainstones which constitutes the transition to carbonate platform sedimentation. Cafias (1988) described a hardground and an erosional unconformity at the base of the overlying car9bonates in the Guandacol area. This unconformity is matched by a type-1 sequence boundary separating the E1 Estero Member from the Soldano Member of the La Laja Formation. The rocks beneath the sequence boundary are white quartz arenites and black shales of a shallow depositional environment. Above the unconformity and above the sequence boundary, two trilobite zones seem to be absent from the sediments of the AP (Palmer pers. comm.). These data indicate that the corresponding erosional event is approximately equivalent in timing to the Hawke Bay event described from the Appalachian margin of Laurentia (Palmer & James 1980). The subsequent Cambrian history of the AP is characterized by rapid but decreasing subsidence until the Dresbachian, and a carbonate factory barely able to keep pace with subsidence (catchup system of Kendall & Schlager 1981). Only during the Late Cambrian is there a keep-up pattern in the sediments. The absence of terrigenous detritus from the Dresbachian onward indicates that the source area had vanished, most probably by onlap of the Upper Cambrian peritidal deposits. Faunal data (Benedetto et al. 1995; Vaccari 1994) indicate that during the Cambrian the AP was faunistically indistinguishable from Laurentia.

Supersequence B (Ibexian-Early Whiterockian) Supersequence B comprises the Lower Ordovician to basal Middle Ordovician carbonate platform sediments attributed to the La Silla Formation and the San Juan Formation. The most striking feature of this supersequence is the total absence of any terrigenous material other than clay.

La Silla Formation (400 m) Lithostratigraphy. The wackestone association is composed of mudstones and wackestones with a scarce and monotonous fauna. Abundant hardgrounds and bedding-parallel trace fossils indicate slow and non-continuous mud accumulation, probably under hypersaline conditions.

42

M. KELLER ET AL.

A more favourable environment is indicated by wackestones and packstones with sponges, receptaculitids, and abundant nautiloids and gastropods. In general, sediments of this association were deposited under low-energy, subtidal conditions. Rocks of the peloidal grainstone association are the most abundant in the La Silla Formation. There are frequent intercalations of intraclast grainstones or intraclast-peloidal grainstones. Most of the intraclasts are of storm origin, the clasts represent the entire spectrum of the microfacies observed in the La Silla Formation. In the oolite association, most of the sediments show abundant cross bedding. Herringbone cross stratification has been observed only locally. Intraclasts are abundant in some of the beds. Thrombolites and microbial mounds are also part of the association. Many of the mounds are onlapped by ooid sands. The presence of herringbone cross-stratification was taken as evidence of a tidal bar origin of the corresponding oolites. Most of the oolites, however, originated as marine sands deposited following major storm events. Storm reworking of semi-lithified sediment is also responsible for the abundant intraclasts found in some of the oolite beds. Microbial laminites and mudstones with abundant mudcracks, pseudomorphs after evaporites, and bird's-eye structures are the most important rocks in the microbial boundstone association. The sediments were formed in higher intertidal and supratidal environments. Intercalated flat pebble breccias are the product of desiccation and local reworking of carbonate layers. Sediments of the La Silla Formation were deposited on a vast platform dominated by shallow subtidal, often restricted environments. In addition, intertidal and rare supratidal areas were present. Although oolites are found throughout the succession, only very few of them can be attributed to shoal systems. The majority of the oolites form sheet-like deposits, typical of redistribution during major storms. Intertidal and supratidal environments are dominated by microbial boundstones and thick storm-induced mudlayers with desiccation cracks and evaporites. Cafias (1995a, b) interpreted the deposits of the La Silla Formation as representing a rimmed shelf. In this model, the oolites are margin-related accumulations. However, there are no indications of a nearby shelf break, nor are the oolites restricted to a single facies belt but are found all across the platform. In the westernmost exposures of the La Silla Formation, intertidal sediments are absent and muddy subtidal rocks prevail. This indicates a ramp like configuration of the platform during

the Lower Ibexian which gently dipped towards the west.

Sequence stratigraphy. Within the La Silla Formation three major sequences are developed (sequences 9-11; Fig. 4). They start with subtidal limestones, mainly of the wackestone association. Up section, packstones and peloidal grainstones are developed. The lower two sequences are topped by microbial laminites and rare thrombolites, whereas the uppermost sequence shows bird's-eye mudstones and thrombolites in this interval. At the base of the lowermost succession, silty dolomites, large intraclasts and extraclasts testify to exposure and erosion of underlying strata. Consequently, this boundary is a type-1 sequence boundary. The nature of the subsequent boundaries is less obvious. Erosional features are very rare and there is no obvious basinward shift in facies, hence they are interpreted to represent type-2 boundaries (Keller 1997). Based on their average duration of 3 - 4 M a , all of these sequences are thirdorder sequences. The lowermost and the uppermost successions are relatively thin and contain abundant intertidal deposits. In contrast, in the middle succession, which is relatively thick, subtidal lithologies are most abundant. This tripartite subdivision of the La Silla Formation has a close match in the coeval strata of the northern Appalachian margin. There, this interval, corresponding to the Lower Ibexian, consists of a predominantly subtidal unit sandwiched between peritidal carbonates (Knight et al. 1995). Similarly, the Chepultepec interval of the southern Appalachians shows a succession of peritidal carbonates, subtidal rocks and, again, peritidal strata (Bova & Read 1987). San Juan Formation (330 m) Lithostratigraphy. The most complex facies association is the reef and reef mound association. It consists of biohermal and biostromal accumulations of sponges, receptaculitids, early stromatoporoids, and algae. Non-reef rocks are rudstones, grainstones, wackestones, and in places bird's-eye mudstones. The rocks of this association have been recently described by Cafias & Carrera (1993), Cafias (1995a, b), Cafias & Keller (1993), Keller & Bordonaro (1993), and Keller & Flfigel (1996). The reefs are concentrated in two horizons, one near the base of the formation and one near the top. The packstone-grainstone association is a lateral equivalent of the reef and reef mound

PRECORDILLERA STRATIGRAPHY association. A few isolated biostromes have been found within this association. Packstones and grainstones contain broken and abraded fragments of a varied fauna. Intraclasts are also present, but are only locally abundant. In a few sections cross-bedded, almost monomict pelmatozoan grainstones have been observed. The wackestone-intraclast packstone association and the wackestone-oncolite association both represent the background sedimentation (wackestones and a few mudstones) on the carbonate platform during the Late Ibexian and Early Whiterockian. The sediments and their organic content point to quiet subtidal conditions still within the photic zone. The intraclast packstones and the oncolite beds are the results of storm events affecting different source areas. Tempestites are rare in the sections of the Sierra de Villicum and Sierra Chica de Zonda, but they are very abundant in the J~chal area. There, the beds originating from tempestites are thick and often show an erosional base, grading, and cross bedding. Farther towards the west beds tend to become thinner, erosional features disappear, and oncoids and intraclasts become less abundant. The nodular-wackestone association is dominated by whole-fossil wackestones and few mudstones; tempestites are absent. The association was deposited below storm wave base, but still within the photic zone as indicated by the abundant and diverse fauna. Similarly, the nodularpackstone association records a relatively deep environment; a few graded beds and shallow water biota in these beds, however, indicate some storm activity. Shale partings, thin shale beds, and the total homogeneization of many beds by bioturbation point to reduced sedimentation rates. The mudstone-wackestone-shale association is characterized by dark colours and platy, thinly bedded rocks. The fauna consists of deepwater trilobites and conodonts. The carbonates are typical deep-water, hemipelagic limestones (Wilson 1969). An additional association was described by Cafias (1995a, b) from the Guandacol area. There, rocks similar to the hemipelagic limestones described above alternate with fine- to medium-bedded, poorly sorted packstones and grainstones. The sedimentary succession is interpreted as a deep-water setting, but still within the reach of tempestites. The sediments of the San Juan Formation were deposited on a carbonate ramp dipping gently towards the west. The easternmost outcrops contain the high-energy shoal-water complex which hosts the main reef accumulations.

43

The upper reef horizon can be traced westward into slightly deeper packstones and grainstones with abundant reef-derived material. A similar pattern holds for all sedimentological units described from the San Juan Formation (Cafias 1995a, b; Keller & Flfigel 1996; Keller 1997). The eastern sections always represent a (slightly) more shallow environment than sections farther to the west. Remarkable, however, is the absence of lagoonal, intertidal, and supratidal deposits, indicating that an important part of the platform is not preserved.

Sequence stratigraphy. The San Juan Formation is composed of two major sequences. The lower one starts above the boundary separating the San Juan Formation from the La Silla Formation. Above this type-2 sequence boundary, several small-scale shallowing-upward cycles form a progradational parasequence set, interpreted as a shelf-margin systems tract. Above the overlying transgressive surface, the lower reef mound horizon formed during a rapid relative rise in sea level. Within this transgressive systems tract, several additional flooding surfaces have been identified. The accompanying deepening finally led to the deposition of a characteristic nodular limestones association deposited on the deep ramp. The most marked deepening is observed in the O. evae conodont zone and was caused by a eustatic sea-level rise of global dimensions (Fortey 1984). The top of this lower nodular limestone association is regarded as the zone of maximum flooding, which marks a change from retrogradation to aggradation and the onset of the highstand systems tract. The next sequence boundary is also well defined by the sudden change from mid-ramp deposits to shallow-water grainstone and packstones. In the La Silla section, a hiatus is presumed to exist between the two facies associations. In addition, the Tremadocian conodont Cordylodus cf. angulatus was found just above the sequence boundary, indicating considerable erosion in the more interior parts of the ramp. Again, the transgressive systems tract hosts the main reef accumulation of the upper reef interval. Several flooding surfaces are present above the reef mounds which indicate successive drowning of the carbonate platform, culminating during the Whiterockian. The drowning succession is characterized by black shales and deep-water limestones, some 10m thick. This succession is overlain by graptolitic black shales. The drowning itself is not a uniform process: drowning of the outer part of the ramp (Guandacol area) was earlier than in the other areas and was coeval with the upper sequence boundary in

44

M. KELLER E T AL.

the San Juan Formation south of Jitchal. Another exception is the Las Chacritas section, where some 60 m of spiculitic wackestones rest on the San Juan Formation and form the drowning succession. Finally, in the Las Aguaditas section a carbonate slope and basin developed above the carbonate platform (Keller et al. 1993a). The very different timing of drowning and the different successions themselves are vestiges of the breakdown of the carbonate platform under the onset of crustal extension. Evolution of supersequence B

In comparison to supersequence A, which culminated with the deposition of a laterally extensive tidal-flat complex, supersequence B documents the return to more open-marine conditions. Third-order sequences tend to become thinner towards the Whiterockian and the character changes from strongly progradational during the latest Cambrian to aggradational towards the top of the San Juan Formation. The sequence boundaries themselves also show an evolutionary trend: type-1 boundaries are restricted to the Late Cambrian, whereas type-2 boundaries characterize the Ibexian and Early Whiterockian. In the La Silla Formation, the sequences are topped by microbial laminites and inter- to supratidal facies. However, the thickness of these late highstand deposits progressively becomes thinner and in the San Juan Formation no intertidal or supratidal facies are found at the top of the sequences. This overall trend in the evolution of the sequences and the corresponding boundaries is accompanied by the change from mostly dolomitic rocks of the Late Cambrian to exclusively limestones during the Early Whiterockian. The carbonate rocks of supersequence B reflect various attempts at drowning the platform. Although the carbonate factory was prolific well into the Whiterockian, the sum of all external factors was finally too strong to permit further carbonate production. The Ibexian La Silla Formation shows a rather uniform facies development. Two important events accompanied the deposition of this unit. Near San Rafael (province of Mendoza), Grenvillian basement attributed to the Cuyania terrane (Ramos 1995) was flooded and a carbonate succession similar to the La Silla Formation and basal San Juan Formation was deposited. The cratonic character of this part of the terrane (Fig. 1) is documented by 80m of limestones which correspond to almost 400 m in the AP proper (Bordonaro et al. 1996). The other important event was the incipient evolution of a slope environment to the west of the

platform, although rocks of this depositional setting are only preserved within the Los Tuneles mega-olistolith.

General aspects of supersequences A and B During the formation of supersequences A and B (Cambrian to basal Middle Ordovician), sedimentation in the AP was controlled by the same factors as those responsible for the evolution of the Appalachian margin. These include a thermally subsiding crust, localized tectonic events (e.g., Hawke Bay event), and the increasing importance of eustasy, which exerted major control on the formation of 13 third-order sequences in the AP (Fig. 4). The corresponding qualitative sea-level curve and its correlation with the curve for the southern Appalachians (Read 1989) is shown in Fig. 5. In addition, the benthic faunas of the whole of supersequence A and most of supersequence B (Cambrian through Late Ibexian) cannot be distinguished from the faunal succession in Laurentia (Benedetto et al. 1995; Fig. 6a-c), but are totally different from those of the terranes today adjacent to the AP. The first non-Laurentian fossils are associated with sponge-algal mounds in the lower San Juan Formation (Vaccari 1995). However, the faunal record and the datasets from Laurentia and the AP are not of the same quality. In the AP, there are almost no data on benthic faunas for most of the Early Ibexian. The Laurentian-type conodont succession of the AP (Lehnert 1997; Lehnert et al. 1997) may help to fill this gap. The compilation of the palaeobiogeographical affinities of benthic macrofossil assemblages of the AP (Benedetto et al. 1995; Fig. 6a) includes data on several groups. These data are not always of uniform quality, because some groups have been recorded only from a short interval and/or are represented only by a few taxa (e.g., sponges: Late Ibexian-Early Whiterockian). In addition, there are no data on brachiopods, bryozoans, pelecypods and sponges before the Late Ibexian. Finally, the individual groups may not be of comparable palaeogeographical sensitivity.

Supersequence C (Mid- and Late Ordovician) Sediments of this supersequence were deposited in two different settings. In the western AP, the developing basin accommodated thick successions of a continental margin and adjacent basin

PRECORDILLERA STRATIGRAPHY

45

Fig. 5. Sea-level curve for the Cambro-Ordovician carbonate platform deposits and its correlation with the Appalachians. The Appalachian sea-level curve is modified from Read (1989). A and B show alternative correlation of earliest Ordovician sequence boundaries in the Argentine Precordillera. environment. In the east, this supersequence developed above the former carbonate platform.

The western basin

The most impressive sediments are those of the Los Sombreros Formation. Black graptolitic shales host various types of mass-flow deposits. Debris-flow deposits, turbidite sandstones and greywackes, mega-breccias, and conglomerates

are volumetrically abundant. However, most spectacular are giant olistoliths, which in places are more than 1 km long and more than 300 m thick. It is in this facies that the Middle and Upper Cambrian deep-water limestone olistoliths are present. Other important components of the mass-flow deposits are clasts of metamorphic basement (Banchig et al. 1990). These clasts are interpreted as having been derived locally (Banchig et al. 1990; Keller 1995). Since in some sections basement clasts are found

46

M. KELLER E T A L .

Fig. 6. (a) Changes in the composition of benthic faunas in the Argentine Precordillera on the genus level (modified from Benedetto et al. 1995). During the Cambrian, the fauna is Laurentian. With the onset of extension during the Whiterockian, the first Avalonian and Baltic elements are observed. Note that an influence of Gondwanan elements cannot be observed before Early Caradoc. In addition, from a faunistic point of view, the Caradoc is the time of maximum isolation indicated by a high percentage of endemic faunas and pandemic genera. (b) The trends in palaeobiogeographical affinities are much better visible on a sketch excluding pandemic faunal elements; (e) includes the aspect that there are no reports on other than Laurentian faunas before the Late Ibexian. Faunal composition contradicts any model which claims for a mid-Ordovician collision of the Argentine Precordillera with Gondwana.

together with clasts of the entire carbonate platform succession, their presence implies the formation of a fault scarp more than 2100m high (the cumulative thickness of the carbonate platform succession). Siliciclastic clasts that might have been derived from a rift-related cover of the basement are only present is the Los Tuneles section, where big blocks of quartzpebble conglomerates are exposed. These blocks are lithologically comparable with sediments of the Ocoee Group, rift-related deposits of the Appalachian margin (R. Hatcher pers. comm.). The scarceness of rift-related clasts in most sections indicates that the basement of the AP was mostly covered by carbonate platform sediments (Keller 1995). All mass-flow deposits are proximal, and the giant olistoliths must have been transported by rockfall. In addition, the sediments of the Los Sombreros Formation trace the former margin of the carbonate platform. However, as the siliciclastic succession is younger (latest Llanvirn-Caradoc; see discussion in Keller 1997), there is no sedimentological continuity between the former platform and the newly developed basin. The depositional environment of the Los Sombreros Formation is characterized by local steep scarps which delivered the olistoliths and many of the conglomerates, and a basin area (the continental rise) where these sediments accumulated. Farther to the west, there are several siliciclastic units deposited in a deep basin. The Portezuelo del Tontal Formation contains abundant conglomerates and turbidites (Spalletti et al. 1989) and is interpreted as a deposit of the continental margin. The Don Polo Formation is a succession of proximal to distal turbidites (Fig. 7) and includes some horizons interpreted as contourites (Astini 1991; Keller 1997). The Alcaparrosa Formation consists of graptolitic and aluminiferous black shales. In addition, it contains basaltic pillow lavas (Fig. 7) with MORB-like characteristics (Haller & Ramos 1984; Kay et al. 1984). Most of these sediments are poorly dated, but they all fall in the range from latest Llanvirn to Caradoc. Although the structural relations between these units are not always clear, the spatial distribution of the sediments exhibits an overall fining towards the west (Fig. 7), culminating with basin shales and pillow basalts. West-directed transport has been determined for the proximal deposits and a SWto S-directed transport for the more distal sediments (Spalletti et al. 1989; Astini 199l). All sedimentological data indicate that the sedimentary succession of the western basin is a rift-drift related succession marking the

PRECORDILLERA STRATIGRAPHY

47

Fig. 7. Tectono-sedimentary environments, evolution of the western basin, and sediment distribution in the Argentine Precordillera following continental break-up. Note that there are places where the rift-onset and the break-up unconformities merge. See text for discussion. (Not to scale)

formation of an Atlantic-type continental margin in the AP during the Mid- and early Late Ordovician. This is also evident from the very high sediment accumulation rates (Fig. 8).

The former platform area Above the drowning unconformity of the carbonate platform, most of the deposits are graptolitic black shales. Only locally did carbonate sedimentation continue (Las Chacritas and Las Aguaditas sections) and even in these successions there is a clear tendency towards deep-water limestones. In most sections then, there is a hiatus or an erosional unconformity above which mass-flow deposits dominate. The corresponding event affected the platform area during the earliest Llandeilo. In the Don Braulio section, three fining-upward successions were deposited (La Cantera Formation, c. 150m). The basal succession starts with massive conglomerates locally containing large slabs of the underlying black shales; up-section, turbidites are found. The upper two successions start with coarse sandstones and arkoses, which pass into shales. Another hiatus separates the La Cantera Formation from the Don Braulio Formation, which contains vestiges of the Ashgillian Gondwana glaciation (Peralta & Carter 1990; Buggisch & Astini 1993).

In the Guandacol area, a similar succession is preserved, but with a thickness of more than 1200 m (Trapiche Group). In the basal part of this succession, limestone slabs up to 70 m across are present (Astini 1991). In addition, several high-angle erosional unconformities have been observed in one of these sections (yon Gosen pers. comm.). All these features point to rapid local uplift and the formation of oversteepened fault scarps. The highly varied sediment accumulation rates above the former platform are documented in Fig. 8. A unique succession is preserved at Rio Sassito where about 40m of mainly Caradoc sediments rest on top of deeply eroded San Juan limestones. The lower part of the succession is composed of about 20 m of siliciclastic rocks, the upper 20m are temperate-water limestones. These limestones apparently formed in a horst position because the adjacent areas are characterized by deep-water siliciclastic material with mass-flow deposits. This outcrops demonstrates that a major period of erosion affected the former platform area after the deposition of the Middle Ordovician drowning succession but prior to the Caradoc. Everywhere in the former platform area, Ordovician rocks of supersequence C are separated from the overlying supersequence D by a major unconformity. In many places the basal bed of the younger succession is marked

48

M. KELLER ET AL.

Fig. 8. Plot of cumulative sediment thicknesses in different areas and settings in the Argentine Precordillera. The most marked differences between the platform area and the western basin occurred during Mid- and Late Ordovician time, caused by rapid subsidence of the newly forming basin. Inlay shows history of thermal subsidence. This plot includes only data from the carbonate platform and, during the Late Ordovician to Devonian, from the former platform area (from Gonzfilez Bonorino & Gonz/Llez Bonorino 1991).

by a chert-pebble conglomerate. The onlap of the younger succession onto the eroded platform is strongly diachronous (Astini & Maretto 1996). The overlying strata, however, show a rather uniform development.

Evolution of supersequence C Supersequence C is bounded below and above by two major unconformities. The basal unconformity marks the end of carbonate platform deposition and the onset of a depositional regime which favoured mass-flow sedimentation with highly variable thicknesses; in places, carbonates were deposited. Local unconformities

are common in all successions. The basal unconformity is approximately coeval with the formation of a continental margin and the adjacent western basin. In our interpretation, supersequence C represents major crustal extension related to rifting and the concomitant formation of horst and graben structures, the evolution of a continental rise and an adjacent basin into which MORB-like basalts were extruded. The major fault scarp along which the Los Sombreros Formation was deposited is evidence of the rifting: a scarp more than 2100m high is incompatible with a compressive regime and must be related to large-scale crustal extension. An extensional regime is corroborated by the

PRECORDILLERA STRATIGRAPHY pillow basalts and the formation of Caradoc carbonates on an isolated horst. Various studies have attributed individual sedimentary successions to crustal extension (Spalletti et al. 1989; Loske 1992; von Gosen 1992; Keller 1995). Similarly, the pillow basalts were interpreted as having formed in an extensional regime (Haller & Ramos 1984; Kay et al. 1984). Dalla Salda et al. (1992b) were the first to interpret the Middle and Upper Ordovician strata as representing rifting and in particular the separation of the AP from Laurentia. A similar argument has been used by Keller & Dickerson (1996). Increasing distance from Laurentia is also apparent in the composition of the faunas (Fig. 6). Avalonian, Baltic, and endemic faunas became increasingly abundant during MidOrdovician times. Laurentian faunas were able to settle on the disappearing terrane until the Late Ordovician. In contrast, the first Gondwanan (Mediterranean) brachiopods did not arrive before the Caradoc, which was the time of maximum faunal isolation (Fig. 6c). Falvey (1974) proposed a comprehensive model for rifting, continental break-up, and the evolution of passive continental margins. This model includes two major unconformities: a riftonset unconformity and a break-up unconformity (Fig. 7). Thermal expansion of the crust and concomitant erosion during initial crustal stretching are responsible for the rift-onset unconformity. In a subsequent stage, crustal expansion leads to the formation of horst and graben structures (rift valley stage), whereas the break-up unconformity marks the transition from rifting to drifting. Consequently, the breakup unconformity is almost coeval with the onset of formation of oceanic crust. In the AP, the major unconformity just above the drowning succession, which marks the base of supersequence C, is here interpreted as the rift-onset unconformity (Fig. 7). In the former platform area, the concomitant erosion cut down to highly varied stratigraphical levels in the San Juan Formation (Fig. 9). In the western basin, this unconformity is correlative with the basal strata of the Los Sombreros Formation, which in turn was deposited during the evolution of a continental slope and rise. The rift-valley stage is well expressed on the former platform where highly varied successions of conglomerates, mass-flow sediments, and sandstones were deposited in grabens or half-grabens. Locally, a carbonate slope developed. On corresponding horsts, successions like that at Rio Sassito were laid down. A widespread hiatus is present in the Caradoc, especially on the former platform. This hiatus might have been caused by lateral heat-

49

flow from the active volcanic centres in the basin. This heat-flow caused uplift and erosion on the platform and the resulting unconformity has to be regarded as the break-up unconformity (Fig. 7). The widespread latest Ashgillian/ Silurian chert-pebble conglomerate at the base of supersequence D marks the break-up unconformity on much of the former platform. However, there are successions (Don Braulio Formation, Empozada Formation sensu Keller 1997) where there are relatively thick Ashgillian sediments preserved just above the break-up unconformity.

Supersequence D Silurian deposits are widespread between Jfichal in the north and the Rio San Juan in the south. They are absent in the Guandacol area and seem to be absent in the Mendoza area. Two different environments are evident. Along the eastern margin of the AP (Sierra Chica de Zonda and Sierra de Villicum), a deep graben accommodated mass-flow deposits (Rinconada Formation). Clasts and huge olistoliths mirror the underlying stratigraphy and document erosion down to the San Juan limestones. Consequently, the succession of clasts and olistoliths shows an inverse stratigraphy. In addition, there are abundant clasts with lithologies not found in the platform succession (Loske 1992; yon Gosen et al. 1995), mainly acidic to intermediate magmatic rocks. The thickness of the sediments (conservative estimates are on the order of 1000m, von Gosen et al. 1995), abundant slump and slide structures, and the huge olistoliths document rapid subsidence of a basin and uplift of the adjacent shoulders. The other sedimentary environment is a vast siliciclastic platform with facies spanning the shoreface to areas well below storm wave base (Acefiolaza & Peralta 1986; Peralta 1990; Peralta & Carter 1990; Astini & Maretto 1996). Well into the Devonian, this platform had a clear N-S polarity. Sediments are thickest around Jfichal and get thinner towards the Rio San Juan. In addition, there is an increasing number of hiatuses towards the south, and the time gap represented by the hiatuses increases in the same direction (Benedetto et al. 1992; Herrera 1993; Peralta 1993). The J~chal area was obviously a depocentre, whereas the Rio San Juan area was a structural high. Furque & Caball6 (1990) described N-Strending ridges which subdivided the siliciclastic platform during the Silurian. Although this pattern is indeed visible in several transects

50

M. KELLER E T AL.

from west to east across the platform, the N-S polarity is much more pronounced. The Devonian sediments document a slow transition from siliciclastic shelf sedimentation to the deposition of turbidites. The Lower Devonian Talacasto Formation is predominantly characterized by shales in its lower part; up-section, sandstones dominate. The overlying Punta Negra Formation consists of flysch-like deposits (Gonzfilez Bonorino 1975) attributed to a fan-delta complex. The contact between the Talacasto Formation and the Punta Negra Formation is strongly diachronous (Herrera 1993), although it is not clear whether this is true diachronism or whether it reflects a major episode of erosion prior to the onset of flysch formation. The loss of section from the top down in a southward direction might favour the erosional interpretation whereas the onset of flysch deposition during the Early Devonian in the Mendoza area might be taken to indicate a northward migration of the fan deltas and hence a diachronous evolution. Within supersequence D, there is an interesting evolution in the heavy mineral spectra (Loske 1992). The Silurian (and Ordovician) sediments are dominated by a low-diversity and low-quantity association of weathering-resistant minerals, such as zircons and tourmalines. In contrast, the Devonian clastic rocks received a high-diversity and high-quantity association, with garnet, zoisite, and apatite, among others. Following Loske (1992) this change mirrors the change in the source rocks from a mature hinterland with unmetamorphosed or slightly metamorphosed rocks, to a hinterland where magmatic and metamorphic rocks were exposed. Zircon ages from the Devonian sediments (1.1 Ga, Loske 1995) show that this magmatic/ metamorphic hinterland might have been the basement of the AP itself (the Cuyania terrane).

during the Wenlockian, the AP becomes faunistically indistinguishable from Gondwana (see Benedetto this volume). The Devonian sediments indicate a change from platform sedimentation to the evolution of a flysch trough which received detritus from a rising metamorphic or magmatic hinterland. Palaeocurrent data (Gonzfilez Bonorino 1975) suggest northeastern and southeastern source areas which fed two important fan deltas. Supersequence D also saw the onset of deformation in the AP, marked by a thermal event which affected the western margin of the AP between 425 Ma and 410 Ma (Buggisch et al. 1994). Conodont CAI data (Keller et al. 1993b; Fig. 9) show an increase from 150~ at the eastern margin of the AP to more than 300~ in the continental rise deposits. The concomitant compressional deformation caused west-vergent structures in strata as young as the Devonian Talacasto Formation (von Gosen 1995). Westdirected thrusts and imbrications are also present at the eastern margin of the AP (von Gosen 1992; von Gosen et al. 1995). These structures, which are also visible in seismic lines (Cominguez & Ramos 1990), have a post-Silurian but pre-Late Carboniferous age.

Evolution of supersequence D

Loske (1992) and von Gosen et al. (1995) attributed the sediments of the Rinconada Formation to a depocentre formed by crustal extension. During the Silurian, this extension obviously also affected the remainder of the platform area. The basins and ridges described by Furque & Caball6 (1990) might reflect the formation of tilt blocks striking N-S. The asymmetry between the Jfichal area and the structural high along the Rio San Juan, however, documents an overall tilting of the shelf in a N-S direction. Faunistically, the mid- to Late Silurian marks the change towards entirely Gondwanan elements. With the advent of the Clarkeia fauna

Fig. 9. 'Metamorphism' and level of erosion on the carbonate platform due to block faulting. CAI, conodont colour alteration index. Thermal alteration is weak at the eastern border of the Argentine Precordillera and increases towards the west (locations as in Fig. 2).

PRECORDILLERA STRATIGRAPHY

The plate-tectonic background of sedimentation in the Argentine PrecordUlera It has become widely accepted that the AP is a Laurentia-derived terrane (Keppie 1991; Astini et al. 1995; Dalziel & Dalla Salda 1996; Dalziel et al. 1996; Keller & Dickerson 1996; Thomas 1996) and that it originated in the Ouachita embayment as originally proposed by Dalla Salda et al. (1992a, b). However, the timing of separation of the AP and the plate-tectonic scenario involved are highly controversial (Dalla Salda 1992a, b; Dalziel et al. 1994; Astini et al. 1995; Keller & Dickerson 1996; Dalziel 1997; Keller 1997). The rift-drift transition along the Appalachian margin is supposed to have taken place around the Precambrian-Cambrian boundary (Bond et al. 1984, 1989; Read 1989; Thomas 1996). In contrast, rifting along the Ouachita margin is younger (Thomas 1996) and a passive margin was not established before the MidCambrian. Intracontinental graben systems indicative of incipient rifting are an important feature along the Appalachian-Ouchita margin (Thomas 1991). Typical sediments are finegrained continental redbeds and local evaporites. Sedimentological evidence from the AP records the following steps. (1) Crustal thinning during the Early Cambrian, indicated by the redbeds and evaporites of the Cerro Totora Formation. From the late B o n n i a - O l e n e l l u s chron onwards, carbonate sedimentation prevailed. (2) The occurrence of a widespread hiatus in the AP possibly equivalent to the Hawke Bay regression event and hence confirming its position on Laurentian crust at that time. (3) During the Glossopleura chron of the Mid-Cambrian, a fully developed carbonate factory became established. We interpret the Early/Mid-Cambrian history of the AP as a record of the formation of an intracontinental rift which separated the AP to some extent from the Ouachita margin. (4) The newly formed crustal fragment existed as a Laurentian marginal plateau (Lister et al. 1991). Marginal plateaux are large, relatively unstructured crustal fragments, which owe their characteristics to a mid-crustal detachment and the concomitant pull-out of middle and lower crust from beneath the future marginal plateau (Lister et al. 1991). These plateaux are often separated from the main continent by a ramp syncline or a hanging wall basin, which may be relatively deep. In the AP, carbonate sedimentation started on the newly developed marginal plateau, whereas the deep-

51

water limestones were deposited in the intervening hanging-wall basin. (5) During the Late Dresbachian, much earlier than in other parts of Laurentia, siliciclastic input from a crystalline source area had ceased, testifying to flooding of the plateau. (6) From the Dresbachian to the Early Whiterock9 sedimentation on the carbonate platform proper was almost exclusively controlled by eustasy. The marginal-plateau model helps to explain the absence of margin-derived megabreccias and other mass-flow deposits. The gradient between the platform and the slowly subsiding basin to the west was not great enough to permit gravity-driven processes to operate on a large scale. Supersequence A records the initial intracontinental rifting, the formation of the marginal plateau accommodating the carbonate platform, and the evolution of a hanging wall basin in which deep-water limestones were deposited (Fig. 10). Sedimentology and faunal associations cannot be distinguished from mainland Laurentia. In its higher part, the supersequence documents the loss of a source for terrigenous input and the evolution of a progradational carbonate complex which covered the entire platform. The Early and earliest Mid-Ordovician saw the following events.

my 360 370

~

380

2>

390

I~

__ ~o{~

L

400

D ~

Punta Negra Flysch > i0J ~O

TaLacasto platform

xxxx•215 ~--

410 thermal overprint

,20

O,. "s

g- ~ mid

chertpebbleconglQmerate

Villicurn graben

9Q 9 I9

440

_ _

45O

_~ ,~

470

"~ _ _

4Bo

(~

500

__

~

VVV~VV ~ pillow "basalts

rift onset u ncon f

rifting

o

r

gT

~

"C"

530

(.~

m

Illl I

~ II

510

~

::

A

Hawke Bay event (?)

C

i e

I

~

I

i

i

i

~

540

Fig. 10. Tectonosedimentary evolution of the Argentine Precordillera as discussed in this paper.

i

I

52

M. KELLER E T A L .

(7) Successive return to an aggradational carbonate system which escaped drowning several times. The corresponding increase in relative sea level is probably a combination of lithospheric cooling of the marginal plateau and of eustasy, as many of the Laurentian sea-level events are recognized in the AP (Keller 1997). (8) The carbonate platform was drowned during the Whiterockian. Sea-level rise was an important factor, but facies successions in combination with sequence stratigraphy demonstrate that there was also a tectonic component reflecting crustal extension. Supersequence B as a whole reflects the demise of the carbonate platform until its complete drowning. During deposition of supersequences A and B, the history of thermal subsidence and of sediment accumulation (Figs 8 & 10) is characteristic of passive-margin evolution. The Mid- and Late Ordovician deposits mirror a climax of tectonic activity. (9) The onset of crustal extension occurred during the earliest Mid-Ordovician. (10) The climax of rifting occurred during the late Mid- and the earliest Late Ordovician. It resulted in major block movements in the former carbonate platform area and the evolution of an Atlantic-type continental margin with fault scarps locally more than 2100 m high. The pillow basalts in the western basin are interpreted as remnants of the adjacent ocean floor. (11) During the Ashgillian, after its separation from Laurentia, the AP occupied a palaeogeographical position in higher latitudes within the reach of icebergs, and hence ice-rafted sediments. According to Brenchley & Newall (1984), icebergs may have drifted as far north as 30~ during the Late Ordovician glaciation. Consequently, ice-rafted sediments alone (such as the diamictites of the Don Braulio Formation) cannot be taken as evidence the accretion of the AP to Gondwana. Faunal data and the change of the palaeogeographical affinities of the benthic associations display strikingly the timing of separation and the subsequent drift towards isolation of the AP (Fig. 6a). This is even more evident if pandemic faunal elements are excluded from the diagrams (Fig. 6b-c). Supersequence C reveals the continental break-up and the final separation of the AP from Laurentia and the complete opening of the Ouachita basin (see Dickerson & Keller this volume). The supersequence is bounded at the base by the rift-onset unconformity and at the top by the break-up unconformity (Fig. 10). The contrasting sediment accumulation rates

in the former platform area and the newly developing basin are documented in Fig. 8. The Silurian and Devonian history include the following. (12) The drift to high latitudes and the approach to Gondwana. This is shown by the Mid-Late Silurian Clarkeia fauna, because this fauna is only present in a few areas of highlatitude Gondwana (Cocks & Fortey 1990). (13) There is a marked change from siliciclastic platform deposition to flysch sedimentation during the Devonian. (14) Heavy-mineral populations in the Lower Palaeozoic strata indicate that a predominantly sedimentary hinterland was recycled prior to the (Mid-) Devonian. Subsequently, however, a source area was present which was composed of granitoids and metamorphic rocks (Loske 1992; Kury 1993; Astini & Maretto 1996). This change of source rock exposure is here interpreted as an effect of the final accretion of the AP to Gondwana. (15) The Late Silurian-Early Devonian metamorphism and the corresponding deformation in the western part of the AP indicate the onset of structural deformation (Fig. 10). They are interpreted as reflecting closure of the western ocean and collision with the Chilenia terrane (Ramos et al. 1984, 1986), whereas concomitant structures in the east may be related to the accretion of the AP to Gondwana (von Gosen 1992). This scenario is supported by the magma evolution and igneous activity in the Sierras Pampeanas indicating continuous subduction into the Silurian, changing gradually towards collision during Late Silurian-Devonian times (Rapela et al. 1992). It seems logical that the separation of the AP from Laurentia and the formation of oceanic crust must have been accompanied by subduction of the opposite side of the terrane. This in turn may have been subduction beneath the Sierras Pampeanas, causing the corresponding magmatism there (see Saavedra et al. this volume and Pankhurst et al. this volume). (16) The accretion of the AP to Gondwana was terminated prior to the deposition of the Late Carboniferous molasse sediments. Supersequence D is an expression of the final approach and the subsequent accretion of the AP to Gondwana, but also of the approach of Chilenia in the west. The Early Palaeozoic history of the AP resulted from the complex plate interactions between Laurentia, western Gondwana, and the Chilenia terrane (Ramos et al. 1984, 1986; Astini et al. 1996). We are convinced with some certainty that sedimentological evidence and the faunal data from the AP exclude models

PRECORDILLERA STRATIGRAPHY favouring M i d - O r d o v i c i a n collision of the A P terrane (Astini et al. 1995), of the C u y a n i a terrane ( R a m o s 1995), or of the Occidentalia terrane (Dalla Salda et al. 1992a, b) with the eastern Sierras P a m p e a n a s . A scenario as presented by Astini et al. (1995, 1996) and especially its timing is also unlikely. Based on our sedimentological and faunistic data a n d considering the structural evidence (von G o s e n 1992, 1995; von G o s e n et al. 1995; C o m i n g u e z & R a m o s 1990), we favour a m o d e l in which the A P was separated from L a u r e n t i a during the Ordovician and afterwards slowly a p p r o a c h e d G o n d w a n a during the Silurian and D e v o n i a n . This paper is an output of a large project of the University of Erlangen on the evolution of the AP. We thank W. von Gosen and S. Krumm for many helpful discussions and support. We also thank all our Argentinian and non-Argentinian colleagues and friends, who participated in one form or another, for their co-operation and hospitality. The support of project funds to all three of us from the Deutsche Forschungsgemeinschaft is gratefully acknowledged. This paper benefited from reviews by S. Flint (Liverpool) and an anonymous reviewer.

References ABBRUZZI, J. M., KAY, S. M. & BICKFORD, M. E. 1993. Implications for the nature of the Precordilleran basement from the geochemistry and age of Precambrian xenoliths in Miocene volcanic rocks, San Juan province. XII Congreso Geol6gico Argentino y II Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, III, 331-339. ACElqOLAZA, F. G. & PERALTA, S. H. 1986. Interpretaci6n del comportamiento de la cuenca silflrica en funci6n de sus ichnofacies en el area de Talacasto, Precordillera de San Juan. Primera Jornadas sobre Geologla de Precordillera, Actas, 151-156. ARMELLA, C. 1989a. Microfacies tromboliticas de un biociclo ideal de la Formaci6n La Flecha, Precordillera Oriental, Argentina. Correlaci6n Geo16gica, 5, 45-52. - - 1 9 8 9 b . Grupos morfol6gicos y ffibrica de trombolitos del limite Cfimbrico-Ordovicico de Argentina. Correlaci6n Geol6gica, 5, 53-66. ASTINI, R. A. 1991. Paleoambientes sedimentarios y secuencias depositacionales del Ordovicico cl6stico de la Precordillera Argentina. PhD thesis, Universidad Nacional de C6rdoba, Argentina. - - 1 9 9 2 . Tectofacies ordovicicas y evoluci6n de la cuenca eopaleozoico de la Precordillera Argentina. Estudios Geol6gicos, 48, 315-327. -& MARETTO, H. M. 1996. Amilisis estratigrfifico del Silflrico de la Precordillera Central de San Juan y consideraciones sobre la evoluci6n de la cuenca. XIII Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, I, 351-368.

53

VACCARI,N. E. 1996. Sucesibn evaporitica del Cfimbrico inferior de la Precordillera: significado geolbgico. Revista de la Asociacidn Geoldgica Argentina, 51, 97-106. , BENEDETTO,J. L. 8~; VACCARI, N. E. 1995. The early Paleozoic evolution of the Argentine Precordillera as a Laurentian rifted, drifted, and collided terrane: a geodynamic model. Geological Society of America Bulletin, 107, 253-273. --, RAMOS, A., BENEDETTO, J. L., VACCARI, N. E. 8~; CAlqAS, F. L. 1996. La Precordillera: Un terreno ex6tico a Gondwana. XIII Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, V, 293-324. BALDIS, B. A., BERESI, M., BORDONARO, O. L. & VACA, A. 1982. Sintesis evolutiva de la Precordillera Argentina. V Congreso Latinoamericano de Geologia, Actas, 4, 399-445. , PERALTA, S. H. & VILLEGAS, R. 1989. Esquematizaciones de una posible transcurrencia del terrane de Precordillera como fragmento continental procedente de areas pampeano-bonaerenses. Correlaci6n Geol6gica, 5, 81-100. BANCHIG, L. A., KELLER, M. & MILANA, J. P. 1990. Brechas calcgreas de la Formacidn Los Sombreros, Quebrada Ojos de Agua, Sierra del Tontal, San Juan. XI Congreso Geol6gico Argentino, San Juan, Actas, II, 149-152. BENEDETTO, J. L. 1998. Early Palaeozoic brachiopods and associated shelly faunas from western Gondwana: their bearing on the geodynamic history of the pre-Andean margin. This volume. - - , RACHBOEUF,R., HERRERA, Z., BRUSSA,E. D. & TORO, B. A. 1992. Brachiopodes et biostratigraphie de la Formation Los Espejos, SiluroD6vonien de la Pr6cordill~re (NW Argentine). Geobios, 25, 599-637. - - , VACCARI, N. E., CARRERA, M. G. & SANCHEZ, T. M. 1995. The evolution of faunal provincialism in the Argentine Precordillera during the Ordovician: New evidence and paleogeographic implications. In: COOPER, J. D., DROSER, M. L. & FINNEY, S. C. (eds) Ordovician Odyssey. Pacific Section, Society of Economic Paleontologists and Mineralogists, Book 77, 181-184. BOND, G. C., NICKESON, A., & KOMINZ, M. A. 1984. Breakup of a supercontinent between 625 Ma and 555Ma: New evidence and implications for continental histories. Earth and Planetary Science Letters, 70, 325-345. - - , KOMINZ, M. A. & STECKLER, M. S. 1989. Role of thermal subsidence, flexure, and eustasy in the evolution of Early Paleozoic passive-margin carbonate platforms. In: CREVELLO, D., WILSON, J. L., SARG, J. F. & READ, J. F. (eds) Controls on carbonate platform and basin development. Society of Economic Paleontologists and Mineralogists Special Publications, 44, 39-61. BORDONARO, O. L. & BANCHIG, A. L. 1996. Estratigrafia de los olistolitos Cfimbricos de la Precordillera Argentina. XIII Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, V, 471-479.

54

M. K E L L E R ET AL.

, KELLER, M. & LEHNERT, O. 1996. E10rdovicico de Ponon Trehue en la Provincia de Mendoza (Argentina) redefiniciones estratigrfificas. XIII Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, I, 541-550. BORRELLO, A. 1971. The Cambrian of South America. In: HOLLAND, C. H. (ed.) Cambrian of the New World. Wiley Interscience, New York, 385-438. BOVA, J. A. & READ, J. F. 1987. Incipiently drowned facies within a cyclic peritidal ramp sequence, Early Ordovician Chepultepec interval, Virginia Appalachians. Geological Society of America Bulletin, 98, 714-727. BRENCHLEY, J. & NEWALL, G. 1984. Late Ordovician environmental changes and their effect on faunas. In: BRUTON, D. L. (ed.) Aspects of the Ordovician System. Universitetsforlaget Oslo, 295, 65-79. BUGGISCH, W. & ASTINh R. A. 1993. The Late Ordovician Ice age - new evidence from the Argentine Precordillera. In: FINDLAY, R. H. (ed.) Gondwana Eight, 439-447. , GOSEN, W. VON, HENJES-KUNST, F. & KRUMM, S. 1994. The age of early Paleozoic deformation and metamorphism in the Argentine Precordillera - Evidence from K-Ar Data. Zentralblattfiir Geologie und Paldontologie, I, 275-286. BURCHETTE, T. P. & WRIGHT, 1992. Carbonate ramp depositional systems. Sedimentary Geology, 79, 3-57. CAlqAS,F. L. 1988. Facies perimareales del C~imbrico inferior en el firea de Guandacol. 2. Reuni6n de Sedimentologla Argentina, Actas, 46-50. - - 1 9 9 5 a . Estratigrafia y evoluci6n paleoambiental de las sucesiones carbondticas del Cdmbrico tardio y Ordovlcico temprano de la Precordillera Septentrional, Rephblica Argentina. PhD thesis, Universidad Nacional de C6rdoba, Argentina. - - 1 9 9 5 b . Early Ordovician carbonate platform facies of the Argentine Precordillera: Restricted shelf to open platform evolution. In: COOPER, J. D., DROSER, M. L. & FINNEY, S. C. (eds) Ordovician Odyssey. Pacific section Society of Economic Paleontologists and Mineralogists, Book 77, 221-224. & CARRERA, M. G. 1993. Early Ordovician Microbial-Sponge-Receptaculitid Bioherms of the Precordillera, Western Argentina. Facies, 29, 169-178. & KELLER, M. 1993. "Reefs" y "Reef Mounds" en la Formaci6n San Juan (Precordillera Sanjuanina, Argentina. Los arrecifes m~s antiguos de Sudam6rica. Boletln de la Real Sociedad de Historla Natural (Geologia), 88, 127-136. COCKS, L. R. M. & FORTEY, R. A. 1990. Biogeography of Ordovician and Silurian faunas. In: MCKERROW, W. S. & SCOTESE, C. R. (eds) Palaeozoic Palaeogeography and Biogeography. Geological Society, London, Memoirs, 12, 97-104. COMiNGUEZ, A. H. & RAMOS, V. A. 1990. Sismica de reflexi6n profundaentre Precordillera y Sierras Pampeanas. XI Congreso Geol6gico Argentino, San Juan, Actas, If, 311-314.

-

-

DALLA SALDA, L. H., CINGOLANI,C. A. & VARELA, R. 1992a. Early Paleozoic orogenic belt of the Andes in southwestern South America: Results of Laurentia-Gondwana collision? Geology, 20, 617-621. , DALZIEL, I. W. D., CINGOLANI, C. A. & VARELA, R. 1992b. Did the Taconic Appalachians continue into southern South America? Geology, 20, 1059-1062. --, CINGOLANI, C. A. & VARELA, R. 1993. A preCarboniferous tectonic model in the evolution of southern South America. International Congress on the Carboniferous-Permian, Comptes Rendus, 12, 371-384. DALZIEL, I. W. D. 1997. Neoproterozoic-Early Paleozoic geography and tectonics: Review, hypothesis, environmental speculation. Geological Society of America Bulletin, 109, 16 42. -& DALLA SALDA, L. H. 1996. The early Paleozoic evolution of the Argentine Precordillera as a Laurentian rifted, drifted, and collided terrane: A geodynamic m o d e l - Discussion. Geological Society of America Bulletin, 108, 372-373. & GAHAGAN, L. M. 1994. Paleozoic Laurentia-Gondwana interaction and the origin of the Appalachian-Andean mountain system. Geological Society of America Bulletin, 106, 243-252. - - , CINGOLAN1, C. A. & PALMER, A. R. 1996. ~l'he Argentine Precordillera: A Laurentian Terrane? Penrose Conference Report. GSA Today, 16-18. DICKERSON, P. W. & KELLER, M. 1998. The Argentine Precordillera: its odyssey from the Laurentian Ouachita margin towards the Sierras Pampeanas of Gondwana. This volume. FALVEY, D. A. 1974. The Development of Continental Margins in Plate Tectonic Theory. Australian Petroleum Exploration Association Journal, 14, 95-106. FORTEY, R. A. 1984. Global earlier Ordovician transgressions and regressions and their biological implications. In: BRUTON, D . L. (ed.) Aspects of the Ordovician System. Universitetsforlaget Oslo, 295, 37-50. FURQUE, G. & CABALLt~, M. F. 1990. Dep6sitos marinos del Paleozoico Medio en la Precordillera Central de San Juan, Argentina. XI Congreso Geol6gico Argentino, San Juan, Actas, II, 81-84. GONZALEZ BONORINO, G. 1975. Sedimentologia de la Formacidn Punta Negra y algunas consideraciones sobre la geologia regional de la Precordillera de San Juan y Mendoza. Revista de la Asociacidn Geol6gica Argentina, 30, 223-246. & GONZALEZ BONORINO, F. 1991. Precordillera de Cuyo y Cordillera Frontal en el Paleozoico temprano: terrenos bajo sospecha de ser aut6ctonos. Revista Geol6gica de Chile, 18, 97-107. HALLER, M. J. & RAMOS, V. A. 1984. Las ofiolitas famatinianas (eopaleoz6ico) de la provincia de San Juan. IX Congreso Geol6gico Argentino, Buenos Aires, Actas, II, 66-83.

-

-

PRECORDILLERA HERRERA, Z. A. 1993. Nuevas precisiones sobre la edad de la Formaci6n Talacasto (Precordiltera Argentina) en base a su fauna de braqui6podos. XII Congreso Geol6gico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, II, 289-295. KAY, S. M., ORRELL, S. 8,~ ABBRUZZI, J. M. 1996. Zircon and whole rock Nd-Pb isotopic evidence for a Grenville age and Laurentian origin for the basement of the Precordilleran terrane in Argentina. Geological Society of America South-Central section, Abstracts with program, 28, 21-22. , RAMOS, A. & KAY, R. 1984. Elementos mayoritarios y trazas de las vulcanitas ordovicicas de la Precordillera occidental: basaltos de rift ocefinico temprano(?) proximos al margen continental. IX Congreso Geol6gico Argentino, Buenos Aires, Actas, II, 48-65. KELLER, M. 1995. Continental slope deposits in the Argentine Precordillera: Sediments and geotectonic significance. In: COOPER, J. D., DROSER, M. L. & FINNEY, S. C. (eds) Ordovician Odyssey. Pacific section Society of Economic Paleontologists and Mineralogists, Book 77, 211-215. - - 1 9 9 7 . The Argentine Preeordillera- sedimentary and plate-tectonic history of a Laurentian crustal fragment in South America. Habilitation thesis, University of Erlangen, Germany. & BORDONARO, O. L. 1993. Arrecifes de estromatop6ridos en el Ordovicico inferior del Oeste Argentino y sus implicaciones paleogeogr/tficas. Revista Espaffola de Paleontologla, 8, 165-169. & in press. Cyclic carbonate platform evolution controlled by major sea-level changes: The Lower/Middle Cambrian La Laja Formation, Western Argentine Precordillera. Journal of South American Earth Sciences. & DICKERSON, W. 1996. The missing continent of L l a n o r i a - W a s it the Argentine Precordillera? XIII Congreso Geoldgico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, V, 355-367. & FL~GEL, E. 1996. Early Ordovician reefs from Argentina: Stromatoporoid vs Stromatolite origin. Facies, 34, 177-192. --, BUGGISCH, W. & BERCOWSKI, F. 1989. Facies and sedimentology of Upper Cambrian shallowing-upward cycles in the La Flecha Formation (Argentine Precordillera). Zentralblatt ffir Geologie und Paldontologie, 1, 999 1011. --, EBERLEIN, S. & LEHNERT, O. 1993a. Sedimentology of Middle Ordovician Carbonates in the Argentine Precordillera: Evidence for regional relative sea-level changes. Geologische Rundschau, 82, 362-377. - - , BUGGISCH,W. & LEHNERT, O. 1993b. The transition from diagenesis to low grade metamorphism in the Argentine Precordillera: An application of the conodont alteration index. XII Congreso Geoldgico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, I, 294-299. KENDALL, G. C. ST. C. & SCHLAGER, W. 1981. Carbonates and relative changes in sea level. Marine Geology, 44, 181-212. -

-

-

-

STRATIGRAPHY

55

KEPPIE, J. D. 1991. Avalon, an exotic AppalachianCaledonide terrane of western South American provenance. Departamento de Geologia, Universidad de Chile, Santiago, Comunicaciones, 42, 109-11. KNIGHT, I., JAMES, N. P. • WILLIAMS, H. 1995. Cambrian-Ordovician carbonate sequence. In: WILLIAMS, H. (ed.) Geology of the AppalachianCaledonian in Canada and Greenland. Geological Survey of Canada, The Geology of Canada, 6, 67-87. KURY, W. 1993. Caracteristicas composicionales de la Formaci6n Villavicencio, Dev6nico, Precordillera de Mendoza. XII Congreso Geol6gico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, I, 321 328. LEHNERT, O. 1995. Ordovizische Conodonten aus der Prdkordillere Westargentiniens: Ihre Bedeutung fiir Stratigraphie und Paliiogeographie. Erlanger Geologische Abhandlungen, 125. 1997. A Lower/Middle Ibexian conodont succession in the Early Paleozoic carbonate platform of the Argentine Precordillera. Geological Society of America, Abstracts with Programs, 29, 19-20. --, MILLER, J. F. & REPETSKI, J. E. 1997. Paleogeographical significance for the occurrence of Clavohamulus hintzei (Conodonta) MILLER in the Early Paleozoic carbonate platform of the Argentine Precordillera. Geological Society of America Bulletin, 109, 429 443. LISTER, G. S., ETHERIDGE, M. A. & SYMONDS,A. 1991. Detachment models for the formation of passive continental margins. Tectonics, 10, 1038-1064. LoSICE, W. P. 1992. Sedimentologie, Herkunft und geotektonische Entwicklung pal~iozoischer Gesteine der Pr/ikordillere West-Argentiniens. Mitnchener Geologische Hefte, 7, 1 155. 1995. 1.1 Ga old zircons in W Argentina: implications for sedimentary provenance in the Palaeozoic of Western Gondwana. Neues Jahrbuch ffir Geologie und Palgiontologie, Monatshefte 1995]1, 51-64. MCDONOUGH, M. R., RAMOS, A., ISACHSEN, C. E., BOWRING, S. A. & NUJOVICH,G. I. 1993. Nuevas edades de circones del basamento de la sierra de Pie de Palo, Sierras Pampeanas Occidentales de San Juan: sus implicancias para los modelos del supercontinente proterozoico de Rodinia. XII Congreso Geol@ico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, III, 340 342 PALMER, A. R. & JAMES, N. P. 1980. The Hawke Bay event: A circum-Iapetus regression event near the Lower to Middle Cambrian boundary. In: WONES, D. R. (ed.) Proceedings- Caledonides in the, U.S.A. Polytechnic Institute and State Memoirs, 2, 15-18. PANKHURST, R. J., RAPELA, C. W., SAAVEDRA, J., BALDO, E., DAHLQUIST, J., PASCUA, I. FANNING, C. M. 1998. The Famatinian magmatic arc in the central Sierras Pampeanas. This volume. PERALTA, S. H. 1990. Silfirico de la Precordillera de San Juan - Argentina. XI Congreso Argentino de Geologia, San Juan, Relatorio, 48 64.

56

M. K E L L E R E T AL.

- - 1 9 9 3 . Estratigrafia y consideraciones paleoambientales de los dep6sitos marino-clfisticos eopaleozoicos de la Precordillera Oriental de San Juan. XII Congreso Geol6gico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, I, 128-137. - - ~ & CARTER, C. H. 1990. La glaciaci6n Gondwanica del Ordovicico tardlo: Evidencias en fangolitas guijarrosas de la Precordillera de San Juan. XI Congreso Argentino de Geologla, San Juan, Actas, II, 181-185. RAMOS, V. A. 1995. Sudam6rica: un mosaico de continentes y oc6anos. Ciencias Hoy, 6, 24-29. , JORDAN, T. E., ALLMENDINGER, R. W., KAY, S. M., CORTES,J. M. & PALMA,M. 1984. Chilenia: un terreno albctono en la evoluci6n paleozoica de los Andes centrales. IX Congreso Geol6gico Argentino, Buenos Aires, Actas, II, 84-106. , - - , MPODOZIS, C., KAY, S. M., CORTES, J. M. & PALMA, M. 1986. Paleozoic terranes of the Central Argentine-Chilean Andes. Tectonics, 5, 855-880. RAPELA, C. W., COIRA, B., TOSELLI,A. & SAAVEDRA,J. 1992. E1 magmatismo del Paleozoico inferior en el sudoeste de Gondwana. In: GUTII~RREZ-MARCO, J. C., SAAVEDRA,J. & RABANO, I. (eds) Paleozoico Inferior de Ibero-AmOrica. Publicaciones Universidad de Extremadura, Spain, 21-68. READ, J. F. 1982. Carbonate platforms of passive (extensional) continental margins: types, characteristics and evolution. Tectonophysics, 81 195-212. - - 1 9 8 5 . Carbonate platform facies models. American Association of Petroleum Geologists Bulletin, 66, 860-878. --1989. Controls on evolution of CambrianOrdovician passive margin, U.S. Appalachians. In: CREVELLO, D., WILSON, J. L., SARG, J. F. READ, J. F. (eds) Controls on carbonate platform and basin development. Society of Economic Paleontologists and Mineralogists Special Publications, 44, 167-186. SAAVEDRA,J., TOSELLI,A., RossI, J., PELLITERO, E. 8s DURAND, F. 1998. The Early Palaeozoic magmatic record of the Famatina System: a review. This volume.

SPALLETTI, L. A., CINGOLANI, C. A., VARELA, R. & CUERDA,A. J. 1989. Sediment gravity flow deposits of an Ordovician deep-sea fan system (western Precordillera, Argentina. Sedimentary Geology, 61, 287-301. THOMAS,W. A. 1991. The Appalachian-Ouachita rifted margin of southeastern North America. Geological Society of America Bulletin, 103, 415-431. 1996. Rifting of the Argentine Precordillera from the Ouachita Embayment of the Laurentian margin. Geological Society of America SouthCentral section, Abstracts with Program, 66. VACCARI, N. E. 1994. Las faunas de trilobites de las sucesiones carbondticas del Cdmbrico y Ordovicico temprano de la Precordillera Septentrional Argentina. PhD thesis, Universidad Nacional de C6rdoba, Argentina. - - 1 9 9 5 . Early Ordovician trilobite biogeography of Precordillera and Famatina, Western Argentina: Preliminary results. In: COOPER, J. D., DROSER, M. L. & FINNEY, S. C. (eds) Ordovician Odyssey. Pacific section Society of Economic Paleontologists and Mineralogists, Book 77, 193-196. VAIL, P. R., MITCHUM, R. M. JR & THOMPSON, S. III 1977. Seismic stratigraphy and global changes of sea level, Part 4: Global Cycles of relative changes of sea level. In: PAYTON, C. E. (ed.) Seismic stratigraphy - applications to hydrocarbon exploration. American Association of Petroleum Geologists, Memoirs, 26, 83-98. VON GOSEN, W. 1992. Structural evolution of the Precordillera (Argentina). The Rio San Juan section. JournalofStructural Geology, 14, 643-667. - - 1 9 9 5 . Polyphase structural evolution of the southwestern Argentine Precordillera. Journal of South American Earth Sciences, 8, 377-404. --, BUGGISCH, W. & LEHNERT, O. 1995. Evolution of the Early Paleozoic m61ange at the eastern margin of the Argentine Precordillera. Journal of South American Earth Sciences, 8, 405-424. WILSON, J. L. 1969. Microfacies and sedimentary structures in "deeper water" lime mudstones. In: FRIEDMAN, G. M. (ed.) Depositional environments in carbonate rocks. Society of Economic Paleontologists and Mineralogists Special Publications, 14, 4-19.

Early Palaeozoic brachiopods and associated shelly faunas from western Gondwana: their bearing on the geodynamic history of the pre-Andean margin JUAN

L. B E N E D E T T O

Cdtedra de Estratigrafia y Geologia Histdriea, Facultad de Ciencias Exactas, Fisicas y Naturales, Universidad Nacional de C6rdoba, CONICET. Av. VOlez Sarsfield 299, 5000 C6rdoba, Argentina Abstract: The strong provincialism exhibited by early Ordovician to Devonian brachiopod

faunas provides an independent tool for testing palaeogeographical hypotheses. Patterns of biogeographical affinities of early Palaeozoic brachiopods from the Argentine Precordillera, the Sierra de Famatina and the Central Andean basin (NW Argentina, Bolivia, southern Peru) suggest that the pre-Andean margin was linked to northern Iapetus Ocean history. The low-grade Vendian-Early Cambrian Puncoviscana Formation and the broadly coeval rocks from the Carolina and Gander terranes may represent sedimentation in a narrow basin developed between Laurentia and Gondwana during rifting. Palaeontological evidence suggests that the Precordillera developed on the passive margin of Laurentia but moved away as an independent plate during the Ordovician, the hypothesis of the Precordillera as a Laurentian plateau does not explain faunal differences between their Late Arenig-Llanvirn brachiopods. Ordovician subduction beneath Gondwana resulted in formation of the Famatina-Puna-Avalonia volcanic-arc system. Affinities between Celtic brachiopods of this age in volcaniclastic rocks from South America (Famatina) and eastern North America (Maine, Gander, Central Newfoundland) suggest some geographical continuity between them, consistent with palaeomagnetic evidence. It is proposed that accretion of allochthonous terranes to eastern Laurentia was related to collision with the northwestern corner of South America in late Ordovician time, an idea supported by the affinities of Silurian and Devonian brachiopods from Venezuela and Colombia.

The term 'Southern Proto-Atlantic' was first used by Benedetto & Sfinchez (1979) for the Cambro-Ordovician sea-way between the N W corner of Gondwana (western Venezuela and Colombia) and southern Laurentia. This name, and the more widely accepted 'Gondwanan Iapetus' or 'Southern Iapetus' (Dalziel 1992), implicitly admitted the existence of a palaeocontinent to the west of the Andean region of Gondwana (cf. Dalmayrac et al. 1980). In searching for the counterpart of the early Palaeozoic Appalachian rifted margin, Bond et al. (1984) suggested, on the basis of the striking similarities between the Cambrian trilobites of the central Appalachians and western Argentina (Precordillera basin), that the two areas could have been juxtaposed. Similarities between the Cambrian stratigraphy and trilobite faunas of the northern Appalachians and the Argentine Precordillera, and their stratigraphical divergence since the early mid-Silurian, were also noted by Ramos et al. (1986), who interpreted the Precordillera as a displaced terrane sutured with the Pampeanas terrane before the Carboniferous.

A new palaeobiogeographical picture was proposed by Benedetto (1985) and Herrera & Benedetto (1991), using the information from the rich Arenig-Llanvirn faunas of the San Juan Limestone in the Precordillera. The peculiar combination of genera belonging to the Toquima-Table Head, Baltic and Celtic realms exhibited by these brachiopod faunas, led us to propose a model (Fig. la) in which the Andean margin of Gondwana (including the Precordillera basin) is at nearly the same longitude as the southern Appalachians. This configuration (achieved from the classic reconstructions by means of a palaeomagnetically acceptable clockwise rotation of Gondwana of c. 70 ~ explains faunal exchange between the Precordillera and the Laurentian and Baltic-Avalonian margins of the Iapetus Ocean. On the basis of nonbiological evidence, a very similar CambroOrdovician configuration was proposed by Dalziel (1991), who delineated a wander path for Laurentia clockwise around the protoAndean margin from its original hypothetical juxtaposition with the east Antarctica-Australia margin in the Neoproterozoic (Moores 1991).

BENEDETTO, J. L. 1998. Early Palaeozoic brachiopods and associated shelly faunas from western Gondwana: their bearing on the geodynamic history of the pre-Andean margin. In: PANKHURST,R. J. & RAPELA,C. W. (eds) The Proto-Andean Margin of Gondwana. Geological Society, London, Special Publications, 142, 57-83.

58

J. L. B E N E D E T T O

)

...... . .............. ..'" 9"""

"'"'" "'...

~..." ~, 9

.."t 1

*9

,,,~ (~ ~

~

~-~A

2+:;-'";','

9

I/

.....

!I

_-L

-,C ..

~'le.a

-.

"-

/

.P

55"

~ ::i:~-/ . "" ' ," TORNQUIST I LEGEND ,~ ~:.:-:;-{ 9 t ~ SEA I ":~";".-/ ." ,.-.,=:~-" . . . . . . . .r ,, I ~ A~erican ~eolm ,,. ~..TJ : ~ ~-~ :'::p'.".::~" ~': J ; kl~l'xz::~,~l~---------! Anglo-French Equ~t~ . "(. . ~ .30o ~ ' ~)o~,~--'~i~le :~s ::~ !~. I:::Z:I Province

i

-.

"

......... ~ - ~ ,

/,,,~.~I

="

i,..'~"~

: ?':':':'~

S ~ , t

~:{!i!i::~,.

" ..

";:.i:'."-.! 9

~o~,cP,ov~nce

.."L_

..~

....... "'"L-I'~::~ I ~ "'-...

,,

o

CelticProvince

,..."

.

. . . . . . .; ;'

~"

~

/

.."

"".....5 "2~--,,~-:';

.0.~ i

|176 .'.'~

;;

;"

[::)0~

"'" ..,.~.,~

.0..~ |176 /~ b

/ _o /

"'-'-a "~"

~

.'I-'."

..'I

,,.-.-.i':-"

~:

"4 ,-.K

".>'.;, ,,...

,oi:, ',

:

b',.;

?(

i/ /f

Fig. 1. Some palaeogeographical hypotheses mentioned in the text: (a) Early Ordovician reconstruction after Herrera & Benedetto (1991); (b) 'Microcontinental model' (A-B) and 'paired rift margins' model (C-D), from Benedetto (1993); (e) Early Devonian configuration after Benedetto (1984) and distribution of selected brachiopod genera.

BRACHIOPODS AND PRE-ANDEAN MARGIN In biogeographical analysis of the Early Palaeozoic, I pointed out that faunal affinities between Laurentia and the Precordillera are particularly strong in the Cambrian-earliest Ordovician, but show a clear pattern of divergence through the Ordovician (Benedetto 1993). Conversely, non-Laurentian forms (including Celtic, Baltic and Mediterranean brachiopod taxa) progressively increase in number from Late Arenig to the Ashgill. This biogeographical pattern seems to be compatible with only two palaeogeographical scenarios (Fig. lb, from Benedetto 1993), characterized by the following facts: (1) the Precordillera and Laurentia have moved apart since the Cambrian, and (2) the Precordillera represents the conjugate riftmargin to that of the southern Appalachians. In one of these models (Fig. lb, A-B), the Precordillera is interpreted as a far-travelled microplate, rifted off from the southern Appalachians margin in the early Cambrian and colliding with Gondwana by the middle Ordovician. This was the first hypothesis attempting to explain both the changes in provincialism and the transition from a tropical belt in the CambrianEarly Ordovician to a peri-glacial, high latitude zone in the Late Ordovician, suggested by the lithological succession. Finally, this model established a causal relationship between the Early Ordovician Famatinian volcanic belt and the subduction and subsequent accretion of the Precordillera. This geodynamic interpretation, summarized by Benedetto & Astini (1993) and refined and amplified by Astini et al. (1995) and Thomas & Astini (1996), is known as the 'microcontinental model', or more informally the 'funeral ship' scenario (Dalziel 1997). The other biogeographically compatible reconstruction (Fig. l b, C-D) mainly differs in that the Precordillera is interpreted as an autochthonous part of the Gondwana continent, which rifted away as a whole from the Laurentian margin. The 550 Ma configuration is quite similar to the 545 Ma geologically constrained reconstruction proposed by Dalziel (1997, fig. 15), although in the latter the Precordillera is considered as part of the Laurentian 'Texas Plateau'. In this paper Palaeozoic palaeogeographical models of the southern Iapetus ocean are tested against the palaeontological evidence accumulated from the Cambrian to Silurian successions of western Argentina and Bolivia (Precordillera, Famatina Range and Central Andean basin). The goal of this analysis is to determine whether the known fossil record forms a pattern consistent with the reconstructions based on geophysical and/or geological data. Another objective of

59

the work is to try to integrate the evolution of the Precordillera terrane to the early Palaeozoic history of the NW Argentina-Chile-Bolivia segment of the pre-Andean margin and to search for a biogeographically more consistent disposition of volcanic islands bearing Celtic Realm brachiopods.

Brachiopod dispersion and continental separations Brachiopods are benthic organisms with a planktonic larval stage. Recent articulate brachiopods possess larvae with a very short, nonpelagic stage which may last from a few hours to not more than 2 days (Scheltema 1977). Dispersal potential is greater in groups having teleplanic larvae, such as inarticulate brachiopods and several gastropods, bivalves, crustaceans, and echinoderms. In these benthic invertebrates, pelagic development is commonly of 4-6 weeks, so that with an average current velocity of 1 k m h -1, larvae may be transported 670 to 1000 km. The wide geographical distribution of many species of Palaeozoic Brachiopoda strongly suggests that they may have had longdispersal, planktotrophic larvae (Rong et al. 1995); such organisms can be dispersed over long distances along shorelines and throughout epicontinental seas. An important barrier to the colonization of new areas, however, is the variation of water temperature across the latitudinal belts (Scheltema 1977). Dispersion across ocean basins has been demonstrated for some living echinoderms, bivalves and gastropods, which may attain amphi-Atlantic distribution. Such long-dispersal capability is directly related to the duration of larval development and current velocity. However, many widespread benthic invertebrates have a very short planktonic stage or completely lack them. If this was the case for Early Palaeozoic brachiopods, their dispersion could be achieved by rafting on drift materials (e.g., pumice, pyroclastic rocks or seaweed), as has been demonstrated for many groups of marine invertebrates (Jokiel 1990) and was probably important for shallow water organisms in the Ordovician (Neuman 1984). Also important is the presence in the open ocean of 'standing points', such as volcanic islands, which encourage larvae settlement and enlargement of the area occupied by a population of organisms. The possibility of migration from one island to another depends on the sense of oceanic currents, so that understanding the distribution of benthic organisms in the past

60

J. L. BENEDETTO

requires estimation of palaeo-circulation patterns for each time interval.

Palaeobiogeographical evidence Until the early 1980s, brachiopods from the wide-spread Ordovician successions of the Precordillera basin were virtually unknown. Now, more than 60 Ordovician genera have been described, 34 of which come from the ArenigLlanvirn San Juan Limestone, while the remaining taxa are from the overlying mid-late Ordovician mixed carbonate-clastic or terrigenous successions (Las Plantas, Las Aguaditas,

Trapiche, La Cantera and Don Braulio formations; see correlation chart Fig. 2). About 20 genera are known from the early Ordovician volcaniclastic rocks of the Sierra de Famatina and 11 genera from the less-known, thick siliciclastic sequences of NW Argentina and Bolivia. This background now permits a more confident assessment of the palaeontological constraints on southern Iapetus reconstructions. Genera and species recorded from each basin and the corresponding bibliographical source are summarized in Tables 1, 2 and 3, which include some recently discovered, unpublished taxa and re-assignement of previously described forms. The geographical location of early Palaeozoic sedimentary basins and other geological features

Fig. 2. Correlation chart of main early Palaeozoic stratigraphic units of the Precordillera, Famatina and Central Andean basins. Conodont zones from Albanesi et al. (1995a), brachiopod biozones from Herrera & Benedetto (1990).

BRACHIOPODS AND PRE-ANDEAN

MARGIN

61

Table 1. Arenigian and early Llanvirn brachiopods from the Precordillera basin Taxa

Biozones A

Orthidiella sp.* Orthidium geniculatum Herrera & Benedetto Orthidium sp. Hesperonomia sp Notorthis termalis (Herrera & Benedetto)* Nanorthis ornatus Herrera & Benedetto Nanorthisfragilis Benedetto Paralenorthis vulgaris Herrera & Benedetto Paralenorthis sp. Archaeorthis sp. Oligorthis sp.* Productorthis cienagaensis Herrera & Benedetto Platystrophia minuta Benedetto & Herrera Platystrophia sp. Phragmorthis sp. Skenidioides sp.* Paurorthis ellipticus Herrera & Benedetto Pomatotrema mesocosta Benedetto Pomatotrema talacastoensis Benedetto Tritoechia azulensis Benedetto Tritoechia inaequicostata Benedetto Tritoechia sp. Acanthotoechia n.sp.* Ahtiella argentina Benedetto & Herrera Sanjuanella plicata Benedetto & Herrera Inversella (Reinversella) arancibiai Herr. & Bened. Aporthophyla sp.* Taffia anomala Benedetto & Herrera Leptella (Petroria) rugosa Wilson Leptetta (Leptella) variabilis Benedetto & Herrera Leptella (Leptella) alata Benedetto & Herrera Leptella (Leptella) costellata Benedetto & Herrera Leptella (Leptella)plana Benedetto & Herrera Huacoella radiata Benedetto & Herrera Niquivilia extensa Benedetto & Herrera Leptellina spp. Syntrophia sp.* Rugostrophia sp.* Cuparius sp.* Camerella sp.* Idiostrophia cf. I. perfecta Cooper*

H

N

M

Ah x x

x

• x

x x

x

x x x

x

x x

x

x x • x x x

x

x

x x x x x x x

• • x • • • x • •

• •

•

x x

• x x • x • • •

• x •

Biozones: A, Archaeorthis; H, Huacoella; N, Niquivilia; M, Monorthis; Ah, Ahtiella. Data from: Benedetto (1987), Benedetto & Herrera (1986, 1987a, b, 1993) and Herrera & Benedetto (1987, 1988). * Unpublished taxa.

discussed in this p a p e r are s h o w n in Fig. 3. W i t h s o m e m o d i f i c a t i o n s , the E a r l y O r d o v i c i a n biost r a t i g r a p h i c a l f r a m e w o r k (Fig. 2) is t h a t p r o p o s e d b y H e r r e r a & B e n e d e t t o (1991), w h o r e c o g n i z e d five b r a c h i o p o d a s s e m b l a g e zones c a l i b r a t e d b y m e a n s o f the a s s o c i a t e d c o n o d o n t s ( A l b a n e s i et al. 1995a; L e h n e r t 1995 a n d references therein).

P a l a e o b i o g e o g r a p h i c a l evidence discussed in this p a p e r is largely b a s e d o n the studies by H e r r e r a & B e n e d e t t o (1991), B e n e d e t t o et al. (1995), Astini et al. (1995) a n d B e n e d e t t o (1995a, b) c o n c e r n i n g the P r e c o r d i l l e r a n brac h i o p o d s . T h e b i o g e o g r a p h i c a l affinities o f N W A r g e n t i n a a n d F a m a t i n a b r a c h i o p o d s a n d associated shelly f a u n a s w e r e a n a l y s e d b y V a c c a r i

62

J. L. BENEDETTO Table 2. Arenigian brachiopods from the Famatina and Central Andean basin (CAB) Famatina Hesperonomia orientalis Benedetto Hesperonomia sp. Hesperonomiella sp.* Monorthis cf. M. menapiae (Davidson) Monorthis coloradoensis Benedettot Nanorthis spp. Paralenorthis riojanus (Levy & Nullo) Paralenorthis altiplanicus Benedetto Paralenorthis immitatrix Havli~ek & Branisa Incorthis boliviana Havli6ek & Branisa Incorthis aft. L maroccana Mergl Incorthis sp. Famatinorthis turneri Levy & Nullo Desmorthis segnis Havli6ek & Branisa Glyptorthis imbrex Havli~ek & Branisa Glyptorthis cf. G. imbrex Havli6ek and Branisa Pleurorthis ? sp. Ffynnonia sp.* Prantlina ? sp.* Skenidioides sp. Crossiskenidium sp.* Salopia ? lipanensis Benedetto Tritoechia sp. Euorthisina sp. Camerella sp.*

CAB x

• x • • • x • • • • x • • x • • x • • • x x •

x •

* Unpublished taxa. t Probably younger. Data from Levy & Nullo (1973), Havli6ek & Branisa (1980) and Benedetto (1994, in press b).

al. (1993), Benedetto (1994), Benedetto & Sfinchez (1996a) and Benedetto (in press b). A complete summary of the Precordilleran benthic faunas, including sponges, bryozoans, brachiopods, molluscs and trilobites was presented by Benedetto et al. (1995) and a more comprehensive review of provincial affinities of the Precordillera terrane and related areas was recently concluded (Benedetto et al. in press). Provincialism of Cambrian-Ordovician trilobite faunas from Argentina was reviewed by Bordonaro (1992), Vaccari et al. (1993), Waisfeld (1995), Sfinchez & Waisfeld (1995), Vaccari (1994, 1995) and Astini et al. (1995). Brachiopod biogeographical entities employed in this paper are those recognized by Williams (1973). The term 'Toquima-Table Head Realm' (Ross & Ingham 1970) is used here as equivalent to Williams's 'American Realm', the latter being used in a wider sense than that of Neuman & Harper (1992) who proposed it to replace the William's 'NW American Province'. The following discussion reviews the biogeographical links that constrain the positions and et

motions of island arcs, microplates and continents involved in the southern Iapetus palaeogeography. Trilobite and conodont data are considered, especially for the Cambrian Period, in which brachiopods are nearly absent, and in the time-intervals where the brachiopods are scarce or still poorly known. Data are reviewed separately for each geological region in ascending stratigraphical order. Following Fortey et al. (1995) the Llandeilo is considered as the upper stage of the Llanvirn Series. P r e c o r d i l l e r a basin

(1) Early to Late Cambrian trilobite faunas from the Precordillera basin and Laurentia belong to the same province (Borrello 1971). Recent studies by Vaccari (1994, 1995) demonstrate a strong affinity not only at generic level but also at specific level. This is particularly evident in the Dreisbachian faunas (Crepicephalus Zone) of the northern Precordillera, which share several species with the Appalachians (Vaccari 1994, 1995).

BRACHIOPODS AND PRE-ANDEAN MARGIN

63

Table 3. Caradocian brachiopods from the Precordillera (Precord.) and the Central Andean basin (CAB) Precord. Hesperorthis sp. Dinorthis sp. Dinorthis flabellum (Sowerby) Howellites cf. H. macrostoma (Barrande) Drabovia ? sp. Campylorthis gualcamayensis Benedetto Oanduporella alamensis Benedetto Oanduporella sp. Tissintia robusta Benedetto Tissintia canalifera Havli~ek & Branisa* Eorhipidomella cardocanalis Havli6ek & Branisa Destombesium pacochicoense Havli~ek & Branisa Drabovinella curiosa Havli~ek & Branisa Drabovinella cf. D. erratica (Davidson) Drabovinella sp. Hirnantia? sp. Bicuspina deflecta Benedetto Bicu~pina riojana Benedetto Anchoramena cristata Benedetto Anchoramena? sp. S. (Sowerbyella) cf. S. (S.) sericea (J. de C. Sowerby) Aegiromena glacialis Benedetto Aegiromena corolla Havli6ek & Branisa* Anoptambonites sp.t Oepikoides notus Benedetto Bystromena? protegula Benedetto Camerella sp. Rostricellula sp.

CAB

• • x x • x x

x x • • • x x • x •

* Considered as late Llanvirn by Havli6ek & Branisa (1980). t Unpublished. Data from Benedetto (1995a, 1997a, in press a), Su&rez Soruco & Benedetto (1996) and Havli6ek & Branisa (1980).

(2) The unique trilobite species described from the earliest Ordovician strata of the Precordillera (La Silla Formation), Plethopeltis obtusus, is restricted to the Laurentian shallowwater carbonate platform (Vaccari 1994). The overlying early Ibexian (Skullrockian) limestones contain the conodont Clavohamulus hintzei, a typical form of the North American MidContinent Province; outside Laurentia it is only recorded from Georgina basin of Queensland, eastern Australia (Lehnert et al. 1997). (3) The earliest Arenig faunas from the Precordillera (Archaeorthis Zone) include the typical Laurentian brachiopods Syntrophia, Orthidium and Nothorthis. Syntrophia is recorded from nearly coeval limestones of Arkansas, Vermont and Quebec, as well as from the allochthonous Cow Head Group of western Newfoundland (Ross & James 1987) and from the central Newfoundland volcaniclastic rocks (Neuman 1984). Outside North America, Orthidium was

recorded from the Tourmakeady Limestone of Co. Mayo, Ireland (Williams & Curry 1985), which was interpreted as a dismembered part of Laurentia, and from northeastern China in the Altai-Hinggan mobile belt (Chen& Rong 1992). The Chinese assemblage was considered by Neuman & Harper (1992) as closely related to those of the Celtic province. Nothorthis is also present in western Newfoundland, in the Fort Cassin Formation of Vermont (where it is associated with Syntrophia), in the L6vis Shale of Quebec (Cooper 1956), and in the abovementioned Tourmakeady Limestone of Ireland. The Baltic occurrence of Nothorthis in late Arenig (B. naris Zone) rocks of Estonia (Rubel 1961), is the only one known outside Laurentia. (4) Mid-Late Arenig bachiopod assemblages (Huacoella and Niquivilia zones, see Table 1) are characterized by the abundance of Leptellinae, the pentamerid Cuparius (a typical Laurentian form, but probably present in Kazahkstan), and

64

J. L BENEDETTO

Fig. 3. Location map of early Palaeozoic sedimentary basins of Central and Southern Andes and Puna volcanic belt ('Faja eruptiva de la Puna').

the first record of Platvstrophia (a genus diagnostic of the Baltic Province but widespread in beds younger than Llanvirn). Precordilleran leptellinines include Leptella (Leptella), present in Laurentia and Ireland, and the exclusively Laurentian subgenus Leptella (Petroria), which is represented in the Precordillera by P. (P.) rugosa, the same species founded in the British Columbia. Another distinctive feature is the development of an endemic stock of leptellinines represented by the genera Huacoella and Niquivilia. The species Leptella corbetti, described by Laurie (1991) from coeval rocks of Tasmania (0. evae zone), may in fact belong to Niquivilia (Benedetto & Sfinchez 1996a). Another biogeographical link with the Scoto-Appalachian faunas is Acanthotoechia, a genus recorded only from the Irish Tourmakeady Limestone. (5) The dominance of the genus Monorthis and the addition of Oligorthis and Aporthophyla to several forms recorded from the underlying zones are the most distinctive features of the latest Arenig. Monorthis is restricted to the Ogof Hen and Treiorwerth formations of Anglesey (Bates 1968), the type-locality of the Celtic Province. The species Monorthis noctilio, mentioned by Gutierrez-Marco et al. (1984) from the

Llanvirn of Spain probably belongs to Hesperonomia. The true Oligorthis has only been reported from the Arbuckle Limestone of Oklahoma (Cooper 1956), but Oligorthis? sp. was recently discovered in early Arenig limestones of the Canning Basin, Western Australia (Laurie 1997). (6) As Herrera & Benedetto (1991) pointed out, the uppermost brachiopod assemblage of the San Juan Limestone (Ahtiella Zone, early Llanvirn) is characterized by a mixture of Celtic, Baltic and North American taxa, associated with few endemic and several pandemic forms (Table 1, Fig. 4). Typical genera of the Celtic Province are Ahtiella, Platystrophia, Productorthis, Inversella (Reinversella) and Rugostrophia. All of them, with the exception of Rugostrophia, also occur in the Baltic Province. On the other hand, Pomatotrema,

Taffia, Leptella (Petroria), Camerella, Idiostrophia and the recently discovered Orthidiella (unpublished data) all are distinctive Laurentian forms. In summary, of 20 genera recorded in the early Llanvirn Precordilleran limestones, 34% belong to the Celtic-Baltic provinces and 31% are Laurentian forms (ToquimaTable Head Realm). Applying correspondence

BRACHIOPODS AND PRE-ANDEAN MARGIN

65

0 "-~.............................................................

Z

50 .................................................. 40

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

9 [-,

I........................................... i

!:!iii! iiiii!iiiii iiii

z

10

ARENIG

LLANVIRN

CARADOC I

Fig. 4. Variation of affinities of brachiopod faunas from the Precordillera terrane expressed in percentage of genera shared with different biogeographical provinces: AM, American Realm (Scoto-Appalachian Province); ANB, Anglo-Welsh and Baltic Provinces; B, Baltic Province; C, Cosmopolitan; Ce, Celtic Province; E, Endemics; EU, European Realm (includes Anglo-Welsh, Baltic and Bohemian Provinces); L, Laurentian taxa (Toquima Table-Head Realm); M, Mediterranean (Bohemian) Province.

analysis, Neuman & Harper (1992) concluded that the Precordilleran assemblage considered as a whole (that is, including Arenig-Llanvirn faunas), is closer to the Toquima-Table Head cluster than any of the Celtic faunas. (7) By the late Llanvirn (P. serra Zone) the only biogeographically significant evidence is from the early vertebrates, represented in the Precordillera basin by the agnathan Sacabambaspis janvieri, a species recorded in nearly coeval beds from Bolivia (Albanesi et al. 1995b). (8) Brachiopod assemblages of early Caradoc age include 17 genera (Table 2), of which only Campylorthis is restricted to the Scoto-Appalachian (Laurentian marginal) Province. The majority of taxa (seven genera, representing the 41%) belong to the European Reahn, showing affinities with both the Anglo-Welsh and Mediterranean provinces. A particularly interesting feature is the presence in the Precordillera basin of Aegiromena associated with Drabovia? and Tissintia, which indicates a strong link with the Berounian Aegiromena-Drabovia Fauna (Havli~ek 1989) of Bohemia and northern Africa (Benedetto 1995a; in press a; Benedetto & S/mchez 1996a). Of the remaining genera, two are endemic (Oepikoides and Anchoramena) and seven have a cosmopolitan or widespread distribution. Evidence from other

taxonomic groups, particularly molluscs and trilobites (Benedetto et al. 1995; Benedetto & S/mchez 1996a; Sfinchez in press; Baldis & P6the 1995), indicates that endemicity was higher during the Caradoc than during any other Ordovician epoch. Cladistic analysis of some trilobites from the early Caradoc Las Aguaditas Formation of Precordillera suggest affinities with Laurentian forms (Chatterton et al. 1997; Edgecombe et al. 1997). (9) Latest Ashgill (Hirnantian) brachiopod faunas are known from the glacigenic Don Braulio Formation (Benedetto 1986, 1990). The taxa recorded are Hirnantia sagittifera, Dalmanella testudinaria, Plectothyrella crassicosta, Cliftonia oxoplecioides, Paromalomena polonica and Eostropheodonta hirnantensis, all of which are characteristic of the Kosov Province (Rong & Harper 1988) developed in central-southern (Gondwanan) Europe, southern China, Kazakhstan, Baltica and Avalonia. Laurentian records of Kosov taxa are restricted to Perc6 and probably Maine.

Sierra de Famatina (1) Early Tremadoc (including probably latest Cambrian) times are represented by black shales

66

J. L. BENEDETTO

(Volcancito Formation) containing the distinctive trilobite Parabolina (Neoparabolina)frequens and other olenids. This trilobite association is similar to those of NW Argentina, assigned by Shergold (1988) to the Baltic Province. (2) The Early Arenig shales have yielded graptolites and some trilobites including pelagic forms such as the cyclopygid Degamella (Esteban 1996). Brachiopods are very abundant in the overlying volcaniclastic sequence represented by the Suri and Molles formations (Astini & Benedetto 1996). The fossiliferous beds fall within the middle Arenig (Whitlandian, B. naris Zone) (Albanesi & Vaccari 1994; O. Lehnert, pers. comm.). The Famatinian fauna is a Celticlike assemblage with weak Bohemian affinities (Benedetto 1994). Typical genera belonging to the Celtic Province are Skenidioides, Monorthis and the recently discovered Ffynnonia, which is recorded for the first time outside Anglesey, NW Wales (Neuman & Bates 1978). Also characteristic is the genus Famatinorthis, which has recently been reported from the coeval Shin Brook Formation of Maine (Neuman 1997), a classic 'Celtic' locality. Finally it is interesting to note that specimens of 'aft. Eostrophomena sp' from the Davidsville Group of New World Island, illustrated by McKerrow & Cocks (1986, fig. 2b-f), are very similar to the Famatinian Monorthis aft. menapiae (Davidson) (cf. Benedetto 1994, pl. 1, figs 1-14). Bohemian links are suggested by the genus Prantlina, probably present both in the Suri Formation and in the Prague basin. The record of Incorthis represents a biogeographical link with the typical peri-Gondwanan faunas of Morocco (Mergl 1988), NW Argentina (Benedetto in press b) and Bolivia (Havli6ek & Branisa 1980). The record of Crossiskenidium, previously considered to be endemic to Ireland (Tourmakeady Limestone), is significant because it indicates a link with this Scoto-Appalachian locality. The occurrence of clitambonitids and pentamerids in the Famatinian volcaniclastic rocks is another feature that suggests affinity with the warm-water Precordilleran faunas. It is important to note that Famatinian brachiopods are associated with trilobites of the peri-Gondwanan Neseuretus Fauna (Vaccari et al. 1993; Vaccari 1995). These assemblages include western Gondwana (Australian) taxa, such as Hungioides and Gogoella and southern European-Avalonian genera such as Merlinia and Illaenopsis, associated with at least two endemic taxa (Famatinolithus, Pliomeridius) shared with the NW Argentina-southern Bolivia basin. The affinities between the warm-water conodont faunas of the Sierra de Famatina and Australia

were noted by Albanesi & Vaccari (1994). The peri-Gondwanan relationships are also reinforced by the record in the Famatinian basin of the bivalve Redonia, known from southern Europe, Morocco and NW Argentina (Sfinchez & Babin 1994).

Central Andean basin We include here the vast Early Palaeozoic sedimentary basin developed approximately between 10~ and 23~ in the southern Central Andes, comprising outcrops in NW Argentina, N Chile, Bolivia, Paraguay (to the west of the Asunci6n Arch) and south-central Peru (Fig. 3; Benedetto et al. 1992). (1) With the exception of a few inarticulate brachiopods, Early and mid-Cambrian faunas are unknown from this basin. Benedetto (1977) presented evidence for a latest Cambrian age for trilobite faunas from the lower part of the Santa Victoria Group of NW Argentina. Trilobites recovered from these beds, the Parabolina (Neoparabolina) frequens Zone, as well as those from the earliest Tremadoc (including Plicato-

lina, Onychopyge, Beltella, Protopeltura, Shumardia and Jujuyapis) were referred by Shergold (1988) to the cold-water Baltic Province (Olenid biofacies). This province extends across the East European, Baltic and Avalonian platforms and northern Siberia; it has also been recognized in Oaxaca (Mexico), Venezuela (El Baul Massif), Bolivia and NW Argentina. The diverse late Tremadoc trilobite assemblages of NW Argentina and Bolivia show the same biogeographical signature (Harrington & Leanza 1957). Articulate brachiopods are represented by the widespread genera Apheorthis and Nanorthis and the endemic Notorthisina (Havli6ek & Branisa 1980). (2) Arenig brachiopod faunas have been recovered mainly from the Whitlandian beds (middle Arenig) of the Cordillera Oriental and Subandean Ranges in NW Argentina and Bolivia (Havli6ek & Branisa 1980; Benedetto in press b) (Table 2). Of special interest is the occurrence in these regions of Euorthisina, known from the Mytton Flags of Shropshire, England (Williams 1974), the Moroccan Anti-Atlas (Havli6ek 1971) and Bohemia (Havli6ek 1967). Another genus in common with Morocco is Incorthis, represented in the Sepulturas Formation of Argentina by a form very close to L maroccana. The Central Andean Arenig brachiopod faunas are linked to NW African assemblages which, in turn, are related to the Bohemian faunas. Based on the associated trilobites, Waisfeld (1995) also noted provincial affinities with Avalonian

BRACHIOPODS AND PRE-ANDEAN MARGIN faunas, which is consistent with the probable presence in NW Argentina of the Anglo-Welsh genus Salopia. Brachiopod assemblages from the Sierra de Famatina and the Central Andean basin have in common four genera, of which Incorthis is of particular significance because of its more restricted geographical range (northwestern and western Gondwana). (3) Little can be said about the Llanvirn faunas, due to the almost complete absence of sedimentary rocks of that age. Tissintia has been described from Bolivia, where is associated with Aegiromena (Havli~ek & Branisa 1980), and from Peril (Hughes et al. 1980). The Bolivian species T. canalifera is very similar to T. convergens from Morocco and T. prototypa from Shropshire, and indicates close affinities with Anglo-French and Mediterranean faunas. (4) Caradoc faunas are represented in the Central Andean basin by a suite of typically Mediterranean taxa including Eorhipidomella, Destombesium, Heterorthis, Drabovinella and Hirnantia? (Havli~ek 1990; Sufirez Soruco & Benedetto 1996; Benedetto 1997a). The record of Oanduporella in supposed early Caradoc beds of Bolivia suggests a connection with Precordilleran faunas. (5) No typical members of the late Ashgill Hirnantia Fauna have been described from this basin, but some specimens assigned to Schizophoria from the Cancafiiri Formation probably belong to H. sagittifera (Benedetto et al. 1992).

67

Palaeogeographical models and biogeographical constraints A number of geodynamic models have been published in the last 15 years in order to explain the evolution of the pre-Andean margin of Gondwana and its palaeogeographical relationships with other supposedly associated continents and terranes (Herrera & Benedetto 1991; Benedetto 1993; Benedetto & Astini 1993; Astini et al. 1995; Thomas & Astini 1996; Dalla Salda et al. 1992a, b; Dalziel et al. 1994; Dalziel 1994, 1997; Keller & Dickerson 1996; Ramos 1988, 1995; Ramos et al. 1984, 1986; Bahlburg 1990; Conti et al. 1996). Each of these reconstructions fall within one of four basic geodynamic models: (a) unrelated margins, (b) paired rift margins, (c) far-travelled microplate, or (d) collision, with transfer or 'exchange' of plate fragments (Fig. 5).

Unrelated margins

In this category are the classic reconstructions in which Laurentia and the pre-Andean margin of Gondwana are placed at very different longitudes in the earliest Palaeozoic (e.g., Scotese & McKerrow 1990). Faunal migration from Laurentia to the western Gondwana margin and a similar peri-equatorial latitudinal belt have been argued to explain stratigraphical and faunal similarities during the Cambrian (Baldis &

Fig. 5. Sketch of the three palaeogeographical hypotheses for the origin of the Precordillera mountain belt (black rectangle) analysed in the text.

68

J. L. BENEDETTO

Bordonaro 1982). The species-level differentiation of certain Precordilleran trilobites led Bordonaro (1989, 1992) to recognize a separate South American Province, suggesting a high degree of faunal isolation of the Precordillera at that time. Faunal evidence does not support such a reconstruction. On the contrary, there is no doubt that Laurentia and Precordillera are faunistically indistinguishable and belong to the same Cambrian palaeobiogeographical entity. In other words, by Cambrian times there is no evidence of barriers to faunal exchange separating these regions, so that they were probably geographically close.

P a i r e d rift margins

This category includes reconstructions in which western (South American) Gondwana was initially contiguous with the Laurentia plate, the posterior separation resulting from an early Palaeozoic rifting event. As a result, Precordilleran and Appalachian carbonate platforms developed as conjugate rift-margins (Bond et al. 1984; Herrera & Benedetto 1991; Benedetto 1993 model, see Fig. l b, C-D). The 'faced margins' configuration (Fig. l a; Herrera & Benedetto, 1991) represents a stage after the Neoproterozoic Pangaea configuration of Bond et al. (1984). Two presumptions characterize these scenarios: (1) the Precordillera was always a part of the Gondwana plate, which rifted away from Laurentia after the Cambrian, and (2) neither Laurentia nor the Precordillera collided with the Gondwana margin after the break-up of Pangea. Biogeographical evidence supports a direct connection only between Laurentia and the Precordilleran region of South America, not with the entire 'Pacific' margin of Gondwana such as depicted by Bond et al. (1984). Much uncertainty exists about the affinities of the early-mid-Cambrian faunas developed outside the Precordillera. The only biogeographically significant locality is in the Sierra de La Macarena, Eastern Cordillera of Colombia, where a trilobite association including the genera Asaphiscus, Ehmania and Paradoxides was recorded (Harrington & Kay 1951; Rushton 1963). The last genus is distinctive of the Acado-Baltic Province and its record in North America is restricted to the Carolina and Avalon accreted terranes (Secor et al. 1983, N e u m a n et al. 1989). The complete absence of Paradoxides in the autochthonous basins of North America makes unlikely the juxtaposition of Laurentia and the northern Andean margin of Gondwana in the mid-Cambrian.

This assumption is supported by the 'European' (Baltic Province) biogeographical signature of the latest Cambrian-Tremadoc trilobite faunas founded from NW Argentina to the E1 Bafil Massif of Venezuela, even if the palaeogeographical significance of such olenid faunas has been seriously questioned (Fortey & Cocks 1986). The pattern of Cambrian to earliest Ordovician trilobite affinities indicate that a sharp faunal boundary exists between the Precordillera and the remaining Andean basins. The record of Asaphiscus in the Lizoite Formation of the Central Andean basin mentioned by Acefiolaza & Bordonaro (1988) needs confirmation by further studies because the poor preservation of this material prevents an accurate generic identification (Vaccari pers. comm. 1996). Such a biogeographical boundary also coincides with strong contrasts in the lithological succession, the Precordillera being characterized by carbonate rocks, whereas the Famatina and Central Andean basins are entirely filled by siliciclastic rocks.

Far-travelled microplate

According to this model (Fig. lb, A-B), the Precordillera and surrounding 'Grenvillian' metamorphic rocks represent a far-travelled microplate (Precordillera, or Cuyania, terrane) rifted off from probably the southeastern margin of Laurentia and later accreted to the Gondwana margin (Benedetto 1993; Benedetto & Astini 1993; Benedetto et al. 1995; Astini et al. 1995, 1996; Thomas & Astini 1996; Ramos 1995). This reconstruction is consistent, not only with the identity of Cambrian trilobites from both regions, but with the progressive change in faunal provinciality of the Precordilleran brachiopods through the Ordovician (Fig. 4). Although the rifting phase probably started in the Early Cambrian, the Precordillera became faunistically distinguishable from Laurentia only from the earliest Arenig. At that time, and especially during the mid-late Arenig and early Llanvirn, several brachiopod genera unknown from the Laurentian carbonate platform appear. By a trans-latitudinal trajectory across the southern Iapetus, the microcontinental model provides an elegant explanation for the faunal exchange between the Cuyania terrane and the northern Iapetus volcanic islands and subduction-related volcanic arcs inhabited by Celtic faunas (Neuman & Harper 1992). The strong Baltic signature of the Precordilleran brachiopods remains ambiguous because in our reconstruction (Astini et al. 1995, fig. 14c) the Baltica

BRACHIOPODS AND PRE-ANDEAN MARGIN palaeocontinent is too far away from the CuyaDia terrane. This is also the case of the northern Iapetus volcanic islands inhabited by 'Celtic' brachiopods (Neuman 1984). Estimation of the distance at which the Precordillera was able to attain faunal distinctiveness largely depends on the inferred effectiveness of larval migration and patterns of palaeoceanic circulation within Iapetus. Assuming a current velocity of c. l k m / h - : and a planktonic larval stage of 6 weeks, the maximum distance at which 'Laurentian' larvae

69

could reach the Precordillera, or vice-versa, is c. 1000-1500km. This range is consistent with separation at a standard sea floor spreading rate during c. 50 Ma (i.e., Early Cambrian to Early Arenig). An unanswered question is why Laurentian forms such as Syntrophia, Nothorthis, Cuparius, Leptella (Petroria) and Oligorthis migrated from Laurentia to the Precordillera, but none of the Celtic, Baltic and endemic genera of the Precordillera colonized the Appalachian carbonate platform. It seems that this, as well as the strong faunal exchange between the

L

oo,, n

12 ,9 .... ,

+

[

uric

i

Early Ordovician --] (Late Arenig-Eady L!anvirn)l

9

-.

1000 km

Fig. 6. Early Ordovician schematic palaeogeographical reconstruction of the Southern Iapetus. C, Inferred original location of the Carolina terrane; EA, Eastern Avalonia; EP, Eastern Sierras Pampeanas ; F, Famatina volcanic arc; FEP, Puna volcanic belt (includes the 'Faja Eruptiva de la Puna Occidental' and 'Oriental'); GB, Guyana/Brazilian craton; KN, Kola nappe; P, Precordillera terrane (Cuyania); Punc, Puncoviscana folded belt; T, Northwestern Ireland terrane (includes Tourmakeady Limestone.); WA, Western Avalonia.

70

J . L . BENEDETTO

Precordillera terrane and the northern Iapetus islands, can only be explained by ocean currents controlling the larval dispersal (Benedetto et al. in press). The timing of Precordillera accretion to Gondwana is still poorly constrained. Originally we assumed that by the Arenig, a narrow seaway separated the Precordillera terrane from the Famatinian arc-related volcanic rocks, and that docking of the Precordillera with Gondwana occurred during the Llanvirn (Benedetto & Astini 1993). The Famatinian faunas combine taxa from the Gondwana continental shelf and Celtic and Laurentian forms from the Precordillera terrane. This two-way colonization is what might be expected if the Famatinian volcanic arc lay, as we suppose, between Gondwana and the Precordillera (Fig. 6). The first 'Gondwanan immigrant' into the Precordillera is the shallow-water agnathan Sacabambaspis, recorded in the Llandeilo, followed in the Caradoc by brachiopods of the Mediterranean (and Gondwanan) DraboviaAegiromena Fauna. At this time, Cuyania received only a few Laurentian colonizers and attained the highest degree of endemism since its separation from Laurentia. The relatively high proportion of strophomenides and the AngloBaltic signature of several taxa, together with the persistence of carbonate deposits, suggest warmer waters than those in the autochthonous Gondwanan basins (Benedetto & Sfinchez 1996b). By Caradoc times, the Precordillera was probably not yet accreted or, if it was, some kind of barrier separated it from the Central Andean basin. Palaeontological evidence for the accretion is conclusive only from the early Wenlock, when the typical Clarkeia Fauna (Afro-South American Realm) became widely distributed across the basins of southern South America (Benedetto & Sfinchez 1996b). The widespread late Ordovician Hirnantia Fauna, well documented in the Precordillera, does not necessarily indicate geographical continuity with Gondwana; but merely, because of its association with glacigenic horizons, suggests a mid- to highlatitude position at the end of the Ordovician.

CoO&ion with exchange o f plate fragments The reconstructions included here differ significantly from the preceding ones, in that the Precordillera was transferred to the pre-Andean margin through Laurentia-Gondwana collision followed by new rifting. Such a rift system was not exactly coincident with the former collisional suture, so that the Precordillera remained

attached to Gondwana in a similar way to that by which the eastern Newfoundland was transferred from Baltica to Laurentia when Iapetus closed and re-opened (the Wilson Cycle archetype). Two different models have been proposed, which basically differ in the size and location of the transferred 'allochthonous' area in relation to the Laurentia plate: (1) the 'Occidentalia' model and (2) the 'Texas Plateau' model. The first hypothesis, based in the Cambrian palaeogeographical reconstruction of Dalziel (1991), was proposed by Dalla Salda et al. (1992a, b), who interpreted the Famatinian orogen as the result of the Laurentia-Gondwana collision, with the Occidentalia terrane representing a continental sliver detached from eastern North America following the collision. According to this reconstruction, the Precordilleran carbonate belt represents the continuation in South America of the Appalachian platform. If correct, faunal exchange between the two areas must have been free and continuous at least until the late Ordovician rifting event. However, it is now clear that biogeographical divergence increased after the early Ordovician, and that by the mid-Ordovician the Precordilleran faunas became markedly different from those of the Appalachians. This and the latitudinal incompatibility of climatically-sensitive sediments (late Ordovician carbonates and evaporites in Laurentia versus glacigenic diamictites in Gondwana) leads to rejection of this reconstruction (Benedetto 1993). Recently, Dalziel (1997, p. 34) pointed out that the 'Occidentalia' model of Dalla Salda et al. (1992a, b) contradicts the faunal and palaeoenvironmental differences between Gondwana and Laurentia during the Ordovician. According to his 'Texas Plateau' hypothesis (Dalziel 1997), the Precordillera was part of a continental plateau outboard of the Ouachita embayment, whose tectonic, geodynamic and sedimentary history was analogous to the present-day 'Malvinas/Falkland Plateau'. This Laurentian promontory was originally rifted from Gondwana near the Proterozoic-Cambrian boundary, subsequently collided with the proto-Andean margin by the mid-Ordovician forming the 'Artejia' megacontinent (Laurentia plus Gondwana) and finally detached from Laurentia during a Caradoc rifting event. Although this hypothesis attempts a better integration of plate motions and faunal data than the 'Occidentalia' model, a number of biogeographical objections can be made to this scenario: (1) If, as Dalziel (1997, fig. 15) postulates, the Precordillera was a part of the 'Texas continental plateau' by the early Ordovician (475 Ma), then a nearly complete identity might

BRACHIOPODS AND PRE-ANDEAN MARGIN be expected between the 'Laurentian Plateau' faunas and those on the Laurentia plate (Great Basin, Appalachians). This is based on the fact that the benthic faunas of both the Malvinas/ Falkland plateau (the Mesozoic-Tertiary analogue) and the contiguous Patagonian shelf belong unequivocally to the Magellanic Province (Ringuelet 1954; see also Valentine 1973). However, in Early Ordovician (Canadian and Early Whiterock) times, Precordilleran brachiopods are markedly different from those of Laurentia: as stated above, only 31% are typical ToquimaTable Head genera while the other associated forms are from other realms or endemic. (2) After the mid-Ordovician collision and the subsequent closure of the 'Western Iapetus', a bridge of about 1000kin of submerged continental crust was established between Laurentia and Gondwana (Dalziel 1997, fig. 16). If correct, the progressive interchange of brachiopods along this migratory route should have produced a break-down of provinciality. In contrast, there are in reality few mid-Ordovician genera in common between the Precordillera and Laurentia and between the Precordillera and Gondwana. Likewise, in spite of the supposed relative proximity of both continents by the early Caradoc, no Gondwanan brachiopods have ever been recovered from Laurentia. In summary, from a dynamic perspective, the 'Texas Plateau' model only explains the increasing affinities between Precordilleran and Gondwanan faunas through the Ordovician, but does not account for the corresponding increase in the faunal differences between the Precordillera and Laurentia.

The Precordillera, a microplate or a Gondwanan plateau? From the faunal evidence summarized above it is clear that in Cambrian times, the Precordillera and Laurentia were geographically close. With the exception of the 'unrelated margins' hypothesis, all of the proposed reconstructions agree on this point. The fact that Laurentian and Precordilleran faunas diverge after the early Ordovician leads to the conclusion that the Precordillera gradually moved away from Laurentia as an independent continental plate and oceanic-crust floored sea-way separated both plates. It should be noted, however, that such faunal divergence is also biogeographically compatible with the 'paired rift margins' hypothesis. In the Benedetto (1993) scenario (reproduced in Fig. 1b, C), the Precordillera is represented as a Gondwanan promontory which could have

71

been genetically similar to the Laurentian 'Texas plateau' envisaged by Dalziel (1997). From the data reviewed here, the only palaeontological support for this reconstruction is the relatively high percentage of shared forms with the Precordillera: of the 13 genera identified in the Famatina basin, 7 (54%) have also been recorded from coeval beds in the Precordillera, suggesting a closer relationship between both regions than was previously accepted. Moreover, the Famatina brachiopod fauna seems intermediate between the mixed LaurentianCeltic associations of the Precordillera and the nearly pure Gondwanan ('Mediterranean') faunas of the Central Andean basin. It is clear that a faunal gradient existed from the Central Andean to the Precordillera basins. The problem is to determine if such a gradient was the result of a gradual climatic change from temperate to warm waters along a nearly continuous continental margin across palaeo-latitudes (a geographical and climatic cline) or, in contrast, caused by other factors such as oceanic barriers and/or a greater geographical separation than that seen today. Because of its analogy with the Ordovician basins, a better example of climatically controlled facies and faunas is that of the autochthonous Permian basins of Gondwana. Warm-water carbonate facies are widely distributed from Venezuela to southern Bolivia (Titicaca Group), approximately between the Permian palaeo-equator and 30~ These Central Andean successions have yielded highly diverse, warm water brachiopod faunas belonging to the Texas Province of the Tethyan Realm (Shi & Waterhouse 1991). Conversely, coeval rocks of the Precordillera basin (Paganzo Group) are nearly exclusively clastic and contain temperate to cold-water brachiopods of the Gondwanan Realm. Moreover, the Tepuel basin (Argentine Patagonia) and the ParaM basin (Brazil), located to the south and east respectively of the Precordillera basin, both contain early Permian glacigenic sediments. Present-day separation between the Central Andes and the Precordillera is about 12~ of latitude, and between the former and the Patagonian and Brazilian basins reaches about 18-20 ~. According to this analogy, the early Ordovician faunal-lithological gradient could be explained by means of a latitudinal separation like that inferred in the early Permian model, which is comparable to the present-day. To account for the absence of warm water indicators in the early Ordovician rocks of NW Argentina and Bolivia, the Precordillera would have to have been located at about 30 ~ and the

72

J. L. BENEDETTO

Central Andean basin at c. 40-45 ~ Waisfeld & Astini (1993) pointed out that both sedimentary regimes, siliciclastic in NW Argentina and carbonate in the Precordillera, could have coexisted along the same Gondwanan margin, as occurs in the Recent southern coast of North America, where the Florida peninsula separates deltaic sedimentation to the west and carbonate deposition to the east. Although the 'Gondwana Plateau' model seems to be consistent with development of the three regions (Central Andes, Famatina, Precordillera) along the same pre-Andean continental margin, the following evidence favours the 'microplate' model. (1) No Cambrian trilobites of Laurentian signature have been encountered in the autochthonous pre-Andean basins. Certainly not enough is known yet about Early-Mid Cambrian faunas, but those recorded from Colombia, as stated above, indicates European affinities. (2) Latest Cambrian and Early Tremadoc trilobite associations from the outer shelf black shales of NW Argentina and coeval units in Bolivia are represented by 'Olenid biofacies' (Fortey 1975) which are generally thought to be independent of palaeogeography. However, the shallow water facies have also yielded only Avalonian and Baltic trilobites. Likewise, shallow-water communities from late Tremadoc beds of the Central Andean basin also lack Laurentian trilobites. (3) The high endemism exhibited by the Arenig brachiopods of the Famatina basin, as well as petrological evidence, is compatible with a volcanic arc system rather than an epicontinental basin. (4) Differences in tectono-stratigraphical history preclude the geographical continuity of the Famatina and the Precordillera during the Cambrian-mid Ordovician time interval and suggest that they were somewhat isolated by oceanic barriers and probably more separated than at present. In summary, although the 'Gondwana plateau' is consistent with the distribution of climatically sensitive sediments and, at least partially, with the early Ordovician faunal evidence, the 'far-travelled microplate' model fits better with all the available palaeontological and geological data.

The Ordovician Famatina-Puna-Avalonia volcanic island-arc system Since the seminal synthesis of northern Iapetus geology and biogeography by Neuman (1984), Ordovician volcaniclastic successions within the

Appalachian-Caledonide orogen are generally interpreted as insular in origin. Such volcanic islands are thought to have developed at midto high latitudes, in both mid-oceanic and peri-continental settings. From a biological viewpoint, they were inhabited by distinctive brachiopod assemblages, constituting the Celtic Province characterized by a relatively high proportion of endemics (c. 46% according to Neuman 1984), some genera that occur later on the Laurentian platform, and several Baltic forms. These Celtic assemblages were first described from Arenigearly Llanvirn rocks of Anglesey and the Irish Rosslare terrane and further recognized in volcaniclastic rocks from Central Newfoundland, Maine, New Brunswick, Norwegian Caledonides, the Argentine Precordillera, and probably NE China (Neuman & Harper 1992 and references therein). The new palaeontological evidence indicates that the Famatina volcanic sequence must also be included in the Celtic Province. According to the interpretation of Neuman and Harper (1992, fig. 3) this developed on the periphery of Gondwana, whereas the Toquima-Table Head Realm (North American) represented peri-equatorial passive-margin settings and related low-latitude terrains. Petrological and geochemical evidence suggest that the Famatina System was a magmatic arc related to an active margin (Acefiolaza & Toselli 1988; Toselli et al. 1993). These volcanic rocks were interpreted by Manheim (1993) as islandarc complexes and the associated volcanosedimentary succession as a back-arc setting. Genesis of this volcanic chain seems related to the gradual closure of the ocean separating the Precordillera and Gondwana. We previously speculated that east-dipping subduction was initiated in the Tremadoc and ended near the Arenig-Llanvirn transition with the accretion of the Precordillera terrane (Benedetto 1993, Benedetto & Astini 1993). The back-arc basin was filled by a c. 2000m thick succession (Suri and Molles formations) in which Astini & Benedetto (1996) recognized a first stage of rapid thermal subsidence, characterized by starved-basin graptolitic black shales, followed by a shallowing upward platform succession punctuated by storm layers and prograding volcaniclastic wedges (Mfingano & Buatois 1994). Deposition culminated with shallow-water reddish siltstones and sandstones interbedded with volcanic breccias and fossiliferous tufts. The mid-Arenig volcaniclastic rocks of the Famatina basin are richly fossiliferous but unfortunately their study is not concluded. However, their brachiopod faunas are characterized by a dominance of Celtic forms associated

BRACHIOPODS AND PRE-ANDEAN MARGIN with few typical Gondwanan genera and at least four endemics. Particularly relevant to the biogeographical interpretation is the presence of the Welsh genus Ffynnonia in the Suri Formation and of Famatinorthis in the Shin Brook Formation of Maine (Neuman 1997). This new palaeontological information confirms our inferred palaeogeographical connection between the Famatina island-arc complex and northern Iapetus islands and microplates bearing Celtic brachiopods (cf. Astini et al. 1995, fig. 14b). The Famatina System probably continues farther north into the Puna Region of NW Argentina, Chile and Bolivia, where thick volcano-sedimentary successions consisting of rhyolitic lavas, siliceous tuff breccias, spilites and volcaniclastic turbidites are recorded ('Faja Eruptiva de la Puna' volcanic belt, Fig. 3). Associated fossils are Tremadoc to Arenig in age; Arenig beds have yielded the trilobites Annamitella, Platillaenus and Illaenus (Koukharsky et al. 1996) and several articulate brachiopods which have not been studied. This volcanism has been related by several authors (Coira et al. 1982, Koukharsky et al. 1988) to an east-dipping subduction zone which was active until the late Arenig. Recent palaeomagnetic measurements from the early Ordovician rocks of Famatina and Puna Oriental (Conti et al. 1996) provide independent evidence that these two regions may constitute a separate terrane located near the Gondwana margin and geographically related to the early Ordovician Central Newfoundland volcanic sequence. The peri-Amazonian placement of West Avalonia is also supported by Nd-isotope data and the age of detrital zircons (Nance & Murphy 1994). To account for this extended F a m a t i n ~ P u n a Avalonia volcanic arc, a nearly continuous subduction zone is required along the preAndean Gondwana margin, similar to the Puna-Arequipa-Perij~ east-dipping subduction zone inferred by Scotese & McKerrow (1991). Another continental fragment involved in the Iapetus Ocean is the Irish Northwestern terrane. Shelly faunas from the Arenig Tourmakeady Limestone described by Williams & Curry (1985) as well as those from late Llanvirn beds (Harper & Parkes 1989) can be assigned unequivocally to the Toquima-Table Head Realm. The Tourmakeady Limestone assemblage is remarkable because of 23 articulate brachiopod genera, 9 (39%) are endemics. This unusually high endemism strongly suggests a peri-insular setting for the Tourmakeady fauna. One of these Irish 'endemics' (Acanthotoechia), has been recently discovered in Arenig limestones near Talacasto,

73

central Precordillera, and another (Crossiskenidium) seems to be represented in the Suri Formation (mid-Arenig) of the Sierra de Famatina. The Tourmakeady and San Juan formations have 12 genera in common, which represent 52% of the Irish fauna. On the basis of these affinities, it might be inferred that the Northwestern terrane of Ireland, which is usually placed near the northernmost Laurentian margin (cf. Neuman & Harper 1992), could have been geographically close to both the Precordillera terrane and the Famatina island arc. Inferred positions of the Precordillera, Famatina, Puna, east and west Avalonia and related terranes are shown in Fig. 6.

Iapetus geodynamics and the early evolution of the pre-Andean margin L a t e Proterozoic to Early Ordovician

Assembly of palaeocontinents, microplates and volcanic islands prior to the early Palaeozoic opening of the ]apetus Ocean is a difficult task because of the absence of palaeontological criteria. On the basis of the biogeographically better constrained early Palaeozoic reconstructions, however, some inferences about Neoproterozoic palaeogeography can be made by means of back-stepping analysis. The first point concerns the identification of the Laurentian margin counterparts. Since the hypothetical reconstruction of Bond et al. (1984) and the assembly proposed by Hoffman (1988), the preAndean margin of South America seems a better candidate than that of northwestern Africa. In the late Proterozoic (750Ma) reconstructions made by Hoffman (1991) and Dalziel (1991), eastern Laurentia is juxtaposed with Amazonia, and Ramos et al. (1993) add the Pampia and Patagonia plates. All of these continental blocks are thought to have been amalgamated during the Grenville orogeny (c. 1.1 Ga), even if the final assembly of Gondwana did not occur until the Brazilian (Pan-African) orogeny. Dalziel (1993, 1994) suggested, based on geochronological evidence, that the conjugate margin of the Hebridean shield of NW Scotland and Ireland may have been the South American Arica embayment. Both the Arequipa-Antofalla massifs and the Labrador-Scotland-Greenland promontory have yielded 'old' isotopic ages of c. 1.8-1.9 Ga and 'Grenvillian' ages. On the basis of the similarities between the brachiopod faunas of northwesternmost Ireland (South Mayo Trough) and the Precordillera, I suggest that both terranes could have been relatively close during the early Ordovician and that this

74

J. L. BENEDETTO

geographical proximity may have been due to a similar history of rifting and drifting from the Laurentian margin. In the reconstructions above, Laurentia rifted from Gondwana by the end of the Proterozoic. Development of the Appalachian passive margin is well documented by a complete succession of late Precambrian rift-facies, followed by clastic and carbonate sediments recording transition to the drifting phase in the early Cambrian (Hatcher 1989 and references therein). On the pre-Andean margin, no conclusive evidence of a coeval rifted margin has been encountered. The thick turbidite-like sequence of the Puncoviscana Formation and its associated bimodal magmatism is interpreted as being developed on a passive margin facing the Iapetus (Ramos 1988) or as an intracratonic basin developed on thinned crust (Omarini & Sureda 1993). The Puncoviscana rocks contain a varied ichnofauna of Vendian, Tommotian and earliest Cambrian age, the latter dominated by Oldhamia (Acefiolaza & Durand 1986). Late Precambrian to early Cambrian Oldhamia-bearing slates and sandstones (Grand Pitch Formation) crop out in the Gander terrane of the northern Appalachians. These rocks were deformed by the Late Cambrian Penobscot orogeny, which has also been recognized along the Scottish-Irish Caledonides (Grampian orogeny) and in the Scandinavian Caledonides (Finmarkian orogeny) (Neuman &

Max 1989). It is important to remember that metamorphic rocks of the Gander terrane in the Penobscot County of Maine are unconformably overlain by the volcano-sedimentary Shin Brook Formation in which Famatinorthis occurs (Neuman 1997). Another allochthonous terrane to the east of the former Appalachian margin is the Carolina terrane, composed of thick marine metasedimentary and volcanic rocks, probably deposited on a thin continental crust (Horton et al. 1989). Fossils include soft-bodied Ediacaran-like metazoans and the trilobite Paradoxides (Secor et al. 1983). The Carolina slate belt rocks were affected by Late Precambrian-Early Cambrian? deformation, but the age of regional metamorphism suggests that the Carolina terrane was accreted to Laurentia during the Taconian orogeny (Horton et al. 1989). The Avalon terrane of New England and eastern Newfoundland also has yielded early-mid-Cambrian trilobites similar to those of the Carolina terrane. Palaeontological evidence suggest that these Appalachian terranes have stronger affinities with peri-Gondwanan terranes than with Baltica (Neuman et al. 1989). Although the Neoproterozoic geology is highly complex and the available evidence is not enough to propose an accurate reconstruction, the data reviewed above suggest the following (Fig. 7).

"'--.

anl

1000 km -

Fig. 7. Schematic palaeogeographical reconstructions for the early opening of the Southern Iapetus Ocean. Late Precambrian map, modified from Dalziel (1994).

BRACHIOPODS AND PRE-ANDEAN MARGIN (1) The low-grade Vendian-Early Cambrian Puncoviscana Formation passive margin deposits (Je~ek & Miller 1987) and the broadly coeval similar rocks of the allochthonous Carolina and Gander terranes were probably deposited in a narrow, rapidly subsiding basin developed between Laurentia and Gondwana during an early rifting phase (cf. Dalziel et al. 1994) (2) After break-up, Laurentia moved from high to low latitudes. The Precordillera carbonate platform and the Northwestern Irish terrane developed on the eastern margin of Laurentian. (3) Both microplates detached from Laurentia, probably at different times: initial rifting of the Precordillera began in the early Cambrian, but final separation from the Ouachita embayment did not occur until the mid-late Cambrian (4) Folding and metamorphism of the Puncoviscana Formation and associated plutonism (Tilacaric orogeny), which is by far the most important tectonic event in the Central Andean basin, as well as the broadly coeval Penobscot orogeny, could be related to the start of the eastward subduction of the Iapetus oceanic crust beneath the Gondwana plate. The absence of such a disconformity in the Precordillera basin is another indication that they were not on the same margin in the Early-Mid-Cambrian. As Miller (1996) has pointed out, the change

/

/

75

from a passive to an active margin occurred when Laurentia was far away from the Gondwana margin (5) This subduction zone probably extended along the greater part of the pre-Andean margin and continued northwards into northwestern Africa and the margins of Baltica. When subduction began, arc-related volcanism developed extensively along the active margin, forming the Famatina-Puna-Avalonia volcanic arc. It is conceivable that the Late Arenig to early Llanvirn K-bentonites recorded in the Precordillera basin resulted from this volcanic activity (Huff et al. 1995) (Figs 6, 7).

Late Ordovician

In the Southern Andes, the Precordillera gradually approached the Gondwana margin before final accretion. However, faunal evidence cannot provide constraints on the timing of the collision. It seems likely, however, that by the Caradoc the Precordillera was near, or accreted to, the Famatina arc, resulting in closure of the back-arc basin (Fig. 8). This is also reflected in a rapid change in sedimentary facies of the Precordillera, with drowning of the carbonate platform and widespread black shale sedimentation (Astini et al. 1995). In the Puna region,

S

/ //

JVA

/

///////////"

//

/ i/

//

// // /

oooo \\

/

// /

k

/

t

/

/

/

/

./

/

Late~

"'~'0o

LateSilurian 1 /

Fig. 8. Inferred Laurentia and Gondwana interactions during the Late Ordovician and Silurian. Key as in Fig. 6.

76

J. L. BENEDETTO

volcanic activity ceased when the ArequipaAntofalla terrane collided with the Gondwana margin transforming the Puna back-arc into a foreland basin (Palma 1990; Bahlburg 1990: Bahlburg & Breitkreutz 1991). During the Arenig-Llanvirn, Laurentia moved along the equatorial belt around Gondwana from the 'western' to the 'eastern' Iapetus. The main modification that I introduce to the Dalziel (1991) wander path is that the site of the Late Ordovician collision between the eastern margin of Laurentia and the Avalonian island arcs is the northwestern corner of South America (Fig. 8) instead of the opposite margin of the 'Pacific' Ocean (cf. Dalziel 1997, fig. 17). This is supported by the following: (a) The Taconian orogeny along the Appalachian fold belt has been related to accretion of terranes (e.g., Hatcher 1987). One, the Carolina terrane, is generally thought to have originated in a volcanic-arc setting above an east-dipping subduction zone (Hatcher 1989); the age of metamorphism suggests that it was accreted during the Taconian orogeny (Horton et al. 1989). (b) Because of the strong lithological similarity to the Puncoviscana belt and their Atlantic fauna, it is not unreasonable to locate the Carolina terrane as bordering the Gondwana margin, as was suggested by Dalziel (1997), and near the localities with the Paradoxides fauna (La Macarena Massif and Cordillera Oriental of Colombia) (Bordonaro 1992). The Late Ordovician palaeolatitude of about 22 ~ suggested for the Carolina terrane by Vick et al. (1987) is consistent with the presence of mid-Cambrian carbonate rocks in Colombia. Therefore it is conceivable that transfer of this terrane to Laurentia occurred as the result of the late Ordovician collision. (c) Similarly, the late Precambrian volcanosedimentary succession and overlying stableplatform shales and sandstones with an Atlantic Cambrian fauna in the Avalon Zone of Newfoundland and related zones of New Brunswick might have formed along the NW Gondwana margin and been transferred by the same mechanism. (d) There is strong evidence to support a 'Caledonian' orogeny in NW South America, including widespread magmatism of 475 Ma to 439 Ma in the Cordillera Oriental and the Santa Marta Massif of Colombia as well as in the M6rida Andes of Venezuela (460 Ma to 430 Ma) (Gonz/tlez de Juana et al. 1980 and references therein). Stratigraphical units of the Sierra de Perijfi and Cordillera Oriental of Colombia include early Palaeozoic, low to high-grade

metamorphic rocks unconformably overlain by undeformed Early Devonian sedimentary rocks (Benedetto 1992; Pimentel de Bellizzia 1992). I also infer that the volcaniclastic Shin Brook Formation of Maine and the volcanic rocks of central Newfoundland developed originally on the periphery of NW South America and were transferred to Laurentia during the Taconian collision and the subsequent break-up. If the NW African margin is replaced by the NW corner of South America, the reconstruction proposed here is very similar to that presented by Pickering et al. (1988), based on a review of palaeomagnetic, petrological, stratigraphical and structural evidence. According to this model, by the Late Ordovician, Eastern Avalonia and Laurentia were geographically close and the Iapetus Ocean was nearly closed, but with a narrow sea-way persisting until the end of the Silurian. Faunal exchange between the Mediterranean and the Central Andean basins during the Caradoc might have occurred along the epicontinental sea developed outboard of the West Africa, Guayana and Amazon cratons.

Silurian and Devonian

The distribution of post-Taconian sedimentary rocks in the northern Appalachians is unrelated to the early Palaeozoic tectono-stratigraphical zonation. Silurian and Devonian rocks have been interpreted as deposits of successor basins formed across the vestiges of the Iapetus margins (Williams 1979). Authors such as Arnott et al. (1985) and Pickering et al. (1988) have suggested that late Caradoc to Wenlock basins were bounded and controlled by thrusts in the Dunnage zone, whereas in the Gander zone lithospheric loading led to the development of a deep marine foreland system. Evidence from brachiopod distribution can be used to test the scenario proposed here (Fig. 8). Brachiopod faunas from the E1 Horno Formation of the M&ida Andes, northern Venezuela, were described in detail by Boucot et al. (1972). These faunas can be unequivocally assigned to the North Atlantic Region of the North Silurian Realm. In particular, the late Llandovery-early Wenlock Antirhynchonella Fauna shows close relationships with the Back Bay fauna of southern New Brunswick (Boucot et al. 1966). Recently Boucot et al. (in press) stated that the Venezuelan Antirhynchonella Community has a slightly more European than North American aspect. I suggested that during the Silurian, the Venezuelan basin was closely related to the

BRACHIOPODS AND PRE-ANDEAN MARGIN Appalachian-Avalonian basins (Benedetto 1984, fig. 39). On the other hand, the record of Eocoelia sp. in the Llandovery of Paraguay (unpublished data) and the discovery in the Precordillera of several taxa showing strong affinities with the earliest Llandovery faunas of Wales (Benedetto 1997b) suggest that, as in the Late Ordovician, a faunal exchange occurred along the marine platforms surrounding the Guyana and Amazon cratons. A gradual isolation of Bolivian, Argentinian and especially of the intracratonic Brazilian basins, however, led to the differentiation of the cold water, AfroSouth American brachiopod Realm (Benedetto & Simchez 1996b). Provincialism of the highly diverse Devonian brachiopod faunas from the Sierra de Perij'fi of Venezuela provides better evidence to support the juxtaposition of Laurentia and NW South America (Benedetto 1984). It is proposed that such a continental assembly during the Devonian was inherited from the Late Ordovician/ Silurian ('Caledonian') continental collision previously suggested. Of the ten species recorded in the Carlo Grande Formation (Siegenian) of Venezuela, nine are identical to species from the Oriskany Formation of the Appalachian basin. Likewise, the Emsian-Eifelian-Givetian rocks of the Sierra de Perijfi (Rio Cachiri Group) have yielded more than 50 genera, all of which are also present in the Appohimchi Subprovince of the Eastern Americas Realm. Moreover, herbaceous lycophyte plants from the upper part of the Venezuelan succession show the greatest similarity to those from New York State (Edwards & Benedetto 1985; Berry et al. 1993). Similarities with the Appalachians are not only evident in the palaeontological content but also in the stratigraphical sequence: the succession of Carlo Grande (shallow water sandstones and biostromal limestones), Carlo del Oeste (platform shales and siltstones) and Campo Chico (prograding delta front sandstones) formations is nearly identical to the Oriskany-Onondaga-Marcellus-Catskill sequence of New York State, including the record of volcaniclastic rocks coeval with the Tioga Bentonite. To account for these facts, I postulate that the sedimentary basins of the Central Appalachians and NW Colombia-Venezuela were geographically continuous. The resulting reconstruction (Benedetto 1984), reproduced here in Fig. lc, is very similar to the palaeogeographical model by Dalziel et al. (1994) using palaeomagnetic data from Van der Voo (1993, fig. 4). In relation to the Devonian assembly, Dalziel et al. (1994) pointed out that the Acadian orogeny involved not only arc collision but also strike-slip movements. Similar

77

displacements were also detected in Central Newfoundland (Pickering et al. 1988). Assuming that such dextral movement continued during the Carboniferous, placement of the NW corner of South America at or near the southern margin of North America (approximately in the present-day Gulf of Mexico) can be expected in the late Palaeozoic, as is depicted in current palaeomagnetically controlled reconstructions. A corollary of the plate interactions proposed in this paper is that from Late OrdovicianSilurian times, Gondwana, Laurentia, Baltica and Avalonia became merged in a unique continent. To the south of the present-day Gulf of Guayaquil, the pre-Andean margin opened onto a wide ocean at least partially analogous to the Pacific Ocean, for which I propose the name 'Notus Ocean' (Fig. 8).

Concluding remarks The high level of provincialism displayed by articulate brachiopods through the major part of the Palaeozoic Era provides invaluable evidence to establish spatial relationships between macroand micro-plates and is therefore a useful tool to test different palaeogeographical hypotheses. However, the criterion of concordance of biological evidence with independent geological and geophysical data is the only way to construct satisfactory palaeogeographical maps. The biogeographical affinities of Early Palaeozoic brachiopods and associated shelly faunas published in the last 15 years from western Argentina, Bolivia, and Venezuela suggest that the preAndean margin of Gondwana was the site of complex plate interactions. One of the most important conclusions is that the history of this margin was linked to the evolution of the Iapetus Ocean, formerly restricted to the present-day North Atlantic continents. It is important to remark that there is significant agreement between the biogeographically-controlled maps proposed here and the sequential set of reconstructions proposed by Dalziel in successive papers since 1991, but in particular with his last interpretation of 1997. By different methods, and based on qualitatively different data, both of us have suggested that East Laurentia faced western South America, especially during the Cambrian and Ordovician, and that some kind of interaction between the two continents was established during the Devonian Recently, Dalziel (1997) stated that the 'microcontinental' model of Benedetto (1993) does not account for the relative positions

78

J. L. BENEDETTO

of Laurentia and Gondwana, concluding that 'In the case of Precordillera, if attached to Laurentia at the time of Ordovician docking in Gondwana, it is a palaeogeographical key, defining the relative palaeolongitudes of those major continental entities at certain times prior to the amalgamation of P a n g e a . . . ' . I agree completely with this possibility, but the critical question is what evidence is there to support the idea that the Precordillera was in such a position at the time of the collision? Trilobite evidence, as it is known, indicates that the Precordillera carbonate platform developed on the eastern margin of Laurentia. The subsequent Ordovician history of the Precordillera terrane can be now reconstructed accurately on the basis of a very substantial amount of palaeontological data, including not only articulate brachiopods but sponges, bryozoans, bivalves, ostracods, trilobites and graptolites. In this respect, evidence seems to be conclusive that the Precordillera moved away from Laurentia as an independent plate. Change of faunal affinities through time, from exclusively Laurentian in the Cambrian to dominantly Gondwanan in the Silurian, with intermediate 'mixed' LaurentianCeltic-Gondwanan associations during the Ordovician does not support, in my opinion, the hypothesis that the Precordillera was a Laurentian plateau until the mid-Ordovician. In general, displacement of the Precordillera fits well in the theoretical model (case C) proposed by Fortey & Cocks (1986) in which two opposing continental margins lie at different latitudes and the 'island' occupies an intermediate latitudinal and geographical position. Until recently, it has generally been assumed that the East Laurentian terranes accreted during the TaconianAcadian orogenies, such as Avalonia, Indian Bay, Maine, and the volcanic sequences of Central Newfoundland, were originally located near the northern edge of Gondwana, relatively close to Baltica, forming microcontinents, volcanic arcs or intra-Iapetus islands (e.g., the Central Newfoundland volcanic rocks) (Neuman 1988). However, new palaeontological evidence suggests that, by the Arenig, 'North Iapetus' and the Famatina volcanic sequence bearing Celtic brachiopods were geographically closer than had been generally acknowledged. If correct, a new set of plate interactions is needed to explain the evolution of the Appalachian margin. From a South American perspective, much additional geological and palaeontological data from the Palaeozoic rocks of the Andean Cordillera will be necessary to clarify the early history of the pre-Andean margin.

I thank T. M. S/mchez for her ideas during the writing of the paper and D. Harper for helpful comments on the manuscript. I also benefited from comments and discussions with many colleagues, in particular M. Carrera and M. J. Salas. I am specially indebted to R. J. Pankhurst and the reviewers for revision of the English translation. The field work was supported by the Secretarla de Investigaciones Cientificas, C6rdoba University (Project 'The Early Palaeozoic of the Famatina Range') and by grants from Consejo de Investigaciones Cientificas y Tecnol6gicas de la Provincia de C6rdoba (CONICOR) and Consejo de Investigaciones Cientificas y Tecnol6gicas (CONICET). References

ACENOLAZA, F. G. & BORDONARO,O. 1988. Presencia de Asaphiscus (Asaphiscidae, Trilobita) en la Formacion Lizoite, Salta y su significado geologico. Serie Correlaci6n Geol6gica, Universidad Nacional de Tucumfin, Argentina, 5, 21-28. - - & DURAND, F. 1986. Upper Precambrian-Lower Cambrian biota from Northwest of Argentina. Geological Magazine, 123, 357-374. - & TOSELL1,A. J. 1988. E1 sistema de Famatina: Su interpretacidn como or6geno de margen activo. 5 ~ Congreso Geoldgico Chileno, Santiago, Actas, I, 55-65. ALBANESI, G. L. & VACCARI,N. E. 1994. Conodontes del Arenig en la Formacidn Suri, Sistema del Famatina, Argentina. Revista Espagola de Micropaleontolog[a, 26, 125-146. - - , HI3NICKEN,M. A. & ORTEGA, G. 1995a. Review of Ordovician conodont-graptolite biostratigraphy of the Argentine Precordillera. In: COOPER, J. D., DROSER, M. K. & FINNEY, S. C. (eds) Ordovician Odyssey. Society Economic Paleontologists and Mineralogists, Book 77, 31-36. - - , BENEDETTO,J. L. & GAGNIER,P. G. 1995b. Sacabambaspis janvieri (Vertebrata) y conodontes del Llandeiliano temprano en la Formaci6n La Cantera, Precordillera de San Juan, Argentina. Boletin de la Academia Nacional de Ciencias, 60, 519-544. ARNOTT, R. J., MCKERROW,W. S. & COCKS, L. R. M. 1985. The tectonics and depositional history of the Ordovician and Silurian rocks of Notre Dame Bay, Newfoundland. Canadian Journal of Earth Sciences, 22, 607-618. ASTINI, R. A. & BENEDETTO, J. L. 1996. Paleoenvironmental features and basin evolution of a complex volcanic arc region in the protoAndean western Gondwana. 3rd International Symposium on Andean Geodynamics, St Malo, France, R~sum~s 6tendus, 755-758. & VACCARI, N. E. 1995. The early Paleozoic evolution of the Argentine Precordillera as a Laurentian rifted, drifted, and collided terrane: A geodynamic model. Geological Society of America Bulletin, 107, 253-273. - - , RAMOS, V., BENEDETTO,J. L., CAIqAS, F. L. & VACCARI,N. E. 1996. La Precordillera: Un terreno exdtico a Gondwana. J(III Congreso Geol6gico Argentino y H Congreso de Exploraei6n de Hidrocarburos, Buenos Aires, Actas, V, 293-324.

BRACHIOPODS AND PRE-ANDEAN MARGIN BAHLBURG, H. 1990. The Ordovician basin in the Puna of NW Argentina and Chile: geodynamic evolution from back-arc to foreland basin. Geotektonische Forschungen, 75, 1-107 & BREITKREUZ, C. 1991. The evolution of marginal basins in southern Central Andes of Argentina and Chile during the Paleozoic. Journal of South American Earth Sciences, 4, 171-188. BALDIS, B. A. & BORDONARO, O. 1982. Comparacidn entre el Cfimbrico de la "Great Basin" norteamericana y la Precordillera de San Juan, Argentina, su omplicancia intercontinental. V Congreso Latinoamericano de Geologia, Buenos Aires, 1, 97-108. -& POTHE DE BALDIS, E. D. 1995. Trilobites ordovicicos de la Formaci6n Las Aguaditas (San Juan, Argentina) y consideraciones estratigr~ficas. Boletln de la Academia Nacional de Ciencias, C6rdoba, 60, 409-448. BATES, D. E. B. 1968. The Lower Paleozoic brachiopod and trilobite faunas of Anglessey. Bulletin of the British Museum (Natural History) Geology, 16, 127-199. BENEDETTO, J. L. 1977. Algunas consideraciones acerca del limite Cambro-Ordovicico en Am+rica del Sur. Geos, 25, 13-18. - - 1 9 8 4 . Les Brachiopods Devonibns de la Sierra de Perij~ (Venezuela). Syst~matique et implications paleog6ographiques. Biostratigraphie du Pal~ozoi" que, 1, 1-191. - - 1 9 8 5 . Early Ordovician brachiopods of Argentine Precordillera and their paleogeographic significance. First International Congress on Brachiopods, Brest, Abstracts, 14. - - 1 9 8 6 . The first typical Hirnantia Fauna from South America (San Juan Province, Argentine Precordillera). In: RACHEBOEUF,P. R. & EMIG, D. (eds) Les Braehiopodes fossiles et actuels. Biostratigraphie du Pal6ozoique, 4, 439-477. - - 1 9 8 7 . Braqui6podos Clitambonitficeos de la Formaci6n San Juan (Ordovicico temprano), Precordillera de San Juan, Argentina. Ameghiniana, 24, 95-108. - - 1 9 9 0 . Los g6neros Cliftonia y Paromalomena (Brachiopoda) en el Ashgilliano tardio de la Sierra de Villicum, Precordillera de San Juan. Ameghiniana, 27, 151-159. - - 1 9 9 3 . La hip6tesis de la aloctonia de la Precordillera Argentina: Un test estratigrfifico y biogeogrfifico. XII Congreso Geol6gico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, III, 375-384. - - 1 9 9 4 . Braqui6podos ordovicicos (Arenigiano) de la Formaci6n Suri en la regi6n de Chaschuil, Sistema de Famatina, Argentina. Ameghiniana, 31, 221-238. - - 1 9 9 5 a . La fauna de braqui6podos de la Formaci6n Las Plantas (Ordovicico tardio, Caradoc), Precordillera Argentina. Revista Espa~ola de Paleontologia, 10, 239-258. - - 1 9 9 5 b . Braqui6podos del Silurico temprano (Llandoveriano) Malvinoc/ffrico, Formaci6n La Chilca, Precordillera Argentina. Geobios, 28, 425-457. - - 1 9 9 7 a . E1 g6nero Drabovinella (Brachiopoda) en el Caradociano temprano de la Sierra de Mojotoro,

79

Provincia de Salta. Asociaci6n Paleontol6gica Argentina, Reuni6n Anual de Comunicaciones. Ameghiniana, 34, 115. - - 1 9 9 7 b . Una nueva asociaci6n de braqui6podos del Llandoveriano de la Formaci6n La Chilca, Mogotes Azules, Precordillera Argentina. Asociaci6n Paleontol6gica Argentina, Reuni6n Anual de Comunicaciones. Ameghiniana, 34, 115. - in press a. Braqui6podos Caradocianos en los bloques de la diamictita glacig~nica de la Formaci6n Don Braulio (Ashgilliano), Sierra de Villicum, Precordiltera Argentina. Ameghiniana. -in press b. Early Ordovician (Arenig) brachiopods from the Acoite and Sepulturas formations, Cordillera Oriental, northwestern Argentina. Geologica et Palaeontologica. & ASTINI, R. A. 1993. A collisional model for the stratigraphic evolution of the Argentine Precordillera during the Early Paleozoic. 2nd International Symposium on Andean Geodynamics, Oxford, 501-504. -& HERRERA, Z. A. 1986. Braqui6podos del Suborden Strophomenidina de la Formaci6n San Juan (Ordovlcico temprano), Precordillera de San Juan, Argentina. Congreso Argentino de Paleontologia y Bioestratigrafia, Mendoza, Actas, IV, 103-11.

& 1987a. E1 g6nero Platystrophia King (Brachiopoda en la Formaci6n San Juan de la Precordillera Argentina. Ameghiniana, 24, 51-59. -& - - 1 9 8 7 b . Sanjuanella, un nuevo g6nero de la subfamilia Ahtiellinae (Brachiopoda, Plectambonitacea) del Ordovicico de la Precordillera Argentina. Congreso Latinoamericano de Paleontologia, Santa Cruz de la Sierra, Actas, IV, 97-109. & - - 1 9 9 3 . New Early Ordovician Leptellinidae (Brachiopoda) from the San Juan Formation, Argentina. Ameghiniana, 30, 39-58. -& SANCHEZ, T. M. 1979. Modelo de desarrollo del Oc6ano Protoatlfintico en la regi6n Norte de Sudambrica. IV Congreso Latinoamericano de Geologia, Trinidad & Tobago, 2, 825-844. -& 1996a. Paleobiogeography of brachiopod and molluscan faunas along the South American margin during the Ordovician. In: BALDIS, B. & ACElqOLAZA, G. (eds) El Paleozoico inferior del Noroeste de Gondwana. Serie Correlaci6n Geol6gica, Universidad Nacional de Tucumfin, 12, 23-38. & 1996b. The "Afro-South American Realm", and Silurian "Clarkeia Fauna". In: COOPER, P. & JIN, JISUO (eds) Brachiopods, A. A. Balkema, Amsterdam, 29-33. & BRUSSA,E. 1992. Las cuencas silfiricas de Am6rica Latina. In: GUTIERREZ-MARCO, J. C., SAAVEDRA, J. & R.~.BANO, I. (eds) Paleozoico Inferior de Ibero-AmOrica. Universidad de Extremadura, Spain, 119-148. --, SANCHEZ,T. M., CARRERA, M., BRUSSA, E. D. & SALAS, M. J. in press. Paleontological constraints on successive paleogeographic positions of Precordillera terrane during the Early Paleozoic. In: KEPPIE, D. & RAMOS, V. A. (eds),

80

J. L. B E N E D E T T O

Laurentia-Gondwana connections before Pangea. Geological Society of America, Special Paper. , VACCARI, N. E., CARRERA, M. & SANCHEZ, T. M. 1995. The evolution of faunal provincialism in the Argentine Precordillera during the Ordoviclan: New evidence and paleogeographic implications. In: COOPER, J. D., DROSER, M. K. & FINNEY, S. C. (eds) Ordovician Odyssey. Society of Economic Paleontologists and Mineralogists, Book 77, 181-184. BERRY, C. M., CASAS, J. E. & MOODY, J. M. 1993. Diverse Devonian plant assemblages from Venezuela. Documents Laboratoire G~ologique Lyon, 125, 29-42. BOND, C. G., NICKESON, A. & KOMINZ, M. A. 1984. Breakup of a supercontinent between 625 Ma and 555 Ma: new evidence and implications for continental histories. Earth Planet Science Letters, 80, 325-345. BORDONARO, O. 1989. Southamerican trilobites provincialism during the Early Paleozoic. 28th International Geological Congress, Washington, Abstracts, 1, 173. - - 1 9 9 2 . E1 Cfimbrico de Sudam6rica. In: GUTIERREZ-MARCO, J. C., SAAVEDRA,J. & R,~BANO, I. (eds) Paleozoico Inferior de Ibero-Am~rica, Universidad de Extremadura, Spain, 69-84. BORRELLO, A. V. 1971. The Cambrian of South America. In: HOLLAND, D. H. (ed.) Cambrian of the New World. Wiley-Interscience, New York, 385-438. BOUCOT, A. J., HARPER, C. W., JOHNSON, J. G. & WALMSEEY; V. G. 1966. Silurian Brachiopods and Gastropods of southern New Brunswick. Geological Survey of Canada, Bulletin, 140, 1-45. , BENEDETTO, J. L., GRAHN, Y., GONt~ALVEZDE MELO, J. E., S,~NCHEZ,T. M. & WAISFELD, B. G. in press. South American marine Silurian communities. Ecostratigraphic Project, Final Report, Cambridge University Press, chapter 51,841-847. , JOHNSON, J. G. & SHAGAM, R. 1972. Braquibpodos silfiricos de los Andes Meridefios. V Congreso Geoldgico Venezolano, Memorias, 2, 585-727 CHATTERTON, B. D. E., EDGECOMBE,G. D., VACCARI, N. E. & WAISFELD, B. G. 1997. Ontogeny and relationships of the Ordovician odontopleurid trilobite Ceratocara, with new species from Argentina and New York. Journal of Paleontology, 71, 108-124. CHEN XU & RONG JIA-YU. 1992. Ordovician plate tectonics of China and its neighbouring regions. In: WEBBY, B. D. & LAURIE, J. R. (eds) Global Perspectives on Ordovician Geology, A. A. Balkema, Rotterdam, 277-292. COIRA, B., DAVIDSON,J., MPODOZIS, C. ~r RAMOS, V. 1982. Tectonic and magmatic evolution of the Andes of northern Argentina and Chile. Earth Science Reviews, 18, 303-332 CONTI, C. M., RAPALINI, A. E., COIRA, B. & KOUKHARSKY, M. 1996. Paleomagnetic evidence of an early Paleozoic rotated terrane in northwest Argentina: A clue for Gondwana-Laurentia interaction? Geology, 24, 953-956.

COOPER, G. A. 1956. Chazyan and related brachiopods. Smithsonian Miscellanea Collection, No. 127, 1-1245. DALLA SALDA, L. H., CINGOLAN1,C. A. & VARELA,R. 1992a. E1 or6geno colisional Paleozoico de Argentina. Serie CorrelaciOn Geoldgica, Universidad Nacional de Tucumfin, 9, 165-178. - - - , DALZIEL,I. W. D., C1NGOLANI,C. A. & VARELA, R. 1992b. Early Paleozoic orogenic belt on the Andes in South America: results of LaurentiaGondwana collision? Geology', 20, 616-620. DALMAYRAC, B., LAUBACHER, G., MAROCCO, R., MARTiNEZ, C. & TOMASI, P. 1980. La cha~ne hercynienne d'Am6rique du sud; structure et evolution d'un orog+ne intracratonique. Geologische Rundschau, 69, 1-21. DALZIEL, I. W. D. 1991. Pacific margins of Laurentia and East Antarctica as a conjugate rift pair: Evidence and implication for an Eocambrian supercontinent. Geology', 19, 598 601. - - 1 9 9 2 . On the organization of American plates in the Neoproterozoic and the breakout of Laurentia. GSA Today', 2, 237-241. - - 1 9 9 3 . Tectonic tracers and the origin of the protoAndean margin. XII Congreso Geoldgico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, lII, 367-374. 1994. Precambrian Scotland as a LaurentiaGondwana link- Origin and significance of cratonic promontories. Geology, 22, 589-592. - - 1 9 9 7 . Neoproterozoic-Paleozoic geography and tectonics: Review, hypothesis, environmental speculation. Geological Society oJ America Bulletin, 109, 16 42 , DALLA SALDA, L. H. • GAHAGAN, L. M. 1994. Paleozoic Laurentia-Gondwana interaction and the origin of the Appalachian-Andean mountain system. Geological Society of America Bulletin, 106, 243 252. EDGECOMBE, G. O., CHATTERTON, B. O. E., VACCARI, N. E. & WAISFELD, B. G. 1997. Ontogeny of the proetoid trilobite Stenoblepharum, and relationships of a new species from the Upper Ordovician of Argentina. Journal of Paleontology, 71, 419-433. EDWARDS, J. & BENEDETTO, J. L. 1985. Two new species of herbaceous lycopods from the Devonian of Venezuela with comments on their taphonomy. Palaeontology, 28, 599-618 ESTEBAN, S. B. 1996. Los primeros trilobites ciclopigidos en el Ordovicico de Argentina (Formaci6n Volcancito, Sistema de Famatina). Ameghiniana, 33, 57-64. FORTEY, R. A. 1975. Early Ordovician trilobite communities. Fossils and Strata, 4, 331-352 - & CocKs, L. R. M. 1986. Marginal faunal belts and their structural implications, with examples from the Lower Palaeozoic. Journal of the Geological Society, London, 143, 151-160. --, HARPER, D. A. T., INGHAM, J. K., OWEN, A. W. & RUSHTON, A. W. A. 1995. A revision of Ordovician series and stages from the historical type area. Geological Magazine, 132, 15-30.

BRACHIOPODS AND PRE-ANDEAN MARGIN GONZALEZDE JUANA, C., Iturralde de Arozena, J. M. & PICCARD, X. 1980. Geologia de Venezuela y de sus cuencas petrolijbras. Ediciones Foninves, 1. Caracas. GUTIERREZ-MARCO, J. C., R.h,BANO, I., PRIETO, M. & MARXIN, J. 1984. Estudio bioestratigr~fico del Llanvirn y Llandeilo (Dobrotiviense) de la parte meridional de la zona Centroib~rica (Espafia). Cuadernos Geol6gicos Ib~ricos, 8, 287-319. HARPER, D. A. T. & PARKES, M. A. 1989. Paleontological constraints on the definition and development of Irish Caledonide terranes. Journal of the Geological Society, London, 146, 413-415. HARRINGTON, H. J. & KAy, M. 1951. Cambrian and Ordovician faunas of eastern Colombia. Journal of Paleontology, 25, 655-668. -& LEANZA, A. F. 1957. Ordovician trilobites of Argentina. University of Kansas, Department of Geology, Special Publication, 1. HATCHER, R. D. JR. 1987. Tetonics of the southern and central Appalachian internides. Annual Review of Earth and Planetary Sciences, 15, 337-362. - - 1 9 8 9 . Tectonic synthesis of the US Appalachians. In: HATCHER, R. D. JR,, THOMAS,W. A. & VIELE, G. W. (eds) The Appalachian-Ouachita Orogen in the United States. Geological Society of America, The Geology of North America, F-2, 511-535. HAVLi~;EK, V. 1967. Brachiopoda of the Suborder Strophomenidina in Czechoslovakia. Rozpravy Ustrednio {tstavu geologickOho, 36, 1-226. - - 1 9 7 1 . Brachiopodes de l'Ordovicien du Maroc. Notes et Memoires, Service des Mines et de la Carte Gdologique du Maroc, 230, 1-132. - - 1 9 8 9 . Climatic changes and development of benthic communities through the Mediterranean Ordovician. Sbornik geologickych red Geologie, 44, 79-110. 1990. Mediterranean and Malvinokaffric Provinces: New data on the Upper Ordovician and Lower Silurian brachiopods. Casopis pro Mineralogii a Geologii, 35, 1-13. - - & BRANISA, L. 1980. Ordovician brachiopods of Bolivia. Rozpravy Ceskoslovenske Akademie ved, 90, 1-54. HERRERA, Z. A. & BENEDETTO, J. L. 1987. E1 g6nero Reinversella (Brachiopoda) en el Ordovicico temprano de la Precordillera Argentina. X Congreso Geol6gico Argentino, Tucum~m, Actas, III, 73-76. - & 1989. Braqui6podos del Suborden Orthidina de la Formaci6n San Juan (Ordovicico temprano) en el area de Huaco-Cerro Viejo, Precordillera Argentina. Ameghiniana, 26, 3 22. & --1991. Early Ordovician brachiopod faunas of the Precordillera Basin, Western Argentina: Biostratigraphy and paleobiogeographical affinities. In: MACK1NNON, D. I., LEE, D. E. & CAMPBELL, J. D. (eds) Brachiopods through time. A. A. Balkema, Rotterdam, 283-301. HOFFMAN, P. F. 1988. United plates of America, the birth of a craton: Early Proterozoic assembly and growth of Laurentia. Annual Review of Earth and Planetary Sciences, 16, 543-603.

81

1991. Did the breakout of Laurentia turn Gondwanaland inside-out? Science, 252, 140%1412. HORTON, J. W. JR., DRAKE, A. A. JR. & RANKIN, D. W. 1989. Tectonostratigraphic terranes and their Paleozoic boundaries in the central and southern Appalachians. Geological Society of America Special Paper, 230, 213-245. HUFF, W. D., BERGSTROM, S. M., KOLATA, D. R., CINGOLANI, C. & DAVIS, D. W. 1995. Middle ordovician K-bentonites discovered in the Precordillera of Argentina: Geochemical and paleogeographical implications. In: COOPER, J. D., DROSSER, M. K. & FINNEY, S. C. (eds) Ordovician Odyssey. Society of Economic Paleontologists and Mineralogists, Book 77, 343-350. HUGHES, C. P., RICKARDS, R. B. & WILLIAMS, A. 1980. The Ordovician fauna from the Contaya Formation of eastern Peril. Geological Magazine, 117, 1-21. JE~EK, P. & MILLER, H. 1987. Petrology and facies analysis of turbiditic sedimentary rocks of the Puncoviscana through (Upper PrecambrianLower Cambrian) in the basement of the NW Argentine Andes. In: MCKENZIE, G. (ed.) Gondwana Six. American Geophysical Union, Geophysical Monographs, 40, 287 293. JOKIEL, P. L. 1990. Long-distance dispersal by rafting: reemergence of an old hypothesis. Endeavour, 14, 66-73. KELLER, M. & DICKERSON, P. W. 1996. The missing continent of Llanoria- Was it the Argentine Precordillera? XIII Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, V, 355-367. KOUKHARSKY, M., COIRA, B., BARBER, E. & HANNING, M. 1988. Geoquimica de vulcanitas ordovicicas de la Puna (Argentina) y sus implicancias tectdnicas. 5 ~ Congreso Geol6gico Chileno, Santiago, Actas, III, 1137-1151. - - - , TORRES CLARO, R., ETCHEVERRiA, M., VACCARl, N. E. & WA1SFELD, B. G. 1996. Episodios volc/micos del Tremadociano y del Arenigiano en Vega Pinato, Puna Saltefia, Argentina. XIII Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, V, 535 542. LAURIE, J. R. 1991. Articulate brachiopods from the Ordovician and Lower Silurian of Tasmania. In: JELL, P. A. (ed.) Australian Ordovician brachiopod studies'. Association Australasian Paleontologists, Memoir, 11, 1-106. 1997. Early Ordovician fauna of the Gap Creek Formation, Canning Basin, Western Australia. Journal of Australian Geology and Geophysics, 16, 701 716. LEHNERT, O. 1995. The Tremadoc/Arenig transition in the Argentine Precordillera. In: COOPER, J. D., DROSSER, M. K. & FINNEY, S. C. (eds) Ordovician Odyssey. Society of Economic Paleontologists and Mineralogists, Book 77, 145-148. --, MILLER, J. F. & REPETSKI, J. E. 1997. Paleogeographic significance of Clavohamulus hintzei Miller (Conodonta) and other Ibexian conodonts in an early Paleozoic carbonate platform facies of

82

J. L. B E N E D E T T O

the Argentine Precordillera. Geological Society of America Bulletin, 109, 429-443. LEVY, R. & NULLO, F. 1973. Braquidpodos ordovicicos de la Sierra de Famatina (Formaci6n Molles), Provincia de La Rioja. Ameghiniana, 10, 139-151. MA,NGANO, M. G. & BUTOIS, L. B. 1994. Historia deposicional de las secuencias Ordovicicas marinas del Sistema de Famatina en el Noroeste de la Sierra de Narvfiez, Catamarca, Argentina. IV Reuni6n Argentina de Sedimentologia, 2, 215-222. MANHEIM, R. 1993. Geodynamic evolution of the Gondwana margin in Western Argentina (Famatina System). 2nd International Symposium on Andean Geodynamics, Oxford, Abstracts, 531-534. MCKERROW, W. S. & COCKS, L. R. M. 1986. Oceans, island arcs and olistostromes: the use of fossils in distinguishing sutures, terranes and environments around the Iapetus Ocean. Journal of the Geological Society, London, 143, 185-191. MERGL, M. 1988. Incorthis (Orthida, Brachiopoda) from the Lower Ordovician (Arenig) of Morocco. Casopis pro Mineralogii a Geologii, 33, 199-201. MILLER, H. 1996. Early Paleozoic geochronology and orogenic events in NW Argentina. 3rd Inter-

national Symposium on Andean Geodynamics, St Malo, France, Abstracts, 787-790. MOORES, E. M. 1991. SOUTHWEST,U.S.-East Antarctic (SWEAT) connection: A hypothesis. Geology, 19, 425-428. NANCE, R. D. & MURPHY, J. B. 1994. Contrasting basement isotopic signatures and the palinspastic restoration of peripheral orogens: Example from the Neoproterozoic Avalonian-Cadomian belt. Geology, 22, 617-620. NEUMAN, R. B. 1984. Geology and paleobiology of islands in Iapetus Ocean: Review and implications. Geological Society of America Bulletin, 95, 1188-1201. - - 1 9 9 7 . Famatinorthis cf. F. turneri Levy and Nullo 1973 (Brachiopoda, Orthida)from the Shin Brook Formation (Ordovician, Arenig) in Maine. Journal of Paleontology, 71, 812-815. & BATES, D. E. B. 1978. Reassessment of Arenig and Llanvirn age (Early Ordovician) brachiopods from Anglesey, Northwest Wales. Palaeontology, 21, 571-613. & HARPER, D. A. T. 1992. Paleogeographic significance of Arenig-Llanvirn Toquima-Table Head and Celtic brachiopod assemblages. In: WEBBY, B. D. & LAURIE, J. R. (eds) Global perspectives on Ordovician Geology. A. A. Balkema, Rotterdam, 241-254. & MAX, M. D. 1989. Penobscottian-GrampianFinnmarkian orogenies as indicators of terrane linkages. Geological Society of America Special Paper, 230, 31-45. , PALMER, A. R. & DUTRO, J. T. JR. 1989. Paleontological contributions to Paleozoic paleogeographic reconstructions of the Appalach ians. In: HATCHER, R. n . JR., THOMAS, W. A. & VIELE, G. W. (eds) The Appalachian-Ouachita Orogen in the United States. Geological Society of America, The Geology of North America, F-2, 375-384. -

-

OMARINI, R. & SUREDA, R. 1993. Evolucidn geodinfimica y configuraci6n paleogeogrfifica en los Andes Centrales del Proterozoico superior al Paleozoico inferior: Modelos, alternativas y problemas. XII Congreso Geol6gico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, III, 291-307. PALMA, M. A. 1990. La evoluci6n tect6nica durante el ciclo Famatiniano en el segmento andino ubicado entre los 24 ~ y 28 ~ de Latitud Sur. XI Congreso Geol6gico Argentino, San Juan, Actas, II, 343-346. PICKERING, K. T., BASSET, M. G. & SIVETER, D. J. 1988. Late Ordovician-Early Silurian destruction of the Iapetus Ocean: Newfoundland, British Isles and Scandinavia- a discussion. Transactions of the Royal Society of Edinburgh. Earth Sciences, 79, 361-382. PIMENTEL DE BELLIZZIA,N. 1992. Paleozoico Inferior: Una sintesis del Noroeste de Am+rica del Sur (Venezuela, Colombia y Ecuador). In: GUTmRREZMARCO, J. C., SAAVEDRA,J. & RABANO, I. (eds) Paleozoico Inferior de Ibero-Amdrica. Universidad de Extremadura, Spain, 203-224. RAMOS, V. A. 1988. Late Proterozoic-early Paleozoic of South A m e r i c a - A collisional history. Episodes, 11, 168-174. - - 1 9 9 5 . Sudam&ica, un mosaico de continentes y oc0anos. Ciencia Hoy, Buenos Aires, 6, (32), 24-29. , JORDAN, T. E., ALLMENDINGER, R. W., KAY, S. M., CORTES, J. M. & PALMA, M. A. 1984. Chilenia, un terreno al6ctono en la evoluci6n Paleozoica de los Andes Centrales. IX Congreso Geol6gico Argentino, Buenos Aires, Actas, II, 84-106. ---, , MPODOZIS, C., KAY, S. M., CORTI~S, J. M. ~ PALMA, M. A. 1986. Paleozoic terranes of the central Argentine Chilean Andes. Tectonics, 5, 855-880. , VUJOVICH, G., KAY, S. M. & MCDONOUGH, M. 1993. La orog~nesis de Grenville en las Sierras Pampeanas Occidentales: La Sierra de Pie de Palo y su integraci6n al supercontinente Proterozoico.

XII Congreso Geol@ico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza,

-

-

Actas, III, 343-357. RINGUELET, R. A. 1954. Ubicaci6n zoogeogrfifica de las Islas Malvinas. Revista del Museo de La Plata, Zoologia, 48, 419-465. RONG JIA-Yu & HARPER, D. A. T. 1988. A global synthesis of the latest Ordovician Hirnantian brachiopod faunas. Transactions of the Royal Society of Edinburgh: Earth Sciences, 79, 383 402. , BOUCOT, A. J., Su YANG-ZHENG & STRUSZ, D. L. 1995. Biogeographical analysis of Late Silurian brachiopod faunas, chiefly from Asia and Australia. Lethaia, 28, 30-60. Ross, R. J. JR. & INGHAM, J. K. 1970. Distribution of the Toquima-Table Head (Middle Ordovician Whiterock) faunal realm in the Northern Hemisphere. Geological Society of America Bulletin, 81, 393-408. & JAMES, N. P. 1987. Brachiopod biostratigraphy of the middle Ordovician Cow Head and Table Head Groups, western Newfoundland. Canadian Journal of Earth Sciences, 24, 70-95.

-

-

BRACHIOPODS AND PRE-ANDEAN MARGIN

-

RUBEL, M. 1961. Lower Ordovician brachiopods of the Superfamilies Orthacea, Dalmanellacea and Syntrophiacea of eastern Baltic. Eesti NSV Teaduste Akadeemia geologia Instituudi, 6, 141-226. RUSHTON, A. W. A. 1963. Paradoxides from Colombia. Geological Magazine, 100, 255-257. SANCHEZ, T. M. in press. New late Ordovician (Early Caradoc) bivalves from the Sierra de Villicum, Argentine Precordillera. Journal of Paleontology. & BABIN, C. 1994. Los g6neros Redonia y Catamarcaia (Mollusca, Bivalvia) de la Formacibn Suri (Ordovicico temprano, Oeste de Argentina) y su intel-res paleobiogeogrfifico. Revista Espat~ola de Paleontologia, 9, 81-90. & WAISFELD, B. G. 1995. Benthic assemblages in the Northwestern of Gondwana: A test of the Ordovician evolutionary radiation model. In: COOPER, J. D., DROSER, M. K. & FINNEY, S. C. (eds) Ordovician Odyssey. Society of Economic Paleontologists and Mineralogists, Book 77, 409-412 SCOTESE, C. R. & MCKERROW, W. S. 1990. Revised world maps and introductions. In: MCKERROW, W. S. & SCOTESE, C. R. (eds) Palaeozoic palaeogeography and biogeography. Geological Society, London, Memoirs, 12, 1-21. & 1991. Ordovician plate tectonics reconstructions. In: BARNES, C. R. & WILLIAMS, S. H. (eds) Advances in Ordovician Geology. Geological Survey of Canada, Paper 90-9, 271-282. SCHELTEMA, R. S. 1977. Dispersal of marine invertebrate organisms: Paleobiogeographic and biostratigraphic implications. In: KAUFFMAN,E. G. & HAZEL, J. E. (eds) Concepts and methods of biostratigraphy. Dowden, Hutchinson & Ross, Pennsylvania, 73-108. SHERGOLD, J. H. 1988. Review of trilobite biofacies distributions at the Cambrian-Ordovician Boundary. Geological Magazine, 125, 363-380. SHI, G. R. & WATERHOUSE, J. B. 1991. Sakmarian (Early Permian) brachiopods biogeography and major associations as related to terrane drift. In: MACKINNON, D. I., LEE, D. E. & CAMPBELL, J. D. (eds) Brachiopods through time., A. A. Balkema, Rotterdam, 355-365. SECOR, D. T. JR., SAMSON, S. L., SMOKE, A. W. & PALMER, A. R. 1983. Confirmation of the Carolina slate belt as an exotic terrane. Science, 221, 649-651. SUAREZ SORUCO, R. & BENEDETTO, J. L. 1996. Primeros datos sobre la fauna de braquidpodos de la Formacidn San Benito en la Cordillera del Tunari, Cochabamba, Bolivia. XII Congreso Geoldgico Boliviano, Tarija, l, 211-216. THOMAS, W. A & ASTINI, R. A. 1996. The Argentine Precordillera: A traveler from the Ouachita

-

-

-

83

embayment of North America Laurentia. Science, 273, 752-757. TOSELLI, A., RossI DE TOSELLI, J., PELLITERO, E. & SAAVEDRA, J. 1993. E1 arco magmafitico del Paleozoico inferior en el Sistema de Famatina, Argentina. XII Congreso Geol6gico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, IV, 7-15. VACCARI, N. E. 1994. Las faunas de trilobites de las sucesiones carbongtticas del Cgtmbrico y Ordovicico temprano de la Precordillera Septentrional, Repfiblica Argentina. PhD Thesis, Universidad Nacional de C6rdoba. - - 1 9 9 5 . Early Ordovician trilobite biogeography of Precordillera and Famatina, Western Argentina: Preliminary results. In: COOPER, J. D., DROSER, M. K. & FINNEY, S. C. (eds) Ordovician Odyssey. Society of Economic Paleontologists and Mineralogists, Book 77, 193-196. , BENEDETTO,J. L., WAISFELD, B. G. & S,h,NCHEZ, T. M. 1993. La fauna de Neseuretus en la Formaci6n Suri (Oeste de Argentina): edad y relaciones paleobiogeogrfificas. Revista Espat~ola de Paleontologia, 8, 185-190. VALENTINE, J. W. 1973 Evolutionary paleoecology of marine biosphere. Prentice Hall, Englewood Cliffs. VAN DER Voo, R. 1993. Paleomagnetism of the Atlantic, Tethys, and Iapetus oceans. Cambridge University Press, Cambridge. VICK, H. K., CHANNEL, J. E. T. & OPDYKE, N. D. 1987. Ordoician docking of the Carolina slate belt; paleomagnetic data. Tectonics, 6, 573-583. WAISFELD, B. G. 1995. Early Ordovician trilobite biofacies in the Argentine Cordillera Oriental, Southwestern Gondwana: Paleoecologic and paleobiogeographic significance. In: COOPER, J. D., DROSER, M. K. & FINNEY, S. C. (eds) Ordovician Odyssey. Society of Economic Paleontologists and Mineralogists, Book 77, 449-452. & ASTINI, R. A. 1993. Anfilisis estratigrfifico y paleoamabiental del Ordovicico inferior (Formaciones Acoite y Sepulturas) al Oeste de Purmamarca, Cordillera Oriental, Argentina. XII Congreso Geol6gico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, I, 96-106. WILLIAMS,A. 1973. Distribution of brachiopod assemblages in relation to Ordovician palaeogeography. Special Papers in Palaeontology, 12, 241-270. 1974. Ordovician Brachiopoda from the Shelve district, Shropshire. Bulletin of the British Museum Natural History (Geology), Supplement 1l, 1-163. WILLIAMS, H. 1979. Appalachian Orogen in Canada. Canadian Journal of Earth Sciences, 16, 792-807. & CURRY, G. B. 1985. Lower Ordovician Brachiopoda from the Tourmakeady Limestone, Co. Mayo, Ireland. Bulletin of the British Museum Natural History (Geology), 38, 183-269. -

-

The Argentine Precordillera: its odyssey from the Laurentian Ouaehita margin towards the Sierras Pampeanas of Gondwana PATRICIA

WOOD

DICKERSON

1 & MARTIN

KELLER 2

1NASA/Lockheed-Martin, Johnson Space C e n t r e - C23, Houston, Texas 77058, USA 2 Institut fu'r Geologie und Mineralogie, Universitdt Erlangen, Schlossgarten 5, D-91054 Erlangen, Germany Abstract: The Argentine Precordillera (Cuyania terrane) and the Marathon/Solitario basin

(west Texas), both underpinned by Grenvillian Laurentian basement, evolved together during Early Cambrian to Mid-Ordovician times. Ages, Pb-isotope and geochemical data for Precordilleran and for west-central Texas basement (including the Llano uplift) are strikingly similar. Carbonate platform sequences developed on both sides (e.g., E1 Paso and Chica de Zonda) of the Marathon/Solitario outer-shelf to slope basin and hosted homologous reef organisms which were unique to the Laurentian Ouachita margin. The Marathon/Solitario basin received sediments from both north and south; much detritus came from the northern shelf, including olistoliths bearing shelf fauna. Erosional vacuities on the platforms correlate with coarse detrital basin deposits. Cuyania constituted the long-sought southern source. Stratigraphy and structures of the Precordillera/Marathon/Solitario basin are consonant with plate reconstructions that place southern Laurentia near western Gondwana from Late Cambrian into mid-Ordovician time. During Caradoc time, Cuyania moved beyond range of faunal exchange with Laurentia, and tholeiitic basalts were intruded into off-shelf turbidites in the western Precordillera. The attenuated, thermally weakened, Laurentian slab broke apart with continued extension and right-oblique separation of Laurentia and Gondwana. Severance of Cuyania from Laurentia was complete before the onset of Taconic or Ocloyic orogenesis.

A large crustal fragment within the Andes (the Precordillera of western Argentina) has recently been interpreted as a Laurentia-derived terrane (Dalla Salda et al. 1992a, b; Astini et al. 1995, 1996; Thomas & Astini 1996). Ramos (1995) considered that the Precordillera proper is part of a much larger 'Cuyania' Terrane which includes the Grenvillian rocks of the westernmost Sierras Pampeanas (e.g., Sierra de Pie de Palo) and a belt of isolated carbonate outcrops from the province of Mendoza (e.g., San Rafael) southeastwards towards the province of La Pampa (LimayMahuida). Thus defined, the Cuyania terrane would be almost 1000 km long and at least 400km wide, if Andean crustal shortening is restored (Figs 1, 2). The most likely place of origin in Laurentia is the Ouachita embayment (sensu Viele 1989; Dalla Salda et al. 1992a, b) of Arkansas, Oklahoma and Texas, which is distinctly different from the adjacent Appalachian mountain chain in its structural and sedimentary record. The Palaeozoic succession up through lowermost Permian strata of the region was deformed during the late Palaeozoic Ouachita orogeny; to date, however, no Late Ordovician (Taconic) contractional event has been documented (King 1937; Viele & Thomas 1989). The Ouachita embayment has notably similar dimensions to the Cuyania

terrane: the distance between its northeastern boundary, the Oklahoma-Alabama transform, and the southwestern boundary, the Texas transform (Thomas 1991), is of the order of 1000 km. At the southwestern end of the embayment is the Marathon/Solitario area of west Texas, the geochemical, litho- and bio-stratigraphical record of which is entirely complementary to that of the Precordillera (Figs 3, 4). Speculations about the provenance of the Precordillera received new impetus with the discovery of meta-bentonites in mid-Ordovician shelf strata of the Precordillera (Kolata et al. 1994; Huff et al. 1995). The premise originally investigated was that the Precordilleran deposits might be correlative with the well-known metabentonites of the central Appalachians and Baltica (Huffet al. 1992; Haynes 1994). However, the Precordilleran meta-bentonites are older (Arenig-Llanvirn) than those of the Appalachians (Caradoc); Huff et al. (1995) reported a zircon U - P b age of 461+%Ma from the Precordilleran suite, whereas Haynes (1994) cited an age of 453-454 Ma for the southernmost dated Caradoc sample from Alabama. Recently, ashrich subaqueous pyroclastic flows and volcaniclastic debris flows of similar age to that of the Precordilleran meta-bentonites have been found in the westernmost outcrops of the Ouachita fold

DICKERSON,P. W. & KELLER, M. 1998. The Argentine Precordillera: its odyssey from the Laurentian Ouachita margin towards the Sierras Pampeanas of Gondwana. In: PANKHURST,R. J. 8r RAPELA,C. W. (eds) The ProtoAndean Margin of Gondwana. Geological Society, London, Special Publications, 142, 85-105.

86

P. W. DICKERSON & M. KELLER

Fig. 1. (a) Geological provinces of western Argentina, showing location of the Precordillera (PR) block: Cordillera de la Costa (CC), Cordillera Frontal (CF), Cordillera Principal (CP), Sierras Pampeanas Occidentales (PO) and Pie de Palo (p), and Famatina (F). (b) Precordillera. Numbered outcrop areas are discussed in text and illustrated on Fig. 7: Guandacol (1), Los Tuneles (2), Portezuelo del Tontal (3), Los Sombreros (4), Cerro Pelado (5), San Isidro (6) (after Astini et al. 1996, figs 1, 3; published with permission, Asociacion Geologica de Argentina).

belt in the Marathon/Solitario area of west Texas (Keller & Dickerson 1996). In the Marathon/ Solitario succession, a notable unconformity spans the time interval during which the younger bentonites were erupted and deposited in Baltica and the central Appalachians (Bergstr6m 1978). Deposits of a two-sided early Palaeozoic sedimentary basin have long been documented in the Marathon/Solitario region (King 1937; Wilson 1954; Young 1970). The Cambrian-Ordovician sedimentary succession is characterized by a variety of siliciclastic and detrital carbonate rocks, which reflect deposition in outer-shelf and upper-slope settings under unstable crustal conditions. Many sediments were derived from the adjacent Laurentian craton to the north; however, felsic volcaniclastic and meta-igneous detritus in the sandstones that could not have been derived from Laurentia (King 1937; Anan 1965; Young 1970; Goter 1973; Calkins 1980)

lends support to the old concept of an enigmatic southern landmass known as 'Llanoria' (e.g., Dumble 1920; Miser 1921; King 1937; Fig. 5). Llanoria has had a chequered past spanning more than 200 Ma, which is discussed in greater detail in Dickerson (in press). Until recently it has been viewed as a single landmass that fed sandto-boulder sized detritus to both the Ouachita and Marathon reaches of the embayment, in both early and late Palaeozoic times (Sellards 1932; King 1937). Llanoria was originally a Carboniferous construct. During Ouachita-Marathon orogenesis, Yucatfin functioned as the late Palaeozoic southern landmass, which encroached and delivered sediment northward to the subsiding Ouachita-Marathon foredeep. The Llanoria concept was subsequently extended back in time to account for a source of non-Laurentian detritus in lower Palaeozoic sedimentary strata in the Ouachita embayment.

THE PRECORDILLERAN ODYSSEY

87

Fig. 2. Northward view of Chile and western Argentina, including the Precordillera. Famatina range (F) and Western Sierras Pampeanas (WSP), southward to near San Rafael. Sierra de Pie de Palo (PdP) in centre. Pacific Ocean (P) at far left; Laguna Llancanelo is the bright area in bottom centre; cities of San Juan (S) and Mendoza (M). (NASA-Mir photograph NM 21-758-085). King (1937) was the first to attribute much of a r61e to Llanoria in the early Palaeozoic; he invoked the landmass as a southerly sediment source for the Cambrian-Ordovician deposits of the Marathon/Solitario basin. Mexico, however, could not have been the early Palaeozoic Llanoria, as it did not arrive on the southern shores of Laurentia until the Devonian (Ruiz et al. 1988; Ortega-Guti6rrez 1994; Keppie & Ortega-Guti6rrez 1995). The presence of Arenig-Llanvirn meta-bentonites in the Precordillera and ash-rich pyroclastic flows in the Marathon/Solitario region lends new support to the ideas of Dalla Salda et al. (1992a, b), Dalziel et al. (1994), and Dalziel (1995, 1997), who claim that Laurentia and western Gondwana were close during Ordovi-

cian time (Fig. 6). In this context, the Cuyania terrane may represent the vanished southeastern continent of Llanoria of early Palaeozoic time.

Cuyania terrane The Cuyania terrane (Fig. 1) consists of two fundamentally different components: (1) metamorphic and igneous basement rocks and (2) the sedimentary cover of Phanerozoic deposits. Those Phanerozoic rocks are the main constituents of the Precordillera of western Argentina. Basement rocks are well exposed in the Sierra de Pie de Palo and related outcrops (Ramos et al. 1993). Their main characteristics are zircon U - P b ages indicative of a major Grenvillian

88

P. W. DICKERSON & M. KELLER

Fig. 3. Ouachita embayment of the Laurentian margin, from Alabama westward through west Texas. The Marathon basin, Solitario (S), and Devils River uplift (DRU) are indicated. The city of E1 Paso (EP) occupies the westernmost corner of Texas (Viele 1989).

tectono-magmatic event (McDonough et al. 1993) and a Pb-isotope signature that is typical of the North American (Laurentian) Grenville belt, particularly of Proterozoic basement of the Llano region of Texas (Kay et al. 1996). Deformational history, metamorphic grade, and composition also correspond well with those of the Llano region. The varied lithotypes include meta-rhyolites and meta-andesites (Vujovich et al. 1994). The metamorphic-igneous complex is covered by meta-quartzites and meta-carbo-

nates of the Caucete Group (Borello 1969) or Caucete Metamorphics (Dalla Salda & Varela 1984), which formed the initial sedimentary cover and were metamorphosed during OrdovicianDevonian times. Within the Precordillera, the contact between the basement and the overlying carbonate platform rocks is not exposed. Within the Phanerozoic sedimentary cover of the Precordillera, two different sequences are distinguished: (1) a Lower Cambrian to lower Middle Ordovician carbonate platform and (2) a

THE PRECORDILLERAN ODYSSEY

89

Fig. 4. Northwestward view of the Big Bend region of west Texas. The sharp bend of the Rio Grande (RG) in Big Bend National Park (bbnp) is within the blade-shaped mountain at bottom centre. The Solitario (S) is the circular feature, a breached dome; the Marathon fold-thrust belt (M) is the rectangular area with crenulated rim in right centre. Orientation of the Marathon/Solitario basin in this photo is fortuitously like its early Palaeozoic eastnortheasterly orientation. (NASA Space Shuttle photograph STS 060-85-OB).

Fig. 5. Early Palaeozoic palaeogeography as sketched by King (1937). The Laurentian carbonate platform (c) lay to the northwest and the off-shelf Marathon/Solitario basin (a) was seaward of the platform. The now-vanished landmass of Llanoria, which formed the south flank of the Marathon/Solitario basin, was the southerly source of the abundant igneous, meta-igneous and metasedimentary detritus (b) in early Palaeozoic sedimentary strata. On the basis of geochemical, litho- and biostratigraphical data, the Argentine Precordillera is a likely candidate for that landmass. variety of siliciclastic rocks which are attributed to shallow sub-tidal to deep basin environments. Passive-margin carbonate platform rocks are present from Early C a m b r i a n time o n w a r d , a l t h o u g h the base of a c o n t i n u o u s c a r b o n a t e succession is no older than late M i d - C a m b r i a n

(A. R. Palmer pers. c o m m . 1995); the relation between L o w e r and Middle C a m b r i a n rocks still has to be established. C o n t i n u o u s sections are present a r o u n d the city of San J u a n and near G u a n d a c o l in the n o r t h e r n Precordillera. The following discussion deals mainly with the sections

90

P. W. DICKERSON & M. KELLER

Fig. 6. Early to Mid-Ordovician global reconstruction suggesting the proximity of western Argentina and the Ouachita embayment of Laurentia. Baltica (B), East Gondwana (EG), Laurentia (L), West Gondwana (WG). South polar perspective (Dalziel 1995). around San Juan (Sierras de Villicum and Chica de Zonda), which are the classical localities of Precordilleran carbonate studies. Lower Cambrian rocks are not widespread; they occur as tectonically isolated outcrops around San Juan, as an olistolith in the Ordovician slope deposits of the Los Tuneles section, and finally as a redbed-evaporite succession near Guandacol. The Middle Cambrian is represented by the majority of the La Laja Formation (Fig. 7). The main features are thick carbonate units, shoaling upward from sub-tidal environments to probable lower inter-tidal units. They are overlain by relatively thin siliciclastic units, which are interpreted as representing a similar depositional environment to that of the top of the underlying carbonates (Keller & Bordonaro in press). Among the siliciclastic rocks, siltstones are most abundant, although near the base of the sections calcareous sandstones and quartzites are present. Rocks of another Middle Cambrian facies are present in carbonate olistoliths within the Ordovician slope facies (Los Sombreros Fro). Dark grey to black, thin- to medium-bedded limestones are mainly mudstones and wackestones. Some graded beds are interpreted as distal turbidites. Slump features are present within most olistoliths, which may be as thick as 300 m and more than 1 km long (Los Tuneles section). The fauna is dominated by agnostid trilobites and some sponge spicules. The depositional environment of this facies, for which lower and upper contacts as well as lateral relationships are

unknown, is interpreted as an outer shelf to slope setting, well below storm wave base and deep enough to allow a rich agnostid fauna to flourish. Upper Cambrian rocks are mainly dolomites, subdivided into two formations. The Zonda Fm (Fig. 7) consists predominantly of microbial laminites and dolo-mudstones. Oolites, intraclast beds and wackestones are of lesser importance. In contrast to the overlying La Flecha Fm, stromatolites are rare in these deposits. The Zonda Fm was deposited mainly in inter-tidal environments (laminites), although sub-tidal mudstones are also present. The La Flecha Fm is a highly cyclical unit, in which environments from the sub-tidal to the supra-tidal are documented (Armella 1989a, b) and even terrestrial calcretes have been described (Keller et al. 1989). The sub-tidal deposits are characterized by oolites, intraclast conglomerates and thrombolites. In addition, bioturbated mudstones are present. The inter-tidal deposits show a variety of stromatolites (Laterally Linked Hemispheroidal, Solitary Hemispheroidal) and microbial laminites, whereas the supra-tidal rocks consist of microbial laminites and storm-induced mud layers which both show desiccation features and evaporites. Chert, chalcedony, and micro-quartz are very abundant and concentrated in oolite beds and stromatolites. An important feature in the Guandacol section is a lithoclast bed with coarse detrital quartz grains. An interesting Upper Cambrian succession is present at Cerro Pelado (Fig. 7), where rocks like those of the La Flecha Fm (Cerro Pelado Fm of Heredia 1990) are overlain by deep-water, agnostid-bearing limestones that grade up-section into predominantly marlstones and shales (El Relincho Fm of Heredia 1990). The contact between the units represents a drowning unconformity (sensu Schlager 1989). A similar event has not yet been documented in the carbonate platform succession proper. Finally, deep-water carbonate olistoliths like those of the Middle Cambrian are present in the Los Sombreros Fm. The Early Ordovician platform succession starts with medium- to thick-bedded limestones (La Silla Fm; Tremadoc) and minor intercalations of microbial laminites and thrombolites. Dominant lithotypes are mudstones and wackestones, the latter containing principally gastropods and nautiloids, and peloidal grainstones and intraclast grainstones. During the Arenig (San Juan Fm) open-marine carbonates prevailed, with an abundant and diverse fauna. Most important are reef-mound zones near the base and near the top of the formation. The lower mounds are composed of sponge-algalreceptacutitid communities, whereas the upper

THE PRECORDILLERAN ODYSSEY

91

Fig. 7. Stratigraphical columns for the Precordillera/Marathon/Solitario basins. See Figs ! and 3 for locations of sections.

reefs are dominated by stromatoporoids with minor amounts of sponges and algae. The upper San Juan Fm contains several meta-bentonite beds (Kolata et al. 1994; Huff et al. 1995). Their intercalation with well-dated fossiliferous Llanvirn strata makes them useful for intercontinental correlation; the oldest bentonites so far recorded are of middle Arenig age (Bergstr6m pers. comm. 1994). During the earliest Llanvirn, carbonate platform evolution was terminated by deposition of graptolitic black shales (Gualcamayo Fm), which cover the carbonates above a drowning unconformity. Near Guandacol, in the northernmost part of the Precordillera, this event was earlier and dated as mid-Arenig. Finally, Lower Ordovician rocks are present within the Los Tuneles olistolith, where Tremadoc debris flows and turbidites erosionally overlie Middle Cambrian agnostid-bearing limestones (Keller 1995).

About 25 m of slope deposits include Tremadoc, Arenig and earliest Llanvirn faunas. Lower Ordovician rocks are also present south of San Rafael at Ponon Trehue (Bordonaro et al. 1996; Keller 1996). There, about 80m of dolomites and limestones represent much of the Tremadoc to middle Arenig record. These strata, which rest on Grenvillian basement (Ramos pers. comm. 1996), are mainly granitoids, gneisses and some meta-siliciclastic rocks. The overlying sediments can be correlated with the Lower Ordovician type sections in the Precordillera at Cerro La Silla. It can be demonstrated that about 100 m of carbonates at Cerro La Silla correspond to about 20 m of the Ponon Trehue Fm. Nevertheless, all major sedimentological elements, including the lower reef mounds, are developed. The Mid- and Late Ordovician were times of major facies differentiation. After the regional

92

P. W. DICKERSON & M. KELLER

drowning of the platform, shallow-marine carbonate sedimentation continued only locally. The Las Aguaditas Fm (Middle to lower Upper Ordovician) is composed of deep-water limestones and marlstones with abundant intercalations of sediment-gravity flows (Keller et al. 1993). The depositional environment is interpreted as a slope apron developed along normal faults. Sediment transport was apparently from east to west. Along the western margin of the former platform a thick succession of continental-slope sediments was deposited (Los Sombreros and Empozada Formations, Fig. 7). The main characteristics of both units are matrices of shales and marlstones which host sediment gravity flows. A unique association in some of the olistostromes are huge Middle Cambrian olistoliths together with basement boulders (Banchig et al. 1990; Keller 1995). The whole platform succession is represented by the clasts within the turbidites and mega-breccias. The formation of olistoliths as thick as 300m and more than l k m long requires steep escarpments, along which the basement of the Precordillera may locally be exposed (Keller 1995). Boulders smaller than the mega-olistoliths can be transported on shalelubricated slopes of 5 ~ to 10~ as documented by Cook (1970) in the Aleutian Islands. Coeval slope and basin sediments include the Portezuelo del Tontal Fm, the clast content of which is similar to that of the Los Sombreros and Empozada Formations and which also indicates a source area to the east (Spaletti et al. 1989). Although the relations among the various Middle and Upper Ordovician basin deposits are not clear, the general tendency is that of fining toward the west. In this context, the Portozuelo del Tontal Fm. (Figs 1, 5) is geographically near the Los Sombreros Fm and probably reflects more distal slope development.

Marathon[Solitario basin, Ouachita embayment Grenvillian igneous and metamorphic basement rocks ranging in age from 1 to 1.3 Ga form the underpinnings of this span of the Laurentian margin (Mosher 1993). Common-lead isotope data indicate that Laurentian crust extends as far south as the Marathon/Solitario region and that it was modified during late Palaeozoic contractional orogenesis (James & Henry 1993). Seismic reflection, potential fields, and field structural data demonstrate that basement rocks are block-faulted (generally down to the south-

east) and that pre-existing fault-block topography exerted marked control on late Palaeozoic fold/thrust geometry (Tauvers 1988; Muehlberger 1995). NE-trending fault-bounded blocks also influenced early Palaeozoic sedimentation (Anan 1965; Young 1969, 1970; Goter 1973; Calkins 1980; Canter 1993). Interpretation of seismic reflection and well data undertaken while the first author of this paper was affiliated with Gulf Research & Development Co. (c. 1980), with Conoco (1985), and later as a consulting geologist for a consortium of petroleum exploration companies ('Controls on Reservoir Development and Distribution in the Ellenburger Group, Val Verde Basin', unpublished report by Coyote Geologic Services 1991), corroborates the cited field observations of fault control on early Palaeozoic sedimentation and on late Palaeozoic fold-thrust geometry. Cambrian to Middle Ordovician rock types for both shelf and off-shelf sequences are broadly comparable, although modes of deposition and fauna differ (Dickerson 1993). At the base is a time-transgressive siliciclastic unit of highly variable thickness. Lower Ordovician lithic suites are mainly limestone and dolomitized limestone, with subordinate calcareous shale and sandstone. Following a hiatus of variable duration, depending upon palaeo-topographical position, voluminous Middle Ordovician conglomerates, sandstones (including calcarenites) and shales were deposited. Much detritus in the off-shelf sediments consists of material derived from nearby shelf strata including large olistoliths bearing shelf fauna. Erosional vacuities on the shelf correlate with coarse detrital deposits in basin strata. A significant difference, however, between the shelf and off-shelf suites is the abundance in offshelf deposits of clasts from a mixed volcanic/ sedimentary/metamorphic source to the south (King 1937; Wilson 1954; Young 1970). Compositions of the commonest felsic volcanic clasts (trachyte, dacite, trachyandesite, felsic porphyry) require a source other than the Devils River uplift (Fig. 3), where the only known southern Laurentian Ordovician volcanic strata (rhyolite) have been analysed (Nicholas & Rozendal 1975). Cambrian volcanic and sedimentary strata of the southern Oklahoma aulacogen (granite/rhyolite, gabbro, unmetamorphosed greywacke) are not of the requisite composition (Muehlberger et al. 1967). Furthermore, neither they nor the Precambrian-Cambrian metasedimentary rocks in northeastern New Mexico and the Texas Panhandle were geographically positioned to satisfy the observed variations in thickness (northward thinning) and abundance (northward decrease).

THE PRECORDILLERAN ODYSSEY The Upper Cambrian (Trempeleauan lower Tremadoc) Dagger Flat Sandstone (Fig. 7) is the oldest unit exposed in the area (King 1937; Wilson 1954). Sands and shales of the formation, which contain metre-scale olistoliths, were deposited in tectonically unstable outer-shelf to upper-slope environments. Sources for the sands lay both to the south and to the north. The northern source supplied metamorphic and granitic clasts; metamorphic rock fragments amount to as much as 42% of Dagger Flat sands in places. Granitic and meta-igneous rocks were exposed along the margin immediately north of the Marathon basin at that time. From the south came volcanic rock fragments (up to 20% of some samples), basic igneous rocks (Anan 1965; no compositional details), and green primary chert clasts. Sporadic increases in volcanic rock fragments from a southerly source could be '... explained by volcanic activity accompanied by some structural activity, such as block-faulting' (Anan 1965). Block faulting has also been proposed by Calkins (1980) as an influence on Dagger Flat deposition farther west in the Solitario. Anan (1965) suggested an array of NEtrending, fault-bounded, high and low blocks to account for lithofacies variations within the Dagger Flat Fm. Concentrations of clean quartz sandstones and flat-pebble conglomerates could reflect shoaling on topographically higher blocks. Older Cambrian shallower-water limestones preserved on downfaulted blocks might also furnish local sources of such boulders which are found in the much younger Woods Hollow Shale (Llandeilo-Caradoc). The disparate interpretations of shallow versus deep marine depositional settings for almost all the Lower Palaeozoic strata of the Marathon succession might be reconciled within an outer-shelf setting of highly variable basin-floor topography. Dagger Flat sandstones and shales grade upward into the shales of the Lower to Middle Ordovician Marathon Fm (Tremadoc-Arenig; Wilson 1954). The upper one-half to one-third of the formation is thin-bedded fossiliferous detrital limestone. Young (1970) reported abundant volcaniclastic material (rhyolite, dacite, trachyte, trachyandesite, some andesite) in the sandier intervals of the Marathon Fm and stated that the volcanic debris came from a southerly source. Dacite, trachyte and trachyandesite sources of appropriate age are unknown in west Texas, New Mexico and Oklahoma, and these lithoclasts become progressively more abundant southward. Thick, mature to supermature quartz arenites are found high in the Marathon Fm in the Solitario and in the south-

93

ern reaches of the Marathon basin but not in the north. Until the recent recognition of ash-rich subaqueous pyroclastic flows in the Fort Pefia Fm (Keller & Dickerson 1996), there was only one mention of bentonite in any of the Lower Palaeozoic strata of the region, that having been from drill cuttings; C. L. Baker & D. O. Carsey (in King 1937) described bentonite interbedded with bluish-purple chert in the upper Marathon limestone that had been penetrated in a well near the town of Marathon. Keller & Dickerson (1996) reported the presence of a 3 cm thick bentonitic bed within the upper Marathon limestone in the Solitario, establishing that the effects of late Arenig explosive volcanism were felt in this region, as they were in the Precordillera (San Juan Fm). Volcanic origin of the Solitario occurrence was established primarily on the basis of its euhedral zircon, 3-quartz, and apatite content. Olistostromes both low and high in the formation attest to unstable conditions during its deposition. Olistoliths of coarse sandstone like that of the Dagger Flat are present in the lower portions, whereas dolomitized carbonateplatform limestone boulders that constitute the Monument Spring Member are found at higher levels. The carbonate boulders appear to have been derived from local fault blocks in the northwestern basin (King 1937; Wilson 1954; Young 1969, 1970; McBride 1989); they are not found farther south in the Marathon region or in the Solitario. There is a minor hiatus of variable duration between the Marathon Fm and the Alsate Shale/ Rodriguez Tank Sandstone (Izold 1993), as there is between the San Juan Fm and younger units in the Precordillera (Baldis et al. 1985). The Alsate Shale of the Marathon basin and the laterally equivalent Rodriguez Tank Sandstone at the type locality in the Solitario are lower Llanvirn. The Alsate consists largely of dark grey graptolitic shale with lesser limestone (graptolite zone 8 - lsograptus caduceus of Berry 1960); limestone becomes more abundant to the south (King 1937). Quartz sand is markedly more abundant in the southern Marathon basin and in the Solitario, at the level of the Alsate/ Rodriguez Tank formations and within the overlying Fort Pefia Fm. Detailed mapping indicates that not all the sands previously mapped as Rodriguez Tank in the Solitario are Alsate-equivalent; several have now been recognized as being within the overlying Fort Pefia Fm (Muehlberger 1996). The Middle Ordovician Fort Pefia Fm (Llanvirn lower Llandeilo) conformably overlies the

94

P. W. DICKERSON & M. KELLER

Alsate and Rodriquez Tank formations (Berry 1960; Goter 1973) and consists of calcarenites, conglomerates, both rounded and flat-pebble, abundant bentonite-derived shale (Weaver 1961), and siliciclastic beds that are both normally and reversely graded. Interbedded with those strata are ash-rich subaqueous pyroclastic flows and volcaniclastic debris flows, discussed in detail below ('Arenig-Llanvirn Volcano-sedimentary Deposits'). All are interpreted as having been deposited in an outer-shelf setting, at or somewhat below normal wavebase (Goter 1973). Isotropic silica is preserved in rare samples from the Solitario and is probably from an igneous source (Calkins 1980); numerous reddish to purplish chert beds, commonly indicative of relict volcanic ash layers (Conkin & Conkin 1983), are characteristic of the formation in the Marathon basin and in the Solitario (King 1937; McBride 1989; Calkins 1980; W. R. Muehlberger pers. comm. 1996). Phosphatic shale and glauconite are common as well (Goter 1973). There are boulder beds with clasts up to 6 m long in the lower part of the Fort Pefia Fm in the Marathon basin (Goter 1973; McBride 1989), but none has yet been noted in the Solitario. The

olistostromes occupy a position roughly equivalent to those of the Los Sombreros Fm of the Precordillera, reflecting unstable depositional conditions and probable syn-depositional faulting in both areas during Llanvirn-Llandeilo time. Boulders consist principally of shallowmarine limestone with subordinate arkosic sandstone; andesite and granite clasts are also present. The Fort Pefia Fm grades upward into the Woods Hollow Shale (Llandeilo-lower Caradoc). The Woods Hollow Fm consists dominantly of silty phosphatic clay shale with subordinate calcareous fine sandstone and limestone (McBride 1989); the clays are bentonite-derived mixed-layer illite/smectite [montmorillonite] (Weaver 1961). Current-rippled beds, flat-pebble conglomerate zones, and other sedimentary structures observed in the Woods Hollow Fm in both the Marathon basin and the Solitario indicate deposition of much of the unit in waters at or above storm-wave base, possibly on a flanking beach of a volcanic edifice (Lajoie & Stix 1992; Fig. 8). The Woods Hollow Fm contains large olistoliths similar to those in coeval Los Azules and La Cantera formations of the Precordillera;

PELI~E-TYPE DEPOSITS~ /

,'

/

B~m /

~

~MASSIVE OR ..... REVERSEL~ GRADED FACaES I~.

~, ~ ~," l~.~f~f"/,

i , ']

/

~

'

~ ~,:~,

9 :..':;", .... ::'-.'f.i~r ~ : "

"~:~:::'.,:,-'("

/

. :'":'i'i.':.:,~ .".

,/

.-~. :. ......

~ I ,I

\,

NUEE-ARDENTE DEPOSITS /

EPICLASTIC DEPOSI

BEACH

////'.

' 'NORUAtLY.GRAOEDr

'~ ~

........,:>':,..~ . .:..::,....x~ i :..J".~.~ "'~.>:..;-:

/GRADED--CROSS BEDDED FACIES

/

Fig. 8. Idealized vertical and lateral facies variations in subaerial and subaqueous environments for a marine volcanic vent. Normally graded, cross-bedded and stratified facies, massive and reversely graded facies, and deposits interpreted as remobilized shoal or possibly beach sediments are represented in the Fort Pefia Fm in the Solitario (from Lajoie & Stix 1992; published with permission, Geological Association of Canada). Subaqueous pyroclastic flows have been observed to retain coherence to distances >250 km from the source vent (Sparks et al. 1980a).

THE PRECORDILLERAN ODYSSEY included are platform limestone blocks with faunas ranging in age from early Late Cambrian (possibly latest mid-Cambrian) to Early Ordovician (late Tremadoc; Wilson 1954). Late Cambrian faunas in the Woods Hollow boulders are older than those in any Cambrian strata now exposed in the Marathon/Solitario region; given the block-faulting that persisted throughout Late Cambrian to mid-Ordovician sedimentation, older Cambrian rocks could well be preserved on down-dropped blocks within the basin. Sandstone, schist and felsic igneous blocks are also present and believed to have come from a southerly source; Devils River uplift could have provided clasts of those compositions but the areal extent and geometry of those strata are unknown. In the outer-shelf to upper-slope sequence of the Marathon/Solitario region there is a significant unconformity at the top of the lowermost Caradoc Woods Hollow Fm/base of the Ashgill Maravillas Fm (Baker & Bowman 1917; King 1937; Wilson 1954; Bergstr6m 1978). On topographically high blocks within the platform sequence the entire mid-Ordovician section is missing at the unconformity; in the Franklin Mountains near E1 Paso, for example, the break spans c. 30 million years (Fig. 9). The unconformity at the top of the El Paso Group is the post-Sauk unconformity recorded in Laurentian platform carbonate suites from Newfoundland to at least eastern New Mexico. In west Texas, Lower Ordovician carbonate strata were blockfaulted and subjected to pervasive karstification prior to deposition of the Ashgill Montoya Fm. In topographically lower areas of the platform, mid-Ordovician shale and sandstone of the Simpson Group fill both fault and karst topography on the Lower Ordovician limestones (Galley 1958; Dyer 1989; Canter 1993); Montoya Fm carbonates were then deposited upon the Simpson siliciclastic strata. In the world beyond the Ouachita embayment, development of the Caradoc unconformity began at the same time that volcanism ceased in the Famatina system. Significantly, inherited zircons from Famatinian rocks yield a protolith age of 1.7Ga (Pankhurst et al. 1996, this volume), which indicates that the continental component there was not the Grenvillian Laurentian continental crust of Cuyania. In addition, the break spans the time of eruption and deposition of the Caradoc meta-bentonites of Appalachia and of Baltica (Kolata et al. 1994; Haynes 1994; Huff et al. 1995). Emplacement of pillow basalts in the lower Caradoc Alcaparrosa Fm of the western Precordillera indicates that separation of that area from Laurentia was then

95

Fig. 9. Post-Sauk unconformity in the Ordovician platform carbonate succession of the Franklin Mountains, El Paso, Texas. The height of the cliff face from road level (top of steps, lower right) to top of dark grey unit is c. 35 m. Darker dolomite and chert of the Montoya Group (Ashgill) directly overlie the lighter limestones and subordinate sandstones of the E1 Paso Group (Arenig). Llanvirn to early Ashgill deposits are absent.

well under way. As discussed in detail by Benedetto and Keller et al. (both in this volume) the Precordillera finally moved beyond the range of faunal exchange with Laurentia during early Caradoc time, thus ending the shared history of the two areas.

Arenig-Llanvirn

volcanogenic deposits

The discovery of meta-bentonites in ArenigLlanvirn shelf-carbonate strata in Cerro Viejo and Sierra de Talacasto of the Precordillera (Bergstr6m et al. 1994; Huffet al. 1995) permitted a first-order test of reconstructions placing the Precordillera and Marathon/Solitario region in proximity at that time (Dalla Salda et al.

96

P. W. DICKERSON & M. KELLER

1992a, b; Dalziel et al. 1994; Dalziel 1995). The Fort Pefia and Marathon formations were selected for examination based on biostratigraphical information from S. Bergstr6m (pers. comm. 1994). Those formations are correlative with meta-bentonite-bearing carbonateshelf strata of the Precordillera, so presence or absence of meta-bentonites could support or refute proximity of the two regions. The presence of ash-rich subaqueous pyroclastic flows and bentonitic debris flows in the Fort Pefia Fm in the Solitario was established in 1996 (Keller & Dickerson 1996). Later that year Keller (in a field party with R. Astini, L. Dalla Salda, I. Dalziel, P. Dickerson, F. Hutson, S. Kay, V. Ramos, & W. Muehlberger) found a 3 cm thick bentonitic bed in the Marathon Fm in the Solitario (Keller & Dickerson 1996). The volcanic origin of that layer was established on the basis of its crystalline constituents: euhedral zircon,/3-quartz and apatite. We regret that the term 'meta-bentonite' was used imprecisely in earlier reports, as the bentonitic material is mixed with normal marine sediment in this outer-shelf to slope setting. We here summarize what is known to date about the pyroclastic deposits of the Marathon/Solitario basin, particularly with regard to questions about their identification (Huff, in Dalziel 1997; Huff et al. this volume).

Previously known occurrences. Prior to this work, known Ordovician volcanic rocks in the Marathon/Solitario region included the bentonite bed in the upper Marathon Fm (Arenig) encountered in a borehole near the town of Marathon. A slightly metamorphosed Ordovician felsic volcanic rock dated at 481 • 20 Ma (Rb-Sr; Nicholas & Rozendal 1975) was penetrated in a borehole on the Devils River uplift. Significantly, with respect to a source for igneous/meta-igneous clasts in Lower Ordovician sedimentary strata, no dacites, trachytes, trachyandesites or andesites were present. King (1980, p. 37) describes andesite intrusive rocks in mid-Ordovician strata in a borehole 3 km SE of the town of Marathon. Stratigraphical assignments differ for the section in sample logs for the well (Marathon v. Fort Pefia); King states that Marathon Fm overlies the Palaeozoic sole thrust (at 4850 or 5980-6100ft) and the intrusive rocks are between 3870 and 3990 ft. The andesites are found above the late Palaeozoic sole thrust and not in the foreland sediments below. The only other known andesites in the region (Campus Andesite, c. 48Ma; Henry & McDowell 1986) are around El Paso and in

sulphur exploration boreholes in the Delaware Basin (G. Foose pets. comm. 1986). In addition, there are numerous intriguing descriptions of green clays, blue to purple cherts, and granular cherts in Lower Ordovician shelf strata of west Texas, which we have not yet examined in the field. Fort Pet~a outcrop. The sequence of yellowishbeige to bluish-grey bentonitic deposits mixed with normal rnarine sediments near the top of the formation in the Solitario is more than 7 m thick and comprises at least 22 event beds; the base of the unit at this locality is a fault contact. The crystalline component, which includes euhedral .%quartz, zircon, apatite, sphene, and minor biotite (commonly altered to chlorite), constitutes up to 80% of some Fort Pefia samples, compared with 10% of the Sierra de Talacasto sample collected and analysed by Dickerson for comparison. The subaqueous pyroclastic flows and volcaniclastic-rich debris flows are at an equivalent to somewhat higher stratigraphical level than are the thin meta-bentonites of the Precordilleran carbonate shelf sequence (S. Bergstr6m pets. comm. 1996); not surprising, given the greater likelihood of wave, storm and faunal reworking on the shelf, as observed by Hanson (1995) and Hanson et al. (1996). Depositional setting. The thickness and crystal content of the Fort Pefia samples, the outer-shelf to slope depositional setting above Carbonate Compensation Depth, the admixture of normal marine carbonate and siliciclastic sediments, and sedimentary structures of varying scales are consistent with deposition as part of the volcaniclastic apron of a submarine to subaerial vent complex. Excellent analogues are illustrated by Lajoie & Stix (1992) and are amply documented in modern volcanic aprons of the Lesser Antilles arc (Carey & Sigurdsson 1984; Reid et al. 1996) and of the New Hebrides arc (Reid et al. 1994), among others. Of particular relevance with respect to the Fort Pefia debris flows is the research of Sparks et al. (1980a, b) on- and offshore Dominica in the Lesser Antilles. They document individual andesitic to dacitic blockand-ash flows as thick as 40 m, which are rich in crystals, poor in shards, entrain boulders up to 10m in diameter, and extend from subaerial to submarine reaches of the vent flanks. Those flows maintained coherence out to 250 km from the source. Composition and geochemist~T. Euhedral /3quartz, zircon, sphene, and apatite of volcanic origin have been separated from all the Fort Pefia samples. Euhedral biotite is also present

THE PRECORDILLERAN ODYSSEY but commonly altered to chlorite. All Fort Pefia samples contained abundant calcium carbonate from marine silt and calcite cements. The crystalline fraction not only establishes the volcanic contribution to the Fort Pefia sediments but should also provide for U-Pb dating (zircon) to refine the graptolite biostratigraphical age. If an analytically significant number of ~-quartz crystals encasing glass inclusions can be extracted, both an accurate age and the composition of the source magma will be determinable. The potential value of glass inclusions in /#quartz is that they can provide vital data on the composition of the melt that was trapped during phenocryst growth within the magma chamber prior to eruption (Delano et al. 1994; Hanson et al. 1996), i.e., under virtually closed-system conditions. Alteration of the bulk composition of a volcanic ash begins the moment it is blown from the vent, starting with physical sorting in air or water, according to its density and aero- or hydro-dynamic properties. In the aqueous environment diagenesis, mechanical and faunal remobilization, and mixing with ambient sediment further modify the composition. Compaction, de-watering, and burial diagenesis contribute. At no stage of meta-bentonite formation do closed-system geochemical conditions prevail. For interpretations of magma-chamber processes, for ash-to-ash correlation and geochemical fingerprinting, and for speculations on tectonic setting for volcanism, chemical data from melt inclusions obviate much of the uncertainty and data manipulation required when meta-bentonite bulk composition is used as a surrogate for melt composition (e.g., Huff et al. 1995, 1996; Bergstr6m et al. 1996; further discussion in Hanson et al. 1996). Conditions of closed-system crystallization of granites, for which the tectonic discrimination diagrams of Pearce et al. (1984) were devised, might be more nearly satisfied when melt chemistry derived from glass inclusions in phenocrysts is used. Age. The Fort Pefia beds contain Llanvirn graptolites assigned by Berry (pers. comm. May 1996) to his zone 9, Hallograptus etheridgei of the Marathon succession. Because the later Arenig to Llanvirn was a time of maximum provincialism (Finney & C h e n Xu 1990), isotopic ages for Marathon graptolite zone 9, when determined, will have broader significance for global correlations of Llanvirn strata. Clay mineralogy. Clay mineralogy is neither a unique discriminant for volcanic origin of clays nor a correlation tool (e.g., Conkin & Conkin 1983; Haynes 1994; L. Lynch pers. comm. 1996).

97

Nonetheless, the compositional data are instructive with regard to the diagenetic history of the volcano-sedimentary strata. X-ray diffraction analysis has been carried out on samples from three ash-rich pyroclastic flows from the Fort Pefia Fm in the Solitario and on one sample collected from the thickest meta-bentonite of the San Juan Fm in the Sa. de Talacasto section of the Precordillera. Like those of the San Juan sample, clays of the Fort Pefia are dominantly illite; diffractograms for all three exhibit distinct illite/smectite (~montmorillonite) peaks, whereas the San Juan data showed none. Chlorite is present in two of the three Fort Pefia clays (the peaks persisted after heating the sample to 400~ and to 550~C). Chlorite was observed in thin section as well, in several instances replacing biotite. Alteration of biotite to chlorite commonly occurs during burial diagenesis (e.g., Haynes 1994); in addition, chlorite is a product of weathering of andesitic volcanic rocks (R. L. Folk pets. comm. 1996). From the mineralogical and sedimentological data so far available, we interpret the Fort Pefia Fm volcanogenic sediments as ash-rich pyroclastic flows and debris flows, deposited in an outer-shelf to slope environment. They, and the bentonitic bed in the Marathon Limestone, may be correlated with, or are slightly younger than, meta-bentonite beds in the shelf-carbonate strata of the Precordillera. The Marathon/Solitario volcanogenic sediments may therefore support reconstructions juxtaposing the two regions in Arenig-Llanvirn times.

Co-evolution of the Precordillera] Marathon/Solitario Basin The Marathon/Solitario Basin was a two-sided basin, not merely an embayment in the passive margin. To the north, it was bounded by the circum-Laurentian carbonate platform, from which it received much carbonate detritus. However, a candidate for the southern counterpart that served as a source for much of the siliciclastic sediment has only recently been recognized (Keller & Dickerson 1996). The Cuyania carbonate platform shows a welldeveloped passive-margin succession. Isolated coeval slope sediments do not appear before the Early Ordovician, and regional slope development did not start before the latest Llanvirn (Keller 1995). Lower Ordovician basin deposits are unknown in the Precordillera. The abrupt facies changes throughout the Marathon/Solitario basin sequence, especially the presence of Upper Cambrian carbonate

98

P. W. DICKERSON & M. KELLER

boulders in the Middle Ordovician Woods Hollow Shale, point to considerable syn-sedimentary extension and block faulting of the platform succession. Similarly, the western margin of the Precordillera is characterized by extensional faults from Late Cambrian time onward (e.g., Cerro Pelado section). An important clue to the original platetectonic configuration is provided by the metabentonites (Bergstr6m et al. 1994, 1996; Huff et al. 1995) and ash-rich pyroclastic flows (Keller & Dickerson 1996). In both regions, they are found in the same stratigraphical interval (middle to upper Arenig through to Llanvirn); those of the Solitario persist a little higher into the Llanvirn. Ongoing analyses of the crystalline mineral suite from ash-rich deposits of the Fort Pefia Fm of the Solitario (/3-quartz, apatite, zircon, titanite) will provide meaningful data for testing the correlation of Marathon/Solitario volcanogenic materials in off-shelf strata with those of the Precordillera shelf sedimentary sequence, as well as with the Ordovician rhyolite of the Devils River uplift (Fig. 3). It is noteworthy that both suites of meta-bentonites and ash-rich sediments belong to a distinctly older population (by c. 8Ma) than do those of the central Appalachians and of Baltica (Bergstr6m et al. 1994; Haynes 1994). If the Cuyania terrane and the Marathon/ Solitario area did indeed share a common history from latest Precambrian into mid-Ordovician time, that history would have begun with latest Precambrian to Cambrian rifting, which initiated the opening of an ocean along the eastern Appalachian and southern Ouachita margins of Laurentia. As noted by Thomas (1991), the expression of that margin varies along its length. In contrast with the Appalachian margin, riftedmargin prisms seem to be absent from the Ouachita margin (Viele & Thomas 1989), and they are most probably also absent from the Precordilleran margin (Keller 1995). Their absence is consistent with transtensional separation of the latter two areas as, for example, in the Sea of Japan (e.g., Kong & L a w v e r 1995). Following that, a sedimentary basin developed between the Laurentian margin to the N/NW and the Cuyania continental margin to the S/SE. This basin was flanked on either side by vast carbonate platforms typical of passive margins (e.g., Bond et al. 1984; Gonzfilez Bonorino & Gonzfilez Bonorino 1991). From Mid-Cambrian time onward a deeper water environment persisted on the Precordilleran margin. Disintegration of the carbonate platform along its western margin started during the Late Cambrian (Cerro Pelado section).

The submersion of peri-tidal carbonates beneath deeper water limestones and marlstones, an event which is not observed on the platform itself, points to a sudden foundering of the platform edge, probably along normal faults. The Upper Cambrian succession of the Marathon Basin consists primarily of sandstone. Anan (1965) suggested an array of NE-trending fault blocks to account for lithofacies variations within the Dagger Flat Fm. Boulder beds within the formation attest to unstable conditions during deposition. Onlap of Lower Ordovician carbonates onto exposed basement near San Rafael demonstrates that a source for siliciclastic input into the basin was present during much of the Cambrian and Early Ordovician history. Igneous and metamorphic basement rocks there, as well as in Pie de Palo (Caucete Group), may have provided the exotic siliciclastic detritus to the Dagger Flat, Marathon, Rodriquez Tank and Fort Pefia sandstones in the Marathon/Solitario area. During the Early Ordovician, carbonate sedimentation continued on both platforms and slowly onlapped cratonic basement. In the Cuyania terrane this is documented for the Ponon Trehue section (Bordonaro et al. 1996); on the Laurentian margin an analogous section is the E1 Paso Group near the city of E1 Paso, Texas (LeMone 1988a). Significantly, both platforms

BB continental interior shelf environment

sponge-algal facies

off-shelf Ouaehita facies

hypothetical continuation of carbonate platform, Argentine PrecordiHera

Fig. 10. Distribution of Arenig sponge algal bioherms that rimmed the Ouachita embayment of the Laurentian margin and are found in counterpart strata of the San Juan Fm of the Argentine Precordillera. The star marks the location of E1 Paso, westernmost Texas. (Modified from Alberstadt & Repetski 1989).

THE PRECORDILLERAN ODYSSEY

Marathon-Solitario basin

Laurentian margin

99

western basin

El Paso

eastern basin Zondat

NN

12~

~

NN

BB

carbonate platform succession

ocean floor with basinel shales end pillow basalts

syn-rift deposits

off-shelf siliciclastJc deposits

youngest platform deposits and correlatives

[•

Marathon-Solitario basin

Precordilleran margin

Laurentian margin .

.

.

.

*

+

*

§

. +

l

l'

Et Paso

Zonda

i ,,,i l~m~...:::.x%

~

II

+

§

+

+

+

+

*

+

+

+.

.

*

+

+

+

+

+

+

+

§

Lowerand Middle Cambrian platform succession

+

UpperCambrianl Lower Ordovlcian succession

.

9

+

sponge-algal mound facies

+

..

+

off-shelf siiicictastic deposits

+

+

§

+

+

.

+

1

[

I

I | '

. . . .

+

+

.

+

+

+

+

+

+

+

§

§

+

+

+

+

youngest platform deposits and correlatives

Fig. 11. Palaeogeographical restorations of the Argentine Precordiltera and Marathon/Solitario region of the Ouachita embayment of Laurentia: (a) Early Ordovician, (b) Mid-Late Ordovician. share nearly identical reef structures. In the Franklin Mountains of west Texas the dominant reef organism is Pulchrilamina, a stromatoporoid-like organism (LeMons 1988b), whereas in the Precordillera Zondarella communis is the main organism, also attributed to the stromatoporoids (Keller & Flfigel 1996). The sponge-algal facies that hosts these reefs is mainly distributed around the Ouachita margin of Laurentia (Alberstadt & Repetski 1989; Figs 10, lla). These facies have their clear counterpart in the San Juan Fm, which mimics the Laurentian occurrences even in details like the presence of Nuia and the absence of the Epiphyton/Renalcis community (Cafias & Keller 1993; Keller et al. 1993). Adjacent to the platforms, steep slopes developed which generated sediment gravity flows. Today evidence of them is present within the mega-olistolith in the Los Tuneles section of the Los Sombreros Fro. (Keller 1995) and in the Marathon Fm (McBride 1989).

Carbonate boulders within the Marathon Limestone were locally derived, having come from the block-faulted shelf sequence to the northwest (Wilson 1954; McBride 1989). However, siliciclastic/volcaniclastic sands within the Marathon Fm came from the southeast to southwest (Young 1970) and, as basement rocks in the Cuyania terrane were still exposed in Early Ordovician times, might have been derived from this source, The Arenig-Llanvirn Famatina eruptive complex (Rapela eta/. 1992) could also have constituted a southerly source for fine-grained rhyolitic/dacitic/andesitic detritus in the Marathon and Fort Pefia formations. During Mid- and Late Ordovician times the Precordillera was characterized by large-scale extension. This is documented by the emplacement during Caradoc time of rift-related pillowbasalts within graptolite-bearing (Nemagraptus gracilis zone; Llandeilo) turbidites (Kay etal. 1984) and the formation of horsts and grabens

100

P. W. DICKERSON & M. KELLER

in the former platform area (Fig. llb). Largescale extension is also responsible for the formation of steep fault scarps, the source for the mega-olistoliths and the basement boulders in the Los Sombreros Fm (Keller 1995). The stratigraphically equivalent Fort Pefia Fm and the younger Woods Hollow Shale also contain abundant olistostromes with platform carbonate and exotic clasts (schist, sandstone, felsic porphyry; McBride 1989). The presence of Late Cambrian to Early Ordovician fossils in the carbonate blocks points to considerable uplift in the hinterland. Extension and the block-faulted topography in the Precordillera at that time provides a source for the igneous and metamorphic boulders observed in the Woods Hollow Shale. The Arenig-Llanvirn meta-bentonites described from the Precordillera (Kolata et al. 1994; Huff et al. 1995) and volcanogenic flows from the Marathon/Solitario area (Keller & Dickerson 1996) are also interpreted as related to crustal extension, which was the dominant deformational regime during Cambrian to midOrdovician times in the Precordillera (von Gosen 1992; Loske 1992; Keller 1995; Buggisch 1996), as well as in the platform sequence and in the Marathon/Solitario region of west Texas (Young 1969, 1970; Ammon 1981; Dickerson 1980, 1994, 1995; Dyer 1989). The extensional events reflect ongoing separation of the Cuyania terrane from Laurentia, culminating during the late Llandeilo and Caradoc, which is in agreement with the faunal data and with models presented by Dalla Salda et al. (1992a, b), Dalziel et al. (1994), and Dalziel (1995). Keller et al. (this volume) discuss the biostratigraphical data in detail. Sediment input into the western basin was markedly diminished during the Caradoc; in the Precordillera to the east, sediment was trapped in the platform interior grabens and along the continental slope/continental rise. As a result, the Marathon/Solitario basin was converted to a deep-water oceanic basin shut off from detrital sedimentary input, contributing to the lengthy Caradoc hiatus. Several observations must be reconciled in any plate tectonic scheme proposed to account for rifting and transfer of the Cuyania terrane from Laurentia to Gondwana. Based upon the palaeomagnetic reconstructions of Dalziel (1995), for example, during the period from 550 to 420 Ma (particularly between c. 485 and 420 Ma), Laurentia was moving steadily northward and clockwise with respect to Gondwana. During the same period Gondwana rotated more than 90 ~ anticlockwise. Such plate trajectories would require

that the mode of deformation of the basin were right transtension rather than near-orthogonal extension. Ramos' (1984) interpretation of dextral offset of elements of the Gondwanan margin would be consonant with that mechanism and timing for deformation. Separation would have been more complex than rifting away of the transform-fault bounded terrane and translating it eastward across the widening sea (Astini et al. 1995; Thomas 1996; Thomas & Astini 1996). Such rifting and transform faulting in the marine realm would result in generation of oceanic crust, of which only small occurrences have been located (e.g., western Precordillera; Kay et al. 1984); none has yet been established on the Ouachita margin, although Thomas (pers. comm. 1997) speculates that serpentinite pods in the Ouachita thrust belt represent fragments of obducted Iapetus crust. Whatever their age, the bodies are not extensive. Plate rotations of the sort described (Dalziel 1995) are particularly problematical for maintaining the integrity of a Laurentian-basement cored plateau spanning the mid-Ordovician ocean (c. 20 ~ of latitude in the 465 Ma reconstructions of Dalziel 1996, 1997). Regarding incorporation of Cuyania into Gondwana, eastward subduction and corresponding magmatism are well documented along the western Gondwanan margin in the Western Sierras Pampeanas (Ramos 1988, 1995; Vujovich et al. 1994; Astini el al. 1996). However, the thermally softened and attenuated Cuyania slab, considered by itself, seems unlikely to have been the colliding body owing to its limited size, diminished coherence, and increased buoyancy relative to the crust of the Gondwanan continental mass (Cloos 1993; see also Buggisch 1996). In addition, the Ordovician contractional tectonism in the Western Sierras Pampeanas occurred during the Ocloyic event, significantly later than Late Cambrian to early Caradoc deformation and magmatism in the Precordillera/Marathon/Solitario basins.

Conclusions The Precordillera, an element of the Cuyania terrane of western Argentina, originated in the Ouachita embayment of the southern Laurentian margin. The Precordillera and the Marathon/ Solitario basin of west Texas, both with fundaments of Grenvillian Laurentian basement, evolved together during Late Cambrian through to mid-Ordovician times. The Marathon/Solitario basin was a twosided basin, not merely an embayment in the

THE PRECORDILLERAN ODYSSEY passive margin. To the north it was bounded by the circum-Laurentian carbonate platform, from which it received much carbonate detritus. Siliciclastic/volcaniclastic sands within the early Palaeozoic succession came from a southerly source, believed to have been recta-igneous basement rocks in Cuyania, with probable contributions from the Famatina eruptive complex. Both Early Ordovician carbonate platforms were home to virtually identical reef organisms. The sponge-algal facies that hosts the Pulchrilamina (west Texas) and Zondarella (Precordillera) reefs was mainly distributed around the Ouachita margin of Laurentia. Syn-sedimentary extensional block-faulting characterized the conjoined basin throughout its existence. Arenig-Llanvirn meta-bentonites and ash-rich pyroclastic flows of the Marathon/ Solitario/Precordillera basin were probable products of that extensional deformation. LlandeiloCaradoc igneous activity reflects the culmination of extension and extrusion of pillow basalts in turbidites of the westernmost Precordillera. Significant unconformities in both the Marathon/ Solitario and Precordilleran sections span the period from later early Caradoc until early Ashgill - the time of final separation. Marathon/Solitario/Precordillera extension, sedimentation, and volcanism are interpreted as having occurred within a right-transtensional regime created by northward translation and clockwise movement of Laurentia with respect to Gondwana. With continued extension and right-oblique separation of the two continental masses, the attenuated and thermally softened Laurentian slab broke apart and oceanic tholeiites were emplaced in turbidites of the western Precordillera. Severance of Cuyania from Laurentia was thus complete before the onset of Taconic or Ocloyic orogenesis. This manuscript has benefited from thoughtful reviews by W, R. Muehlberger, C. W. Rapela, and particularly that ofW. A. Thomas. Lively discussions in the field in the Solitario with V. A. Ramos, L. H. Dalla Salda, R. Astini, I. W. D. Dalziel, W. R. Muehlberger. F. Hutson, and S. M. Kay helped shape the interpretations presented here. Graptolites collected by Ramos, Dalziel and Dickerson on that occasion were identified by W. B. N. Berry, whose collaboration is most gratefully acknowledged. Dickerson thanks S. Tweedy (formerly of Mineral Studies Laboratory, Texas Bureau of Economic Geology), B. Hanson (Coming Laboratories), R. Folk, L. Lynch, D. S. Barker, K. Milliken, F. Hutson (all of University of Texas at Austin) for edification on clay mineralogy/ diagenesis, interpretation of XRD data, melt inclusions, igneous geochemistry, and mineral separation techniques. Fieldwork with W. Muehlberger in the

101

Solitario has been endlessly enlightening. Stimulating exchanges with J. D. Cooper, G. W. Viele, H. Bahlburg, D. L. Amsbury, V. A. Ramos, I. W. D. Dalziel, and O. Lehnert in various venues have expanded Dickerson's views of the Precordillera, west Texas, the Ordovician, Argentina and the world in general.

References ALBERSTADT, L. & REPETSKI, J. E. 1989. A Lower Ordovician sponge/algal facies in the southern United States and its counterparts elsewhere in North America. Palaios, 4, 225-242. AMMON, W. L. 1981. Geology and plate tectonic history of the Marfa basin. Presidio County, Texas. In: Marathon-Ma~fa Region of West Texas, Permian Basin Section. Society of Economic Paleontologists and Mineralogists, Publication 81-20, 75-102. ANAN, F. S. 1965. Petrology and paleocurrent study of the Dagger Flat (Cambrian) sandstone, Marathon basin, Texas. MA thesis, The University of Texas at Austin. ARMEEEA, C. 1989a. Microfacies tromboliticas de un biociclo ideal de la Formaci6n La Flecha, Precordillera oriental, Argentina. Correlacidn Geol6gica, 5, 45 52. - 1989b. Grupos morfol6gicos y ftibrica de trombolitos del limite Cfimbrico-Ordovicio de Argentina. Correlaci6n Geol@ica, 5, 53 66. ASTINI, R. A., BENEDETTO, J. L. & VACCARI, N. E. 1995. The early Paleozoic evolution of the Argentine Precordillera as a Laurentian rifted, drifted, and collided terrane: A geodynamic model. Geological Socieo' of America Bulletin, 107, 253-273. --, RAMOS, V. A., BENEDETTO, J. L., VACCARI, N. E. & CA~AS, F. L. 1996. La Precordillera: Un terrano ex6tico a Gondwana. AT11 Congreso Geol@ico Argentino y IIi Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, V, 293 324. BAKER, C. L. & BOWMAN, W. F. 1917. Geologic exploration of southeastern front range in TransPecos Texas. The UniversiO' of Texas Bulletin, 1753, 107-112. BALDIS, B. A., ARMELLA, C. & CABALERI, N. 1985. Desarollo de la plataforma carbonfitica Ordovicica Argentina. 6~ Congreso Latino-Americano de Geologia, 165-174. BANCHIG, A. L., M1LANA, J. P. & BORDONARO, O. L. 1990. Litofacies de la Formaci6n Los Sombreros (Cfimbrico medio) en la quebrada Ojos de Agua, Sa. de Tontal, San Juan. III ReuniOn Argentina de Sedirnentologla, Actas, 1, 25-30. BENEDETTO, J. L. 1998. Early Palaeozoic brachiopods and associated shelly faunas from western Gondwana: their bearing on the geodynamic history of the pre-Andean margin. This volume. BERGSTROM, S. M. 1978. Middle and Upper Ordovician conodont and graptolite biostratigraphy of the Marathon, Texas graptolite reference zone. Paleontology, 21, 723 -758.

I02

P. W. D I C K E R S O N & M. K E L L E R

, HUFF, W. D., KOLATA,D. R. & CINGOLANI,C. A. 1994. Ordovician K-bentonites discovered in the Precordillera of Argentina. A preliminary comparison with K-bentonite complexes in Laurentia and Baltoscandia. Program and Abstracts, Joint Meeting of IGCP Projects 376 and 319 and 2nd Circum-Atlantic/Circum-Pacific Terranes Conference, Nova Scotia (11/94), 6. , KREKELER, M. P. S., CINGOLANI, C. A . ' & ASTINI, R. 1996. Lower and Middle Ordovician K-bentonites in the Precordillera of Argentina. XII1 Congreso Geoldgico Argentino y III Congreso de ExploraciGn de Hidrocaburos, Buenos Aires, Actas, V, 481-490. BERRY, W. B. N. 1960. Graptolite faunas of the Marathon region, West Texas. The University of Texas at Austin, Publication 6005. BOND, G. C., NICKESON, P. A. & KOMINZ, M. A. 1984. Breakup of a supercontinent between 625 Ma and 555 Ma: New evidence and implications for continental histories. Earth and Planeta~:r Science Letters, 70, 325-345. BORDONARO, O., KELLER, M. & LEHNERT, 0, 1996. El Ordovicico de Ponon Trehue en la provincia de Mendoza (Argentina): Redefiniciones estratigrfificas. XIII Congreso Geol@ico Argentino y III Congreso de ExploraciGn de Hidrocaburos, Buenos Aires, Actas, I, 541 550. BORRELLO, A. V. 1969. Los geosinclinales de la Argentina. DirecciGn Nacional de Geologia y Mineria, Anales, 14. BUGGISCH, W. 1996. The Argentine Precordillera terrane-Critical remarks and remaining questions. XIII Congreso GeolGgico Argentino y III Congreso de ExpIoraciGn de Hidrocaburos, Buenos Aires, Actas, V, 469. CALKINS, C. P. H. 1980. Stratigraphic and structural geology of the northern half of the Solitario, Presidio and Brewster Counties, Texas. Master's thesis, Alpine, Texas, Sul Ross State University. CAt'AS, F. & KELLER, M. 1993. "Reefs" and "reef mounds" en la FormaciGn San Juan (Precordillera Sanjuanina, Argentina): Los arecifes mils antiguos de Sudamerica. Boletin de la Real Sociedad de Historia Natural (Geologia), 88, 127-136. CANTER, K. L. 1992. Paleostructural and related paleokarst controls on reservoir development in the Lower Ordovician Ellenburger Group, Val Verde Basin, Texas. In: CANDELARIA,M. P. & REED, C. L. (eds) Paleokarst, Karst-Related Diagenesis and Reservoir Development." Examples JEom Ordovician-Devonian-Age Strata oJ" West Texas and the Midcontinent. Permian Basin Section, Society of Economic Paleontologists and Mineralogists, Publication 92-33, 79-92. CAREY, S. N. & SIGURDSSON, H. 1984. A model of volcanogenic sedimentation in marginal basins. In: KOKELAAR, P., HOWELLS, M. & ROACH, R. (eds) Marginal Basin Geology: Volcanic and Associated Sedimentar)' and Tectonic Processes in Modern and Ancient Marginal Bashls. Geological Society, London, Special Publications, 16, 37 58.

CLoos, M. P. 1993. Lithospheric buoyancy and collisional orogenesis: Subduction of oceanic plateaus, continental margins, island arcs, spreading ridges and seamounts. Geological Society of America Bulletin, 105, 715-737. CONKIN, J. E. & CONKIN, B. M. 1983. Paleozoic metabentonites of North America." Part 2 - Metabentonites in the Middle Ordovician Tyrone Formation at Boonesborough, Clark County, Kentucky. University of Louisville, Studies in Paleontology and Stratigraphy, 17. DALLA SALDA, L. H. & VARELA, R. 1984. El metamorfismo en el tercio sur de la Sierra Pie de Palo, San Juan. Revista de la AsociaciGn Geol~gica Argentina, 39, 68-93. - - - , CINGOLANI, C. A. & VARELA, R. 1992a. Early Paleozoic orogenic belt of the Andes in southwestern South America: Results of LaurentiaGondwana collision? Geology, 20, 617-621. - - , DALZIEL,I. W. D., CINGOLANI,C. A. • VARELA,R. 19921). Did the Taconic Appalachians continue into South America? Geology, 20, 1059-1062. DALZIEL, I. W. D. 1995. Earth before Pangea. Scientific American, January, 58-63. - - 1 9 9 6 . The "Texas Plateau" Model for the origin of the Argentine Precordillera: A potential key to Paleozoic geography. XIII Congreso Geol@ico Argentino y III Congreso de Exploraci6n de HMrocarburos, Buenos Aires, Actas, V, 543. - - 1 9 9 7 . Neoproterozoic-Paleozoic geography and tectonics: Review, hypothesis, environmental speculation. Geological Society of America Bulletin, 109, 16-42. --, DALLA SALDA, L. H. & GAHAGAN, L. M. 1994. Paleozoic Laurentia-Gondwana interaction and the origin of the Appalachian-Andean mountain system. Geological Society of America Bulletin, 106, 243 252. DELANO, J. W., TICE, S. J., MITCHELL, C. E. & GOLDMAN, D. 1994. Rhyolitic glass in Ordovician K-bentonites: A new stratigraphic tool. Geology, 22, 115-118. DICKERSON, P. W. 1980. Structural zones transecting the southern Rio Grande rift - Preliminary observations. New Mexico Geological Society, 31st Field Conference Guidebook, 63-71. 1993. Eocambrian rifted margin of southwestern North America: Drift-phase miogeoclinal and eugeoclinal strata (abstract). First Circum-Atlantic/Circum-Pacific Terranes Conference, Guanajuato, Mexico, Proceedings, 41. 1994. Ordovician tectonics, sedimentation and volcanism of southern Laurentian m a r g i n Southwestern U.S. & eastern Mexico (abstract). IGCP Project 376- Laurentian-Gondwanan Connections before Pangea, First Cotference, Cape Breton Is., Nova Scotia, Abstracts, 15. 1995. The southern margin of Laurentia prior to Ouachita convergence Influence of earlier Paleozoic orogenic events (abstract). Geological Society of America, Abstracts with Programs, 27, 459. --1998. Co-evolution of the Marathon/Solitario (W. Texas) and Precordillera (W. Argentina)

T H E P R E C O R D I L L E R A N ODYSSEY basins- Implications for the prodigal Llanoria landmass. In: RAMOS, V. A. & KEPPIE, J. D. (eds) Terranes of Western Gondwana, Geological Society of America Special Papers, in press. DUMBLE, E. T. 1920. The geology of West Texas. University of Texas, Bulletin 1869. DYER, R. 1989. Structural geology of the Franklin Mountains, West Texas. In: MUEHLBERGER,W. R. & DICKERSON, P. W. (eds) Structure and stratigraphy of Trans-Pecos Texas. American Geophysical Union, International Geological Congress guidebook, T-317, 65-70. FINNEY, S. C. & CHEN Xu. 1990. The relationship of Ordovician graptolite provincialism to paleogeography. In: MCKERROW, W. S. & SCOTESE, C. R. (eds) Paleozoic Palaeogeography and Biogeography. Geological Society, London, Memoirs, 12, 123-128. GALLEY, J. E. 1958. Oil and geology in the Permian Basin of Texas and New Mexico. In: Habitat o,1" Oil. American Association of Petroleum Geologists, 395-446. GONZALEZ-BONORINO,G. G. & GONZALEZ-BONORINO, F. G. 1991. Precordillera de Cuyo y Cordillera Frontal en el Paleoz6ico temprano: Terrenos "bajo sospecha" de ser autdctonos. Revista GeoIdgica de Chile 18, 97-107. GOTER, E. R. JR. 1973. Stratigraphy and sedimentary petrology of the Ft. Pega Formation, Marathon basin, Texas. MA thesis, The University of Texas at Austin. HANSON, B. 1995. A geochemical study of rhyolitic melt inclusions in igneous phenocrysts from Lower Devonian bentonites. PhD dissertation, The State University of New York at Albany. - - , DELANO, J. W. & LINDSTROM, D. J. 1996. Highprecision analysis of hydrous rhyolitic glass inclusions in quartz phenocrysts using the electron microprobe and INAA. American Mineralogist, 81, 1249-1262. HAYNES, J. T. 1994. The Ordovician Deicke and Millbrig K-bentonites of the Cincinnati Arch and the southern Valley and Ridge Province. Geological Society of America, Special Papers, 290. HENRY, C. D. & McDOWELL, F. W. 1986. Geochronology and magmatism in the Tertiary volcanic field, Trans-Pecos Texas. The University of Texas at Austin, Bureau of Economic Geology, guidebook, 23, 99-122. HEREDIA, H. 1990. Geologia de la cuchilla del Cerro Pelado, Precordillera de Mendoza, Argentina. XI Congreso Geoldgico Argentino, San Juan, Actas, II, 101-104. HUFF, W. A., BERGSTROM, S. M. & KOLATA, D. e. 1992. Gigantic Ordovician volcanic ash fall in North America and Europe: Biological, tectonomagmatic and event-stratigraphic significance. Geology, 20, 875-878. , , CINGOLANI, C. A. & ASTINI, R. A. ]998. Ordovician K-bentonites in the Argentine Precordillera: relations to Gondwana margin evolution. This volume. , , , & DAVIS, D. W. 1995, Middle Ordovician K-bentonites discovered in the Pre-

103

cordillera of Argentina: Geochemical and paleogeographical implications. In: COOPER, J. D., DROSER, M. L. & F~NNEY, S. C. (eds) Ordovician Odyssey. Pacific Section SEPM, Book 77, 343-349. IZOLD, M. D. 1993. Earl), Ordovician shelf-slope conodont biostratigraphy and biofacies differentiation of western Iapetus in west Texas and New York. Master's thesis, The Ohio State University. JAMES, E. W. & HENRY, C. D. 1993. Pb isotopes of ore deposits in Trans-Pecos Texas and northeastern Chihuahua, Mexico: Basement, igneous and sedimentary sources of metals. Economic Geology, 88, 934-947. KAY, S. M., ORRELL, S. & ABBRUZZI, J. M. 1996, Zircon and whole-rock Nd-Pb isotopic evidence for a Grenville age and a Laurentian origin for the basement of the Precordillera terrane in Argentina. Journal of Geology, 104, 637-648. --, RAMOS, V. A. & KAY, R. 1984. Elementos mayoritarios y trazas de las vulcanitas Ordovicicas en la Precordillera occidental: Basaltos de rift oce/micos tempranos(?) prdximos al margen continental. IX Congreso Geol6gico Argentino, Buenos Aires, Actas, II, 48 85. KELLER, M. 1995. Continental slope deposits in the Argentine Precordillera: Sediments and geotectonic significance. In: COOPER, J. D., DROSER, M. L. & FINNEY, S. C. (eds) Ordovician Odyssey. Pacific Section SEPM, Book 77, 211-215. 1996. Anatomy of the Precorditlera (Argentina) during Cambro-Ordovician times: Implications for the Laurentia-Gondwana transfer of the Cuyania Terrane. III International Symposium on Andean Geodvnamics (ISAG 96), St Malo, France, 775-778. & BORDONARO, O. in press. Cyclic carbonate platform evolution controlled by major sea-level changes: the Lower/Middle Cambrian La Laja Formation, western Argentine Precordillera. Journal of" South American Earth Sciences. & DICKERSON,P. W. 1996. The missing continent of Llanoria - Was it the Argentine Precordillera? XIII Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, V, 355-368. -& FLOGEL, E. 1996. Early Ordovician reefs from Argentina: Stromatoporoid vs. stromatolite origin. Facies, Erlangen, 34, 177-192. --, BUGGISCH, W. & BERCOWSKI, F. 1989. Facies and sedimentology of Upper Cambrian shallowing-upward cycles in the La Flecha Formation (Argentine Precordillera). Zentralblatt.[fir Geologie und Paldontologie, 1, 999-1011. & LEHNERT, O. 1998. The stratigraphical record of the Argentine Precordillera and its plate-tectonic background. This volume. - - , EBERLEIN, S. & LEHNERT, O. 1993. Sedimentology of Middle Ordovician carbonates in the Argentine Precordillera: Evidence for regional relative sea-level changes. Geologische Rundschau, 82, 362-377. -

-

104

P. W. D I C K E R S O N & M. K E L L E R

KEPPIE, J. D. & ORTEGA-GUTII~RREZ,F. 1995. Provenance of Mexican terranes: Isotopic constraints. International Geology Review, 37, 813-824. KING, P. B. 1937. Geology of the Marathon region. US Geological Survey Professional Paper 187. - - 1 9 8 0 . Geology of the eastern part of the Marathon Basin, Texas'. US Geological Survey Professional Paper 1157. KOLATA, D. R., HUFF, W. D., BERGSTROM, S. M. & CINGOLANI, C. A. 1994. Ordovician K-bentonite beds discovered in the Precordillera of Argentina (abstract). Geological Society qf America, Abstracts with Program, 26, 503. KONG, F. & LAWYER, L. A. 1995. An overview of the opening of the Japan Sea Timing, pattern and mechanism. The University of Texas at Austin, Institute for Geophysics, Report 101-0695. LAJOIE, J. & STIX, J. 1992. Volcaniclastic rocks. In: WALKER, R. & JAMES, N. (eds) Facies ModelsResponse to Sea-Level Change. Geological Association of Canada, I0t-117. LEMONE, D. V. 1988a. Precambrian and Paleozoic stratigraphy, Franklin Mountains, West Texas. ln: HAYWARD, O. T. (ed.) Centennial Field Guide 4. South-Central Section, Geological Society of America, 387-394. --1988b. Pulchrilamina, the Early Ordovician labechiid stromatoporoid and its mound. In: LEMONZ, D. V. (ed.) Franklin Mountains, Tobosa Mountains, related basins. E1 Paso Geological Society & American Association of Petroleum Geologists, Southwest Section, guidebooks, 142-156. LOSKE, W. P. 1992. Sedimentologie, Herkunft und geotektonische Entwicklung pal~iozoischer Gesteine der Prfikordillere West-Argentiniens. Mu)whener Geologische HeJ?e, 7, 1-155. MCBRIDE, E. F. 1989. Stratigraphy and sedimentary history of pre-Permian Paleozoic rocks of the Marathon uplift. In: HATCHER,R. D. JR, THOMAS, W. A. & VIELE, G. W. (eds) The AppalachianOuachita Orogen in the United States. Geological Society of America, Decade of North American Geology, F2, 603-620, MCDONOUGH, M. R., RAMOS, V. A., ISACHSEN,C. E., BOWRIN6, S. A. & VUJOVICH, G. I. 1993. Nuevas edades de circones del basamento de la Sierra Pie de Palo, Sierras Pampeanas Occidentales de San Juan: Sus implicancias para los modelos del supercontinente proterozoico de Rodinia. XII Congreso Geol6gico Acgentino y H Congreso de Exploraci6n de Hidrocaburos, Mendoza, Actas, III, 340-342. MISER, H. T. 1920. Llanoria, the Paleozoic land area in Louisiana and eastern Texas. American Journal of Science, 5th series, 2, 61-89. MOSHER, S. 1993. Western extensions of Grenvilleage rocks, Texas. In: REED, J. C. JR, B~CKVORD, M. E., HOUSTON, R. S., LINK, P, K., RANKIN, D. W., SIMS, A. K. & VAN SCHMUS, W. R. (eds) Precambrian- Conterminous US. Geological Society of America, Decade of North American Geology, C2, 365-378.

MUEHLBERGER, W. R. 1995. Structure and stratigraphy of Paleozoic rocks exposed in the Solitario, Big Bend region, Texas (abstract). Geological Society of America, Abstracts with Programs, 27, 460. ----1996. Paleozoic Stratigraphy and Paleozoic Structure. In: HENRY, C. D., & MUEHLBERGER, W. R. (eds) Geology of the Solitario Dome, TransPecos Texas." Paleozoic, Mesozoic, and Cenozoic Sedimentation, Tectonism, and Magmatism. The University of Texas, Bureau of Economic Geology, Reports of Investigations, 240, 10-29. - - , DENISON, R. E. & LIDIAK, E. G. 1967. Basement rocks in continental interior of the United States. American Association of Petroleum Geologists Bulletin, 51, 2351 2380. NICHOLAS, R. L. & ROZENDAL,R. A. 1975. Subsurface positive elements within Ouachita foldbelt in Texas and their relation to Paleozoic cratonic margin. American Association of Petroleum Geologists Bulletin, 59 193-216. ORTEGA-GUTII~RREZ, F. 1994. Tectonostratigraphic interpretation of the Paleozoic Acatlan complex of southern Mexico and the problem of its regional correlation (summary). First CircumPacific and Circum-Atlantinc Terrane Conference, Proceedings, 107-109. PANKHURST, R. J., RAPELA, C. W., SAAVEDRA, J., BALDO, E., DAHLQUIST, J. & PASCUA, I. 1996. Sierras de Los Llanos, Malanzfin and Chepes: Ordovician I- and S-type granitic magmatism in the Famatinian orogen. XIII Congreso Geol6gico Argentino y III Congreso de Exploraci6n de Hidrocarburos, Buenos Aires, Actas, V, 415. & FANNING, C. M. '1998. F'amatinian magmatic arc in the central Sierras Pampeanas. This volume. PEARCE, J. A., HARRIS,N. B. W. & TINDLE,A. G. 1984. Trace-element discrimination diagrams for the tectonic interpretation of granitic rocks. Journal of Petrology, 25, 956-983. RAMOS, V. A. 1984, Patagonia: Un continente Paleozoico a la deriva? 1X Congreso Geol6g&o Argentino, Buenos Aires, Actas, II, 3ll 325. ----1988. Late Proterozoic-Early Paleozoic of South America: A collisional history. Episodes, 11, 168-173. - - 1 9 9 5 . Sudamerica: Un mosaico de continentes y oceanos. Ciencia Hoy, 6, 24 29. --, VUJOVICH, G. I., KAY, S. M. & MCDONOUGH, M. R. 1993. La orogenesis de Grenville en las Sierras Pampeanas Occidentales: La Sierra Pie de Palo y su integraci6n al supercontinente Proterozoico. XII Congreso Geol6gico Argentino y H Congreso de Exploraci6n de Hidrocarburos, Mendoza, Actas, III, 343-357. RAPELA, C. W., COIRA, B., TOSELLI,A. & SAAVEDRA,J. 1992. El magmatismo del Paleozoico Inferior en el sudoeste de Gondwana. In: GUT1ERREZ-MARCO,J. C., SAAVEDRA,J. & RABANO, I. (eds) Paleozoico Inferior de Ibero-Am&ica. Universidad de Extremadura, Spain, 21-68.

THE PRECORDILLERAN REID, R. P., CAREY, S. N., STAERKER,T. S. & ASK, M. 1994. Sedimentology of the collision zone between the North D'Entrecasteaux Ridge and New Hebrides island arc (Sites 827, 828, and 829). In: COLLOT, J.-Y., GREENE, H. G., STOKKING, L. B. et al. Scientific Results, Ocean Drilling Program, Leg 134. College Station, Texas, Ocean Drilling Program, 73 88. , & Ross, D. R. 1996. Late Quaternary sedimentation in the Lesser Antilles island arc. Geological Society of America Bulletin, 108, 78-100. RuIZ, J., PATCHETT, P. J. & ORTEGA-GUTII~RREZ, F. 1988. Proterozoic and Phanerozoic basement terranes of Mexico from Nd isotopic studies. Geological Society of America Bulletin, 100, 1197-1200. SCHLAGER, W. 1989. Drowning unconformities on carbonate platforms, hT: CREVECLO, P. D., WILSON, J. L., SARG, J. F. & READ, J. F. (eds) Controls" on carbonate platforms and basins. SEPM, Special Publications, 44, 15 25. SELLARDS, E. H. 1932. Pre-Paleozoic and Paleozoic systems. In: The Geology o[ Texas, Volume 1 Stratigraphy. The University of Texas at Austin, Bulletin, 3232, 235-236. SPALETTI, L. A., CINGOLANI, C. A., VARELA, R. & CUERDA, A. J. 1989. Sediment gravity-flow deposits of an Ordovician deep-sea fan system (western Precordillera, Argentina). Sedimentw3' Geology, 61,287-301. SPARKS, R. S. J., SIGURDSSON, H. & CAREY, S. N. 1980a. The entrance of pyroclastic flows into the sea, I. Oceanographic and geologic evidence from Dominica, Lesser Antilles. Journal of Volcanology and Geothermal Research, 7, 87-96. & 1980b. The entrance of pyroclastic flows into the sea, II. Theoretical considerations on subaqueous emplacement and welding. Journal of Volcanology and Geothermal Research, 7, 97 105. TAUVERS, P. R. 1988. Basement-influenced deformation in the Marathon fold-thrust belt, west Texas. Journal of Geology, 96 197-214. THOMAS, W. A. 1991. The Appalachian-Ouachita rifted margin of southeastern North America. Geological Society q[ America Bulletin, 103, 415-431. ,

-

-

ODYSSEY

105

1996. Rifting of the Argentine Precordillera from the Ouachita embayment of the Laurentian margin. Geological Society of America, Abstracts with Program, 28, 66. - - - & ASTINI, R. 1996. Asymmetric conjugate rift margins of the Ouachita rift and the Argentine Precordillera. Geological Society of America, Abstracts with Program, 28, 66. VIELE, G. W. 1989. The Ouachita orogenic belt. In: HATCHER, R. D. JR, THOMAS, W. A. & VIELE, G. W. (eds) The Appalachian-Ouachita Orogen in the United States. Geological Society of America, Decade of North American Geology, F2, 555-562. & THOMAS, W. A. 1989. Tectonic synthesis of the Ouachita orogenic belt. In: HATCHER, R. D. JR, THOMAS, W. A. & VIELE, G. W. (eds) The Appalachian-Ouachita Orogen in the United States. Geological Society of America, Decade of North American Geology, F2, 695-728.VON GOSEN, W. 1992. Structural evolution of the Precordillera (Argentina): The Rio San Juan section. Journal of Structural Geology, 14, 643-667. VUJOVICH, G. I., MILLER, H. & RAMOS, V. A. 1994. Proterozoic metavolcanics from western Sierras Pampeanas terrane, Argentina. Journal of South American Earth Sciences, 7, 309-323. WEAVER, C. E. 1961, Clay minerals of the Ouachita structural belt and adjacent foreland. In: FLAWN, P. T., GOLDSTEIN, A. JR, KING, P. B. & WEAVER, C. E. (eds) The Ouachita System. The University of Texas at Austin, Bureau of Economic Geology, Publication 6120, 147-162. WILSON, J. L. 1954. Ordovician stratigraphy in the Marathon folded belt, West Texas. American Association of Petroleum Geologists Bulletin, 38, 2455-2475. YOUNG, L. M. 1969. Stratigraphy and sedimentology of the Marathon Formation. In: MCBRIDE, E. F. (ed.) Stratigraphy, sedimentary structures and origin of flysch and pre-flysch rocks of the Marathon Basin, Texas. Dallas Geological Society, guidebooks, 32 42. 1970. Early Ordovician sedimentary history of the Marathon geosyncline, Trans-Pecos Texas. American Association of Petroleum Geologists Bulletin, 54, 2303 2316. -

-

Ordovician K-bentonites in the Argentine Precordillera: relations to Gondwana margin evolution WARREN

D. H U F F ~, S T I G M. B E R G S T R O M 2, D E N N I S

CARLOS

A. C I N G O L A N I 4 A N D

RICARDO

R. K O L A T A 3,

A. A S T I N I 5

1Department o f Geology, University of Cincinnati, Cincinnati, O H 45221, USA 2 Department o f Geological Sciences, The Ohio State University, 155 S. Oval Mall, Columbus, O H 43210, USA 3 Illinois State Geological Survey, 615 E. Peabody Dr., Champaign, IL 61820, USA 4 Centro de Investigaciones Geol6gicas, Universidad Nacional de La Plata, Calle 1, No. 644, 1900 La Plata, Argentina 5 CONICET, C(ttedra de Estratigrafia y Geologia Hist6rica, Facultad de Ciencias Exactas, Fisicas y Naturales, Universidad Nacional de C6rdoba, Av. V~lez S~rsfield 299, CC#395, 5000 C6rdoba, Argentina Abstract: Ordovician K-bentonites have now been recorded from >20 localities in the vicin-

ity of the Argentine Precordillera. Most occur in the eastern thrust belts, in the San Juan Limestone and the overlying the Gualcamayo Formation, but a few ash beds are known also from the central thrust belts. The oldest occur in the middle Arenig I. victoriaelunatus graptolite (Oe. evae conodont) Zone, and the youngest in the middle Llanvirn P. elegans (P. suecicus) Zone. Mineralogical characteristics, typical of other Ordovician K-bentonites, include a matrix of illite/smectite mixed-layer clay and a typical felsic volcanic phenocryst assemblage: biotite, beta-form quartz, alkali and plagioclase feldspar, apatite, and zircon, with lesser amounts of hornblende, clinopyroxene, titanite and Fe-Ti oxides. The proportions of the mineral phases and variations in their crystal chemistry are commonly unique to individual (or small groups of) K-bentonite beds. Glass melt inclusions preserved in quartz are rhyolitic in composition. The sequence is unique in its abundance of K-bentonite beds, but a close association between the Precordillera and other Ordovician sedimentary basins cannot be established. The ash distribution is most consistent with palaeogeographical reconstructions in which early Ordovician drifting of the Precordillera occurred in proximity to one or more volcanic arcs, and with eventual collision along the Andean margin of Gondwana during the mid-Ordovician Ocloyic event of the Famatinian orogeny. The Puna-Famatina terrane northeast of the Precordillera might have served as the source of the K-bentonite ashes, possibly in concert with active arc magmatism on the Gondwana plate itself.

Lower Palaeozoic K-bentonites (altered volcanic ashes) are widely distributed on the continents bordering the former Iapetus Ocean, especially in northwestern Europe and eastern North America. Local and regional studies during the last decade have added a wealth of new information about their geographical and stratigraphical distribution, geochemistry, and mineralogy as well as their tectono-magmatic and palaeogeographical significance (Kolata et al. 1996; Bergstr6m et al. 1997; Huff et al. 1998). In South America, K-bentonites are as yet unknown in the Cambrian and Silurian and were discovered in the Ordovician only a few years ago (Huff et al. 1995a). During the past few years, many Lower and Middle Ordovician K-bentonite beds have been found in the Precordillera of western Argentina; indeed, this region has some of the most abundant ash beds

known from that period anywhere in the world. K-bentonite geochemistry may provide highly significant information about the tectono-magmatic nature of the source area, and the distribution patterns of individual ash beds or complexes of such beds may shed additional light on the former positions of continental plates. Specific chemical and mineralogical features of primary phenocrysts can serve as important criteria for regional stratigraphical correlation of beds and bed sequences, whether expressed as elemental ratios of individual grains or in the immobile element chemistry of bulk K-bentonite samples (Kolata et al. 1987; Huff et al. 1992; Bergstr6m et al. 1995). K-bentonite beds also provide important evidence regarding the timing and location of orogenic events because they typically originate from source volcanoes situated at or near tectonically active plate margins (Kolata et al.

HUFF, W. D., BERGSTR6M, S. M., KOLATA, D. R. et al. 1998. Ordovician K-bentonites in the Argentine Precordillera: relations to Gondwana margin evolution. In: PANKHURST,R. J. & RAPELA,C. W. (eds) The ProtoAndean Margin of Gondwana. Geological Society, London, Special Publications, 142, 107-126.

108

W. D. HUFF E T A L .

1987; Huff et al. 1992; Bergstr6m et al. 1995). Further, dating of K-bentonite beds or complexes of beds provides precise age dates on periods of volcanism associated with continental margin subduction events. Therefore, when trying to unravel the pre-Andean evolution of the western Gondwana margin it is appropriate to consider the various types of evidence provided by the K-bentonites. The first step in any such evaluation is to explore the vertical and horizontal distribution of these ashes, both locally and regionally, and to clarify their age and distribution patterns. Without establishing the position of these beds in a reliable biostratigraphical framework, no meaningful regional comparisons can be made, and the assessment of their palaeogeographical significance is seriously hampered. In view of this, the first part of the present contribution will assess the stratigraphical and geographical distribution of the Ordovician K-bentonites in the Argentine Precordillera. We will also make a comparison with the geographical and stratigraphical distribution of Ordovician K-bentonite complexes in North America (Laurentia) to examine whether or not they are likely to represent the same ash falls. If they do, it would obviously suggest rather close proximity of Laurentia to the Precordillera during the time of the ash fall (Dalziel 1997). The absence of correlative beds on these two land masses would be consistent with models which argue for a wide Iapetus at the time of eruption (Astini et al. 1995; Thomas & Astini 1996), unless prevailing wind direction severely skewed ash distribution. Thus, K-bentonites may serve as a means of testing palaeogeographical reconstructions and models of the Iapetus Ocean. A second part of the present contribution deals with the mineralogy and geochemistry of Precordilleran K-bentonites and their tectonomagmatic significance. We show that melt inclusion composition and the whole rock ratios of immobile elements provide evidence for collision-margin volcanism. A final part summarises the evidence at hand bearing on the evolution of the Gondwana margin, and the Ordovician geological history of the Precordillera. Our investigations of the Argentine Ordovician K-bentonites are continuing and some of the data presented below are of a somewhat preliminary nature.

Occurrence Since their discovery in the Precordillera in 1994, Ordovician K-bentonites have been recorded from more than 20 localities in a region extending about 250 km in a north-south direction across

parts of San Juan and La Rioja Provinces. Some of the principal localities are shown in Fig. l. No ash beds have been observed in the Ordovician outcrops in the San Rafael region some 200 km south of Mendoza. Likewise, as noted by Bergstr6m et al. (1996), no K-bentonite beds are known from the locally rather strongly tectonized slope and basin successions in the western Precordillera, such as the well-exposed Middle Ordovician sequences near the northern end of Sierra de la Invernada about 40km northeast of Talacasto (Fig. 1), or in the outcrops along the mountain front from Calingasta northwards to the Jfichal River. Most of the known K-bentonites occur in the eastern thrusts of the thrust-and-fold belt, where they are quite common in the upper section of the San Juan Limestone, characterized by massive platform carbonates with an early Palaeozoic Laurentian fauna, and in the overlying deep water graptolitic shales of the Gualcamayo Formation. A few ash beds are also known from the central thrusts. For convenience, we will briefly review the geographical and stratigraphical distribution patterns in terms of three main distribution areas, namely the Guandacol region in the northernmost Precordillera, the Jfichal region some 100 km to the south, and the San Juan region about 150kin still farther to the south (Fig. 1). Only a few key sections are discussed here; for further information, see Bergstr6m et al. (1996). The oldest recorded Palaeozoic K-bentonites in the Precordillera occur in the middle Arenig Oe. evae Zone (Htinicken & Sarmiento 1982, 1985) in the topmost part of the Lower Ordovician San Juan Limestone in outcrops along the Gualcamayo River in the Guandacol region. In the overlying, about 200m thick, Gualcamayo Formation, which ranges from the Upper Arenig to at least the Middle and probably Upper Llanvirn, there are numerous ash beds, especially along a small tributary to the Guandacol River (26~ 68~ where more than 170 separate ash beds have been observed. This is by far the most extensive suite of Ordovician K-bentonites recorded from an outcrop anywhere in the world and ranks slightly above the Lower Silurian K-bentonite succession at Dob's Linn in Scotland which contains 135 beds (Merriman & Roberts 1990). Most (c. 85) of the ash beds, which range in thickness from less than 1 mm to more than 50 cm, occur in the upper Arenig U. a u s t r o d e n tatus Zone (Brussa & Astini 1997) but many beds are present also in slightly older and slightly younger strata. Excellent sections of coeval strata, but with fewer K-bentonite beds, are present at the Quebrada de Los Saltitos and

PRECORDILLERA K-BENTONITES L A RIOJA PROVINCE

t> JACHAL

N

fg

.IIJ,"

MOGOTESAZULES JACHAL RIVER

LAS "CHACRITASe,

CERRO LA CHILCA AN ICE rO

TA SAN JUAN SAN JUAN

PAVER

A

~J

..J r~

~

J

i

MENDOZA PROVINCE 0

2o !

1

km

i~vEg

r

9

MENDOZA

109

in the Quebrada Nazareno, a short tributary branch between Quebrada Las Plantas and Quebrada de Potrerillos in the Guandacol area (Fig. 1). In the Jfichal area, there are about 30 individual K-bentonite beds in the Los Azules Formation (equivalent to the Gualcamayo Formation elsewhere) at Cerro Viejo (Huff et al. 1995a; Bergstr6m et al. 1996; Cingolani et al. 1997) and up to a dozen such beds have also been recorded from the upper part of the underlying San Juan Limestone. A concordant U - P b zircon age from Cerro Viejo was reported by Huff et al. (1997b) as 464 i 2 Ma which they considered to be the age of the base of the U. austrodentatus Zone. Beds of the latter complex, which post-date the oldest K-bentonite beds in the Guandacol region, are well exposed at many localities, such as Cerro Potrerillo, Cerro La Chilca, and La Silla (Fig. 1), and also in the still poorly dated successions at Las Chacritas (30~ 68~ (Albanesi & Astini 1994; Astini 1994) and Mogotes Azules (30~ 68~ Multiple K-bentonite sequences are present at all of these localities but none are as extensive as the middle Arenig K-bentonite complex in the Guandacol region. The most accessible K-bentonite locality in the San Juan region is in a highway road cut at Quebrada de Talacasto about 55km northnorthwest of San Juan. As described by Bergstr6m et al. (1996) there are 30 individual ash beds, the thickest ones 10-15 cm, within a 1101n thick succession of the upper San Juan Limestone. At least the upper portion of this complex is probably comparable with the K-bentonite complex in the Gualcamayo Formation at Cerr~ Viejo. The considerable thickness of some beds at the latter locality may be due to the fact that the section is more condensed and some of the thick beds may be composed of several, in time relatively closely spaced, ash falls that are represented by separate ash beds in the thick Talacasto succession. A single bed about 50m below the top of the San Juan Limestone at the Tambolar section in the San Juan River valley (Lehnert 1995; Bergstr6m et al. 1996) is one cf the oldest K-bentonite beds known from the Precordillera. Fig. 1. Location map of the Precordillera, Argentina, showing some of the principal localities where K-bentonites are exposed. The inset map of Argentina shows the position of the Precordillera. The areas in black are the Ordovician outcrop areas of the eastern and central thrust belts, and the stippled areas represent the Ordovician outcrop areas of the western thrust belts. Numbered locations refer to the principal sections used in this study.

110

W. D. HUFF E T A L . or Upper Ordovician strata9 Figure 2 is a n o r t h south transect showing the main patterns of K-bentonite bed distribution. The high concentration of ash beds in the upper Arenig-lower Llanvirn in the Guandacol region may suggest the general direction toward the location of the

In terms of stratigraphical distribution, the oldest K-bentonites currently known in the Precordillera are of middle Arenig age and the youngest are of middle to upper Llanvirn age (Bergstr6m et al. 1996)9 No ash beds have been observed in younger parts of Middle Ordovician

CERRO VIEJO

GUALCAMAYO RIVER GRAP. ZONE TRAPICHE

CON. ZONE

tere-

tlusculus serra

GUALCAM ~1

eteganz"

TALACASTO

AYO

SILURIAN ~ c3 Z

GUALCAMAYO austrodenta -

~

~

i

us

suecicus

~varia~

~

bilis

e~

Onco grap-

us maxl. mu$

2 Jail2

3

evae

SAN JUAN

020 km

1

Fig. 2. A north-south transect from Guandacol to Talacasto showing the stratigraphic position of the K-bentonite sequences. The north to south transgressive nature of the graptolitic shales of the Gualcamayo Formation (Astini et al. 1995) is well documented by the change in facies containing the K-bentonite beds. The column on the right lists the applicable conodont zones.

TALACASTO SILURIAN

SANJUAN

LAS CHACRITAS , ~SILURIAN LAS ~GUADITAS

I

GUAL-

""

i 0 SAN CAMAYO JUAIN I ! ,.~ ...

MOGOTESAZULES ~SILURIAN LAS ~.GUADITAS GUAI.~A-

! ~32

3

Jachai II

1

MAYO I ~2~ SANJUAN ~