2 Be ™) NOZIO Bulletin of the British Museum (Natural History) A Global Analysis of the Ordovician— Silurian boundary Edited by L. R. M. Cocks & R. B. Rickards Geology Vol 43 28 April 1988 The Bulletin of the British Museum (Natural History), instituted in 1949, is issued in four scientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology, and an Historical series. Papers in the Bulletin are primarily the results of research carried out on the unique and ever-growing collections of the Museum, both by the scientific staff of the Museum and by specialists from elsewhere who make use of the Museum’s resources. Many of the papers are works of reference that will remain indispensable for years to come. Parts are published at irregular intervals as they become ready, each is complete in itself, available separately, and individually priced. Volumes contain about 300 pages and several volumes may appear within a calendar year. Subscriptions may be placed for one or more of the series on either an Annual or Per Volume basis. Prices vary according to the contents of the individual parts. Orders and enquiries should be sent to: Publications Sales, British Museum (Natural History), Cromwell Road, London SW7 5BD, England. World List abbreviation: Bull. Br. Mus. nat. Hist. (Geol.) © British Museum (Natural History), 1988 The Geology Series is edited in the Museum’s Department of Palaeontology Keeper of Palaeontology: DrL.R.M. Cocks Editor of the Bulletin: Dr M. K. Howarth. Assistant Editor: Mr D. L. F: Sealy ISBN 0 565 07020 7 ISSN 0007-1471 M3 Ta wef y Geology series British Museum (Natural History): * Vol 43 complete Cromwell Road ‘ we oe London SW7 5BD ser ana Issued 28 April 1988 A Global Analysis of the Ordovician— ae Silurian boundary Edited by L. R. M. Cocks Department of Palaeontology, British Museum (Natural History), Cromwell Road, London SW7 5BD and R. B. Rickards Sedgwick Museum, Downing Street, Cambridge CB2 3EQ Bulletin British Museum (Natural History) Geology Series Vol. 43 International Union of Geological Sciences Sponsored Publication The papers incorporated in this volume represent contributions from the International Work- ing Group on the Ordovician-Silurian Boundary, a constituent body of the International Commission on Stratigraphy within the International Union of Geological Sciences. Contents GRO CECH OTM years, dorsie aise nierere sersneie nierate oie eters ets L. R. M. Cocks & R. B. Rickards The Ordovician-Silurian Boundary and its Working Group ............ L. R. M. Cocks Ordovician—Silurian Boundary Sections EUROPE: Dob’s Linn — the Ordovician-Silurian Boundary Stratotype .......... S. H. Williams Conodonts from the Ordovician-Silurian Boundary Stratotype, Dob’s Linn, Scotland C. R. Barnes & S. H. Williams Preliminary arcritarch and chitinozoan distributions across the Ordovician—Silurian boundary stratotype at Dob’s Linn, Scotland ........................... G. M. Whelan Ordovician—Silurian junctions in the Girvan District, S.W. Scotland ...D. A. T. Harper Base of the Silurian in the Lake District and Howsgill Fells, Northern England R. B. Rickards The Ordovician-Silurian boundary at Keisley, Cumbria ................... A. D. Wright Ordovician-Silurian boundary strata in Wales ..................22..eeeeee eee J. T. Temple La Limite Ordovicien-Silurien en France .... C. Babin, R. Feist, M. Melou & F. Paris The Ordovician-Silurian boundary in the Oslo region, Norway ........ L. R. M. Cocks ast baltiCMRESIOMM ns sise.s0 sect seeneeecicees cee e natets D. Kaljo, H. Nestor & L. Pélma The Ordovician-Silurian boundary in Poland ...................0.0....00eeeeeee L. Teller The Ordovician-Silurian boundary in the Prague Basin, Bohemia .......... P. Storch The Ordovician-—Silurian boundary in the Saxothuringian Zone of the Variscan Orogen H. Jaeger The Ordovician-Silurian boundary in the Carnic Alps of Austria ...... H. P. Schonlaub ASIA: The Ordovician-Silurian boundary in China ....................0000eeeeee ees Mu En-zhi The Ordovician-Silurian boundary beds of the north-east U.S.S.R. T. N. Koren, M. M. Oradovskaya & R. F. Sobolevskaya The Ordovician-Silurian boundary in the Altai Mountains, U.S.S.R. E. A. Yolkin, A. M. Obut & N. V. Sennikov Nature of the Ordovician—Silurian boundary in south Kazakhstan, U.S.S.R. M. K. Apollonov, T. N. Koren, I. F. Nikitin, L. M. Paletz & D. T. Tsai The Ordovician-Silurian boundary in Saudi Arabia ...................... H. A. McClure AFRICA AND AUSTRALASIA: The Ordovician-Silurian boundary in Morocco............ J. Destombes & S. Willefert The Ordovician-Silurian boundary in the Algerian Sahara................... P. Legrand The Ordovician-Silurian boundary in Mauritania .........................05- S. Willefert Ordovician-Silurian boundary in Victoria and New South Wales, Australia A. H. M. Vandenberg & B. D. Webby The base of the Silurian System in Tasmania ..................220e0eeeee eee M. R. Banks AMERICA: Stratigraphy and Palaeontology of the Ordovician-Silurian boundary interval, Anticostiolslands @uebecs @amaday ernest sseleretele-leleielele eteleter= C. R. Barnes Graptolites at and below the Ordovician-Silurian boundary on Anticosti Island, (CANAD 354550000 s0008de neneaadoOdetbasbeccutanon seme HasepoeeaoonunEUunoDTnaapcmrrane J. Riva ered, Oiieloe, (CAME .osoccenss vg duoc nb ocne coocospsaedaas coop padEboooE P. J. Lespérance The Ordovician-Silurian boundary on Manitoulin Island, Ontario, Canada C. R. Barnes & T. E. Bolton Preliminary report on Ordovician-Silurian boundary rocks in the Interlake area, Mianittobam Canad ammeeprrerceereree cere ec Gernecenim cc romances H. R. McCabe nN 145 165 7/7) 183 191 195 211 239 247 255 CONTENTS The Ordovician-Silurian boundary in the Rocky Mountains, Arctic Islands and HudsonuPlatfonmsCanadauyeeereseseraseeo tea eee eee a eee cere B. S. Norford Ordovician-Silurian boundary, northern Yukon, Canada A. C. Lenz & A. D. McCracken The Ordovician-Silurian boundary in the United States S. M. Bergstrom & A. J. Boucot The Ordovician-Silurian boundary in South America ....................... A. J. Boucot The Ordovician-Silurian boundary in Bolivia and Argentina A. Cuerda, R. B. Rickards & C. Cingolani The Ordovician-Silurian boundary in the Sierra de Villicum, Argentine Precordillera B. A. Baldis & E. D. Pothe de Baldis Palaeobiology and Environmental changes Late Ordovician and Early Silurian Acritarchs ...................6...0eeeeeeeeee F. Martin Brachiopods across the Ordovician-Silurian boundary .................. L. R. M. Cocks Chitinozoan stratigraphy in the Ashgill and Llandovery ....................... Y. Grahn Conodont biostratigraphy of the Uppermost Ordovician and Lowermost Silurian C. R. Barnes & S. M. Bergstrom Graptolite faunas at the base of the Silurian ..........................000- R. B. Rickards Land plant spores and the Ordovician-Silurian boundary ....................... J. Gray TODOS Wes 8, aie. tenes Aen eee Ee OE CACM ARSE oN oOnIROE SS P. J. Lespérance Environmental changes close to the Ordovician-Silurian boundary ..... P. J. Brenchley Introduction L. R. M. Cocks & R. B. Rickards The base of the Silurian System was agreed by the I.U.G.S. Executive Committee in May 1985 (published June 1985 in Bassett 1985), and was taken at the base of the acuminatus Zone at Dob’s Linn, Scotland (Cocks 1985). This volume closely reflects the achievements of the Ordovician—-Silurian Boundary Working Group from its formation in 1974 to its disbandment in 1985. A detailed account of the activities of the Group is given in the next chapter, including the procedures followed which led to the decision on the definition of the boundary. We have taken the opportunity to gather in this book a global review of the Ordovician—Silurian boundary. These contributions are partly based on submissions on places and fossil groups made during the lifetime of the Working Groups and circulated by the Secretary, but these, if used in this volume, have been thoroughly updated by the respective authors and their colleagues. In addition we have commissioned a number of papers to give an overview of the many places where the boundary is exposed, as well as others on the global analysis of sedimentary events, and the evolutionary progress of the most important biological groups across the boundary. It has always been clear from discussions that unanimous agreement would never be poss- ible. Different countries have different traditions and philosophies, for example with respect to stratigraphical principles. This is especially true of the concepts of zones, and of the utility of zones for correlative purposes. For example, Mu (this volume) attempts a very detailed correla- tion of what are regarded elsewhere as potential subdivisions of the acuminatus Zone, claiming that an ascensus fauna underlies the acuminatus Zone (as it is, indeed, seen in China). But in some of the most precisely and exhaustively collected sections, such as at Dob’s Linn, Scotland, it seems clear that the two species appear more or less simultaneously, albeit with ascensus more abundant low in the zone, and acuminatus more common in the upper part of the zone and outlasting ascensus. Thus, whilst there is a case for locally subdividing an acuminatus Zone, as Teller (1969) and others have sensibly done, it should be made clear that on current information these subdivisions correlate in total with the acuminatus Zone at Dob’s Linn. In sections where the record is perhaps not very complete, or the fauna not abundant, it may appear that acuminatus follows ascensus. Barnes (this volume) considers that, although the systemic boundary has now been fixed, its ‘reconsideration may be necessary’ (Lespérance et al. 1987). The main grounds for this opinion are that the Anticosti sequence has a future potential for further studies; has all the attributes for a boundary stratotype; and that ‘important stratigraphic principles have been disregarded or overruled in making the final stratotype decision’. It cannot be overemphasized that the procedures adopted by the Working Party Group throughout its life were correct, proper, democratic, and always in accord with I.U.G:S. guidelines and with specific guidance from LU.GS. If some stratigraphical ideas have been disregarded or overruled, then a substantial majority of the Working Group took the decisions to do so: the voting which took place is recorded in the next section. ‘Potential’ is always a difficult commodity to evaluate: and the judged poten- tial of a section cannot delay for ever what will always be arbitrary decisions in the end. By the time a reconsideration was worked through (? ten years) another section would no doubt be vying with Anticosti in terms of its potential. Where then? That Anticosti has most of the attributes necessary for a boundary stratotype is beyond question. That is why it was on a short list of two, voted upon by the Working Group. Other sections were of an almost equally high standard, for example, in China and the Lake District of England. But Anticosti does have one very serious drawback in any current discussions on Bull. Br. Mus. nat. Hist. (Geol) 43: 5—7 6 COCKS & RICKARDS correlations about the boundary, and that is its seemingly poor record of graptolites. It may be that at some future time graptolites may be relatively demoted in value for correlative purposes, but that time is still far away on present information. Dob’s Linn also has most of the attributes of a boundary stratotype, and the Working Group, after eleven years of study, considered it better than Anticosti. In fact, the boundary has now been certainly put at the correct level, using the best group for correlation, the graptolites. Despite the fact that the Hirnantia brachio- pod fauna is very often overlain by persculptus Zone graptolites, unequivocal evidence from both Kazakhstan (Koren et al. this volume) and the Lake District of England (Cocks this volume) shows that it also occurs rarely within the persculptus Zone. There is a strong feeling amongst most biostratigraphers that they prefer to regard the Hirnantia fauna as Ordovician rather than Silurian in age and not straddling the systematic boundary, and this assignment to the Ordovician can be achieved only by a sub-acuminatus Zone boundary, as was eventually decided. A more interesting question is the precise age, in terms of graptolite zones, of the maximum glacio-eustatic drop in sea level, and this is still not yet definitively answered although it was probably about half way through the persculptus Zone—there are some well-dated persculptus bearing post-glacial transgressive beds in parts of North Africa. On the other hand, the precise duration and extent of the glacial episode (Fig. 1) certainly varied from place to place— commencing even in late Caradoc and early Ashgill times in some parts of Gondwana, and certainly continuing into post-Hirnantia fauna times, perhaps into the Rhuddanian, in others, e.g. South Africa. It is also important to note that detailed investigation indicates that the ‘end Hirnantia fauna Mucronaspis tillites etc. Glacial directions Fig. 1 Distribution of the latest Ordovician glacial deposits in Gondwana and adjacent areas (after Cocks & Fortey 1988). INTRODUCTION 7 Ordovician’ faunal extinctions were by no means synchronous. No other faunal or floral group than graptolites yet approaches the sensitivity and exactness of the graptolites during the period in question—for example from the mid-Ashgill (base of the Rawtheyan) to the end of the early Llandovery (Rhuddanian) there are no fewer than eight graptolite zones, as compared with three or four conodont zones, and four successive brachiopod faunas, three or four ostacod faunas, three or four trilobite faunas etc. This, from a period of only perhaps 7 or 8 million years (McKerrow et al. 1985), makes the graptolites compare well with Mesozoic ammonites or Tertiary foraminifera as a precise dating tool. The coverage in this volume of the Ordovician-—Silurian sections themselves cannot be total partly because several regions are little known. However, it is worth drawing attention here to probable additional Ordovician—Silurian boundary sections in Libya (Klitzsch 1981), Burma (Mitchell et al. 1977; Wolfart et al. 1984) and Greenland (e.g. Hurst & Kerr 1982; Surlyk & Hurst 1984). In addition we are aware of preliminary work on strata about the boundary in Vietnam, Thailand, Malaysia and other parts of SE Asia. In the instance of central Nevada, U.S.A., we have not republished a revised preliminary submission because there is nothing yet new to add to the work by Berry (1986). There is also further work in preparation on Scandina- via. We would like to end this introduction with a tribute to the many people involved, both as members of the Working Group and as contributors to the present volume, who patiently took part in the meetings, newsletter, activities and final decision-making, and thank them all for their patience, support, good humour and international friendship; despite the controversy of the eventual scientific conclusion. References Bassett, M. G. 1985. Towards a ‘Common Language’ in Stratigraphy. Episodes, Ottawa, 8: 87-92. Berry, W. B. N. 1986. Stratigraphic significance of Glyptograptus persculptus group graptolites in central Nevada, U.S.A. Spec. Publs geol. Soc. Lond. 20: 135-143. Cocks, L. R. M. 1985. The Ordovician-Silurian boundary. Episodes, Ottawa, 8: 98-100. —— & Fortey, R. A. 1988. Lower Palaeozoic facies and faunas round Gondwana. Geol. Soc. Lond. Spec. Publ. (in press). Hurst, J. M. & Kerr, J. W. 1982. Upper Ordovician to Silurian facies patterns in eastern Ellesmere Island and western North Greenland and their bearing on the Nares Strait lineament. Meddel. om Grgn. Geosci. 8: 137-145. Klitzsch, E. 1981. Lower Palaeozoic rocks of Libya, Egypt, and Sudan. In Holland, C. H. (ed.), Lower Palaeozoic of the Middle East, Eastern and Southern Africa, and Antarctica: 131-163. London. Lespérance, P. J., Barnes, C. R., Berry, W. B. N., Boucot, A. J. & Mu En-zhi 1987. The Ordovician— Silurian boundary stratotype: consequences of its approval by I.U.GS. Lethaia, Oslo, 20: 217-222. McKerrow, W. S., Lambert, R. St.J. & Cocks, L. R. M. 1985. The Ordovician, Silurian and Devonian periods. Mem. geol. Soc. Lond. 10: 73-80. Mitchell, A. H. G., Marshall, T. R., Skinner, A. C., Baker, M. D., Amos, B. J. & Bateman, J. H. 1977. Geology and exploration geochemistry of the Yadanatheingi and Kyaukme-Longtawkno areas. North- ern Shan States, Burma. Overseas Geol. Miner. Resour., London, 51: 1-35, pls 1, 2. Surlyk, F. & Hurst, J. M. 1984. The evolution of the early Palaeozoic deep-water basin of North Greenland. Bull. geol. Soc. Am., New York, 95: 131-154. Teller, L. 1969. The Silurian biostratigraphy of Poland based on graptolites. Acta geol. Pol., Warsaw, 19: 393-501. Wolfart, R. et al. 1984. Stratigraphy of the Western Shan Massif, Burma. Geol. Jb., Hannover, (B) 57: 3-92. December 1986 L. R. M. Cocks, Department of Palaeontology, British Museum (Natural History), Cromwell Road, London SW7 SBD. R. B. Rickards, Sedgwick Museum, Downing Street, Cambridge CB2 3EQ. The Ordovician-Silurian Boundary and its Working Group L. R. M. Cocks Department of Palaeontology, British Museum (Natural History), Cromwell Road, London SW7 5BD Synopsis After a brief history of the study and definition of the Ordovician—Silurian boundary in the nineteenth and early twentieth centuries, the process of setting up the Ordovician-Silurian Boundary Working Group is described, together with its progress, publications and final decision-making during the period 1974-1985. Both the Cambrian and the Silurian Systems were established as formal system names by Sedgwick and Murchison respectively amicably enough in 1835, but during the next thirty years it become clear that the upper part of Sedgwick’s Cambrian occupied the same time and space as the lower part of Murchison’s Silurian. It was not until after the deaths of both men that Charles Lapworth in 1879 established the Ordovician System to occupy the chief overlap- ping ground between the older part of the Silurian and the younger part of the Cambrian. In contrast to the rather generalized earlier definitions of the boundaries of the Cambrian and Silurian, Lapworth’s definition of the limits of the Ordovician was admirably precise: he defined the new Ordovician System as the ‘strata included between the base of the Lower Llandovery formation and that of the Lower Arenig’ (Lapworth 1879: 14). There were subse- quently problems (which are still not entirely resolved today) about the position and interna- tional correlation of the ‘base of the Lower Arenig’, but these are the province of the Cambro—Ordovician Boundary Working Group and will not be further discussed here. “The base of the Lower Llandovery’ has been much less ambiguous, and thus in general any dispute surrounding the definition of the Ordovician—Silurian boundary has always been of a much lesser magnitude than the problems of the Cambro—Ordovician and the Siluro—Devonian boundaries. From the time of Murchison onward, ‘the base of the Lower Llandovery was defined primarily in terms of shelly facies and without much precision, and usually recognized by the incoming of various pentameride brachiopods such as Stricklandia. However, following Charles Lapworth’s classic work on the Ordovician and Silurian rocks of Scotland in the period 1870 to 1880, it became clear that the best national and international correlation tool in rocks of those ages was the sequence of graptolite zones, and these zones were subsequently used in practice, with Lapworth himself, and subsequently the great graptolite monograph of Elles & Wood (1901-1918), using the acuminatus Zone (type locality Dob’s Linn, Scotland) as the de facto base of the Silurian. The acuminatus Zone was poorly developed as such in Wales, and so Jones (1909) erected the persculptus Zone (type locality Pont Erwyd, Wales), which was subse- quently realised to be of the same age as the lower part of Lapworth’s broad acuminatus Zone in Scotland. Thereafter most stratigraphers treated the persculptus Zone as the base of the Silurian, e.g. in the Lexique stratigraphique international (Whittard 1961), and this horizon was also taken as the base of the Silurian by Cocks et al. (1970) when they erected stages for the Llandovery Series, with a basal boundary defined at Dob’s Linn. It was probably the identification of the problems surrounding the Silurian—Devonian boundary and their subsequent illumination and solution that gave impetus to the internation- al effort and will to define properly the exact horizon and identify a type locality for the various systemic divisions of the Phanerozoic. The Siluro-Devonian Boundary Working Group worked formally between 1960 and 1972 (Martinsson 1977), but that work was preceded by a period of uncertainty, during which some of the procedures within the International Geological Congresses and the International Union of Geological Sciences were being developed. Bull. Br. Mus. nat. Hist. (Geol) 43: 9-15 Issued 28 April 1988 10 L. R. M. COCKS And so it was during the Ordovician—Silurian symposium at Brest, France, in 1971 that Claude Babin was the first to identify vocally the need for a group to be formally established to investigate and stabilize the Ordovician—Silurian boundary. This was put to the nascent Com- mission on Stratigraphy at the International Geological Congress in Montreal, Canada, in 1972, who felt that such a boundary working group should be established not by that commis- sion directly, but at a suitable international meeting and through the joint coordination of the then proposed Ordovician and Silurian Subcommissions. These last two bodies were finally established at the Ordovician Symposium at Birmingham, England, in September 1974, and one of their first acts was to arrange the initial meeting of the Ordovician—Silurian Boundary Working Group, which first met at Birmingham on 19th September 1974. Those present at that meeting were C. Babin (France), C. R. Barnes (Canada), S. M. Bergstrom (USA), A. J. Boucot (USA), L. R. M. Cocks (UK), J. Destombes (Morocco), J. K. Ingham (UK), V. Jaanusson (Sweden), P. J. Lespérance (Canada), D. J. McLaren (Canada), L. Marek (Czechoslovakia), F. Martin (Belgium), R. B. Rickards (UK), P. Sartenaer (Belgium), N. Spjeldnaes (Denmark), L. Teller (Poland), J. T. Temple (UK) and E. A. Yolkin (USSR). It was decided that 6 voting members of the Working Group should be nominated by both the Ordovician and the Silurian Subcommissions, plus their two chairmen ex officio, and that 3 voting members from the USSR and | from Czechoslovakia should be nominated by their respective academies of science. Thus the Ordovician Subcommission nominated Barnes, Bergstrom, W. B. N. Berry (USA), Des- tombes, Ingham and Jaanusson, with A. Williams (UK) ex officio as their Chairman, and the Silurian Subcommission nominated Boucot, Cocks, S. Laufeld (Sweden), Lesperance, Rickards and Temple, with Spjeldnaes ex officio as their Chairman. Any interested and active worker on Ordovician-Silurian boundary problems could be accepted as a Corresponding Member. At that first meeting R. B. Rickards was elected by those present as the Chairman of the Working Group, and L. R. M. Cocks as the Secretary. It was also decided that most of the Group’s activities and communication would take the form of circulars to be issued by the Secretary, and this is what subsequently happened, although field and discussion meetings also took place, and that the circulars should include reports on various Ordovician-—Silurian sections or countries and also on the different fossil groups. The first circular was issued in October 1974: it reported the formation of the Working Group, and listed which members had promised to prepare reports. In the next few years many circulars were issued, which included reports on boundary sections in Australia, Austria, Belgium, Canada (many areas), China, Czechoslovakia, England, France, Italy, Morocco, Poland, Scotland, Sweden, Wales, USA and USSR (Altai Mountains, East Baltic, Kazakhstan and NE Siberia), and also on acritarchs, chitinozoa, conodonts, grap- tolites and physical changes near the boundary. Many people became Corresponding Members, and the Voting Members were increased by D. L. Kaljo, T. N. Koren and I. F. Nikitin from the USSR, L. Marek from Czechoslovakia and Mu En-zhi from China, all of these nominations being accepted and ratified at the appropriate times by the I.U.G.S. Commission on Stratigraphy, the parent body of the Working Group. Meetings were held at the Interna- tional Geological Congress at Sydney, Australia, in August 1976 and informal meetings at Alma-Ata, USSR, in May 1977 and at the Ordovician Symposium at Columbus, USA, in August 1977, and it became clear that a more substantial meeting of the Working Group would be valuable so that future plans of action could be formulated. This coincided with an expressed wish by various geologists to see the classic sections of Great Britain, and accord- ingly a meeting was arranged from 30th March to 11th April 1979, jointly with the Silurian Subcommission. By that time R. J. Ross jr and C. H. Holland had taken over the chairman- ships of the Ordovician and Silurian Subcommissions respectively. Those attending the British meeting in 1979 were (Voting Members of the Ordovician— Silurian Boundary Working Group with an asterisk): *C. R. Barnes (Canada), M. G. Bassett (UK), *L. R. M. Cocks (UK), *C. H. Holland (Ireland), *J. K. Ingham (UK), J. S. Jell (Australia), Jin Chun-tai (China), *D. L. Kaljo (USSR), P. Legrand (France), *P. J. Lespérance (Canada), Lin Bao-yu (China), F. Martin (Belgium), A. Martinsson (Sweden), *Mu En-zhi (China), *R. B. Rickards (UK), H.-P. Schénlaub (Austria), B. S. Sokolov (USSR), L. Teller THE ORDOVICIAN-SILURIAN BOUNDARY WORKING GROUP 11 Fig. 1 The British field meeting, April 1979, outside Ludlow Castle, Shropshire. From left to right L. R. M. Cocks, Jin Chun-tai (obscured), B. D. Webby, C. R. Barnes, J. S. Jell, Wang Wei, Lin Bao-yu (obscured), D. Kaljo, Mu En-zhi, D. J. Siveter, F. Martin (obscured), L. Teller, P. J. Lespérance (obscured), D. E. White, A. Martinsson, B. S. Sokolov, P. Legrand, J. T. Temple, H. P. Schonlaub, M. G. Bassett, R. B. Rickards. (Photo C. H. Holland). (Poland), *J. T. Temple (UK), G. B. Vai (Italy) and B. D. Webby (Australia). Thus more than half the Voting Members and a considerable breadth of both stratigraphical and palaeontol- ogical expertise were represented (Fig. 1). Sections were examined in Wales (Llandovery, Meifod, Hirnant and Pont Erwyd), the Lake District of England (Yewdale, Skelgill and Spengill), and Scotland (Dob’s Linn), but, more importantly, business meetings were held in the evenings. Following a long-standing tradition of the Commission on Stratigraphy (whose then Chairman, Martinsson, and Secretary, Bassett, were present) all of the people present were allowed to participate freely in the discussions and also to take part in the informal voting which took place. The various animal and plant groups were discussed and reviewed in turn, and it was agreed that only graptolites, brachiopods, conodonts, and to a lesser extent trilobites, were important in the Ordovician-Silurian boundary discussions in the present state of knowledge. Localities were then considered. Having inspected the type Llandovery area, all members present were unanimous in rejecting that area as the boundary stratotype, large due to the unfossiliferous nature of the A, Sandstone of Jones (1928) at the base of the succession, the sporadic exposure near the base, and the lack of stratigraphically critical fossils, particularly graptolites and conodonts, then known from beds near the boundary (although this situation has been much improved by subsequent work, Cocks et al. 1984). Other localities were graded in turn, with the following scheme: A, a possible section for placing the boundary; B, an important section which may be considered further in discussing the boundary, and C, a section or area unlikely to prove important in boundary definition. The grading was as follows: A Anticosti Island (Canada), Dob’s Linn (Scotland). . tag’ B_ Carnic Alps (Austria), Cornwallis Island (Canada); Hupei (China), Mirnyi Creek (Siberia, 2 L. R. M. COCKS USSR), Missouri (USA), Nevada (USA), Pont Erwyd (Wales), Szechuan (China) and Yewdale Beck (Lake District, England). C Australia, Bala district (including Hirnant area, Wales), Belgium, Bohemia (Czechoslovakia), France, Garth (Wales), Hudson Platform (Canada), Kazakhstan (USSR), Kweichow (China), Lake District (apart from Yewdale Beck, England), Manitoba (Canada), Manitoulin Island (Canada), Morocco, Newfoundland (Canada), North American mid- continent (except Missouri and Nevada), Pembrokeshire (Wales), Perce (Canada), Poland, Scania (Sweden), Shensi (China) and Yukon (Canada). In addition the Working Group then felt that more reports were needed from Algeria, Bornholm (Sweden), Burma, Dalarna and Vasterg6étland (Sweden), Estonia (USSR), India and the Himalayas, Norway, Rae Grain (Scotland), Portugal, South America, Spain and West Nevada: however, although more data on some of these areas were subsequently gathered, none proved to have much extra to offer in the main definition of a stratotype. Because Anticosti Island, Canada, was one of the leading contenders for the definitive boundary section, it was agreed that a further field meeting should be held there. Other briefer meetings were also held in Paris, France, during the 1980 International Geological Congress, and in the Carnic Alps of Austria in late July and early August 1980. Meanwhile the debate persisted as to the best method of correlation across the boundary interval, and whether the actual boundary should be defined by the use of conodonts or graptolites. It was generally agreed that brachio- pods and trilobites should not be used in the definition, except that there was a strong feeling that the widespread Hirnantia brachiopod fauna should be included within the Ordovician rather than the Silurian. The Working Group circulars also contained various discussion and position papers between 1978 and 1982 on the theory and practice of defining the boundary both geographically and biostatigraphically. Opinions differed as to whether or not the stratotype could be satisfactorily placed within a nearly exclusively graptolite sequence such as Dob’s Linn, and, if the boundary was to be defined on graptolites, whether it was to be at the base of the extraordinarius, the persculptus or the acuminatus Zone. There was no real consensus on the answers to these questions. The field meeting to Quebec, which was partly in Anticosti Island and partly in the Gaspé Peninsula, was held in July 1981, again jointly with the Silurian Subcommission. Those attend- ing (apart from various other Canadian hosts) were T. W. Amsden (USA), *C. R. Barnes (Canada), *A. J. Boucot (USA), *L. R. M. Cocks (UK), *C. H. Holland (Ireland), P. Legrand (France), *P. J. Lespérance (Canada), F. Martin (Belgium), G. M. Philip (Australia), *R. J. Ross jr (USA), H.-P. Schonlaub (Austria) and L. Teller (Poland). This was a rather disappointing attendance, particularly of Voting Members, and hence the evening discussion meetings were not as representative of the differing positions of the complete group as they might have been if the attendance had been better. A review was given of each of the relevant biological groups, and general discussions ensued, with the following points noted. There were very favourable general impressions of the simplicity of structure and good exposure at Anticosti, but reser- vations on the lack of graptolites there near the Ordovician—Silurian boundary and the relative lack of work done on groups other than conodonts on the beds near the boundary. Opinions differed about the accessibility of Anticosti Island and also about the importance of the struc- tural complexity of the Dob’s Linn area. At the end of the meeting, two straw votes indicated that those present thought that Anticosti was the best available section across the Ordovician— Silurian boundary in the shelly facies, and that, other things being equal, it would be preferable to have the Ordovocian-Silurian boundary stratotype in the same area as the stratotype area for the lowest series of the Silurian System. The latter point was relevant since at that time Anticosti was one of the three candidates under consideration by the Silurian Subcommission (the other two being Llandovery itself and the Oslo Region, Norway) for the stratotype for the lowest Silurian series. Shortly after this meeting, R. B. Rickards resigned as Chairman of the Working Group, and, because it was clear that the decisions on the boundary were close to being taken, the Commission on Stratigraphy subsequently appointed the Chairmen of the THE ORDOVICIAN-SILURIAN BOUNDARY WORKING GROUP 13 Ordovician and Silurian Subcommissions, R. J. Ross jr and C. H. Holland, as Co-Chairmen of the Group; which they remained until its closure. After the formal circulation of a number of further views on the position and correlation of the future boundary stratotype through the Circular, and informal discussion between inter- ested people, it was agreed that maximum publicity and attendance should be sought for a meeting of the Working Group at the Ordovician Symposium at Oslo, Norway, so that progress would be made on the boundary decision. At that symposium, two meetings of the boundary Working Group were held, as well as seven papers on the boundary being presented within the normal symposium sessions. The meetings, on 20th and 23rd August 1982, attracted 53 and 76 people respectively, including the following Voting Members: Barnes, Bergstrom, Berry, Cocks, Destombes, Holland, Jaanusson, Kaljo, Lespérance, Rickards and Ross. After lengthy discussion, the first decision taken was whether or not the time was yet ripe for a formal vote on deciding the boundary stratotype and horizons, and, despite strong pleas for delays to enable more research to be done from several speakers, it was decided by 47 votes to 14 that the time had now come. The choice of stratotype boundary had been narrowed to three: (i) the first appearance of the conodont Ozarkodina oldhamensis at 50cms above the Oncolitic Platform Bed at Ellis Bay, Anticosti Island, Canada. (11) the base of the persculptus graptolite Zone at Dob’s Linn, near Moffat, Scotland. (iii) the base of the acuminatus graptolite Zone at Dob’s Linn. At the Oslo meeting two informal votes were then taken: (i) Anticosti was preferred to the persculptus Zone at Dob’s Linn by 34 votes to 13, with 25 abstentions; (ii) Anticosti was preferred to the acuminatus Zone at Dob’s Linn by 35 votes to 13, with 26 abstentions. The same questions were also informally voted upon by the 30 Voting and Corresponding Members of the Working Group who were present, and 17 preferred Anticosti against 7 for the per- sculptus Zone (6 abstentions); and 19 preferred Anticosti against 5 for the acuminatus Zone (6 abstentions). Therefore, it was clear that a substantial majority of those at the meeting then preferred to place the base of the Silurian at Anticosti Island using conodonts, and that the Voting Members of the Working Group should take part in a formal postal ballot in the light of this knowledge. Thus Circular No 17 was distributed to the members in October 1982 with a ballot paper to be returned by the end of January 1983. There followed a period during which various letters were informally circulated and lobbying took place, although none formally through the Secretary apart from a paper by P. Legrand which was very critical of the Oslo decision and which was distributed with Circular 17. At the end of the formal voting period, the votes returned stood as follows: (i) Which do you prefer—Anticosti or the persculptus Zone at Dob’s Linn? For Anticosti: Barnes, Bergstrom, Boucot, Holland, Lespérance, Ross: total 6. For persculptus Zone: Berry, Cocks, Destombes, Ingham, Kaljo, Koren, Laufeld, Marek, Nikitin, Rickards, Temple: total 11. No vote received: Jaanusson, Mu: total 2. (iu) Which do you prefer—Anticosti or the acuminatus Zone at Dob’s Linn? The votes received were identical to the persculptus Zone vote. These results were distributed to all members of the Working Group in Circular 18 in March 1983. Since there had been an outright majority on the selection of Dob’s Linn rather than Anticosti, this was accepted by the officers as a decision, and a second formal postal vote was called for, firstly to give Voting Members an opportunity to change their minds, and secondly to decide between the persculptus and the acuminatus Zones at Dob’s Linn for the stratotype horizon. Opportunity was also given to the Corresponding Members to formally express their opinions. The results of this second ballot was announced in Circular No. 19 in August 1983, and were as follows: (i) the place of the stratotype. Voting Members. Dob’s Linn: Berry, Cocks, Destombes, Holland, Ingham, Kaljo, Koren, 14 L. R. M. COCKS Laufeld, Marek, Nikitin, Rickards, Temple: total 12. Anticosti: Barnes, Bergstrom, Boucot, Lespérance, Ross: total 5. Abstain: Jaanusson, Mu: total 2. In addition 14 Corresponding Members voted for Dob’s Linn, 8 for Anticosti, and 4 abstained. (ii) the horizon of the stratotype. Voting Members. Base of acuminatus Zone: Cocks, Holland, Ingham, Jaanusson, Kaljo, Koren, Marek, Nikitin, Rickards, Temple: total 10. Base of persculptus Zone: Berry, Des- tombes, Laufeld, Mu, Ross: total 5. Abstain: Barnes, Bergstrom, Boucot, Lespérance: total 4. 13 Corresponding Members voted for the base of the acuminatus Zone, 9 for the base of the persculptus Zone, and 5 abstained. In addition the question of possible parastratotypes was also voted upon, with the possibility of erecting one parastratotype on Anticosti Island and the other in China, but on this question only 8 Voting Members voted for the erection of these, with 3 against and 8 abstentions, and so the officers decided not to proceed further on that topic, and they were assisted in that decision by informal advice against parastratotypes from the Commission on Stratigraphy. Thus since there was a clear majority for placing the Ordovician-Silurian stratotype bound- ary at the base of the acuminatus graptolite Zone at Dob’s Linn, Scotland, this decision was formally forwarded to the Commission on Stratigraphy for consideration with various other matters at their meeting at the International Geological Congress at Moscow, USSR in August 1984. The decision was endorsed by a postal vote of that committee, who subsequently for- warded it to the I.U.G:S. for ratification. The proposals were reported to a meeting of the full I.U.G.S. Executive Committee in Rabat, Morocco, in February 1985 and submitted to the I.U.G.S. Executive for a postal ballot, whose result was declared in May 1985, and published in June 1985 (Bassett 1985), together with an article describing the Ordovician—Silurian boundary at Dob’s Linn (Cocks 1985). The Ordovician—Silurian Boundary Working Group was finally dissolved in its Circular No. 20, distributed in June 1985. The life of the Ordovician—Silurian Boundary Working Group was therefore somewhat over ten years long, but it was useful not only in determining the position and horizon of the boundary itself, but also in stimulating a great deal of research in various parts of the world, and in encouraging international understanding and cooperation. References Bassett, M. G. 1985. Towards a ‘Common Language’ in Stratigraphy. Episodes, Ottawa, 8: 87-92. Cocks, L. R. M. 1985. The Ordovician-Silurian Boundary. Episodes, Ottawa, 8: 98-100. , Toghill, P. & Ziegler, A. M. 1970. Stage names within the Llandovery Series. Geol. Mag., Cam- bridge, 107: 79-87. , Woodcock, N. H., Rickards, R. B., Temple, J. T. & Lane, P. D. 1984. The Llandovery Series of the type area. Bull. Br. Mus. nat. Hist., London, (Geol.) 38 (3): 131-182. Elles, G. L. & Wood, E. M. R. 1901-18. A monograph of British Graptolites. Palaeontogr. Soc. (Monogr.), London. m + clxxi + 539 pp., 52 pls. Jones, O. T. 1909. The Hartfell-Valentian succession in the district around Plynlimon and Pont Erwyd (North Cardiganshire). Q. Jl geol. Soc. Lond. 65: 463-537, pls 1, 2. Lapworth, C. 1879. On the tripartite classification of the Lower Palaeozoic Rocks. Geol. Mag., London, (dec. 2) 6: 1-15. Martinsson, A. (ed.) 1977. The Silurian—Devonian Boundary. Int. Un. geol. Sci. (A) 5: 1-349. Whittard, W. F. (ed.) 1961. Lexique Stratigraphique International 1 Europe. (3aV: Angleterre, Pays de Galles, Ecosse; Silurien.) 273 pp. Paris, C.N.R.S. THE ORDOVICIAN-SILURIAN BOUNDARY WORKING GROUP 15 Appendix MEMBERSHIP OF THE ORDOVICIAN-SILURIAN BOUNDARY WORKING GROUP Those names with an asterisk* were Voting Members, the remainder were Corresponding Members. Amsden, T. W. USA Lin Bao-yu China Apollonov, M. K. USSR *Marek, L. Czechoslovakia Babin, C. France Martin, F. Belgium *Barnes, C. R. Canada Martinsson, A. Sweden Bassett, M. G. UK McLaren, D. J. Canada Bergstrom, J. Sweden *Mu En-zhi China *Bergstrom, S. USA *Nikitin, I. F. USSR *Berry, W. B. N. USA Norford, B. S. Canada Bolton, T. E. Canada Nowlan, G. S. Canada *Boucot, A. J. USA Oradovskaya, M. M. USSR Brenchley, P. J. UK Poulsen, V. Denmark Bruton, D. L. Norway *Rickards, R. B. UK *Cocks, L. R. M. UK Rong Jia-yu China Cramer, F. H. Spain *Ross, R. J. jr USA *Destombes, J. Morocco Sartenaer, P. J. M. J. Belgium Hamada, T. Japan Schonlaub, H. P. Austria *Holland, C. H. Ireland Sheehan, P. M. USA *Ingham, J. K. UK Sokolov, B. S. USSR *Jaanusson, V. Sweden Spjeldnaes, N. Denmark Jackson, D. E. UK Teller, L. Poland Jaeger, H. East Germany *Temple, J. T. UK Jin Chun-tai China Toghill, P. UK *Kaljo, D. USSR Wang Xiao-feng China Kobayashi, T. Japan Webby, B. D. Australia *Koren, T. N. USSR Williams, A. UK *Laufeld, S. Sweden Williams, S. H. UK Legrand, P. France Wright, A. D. UK Lenz, A. C. Canada Yolkin, E. A. USSR *Lespérance, P. J. Canada Dob’s Linn — the Ordovician-Silurian Boundary Stratotype S. H. Williams Department of Earth Sciences, Memorial University of Newfoundland, St John’s, Newfoundland A1B 3X5, Canada Synopsis Dob’s Linn, north-east of Moffat, southern Scotland, has been designated the Ordovician-Silurian bound- ary stratotype by the Ordovician-Silurian Boundary Working Group of the I.U.G.S. Commission on Stratigraphy. The boundary is placed at the base of the P. acuminatus Zone, marked by the first occurrence of Akidograptus ascensus and Parakidograptus acuminatus, s.l., 1-6m above the base of the Birkhill Shale in the Linn Branch section. The stratigraphical interval covering this boundary consists of richly graptolitic black shale. Occasional metabentonites are also present. The underlying Upper Hartfell Shale is composed predominantly of pale grey-green, non-graptolitic shale and mudstone, with several black graptolitic bands referred to the Complanatus, Anceps and Extraordinarius Bands. The rich faunal assemblage of the Anceps Band reduces to only three diplograptid taxa in the Extraordinarius Band. This major extinction is recorded at an equivalent horizon by all other graptolitic sequences throughout the globe. Sediments of the Upper Ordovician to Lower Silurian Moffat Shale Group were probably deposited entirely by distal turbidites in the abyssal depths of the Iapetus Ocean. Northerly-directed subduction subsequently transported the site of shale deposition into a proximal turbidite environment, resulting in a diachronous transition into coarse clastics of the overlying Gala Greywacke Group. Deformation related to subduction also produced imbricate thrusting and raised the area to a prehnite-pumpellyite facies metamorphic grade. Geophysical evidence indicates that the region is underlain by continental basement; this suggests that the Southern Uplands are allochthonous. Historical introduction Graptolites were first recorded from Dob’s Linn, southern Scotland over one hundred years ago. The earliest publications to include descriptions of the fauna from the Moffat Shales (e.g. Carruthers 1858; Nichoison 1867; Dairon 1869; Hopkinson 1871) paid little or no attention to their stratigraphical importance. Elsewhere during this period, an ever increasing volume of articles on graptolites was being published, including a number which recognized their great potential for both regional and global correlation (Nicholson 1876). These included studies on the Lower Ordovician of northern England (e.g. Nicholson 1870, 1875), South Wales (Hopkinson & Lapworth 1875) and eastern Canada (Hall 1858, 1865; Billings 1865). In 1864 Charles Lapworth obtained a teaching post connected with the Episcopal Church at Galashiels some 30km north-east of Dob’s Linn (Gibson 1921). He had no previous geological training or experience, but soon developed an interest in the local geology of the Southern Uplands. Harkness (1851) had described the repeated, faulted nature of this area composed of thick greywacke and containing a shale sequence termed the “Moffat Series’. Otherwise this structurally complex region still defied satisfactory interpretation despite attempts by several other eminent geologists (e.g. Sedgwick 1850). Lapworth’s first publication on the Lower Palaeozoic (1870) concerned the geology of the Galashiels area. During these early years in his geological career, he recorded graptolites both from within the thick greywacke sequence of the Southern Uplands and from the underlying black shales. A summary of Lapworth’s early lithostratigraphical division was published in 1872. During the next five years he completed an exercise of detailed geological mapping, logging of sections and bed-by-bed faunal collecting throughout the Moffat area. A selection of new graptolite taxa were figured and discussed briefly in 1876, while similar faunas were also illustrated from equivalent strata in northern Ireland (Lapworth 1877). Bull. Br. Mus. nat. Hist. (Geol) 43: 17-30 Issued 28 April 1988 18 S. H. WILLIAMS Lapworth’s major stratigraphical synthesis ‘On the Moffat Series’ was published in 1878, where he established beyond doubt the precise, ordered stratigraphical change in graptolite assemblages through the sequence of black and grey shales. With the exception of the Glenkiln Shale (best developed at Glenkiln Burn, south-east of Moffat) and the lowermost portion of the Lower Hartfell Shale (best exposed at Hartfell Spa, north of Moffat), Lapworth used the Main Cliff and Linn Branch sections of Dob’s Linn as the standard reference for the Moffat Shale. While working at Dob’s Linn, Lapworth stayed at Birkhill Cottage only a few hundred metres above the locality. The uppermost black shale division of the group was named after this cottage. Lithological sections measured at Dob’s Linn, together with graptolite assemblages and biostratigraphical divisions, were figured by Lapworth (1878: figs 27-30) in his major work, where the lithostratigraphical division of the Moffat Shale was also clearly defined. An earlier, less detailed log of the Moffat Shale from Lapworth’s notes, covering the Glenkiln Shale to basal Birkhill Shale, is still preserved in Birmingham University, and is here illustrated for comparison (Fig. 1). Note that Lapworth’s assignment of the lower part of the Upper Hartfell Shale to the “Belcraig Shale’ (after Beldcraig Burn near Moffat) was apparently never published. During this research, Lapworth was appointed in 1875 to an Assistant Mastership at Madras College, St Andrews. In 1881 he was elected to the Chair of Geology at the recently established Mason College, Birmingham, which subsequently became the University of Birmingham. In addition to elucidating the structure of the Southern Uplands, Lapworth established the Ordo- vician System in 1879, solving the embittered feud between the schools of Murchison and Sedgwick (see Bassett 1985). He also made an equally painstaking, detailed stratigraphical study at Girvan on the south-west coast of Scotland (Lapworth 1882). His conclusions regard- Adbef, wh Cie (tore « pe, Ak a thierry thal, WSS a niarey blk bat beck eth Ce sadted fase waned foes pe fv C1 1645 Co hone : a Vg. exraest, pha Puccll sreeps P. Mack auksy, 2 bcheedgta ree ie gaye Meee. Graptolite zone C. vesiculosus 6 Marte that 01 yeler P. linearis ftfb0 Rh Fliaye 3 Alles: hard Mean by Mali LOWER SAL O- clingani Sadao: 3 torcharpiricre” a ON ln lellid Muu retbet Bach Hhrle Diesyan: rarriasis i E Extraordinarius Band A Anceps Bands C Complanatus Bands Fig. 1 Reproduction of an unpublished section through Moffat Shale from Lapworth’s notebook (preserved at Birmingham University), predating his modified version published in 1878. Modern measured section with graptolite zones is included for comparison. THE ORDOVICIAN-SILURIAN BOUNDARY STRATOTY PE 19 ing the relationship between the strata at Girvan and those of the Southern Uplands were published in 1889. The work of Lapworth was drawn upon heavily by Peach & Horne (1899), who also described many confirmatory sections through the Moffat Shale in the Southern Uplands. With the exception of one taxonomic paper published by Lapworth in 1880, most of his new graptolite taxa were first described fully by Elles & Wood (1901-18), whose work he supervised throughout its production. Following this major publication, little taxonomic or stratigraphical work was attempted at Dob’s Linn for half a century. One notable exception was an article by Davies (1929), who included Dob’s Linn in his detailed study of late Ordovician and early Silurian graptolites. A series of recent biostratigraphical and taxonomic papers was initiated by Packham (1962), who described the evolution of Glyptograptus tamariscus and related diplograptids from the Birkhill Shale of Dob’s Linn and from the Lower Silurian of the Rheidol Gorge, mid Wales. Toghill (1968) discussed the evolution of the earliest monograptids and formally established the presence of the G. persculptus Zone at Dob’s Linn. He also gave a biostratigraphical summary of the entire Birkhill Shale with listings of the zonal assemblages (1968a), but none of the fauna was described or illustrated. Toghill (1970) subsequently published a revision of graptolites from the Upper Hartfell Shale and top Lower Hartfell Shale. Brief taxonomic descriptions and illustrations were included; this paper added little in terms of refinement to Lapworth’s biostratigraphical divisions, but was important in demonstrating the presence of previously unrecorded graptolite species. Cocks et al. (1970) used this data to make a premature proposal of Dob’s Linn as the Ordovician— Silurian boundary stratotype, where they placed the boundary at the base of the Birkhill Shale (= ‘base’ of G. persculptus Zone). Rickards (1979) added the record of the C.? extraordinarius Zone, based on the discovery by Ingham (1979) of a black, graptolitic shale band midway between the top Anceps Band of the Upper Hartfell Shale and the basal Birkhill Shale. A geological locality map of Dob’s Linn was given by Toghill (19684), but most geologists visiting the locality were still guided by the remarkably detailed geological map published by Lapworth in 1878. Following several years’ critical work using aerial and ground photographic overlays and modern structural synthesis, Ingham (1979) published a totally revised geological map of Dob’s Linn. During the course of this research, Ingham found that several of Toghill’s measured sections through the Upper Hartfell and lowermost Birkhill Shales were disrupted structurally and measurements were consequently revised. Ingham’s recognition of unbroken sections and several new graptolitic bands in the Upper Hartfell Shale (upper Complanatus Band, Anceps Band A and the Extraordinarius Band), permitted critical faunal recollection of the Moffat Shale. This task was begun by the present author in 1978, leading to a series of taxonomic and biostratigraphical papers covering the top 8m of the Lower Hartfell Shale (Williams 1982a), the Complanatus Bands (Williams & Ingham, in prep.), the Anceps Bands (Williams 1982), the Extraordinarius Band and the basal 2 m of Birkhill Shale (Williams 1983). These papers confirmed Lapworth’s original faith in graptolites as a critical bio- stratigraphical tool and gave more precise definitions of the zonal boundaries. Of particular importance is the revised, unambiguous definition of the boundary between the G. persculptus and P. acuminatus Zones, the horizon now defined as the Ordovician—Silurian boundary. Regional setting, stratigraphy and depositional environment The geology of the Southern Uplands is dominated by a thick package of monotonous, sparse- ly graptolitic greywackes, belonging to the Gala Greywacke Group. This is of unequivocal turbidite origin. The underlying Moffat Shale Group is exposed as a series of elongate, narrow, east-west inliers which Lapworth (1878) and Peach & Horne (1899) considered to represent tight, isoclinal anticlines. It is now considered (Webb 1983) that these structures were formed through progressive shearing of early folds. The appearance of simple, reverse faulting postu- lated by Craig & Walton (1959), Leggett et al. (1979), Eales (1979) and other recent workers is due to almost complete removal of the shorter, south-eastern limbs. 20 S. H. WILLIAMS An overall younging and progressive lateral change in lithologies from north to south was recognized in the Southern Uplands by Peach & Horne (1899). They divided the regions into three tracts, namely the Northern, Central and Southern Belts. The rock types and age ranges of strata characterizing each belt have since been summarized in detail by Leggett et al. (1979), who considered division into ten discrete sequences to be more appropriate. In the most northerly sequences red cherts, siliceous mudstones and pillow basalts of Arenig to Llandeilo age are overlain by Llandeilo-Caradoc greywackes. This succession passes southwards to Llandeilo—Llandovery cherts and black shales overlain by Llandovery greywackes. The diach- ronous base of the greywackes youngs progressively to the south, with consequently extended black shale deposition. The most southerly sequences of the Southern Uplands are composed entirely of Wenlock greywackes. Both the structural pattern of the Moffat Shale outcrops and the diachronous base of the greywackes were explained in a model proposed by Mitchell & McKerrow (1975) and expand- ed by McKerrow et al. (1977) and Leggett et al. (1979). These authors considered the Southern Uplands to have formed as an accretionary prism over a northerly dipping subduction zone on the northern margin of the Iapetus Ocean. The prehnite-pumpellyite metamorphic facies could have resulted from burial and tectonic processes during such accretion (Oliver et al. 1984). Geophysical studies (Powell 1971; Hall et al. 1983), however, indicate crystalline, continental material underlying the area, rather than the oceanic basement required for this model. Bluck (1984) discussed this apparently contradictory evidence; he concluded that the Southern Uplands are probably allochthonous. More recently, Needham & Knipe (1986) reiterated the accretionary prism model, but this was considered inadequate by Murphy & Hutton (1986), who concluded that subduction at both Iapetus margins was complete by late Ordovician times and that the Silurian turbidites were deposited in a successor basin. The Moffat Shale Group is divided into four formations: the Glenkiln Shale, Lower Hartfell Shale, Upper Hartfell Shale and Birkhill Shale (Lapworth 1878). The Glenkiln Shale is com- posed of an unknown thickness of pale grey and black, heavily silicified argillites. At Dob’s Linn the formation is poorly exposed as a series of disconnected, fault-bounded slivers. It is generally unfossiliferous and due to heavy shattering of the competent, siliceous component, even black lithologies rarely yield identifiable graptolites. Useful comparative sections are exposed at the type section of Glenkiln Burn and at several other inliers in the Moffat area (Lapworth 1878; Peach & Horne 1899). The Glenkiln Shale apparently passes gradationally into the almost continuously black Lower Hartfell Shale, which yields a more abundant graptolite fauna and is over 20m thick. The lower half of the formation remains highly siliceous; the proportion of chert to black shale decreases upwards throughout the unit, black shale becoming predominant in the upper 5m. The overlying Upper Hartfell Shale is composed mostly of monotonous, non-graptolitic, pale grey/green shales and mudstones 28m thick (Figs 1, 2). Its lower boundary is marked by a transitional 3cm interval of alternating pale grey and black laminae. Three groups of graptoli- tic, black shale bands occur within the formation, named the Complanatus, Anceps and Extraor- dinarius Bands (Ingham 1974, 1979) after their diagnostic zonal assemblages (Fig. 1). Other atypical lithologies include nodular limestones and one detrital limestone. The latter horizon is a very pale grey, coarse-grained limestone 6:5cm thick, lying 1-5m below the lower Com- planatus Band in the banks of the Linn Branch stream. Unfortunately it has been totally recrystallized and affected by strain-induced pressure solution, but it was presumably of detrital origin. One medium grey nodular limestone, 4cm thick and lying 2m above the base of the Upper Hartfell on the North Cliff section, displays uncompacted bioturbation with horizontal to subvertical simple burrows 1-2mm in diameter. Other nodular horizons present in the Linn Branch section include that known to yield a blind, dalmanitid trilobite 0-1m below the Extraordinarius Band (Ingham 1979) and a second, apparently unfossiliferous bed 0:25m below the base of the Birkhill Shale. Three of these four limestones were not known prior to recent recollecting for conodont samples (Barnes & Williams, this volume) and other similar horizons in the Upper Hartfell Shale probably still await discovery. THE ORDOVICIAN-SILURIAN BOUNDARY STRATOTYPE 21 beg DEL Lee Pas . = y . E>: = a Se 7 1 y V 2 " a F Fig. 2 Sectioned slabs and bedding surfaces from the Moffat Shale at Dob’s Linn (all x 2). A. Lower boundary of Anceps Band C. B. Typical uniformly laminated black shale, lower Birkhill Shale. C. Black shale, thin metabentonite and micaceous horizons from Anceps Band D (note grading shown by metabentonite). D. Uniform pale grey Upper Hartfell Shale lithology, Anceps Bands. E. Irregular laminae and compacted bioturbation, base of lower Complanatus Band. F. Bioturbation above upper Complanatus Band. G. Irregular laminae, compacted bioturbation and low-angle synsedimentary faulting from reversal in Birkhill Shale 0-5m above base. H. Bedding plane section from base of slab shown in Fig. 2G. I. Black shale injection through pale mudstone, Anceps Bands. J. High-angle, post-compactional microfaulting, Anceps Band D. K, L. Bedding surfaces with coarse mica flakes, Anceps Band D. ee 2D, S. H. WILLIAMS The 43m of Birkhill Shale is composed of black, continuously graptolitic shale in the lower part, with the exception of a temporary reversal to an ‘Upper Hartfell’ type lithology 0-46— 0-56 m above its base. The shales become progressively siltier, less fissile and paler towards the top of the formation, culminating in a transition to coarse turbidites of the overlying Gala Greywacke Group. The precise depositional environment of the Moffat Shale Group is still uncertain. Lapworth (1897) envisaged black shale formation in a partially restricted ‘Sargasso Sea’ setting. Later Walton (1963) considered the Moffat Shale to have been deposited on a regional high in a deep ocean environment, explaining the lack of turbidites which are found elsewhere as lateral equivalents and overlying the group. With the recognition of the Lower Palaeozoic Iapetus Ocean in recent years, it has become evident that the Moffat Shale was deposited within a wide, open ocean of complex history. This suffered continued narrowing throughout the Upper Ordovician and Silurian due apparently to subduction on both northern and southern margins (Moseley 1978, Bluck 1984). The signifi- cance of sedimentary features such as postulated winnowing of graptolities, lithological colour alternation, soft-sediment deformation and presence of limited bioturbation (Fig. 2) was dis- cussed by Williams & Rickards (1984). Further observations have emphasized the variation in contacts between pale and black lithologies, from sharp and laminar (Fig. 2A) to gradational and irregular (Figs 2E—G). They have also confirmed the presence of coarse, silty laminae with biotite flakes up to 1mm diameter, particularly within the Anceps Bands of the Upper Hartfell Shale (Figs 2K—L). These strongly suggest a hemipelagic, distal turbidite origin for the sedi- ments, in contradiction to Dewey (1971), Leggett (1980) and Leggett et al. (1979), who con- sidered the shale to be of oceanic, truly pelagic origin formed during periods of high eustatic sea level stands. Several black shale sequences elsewhere are known to have been deposited as distal turbidites, including beds of the Burgess Shale of British Columbia (Piper 1972) and of the Cow Head Group, western Newfoundland (Coniglio 1985). It was therefore evident that during Lower Palaeozoic times black shales could form within an unrestricted oceanic setting lacking any degree of restriction, unlike those deposited during Mesozoic and Recent times (e.g. Jenkyns 1978; Stow & Piper 1984). The presence of metabentonites throughout much of the Moffat Shale indicates sporadic acidic volcanism. Most of these are only laminae or thin beds (Fig. 2C), but they occasionally reach over 5cm thick. Their lateral impersistence was noted by Williams & Rickards (1984), who suggested variable deposition due to a gently undulating sea floor. It seems likely that the metabentonites were transported by a turbidite mechanism in a similar fashion to the remain- ing lithologies; they would not, therefore, have significance in terms of proximity to volcanic activity. The single coarse-grained limestone below the lower Complanatus Band was probably also deposited by a powerful, carbonate-rich, turbidite flow. Such carbonate detritus was prob- ably derived from a northerly source, such as the sites of fore-arc, shelf and slope deposition at Girvan (Bluck 1984). No critical sedimentological studies have been carried out on the Moffat Shale at Dob’s Linn. With recent advances in both understanding of depositional mechanisms in deep-water, hemipelagic sedimentation (Stow & Piper 1984; Coniglio 1985) and development of new tech- niques to assist the study of fine-grained sediments, a detailed review of argillites at Dob’s Linn and at comparitive sections is now warranted. Late Ordovician and Early Silurian graptolite biostratigraphy The following account is based on detailed logging through a trench constructed on the north valley side of the Linn Branch (Figs 3, 4), excepting that of the Lower Hartfell Shale (from the North Cliff trench, Fig. 3) and lower part of the Upper Hartfell Shale, including the Com- planatus Bands (Linn Branch stream bed). The uppermost 5m of the continuously black Lower Hartfell Shale is encompassed within the Pleurograptus linearis Zone (Williams 1982a). Following this level, 9m of unfossiliferous grey shale and mudstone belonging to the Upper Hartfell Shale is present before the black, (9.2) N THE ORDOVICIAN-SILURIAN BOUNDARY STRATOTYPE <= 1) i= o = = — pas oO <£ ~ = /, ° 2: i Fig. 3 Photograph showing northern side of Linn Branch gorge, indicating key collecting localities. Interpretation of geology and structure adapted after Ingham (1974: fig. 25). 24 S. H. WILLIAMS Cel ad : “(UPPER HARTFELL SHALE 7“, Mp aap eta alee ate MH Let C6 ASS (ae UK, ec ag Goi srt Si pacificus ;~ f ae SUlzOMe 2-7 1 4 “ D. anceps /i<- /'= Zone ee AN-°® ,o > C? extraord-*"' \ v, Marius Zone) iy RUN, ce \ . Fig. 4 Photograph of Linn Branch trench with interpretation, photographed from same position as Fig. 2. Notebook lies on position of Ordovician-Silurian boundary. THE ORDOVICIAN-SILURIAN BOUNDARY STRATOTYPE 25 graptolitic Complanatus Bands are reached. Lapworth (1878: 316; fig. 28) originally recorded ‘Dicellograptus forchhammeri, Climacograptus scalaris? and Diplograptus truncatus’. These speci- mens were subsequently recognized as new taxa, described by Lapworth (1880) and Elles & Wood (1901-18) as Dicellograptus complanatus, Climacograptus scalaris miserabilis and Orthograptus truncatus socialis. Davies (1929: 18) relocated the graptolite horizon, as did Toghill (1970). Ingham (1974) proved the existence of a second narrow, black seam about 0:-4m above the 4cm thick, previously recorded band. Williams (1987) records D. complanatus Lap- worth, D. minor Toghill, C. miserabilis Elles & Wood, C. tubuliferus Lapworth and O. socialis (Lapworth) from the lower band. The upper band yields D. complanatus and rare specimens of Orthoretiograptus pulcherrimus (Keble & Harris). Williams & Lockley (1983) described well preserved specimens of an inarticulate brachiopod, both from within and directly above the upper band, which were assigned to a new genus and species Barbatulella lacunosa. Rare, usually fragmented specimens of this brachiopod also occur at several grey mudstone horizons within the following Anceps Bands. The Anceps Bands are separated from the Complanatus Bands by 13m of grey barren shale and mudstone. They comprise a series of alternating black and grey shales with common metabentonites, covering an interval which ranges in thickness from 1:6m on the Main Cliff, to 2:0 m in the Linn Branch trench and 4-5 m in the Long Burn section. The last of these localities is separated from the former two by the Main Fault, and may have been deposited at some distance apart. Other lateral variation in thickness was probably due to deposition on an irregular sea floor and synsedimentary erosion as discussed by Williams & Rickards (1984). Lapworth (1878: 253, 317) erected the Dicellograptus anceps Zone in his major publication on the Moffat Shale, owing to the distinctive nature of the faunal assemblage in the black Anceps Bands. Toghill (1970: 6; fig. 1) recorded four black shales; Ingham (1974) however established the presence of five bands or groups of bands, now referred to Bands A to E. The rich, diverse fauna contained within these black shales (Fig. 5) allowed Williams (1982) to divide the zone into the Dicellograptus complexus and Paraorthograptus pacificus Subzones. In addi- tion to those species’ ranges shown on the range chart, rare specimens of Climacograptus hastatus Hall and Glyptograptus posterus Koren & Tsai have been found in the D. complexus and P. pacificus Subzones respectively. These taxa confirm correlation with the Australian and Chinese graptolite zonal schemes. Ingham (1979) was first to discover the Extraordinarius Band 0-:96m above Anceps Band E. This narrow, dark brown shale contains a sparse graptolite assemblage, identified by Rickards (1979) and Williams (1983) as Climacograptus? extraordinarius (Sobolevskaya), Climacograptus sp. indet. and Glyptograptus? sp. indet. The grey strata separating the Extraordinarius Band from Anceps Band E is unfossiliferous, with the exception of a nodular limestone 0-1 m below the Extraordinarius Bands which yields rare fragmentary specimens of a blind dalmanitid tri- lobite (Ingham 1979). The lower boundary of the Birkhill Shale lies 1:17 m above the Extraordinarius Band. Following a basal, unfossiliferous black shale interval 0:15m thick, an abundant but poorly diverse graptolite fauna is present, including Climacograptus normalis Lapworth, C. miserabilis Elles & Wood and Glyptograptus? ‘venustus cf. venustus’ (Legrand). A temporary reversal to alternating grey/green and black shales occurs at 0-46 to 0-56m above the base. This is followed by black shales yielding a better preserved, more diverse assemblage with the addition of Glyptograptus cf. persculptus (Salter) and Glyptograptus? avitus Davies. Lapworth (1878) referred the basal Birkhill Shale to the P. acuminatus Zone. The G. persculptus Zone was first separated as a biostratigraphical unit underlying the P. acuminatus Zone in central Wales by Jones (1909, 1921), where he also considered it to be lithologically different. Davies (1929) ratified the presence of two distinct zones and recognized the interval equivalent to the G. persculptus Zone in both northern England and southern Scotland. It appears that he referred three ‘horizons’ below the first occurrence of Parakidograptus acuminatus (Nicholson) and Akidograptus ascensus Davies to the G. persculptus Zone at Dob’s Linn (1929: 22; fig. 32), but this is not stated unequivocally in the text. Adoption of the G. persculptus Zone as a formally defined, distinct biostratigraphic unit at Dob’s Linn was not realized prior to Toghill’s revision 26 S. H. WILLIAMS BIRKHILL SHALE Me - j \ D, ornatus” Sey - = q s o > s 0 2 H 28 i , s o oO s iJ ail aT T > wo oO oO m NN i \ metres . pee \ oF (e) — & - « , ° ° 2 D, anceps , AN ow ros | os ——— O90 f a _ | a S > D,. complaxus | o A= o « o o > { ° ° 2 D z =o }/ 2) ce ry) | \ ® ! © re o o ~ < -- — i | oO o oo cu (= D, aff. complexus ; = » oC re) joo / \ o o @ . o > i u I c a a = > / “sd 3 ° BD ' ae “4G 0c = a 3 3 D, minor ‘ 2 5 c nN 2 a o fe} = o KO C. trifills ener KLADAAR MM. C, medius C. normalis aot miserabilis C iF O, abbreviatus G. sp O. fastigatus , G? ‘venustus cf, venustus P. paciticus CT ' es — O, denticulatus G? avitus ees ON “SSOcc> | N. velatus d A. ascensus <= =) | | 1 Prrervecoueye : \ Srnecerssaay P? lautus P. acuminatus P? craticulus \ P. pacificus C? extraordinarius | Subzone \ Zone , D. complexus Subzone G persculptus Zone Fig. 5 Detail of sediments and graptolite ranges for the top Upper Hartfell Shale and basal Birkhill Shale. THE ORDOVICIAN-SILURIAN BOUNDARY STRATOTYPE 27 of the Birkhill Shale in 1968. Rickards (1970) and Hutt (1974) also used this zone as the basal biostratigraphic division of the Lower Silurian Skelgill Formation in northern England. The base of the P. acuminatus Zone is marked by the first appearance of Akidograptus ascensus Davies and Paraorthograptus acuminatus (Nicholson) s.l. at 1-6m above the base of the Birkhill Shale (Fig. 5). It is this level which has now been adopted as the defined Ordovician— Silurian boundary (Cocks 1985), marked by the first occurrence of A. ascensus. Most previous publications (e.g. Toghill 1968a; Cocks et al. 1970) have taken the base of the Birkhill Shale as marking the Ordovician-Silurian boundary. This interval covers a change from grey to black shale, is unfossiliferous and clearly unsuited as a zonal boundary, let alone for an international system boundary stratotype. Similar barren intervals seem to occur at this level in every other graptolitic succession in the world; they are probably related to eustatic sea level changes induced by late Ordovician glaciation in the southern hemisphere (see Rong 1984). In addition to the problem of barren intervals, faunal changes accompanying the transition between the G. persculptus and C.? extraordinarius Zones are poorly understood. Few graptol- ite taxa are present, following the mass extinction at the D. anceps—C.? extraordinarius zonal boundary. Elles (1922, 1925) referred the basal interval of the Birkhill Shale to the ‘Zone of Glyptograptus persculptus and Cephalograptus acuminatus’. In the earlier of these publications (1922: 195) she suggested that this lowest Llandovery zone should perhaps be assigned to the Ordovician owing to the lack of monograptids. It is interesting to note that this proposal has now been partially adopted. Atavograptus ceryx (Rickards & Hutt) occurs 1:9 to 2:3m above the base of the Birkhill Shale. Recent recollecting indicates that it is probably restricted to such a level low in the P. acuminatus Zone at Dob’s Linn. A. ceryx was first recorded from strata referred to the G. persculptus Zone in the English Lake District (Rickards & Hutt 1970), but was later found in the basal P. acuminatus Zone of that area (Hutt 1975), in association with A. ascensus. Monograptus cyphus praematurus Toghill and Atavograptus atavus (Jones) are the next mono- graptids found at Dob’s Linn, marking the boundary between the P. acuminatus and Cysto- graptus vesiculosus Zones (Toghill 1968a). Lapworth (1882: 624) recorded an assemblage of ‘Climacograptus scalaris, Dimorphograptus acuminatus and ?Monograptus tenuis’ from a section through the Lower Silurian at Girvan, south-west Scotland. Jones (1921: 155) remarked that such an assemblage seemed anomalous for the P. acuminatus Zone; it may, however, prove that the monograptid was A. ceryx and that the interval was equivalent to the early P. acuminatus Zone of Dob’s Linn and northern England. Relocation and recollection of Lapworth’s horizon could clearly prove significant as a comparative basal Silurian section. Future research Dob’s Linn has now been adopted as Ordovician-Silurian boundary stratotype, the boundary being set at the base of the P. acuminatus Zone 1:6m above the base of the Birkhill Shale in the Linn Branch trench (Figs 4, 5). This renders necessary ratification and expansion of Williams’ (1983) study of the interval. Other outstanding research still required includes: 1. Detailed study of the basal Birkhill Shale at remaining sections of Dob’s Linn, and at other comparative localities in the Central Belt of the Southern Uplands. 2. Biostratigraphical and taxonomic revision of the Glenkiln Shale, the lower part of the Lower Hartfell Shale and remainder of the Birkhill Shale, employing continuous, bed-by-bed collecting techniques. 3. Critical sedimentological logging and study of the Moffat Shale at Dob’s Linn, with subsequent integration of faunal data, in order to provide a clearer understanding of original depositional setting. Acknowledgements I thank J. K. Ingham for his invaluable supervision of my original work at Glasgow University, which was funded by a NERC postgraduate fellowship. I. Strachan provided efficient guidance through Lapworth’s 28 S. H. WILLIAMS original material stored at Birmingham University and critically read the manuscript. Several colleagues at Memorial University gave fruitful, often lively discussion on biostratigraphical and seimentological problems, particularly C. R. Barnes and M. Coniglio. References Bassett, M. G. 1985. ‘Transition Rocks and Grauwacke’—the Silurian and Cambrian systems through 150 years. Episodes, Ottawa, 8: 231-235. Billings, E. 1861—65. Paleozoic Fossils, 1. 426 pp., 401 figs. Montreal, Canada geol. Surv. Bluck, B. J. 1984. Pre-Carboniferous history of the Midland Valley of Scotland. Trans. R. Soc. Edinb. (Earth Sci.) 75: 275-295. Carruthers, W. 1858. Dumfriesshire graptolites with descriptions of three new species. Proc. R. phys. Soc. Edinb. 1: 466-470. Cocks, L. R. M. 1985. The Ordovician-Silurian boundary. Episodes, Ottawa, 8: 98-100. —— Toghill, P. & Ziegler, A. M. 1970. Stage names within the Llandovery Series. Geol. Mag, Cambridge, 107: 79-87. Coniglio, M. (1985). Origin and diagenesis of fine-grained slope sediments: Cow Head Group (Cambro— Ordovician), western Newfoundland. Unpubl. PhD thesis, Memorial Univ. of Newfoundland (2 vols). Craig, G. Y. & Walton, E. K. 1959. Sequence and structure in the Silurian rocks of Kirkcudbrightshire. Geol. Mag., Hertford, 96: 209-220. Dairon, J. 1869. [Graptolites from the Silurian shales of the Moffat district.] Proc. nat. Hist. Soc. Glasgow 1: 268-269. Davies, K. A. 1929. Notes on the graptolite faunas of the Upper Ordovician and Lower Silurian. Geol. Mag., London, 66: 1—27. Dewey, J. F. 1971. A model for the Lower Palaeozoic evolution of the southern margin of the early Caledonides of Scotland and Ireland. Scott. J. Geol., Edinburgh, 7: 219-240. Eales, M. H. 1979. Structure of the Southern Uplands of Scotland. Spec. Publs geol. Soc. Lond. 8: 269-273. Elles, G. L. 1922. The graptolite faunas of the British Isles. A study in evolution. Proc. geol. Ass., London, 33: 168-200. —— 1925. The characteristic assemblages of the graptolite zones of the British Isles. Geol. Mag., London, 62: 337-347. —— & Wood, E. M. R. 1901-18. A monograph of British Graptolites. Palaeontogr. Soc. (Monogr.), London, m + clxxi + 539 pp., 52 pls. Gibson, W. 1921. Charles Lapworth, M.Sc., LL.D., F.R.S., F.G.S. 1842-1920. Trans. Edinb. geol. Soc. 11: 239-242. Hall, J. 1858. Descriptions of Canadian graptolites. Rep. Prog. geol. Surv. Canada, Toronto, 1857: 111-145. —— 1865. Graptolites of the Quebec Group. Figures and Descriptions of Canadian organic remains, Dec. II. 151 pp. Montreal, Canada geol Surv. Hall, J., Powell, D. W., Warner, M. R., El-Isa, Z. H. M., Adesanya, O. & Bluck, B. J. 1983. Seismological evidence for shallow crystalline basement in the Southern Uplands of Scotland. Nature, Lond. 305: 418-420. Harkness, R. 1851. On the Silurian rocks of Dumfriesshire and Kirkudbrightshire. Q. JI geol. Soc. Lond. 7: 46-58. Hopkinson, J. 1871. On Dicellograpsus, a new genus of graptolite. Geol. Mag., London, (dec. 1) 8: 20-26. —— & Lapworth, C. 1875. Descriptions of the graptolites of the Arenig and Llandeilo rocks of St. David’s. Q. JI geol. Soc. Lond. 31: 631-672. Hutt, J. E. 1974-75. The Llandovery graptolites of the English Lake District. 137 pp., 26 pls. Palaeontogr. Soc. (Monogr.), London. Ingham, J. K. 1979. The Moffat area. In M. G. Bassett, L. R. M. Cocks, C. H. Holland, J. K. Ingham, J. D. Lawson, R. B. Rickards & J. T. Temple, Guidebook to field meeting, Great Britain, March 30—April 11, 1979: 42-46. Subcommission on Silurian Stratigraphy, Ordovician—Silurian Boundary Working Group, LU.GS. —— & Strachan, I. 1974. The Moffat District. In D. A. Bassett, J. K. Ingham & A. D. Wright (eds), Field Excursion Guide to type and classical sections in Britain: 45-49. London, Palaeont. Assoc. (Ordovician System Symposium, Birmingham, 1974). Jenkyns, H. C. 1978. Pelagic environments. In H. G. Reading (ed.), Sedimentary environments and facies: 314-371. Oxford. THE ORDOVICIAN-SILURIAN BOUNDARY STRATOTYPE 29 Jones, O. T. 1909. The Hartfell-Valentian succession in the district around Plynlimon and Pont Erwyd (North Cardiganshire). Q. J! geol. Soc. Lond. 65: 463-537, pls 1, 2. —— 1921. The Valentian series. Q. JI geol. Soc. Lond. 77: 144-174. Lapworth, C. 1870. On the Silurian rocks in the neighbourhood of Galashiels. Geol. Mag., London, (dec. 1) 7: 204-209, 279-284. —— 1872. On the Silurian rocks of the south of Scotland. Trans. geol. Soc. Glasg. 4: 164-174. —— 1876. The Silurian System in the south of Scotland. In J. Armstrong, et al. (eds), Catalogue of western Scottish fossils: 1-28. Glasgow. 1877. On the graptolites of County Down. Rep. Proc. Belf. Nat. Fld Club, 1876-77 (Appendix): 125-144, pls 5—7. —— 1878. The Moffat Series. Q. JI geol. Soc. Lond. 34: 240-346. —— 1879. On the tripartite classification of the Lower Palaeozoic rocks. Geol. Mag., London, (dec. 2) 6: 1-15. — 1880. On new British graptolites. Ann. Mag. nat. Hist., London, (5) 5: 149-178. —— 1882. The Girvan succession. Part 1. Stratigraphy. Q. JI geol. Soc. Lond. 38: 537-666, pls 24-25. —— 1889. On the Ballantrae rocks of the south of Scotland and their place in the upland sequence. Geol. Mag., London, (dec. 3) 6: 20—24, 59-69. —— 1897. Die Lebensweise der Graptolithen. In J. Walther, Ueber die Lebensweise fossiler Meeresthiere. Z. dt. geol. Ges., Berlin, 49: 238-258. Leggett, J. K. 1980. British Lower Palaeozoic black shales and their palaeo-oceanographic significance. J. geol. Soc. Lond. 137: 139-156. — McKerrow, W. S. & Eales, M. H. 1979. The Southern Uplands of Scotland: a Lower Palaeozoic accretionary prism. J. geol. Soc. Lond. 136: 755-770. McKerrow, W. S., Leggett, J. K. & Eales, M. H. 1977. Imbricate thrust model of the Southern Uplands of Scotland. Nature, Lond. 267: 237-239. Mitchell, A. H. G. & McKerrow, W. S. 1975. Analogous evolution of the Burma orogen and the Scottish Caledonides. Bull. geol. Soc. Am., New York, 86: 305-315. Moseley, F. 1978. The geology of the English Lake District. An introductory review. In F. Moseley (ed.), The geology of the Lake District. Occ. Publ. Yorks. geol Soc. 3: 1-16. Murphy, F. C. & Hutton, D. H. W. 1986. Is the Southern Uplands of Scotland really an accretionary prism? Geology, Boulder, Colo. 14: 354-357. Needham, D. T. & Knipe, R. J. 1986. Accretion- and collision-related deformation in the Southern Uplands accretionary wedge, southwestern Scotland. Geology, Boulder, Colo., 14: 303-306. Nicholson, H. A. 1867. Graptolites of the Moffat Shale. Geol. Mag., London, (dec. 1) 4: 108-113. —— 1869. On some new species of Graptolites. Ann. Mag. nat. Hist., London, (4) 4: 231-242. —— 1870. On the British species of Didymograpsus. Ann. Mag. nat. Hist., London, (4) 5: 337-357. —— 1875. On a new genus and some new species of graptolites from the Skiddaw Slates. Ann. Mag. nat. Hist., London, (4) 16: 269-273. — 1876. Notes on the correlation of the graptolitic deposits of Sweden with those of Britain. Geol. Mag., London, (dec. 2) 3: 529-539. Oliver, G. J. H., Smellie, J. L., Thomas, L. J., Casey, D. M., Kemp, A. E. S., Evans, L. J., Baldwin, J. R. & Hepworth, B. C. 1978. Early Palaeozoic metamorphic history of the Midland Valley, Southern Uplands—Longford Down massif and the Lake District, British Isles. Trans. R. Soc. Edinb. (Earth Sci.) 75: 245-258. Packham, G. H. 1962. Some diplograptids from the British Lower Silurian. Palaeontology, London, 5: 498-526. Peach, B. N. & Horne, J. 1899. The Silurian rocks of Britain. I, Scotland. Mem. geol. Surv. U.K., London: 1-749. Piper, D. J. W. 1972. Sediments of the Middle Cambrian Burgess Shale, Canada. Lethaia, Oslo, 5: 169-175. Powell, D. W. 1971. A model for the Lower Palaeozoic evolution of the southern margin of the early Caledonides of Scotland and Ireland. Scott. J. Geol., Edinburgh, 7: 369-372. Rickards, R. B. 1970. The Llandovery (Silurian) graptolites of the Howgill Fells, Northern England. Palaeontogr. Soc. (Monogr.), London. 108 pp., 8 pls. 1979. [New information on some Ordovician-Silurian boundary sections in Great Britain.] Izv. Akad. Nauk kazakh. SSR, Alma-Ata, (Geol.) 4: 103-107 [In Russian]. — & Hutt, J. E. 1970. The earliest monograptid. Proc. geol. Soc., London, 1663: 115-119. Rong Jia-yu 1984. Distribution of the Hirnantia fauna and its meaning. In D. L. Bruton (ed.), Aspects of the Ordovician System: 101-112. Universitetsforlaget, Oslo. 30 S. H. WILLIAMS Sedgwick, A. 1850. On the Geological Structure and Relations of the Frontier Chain of Scotland. Edinb. new phil. J. 51: 250-258. Stow, D. A. V. & Piper, D. J. W. 1984. Deep water fine-grained sediments: facies models. In D. A. V. Stow & D. J. W. Piper (eds), Fine-grained sediments: 611-646. Boston. Toghill, P. 1968. The stratigraphical relationships of the earliest Monograptidae and the Dimorphograp- tidae. Geol. Mag., Hertford, 105: 46-51. 1968a. The graptolite assemblages and zones of the Birkhill Shales (Lower Silurian) at Dobb’s Linn. Palaeontology, London, 11: 654-668. 1970. Highest Ordovician (Hartfell Shales) graptolite faunas from the Moffat area, South Scotland. Bull. Br. Mus. nat. Hist., London, (Geol.) 19: 1-26, pls 1-16. Walton, E. K. 1963. Sedimentation and structure in the Southern Uplands. In M. R. W. Johnson & F. H. Stewart (eds), The British Caledonides: 71-97. Edinburgh. Webb, B. 1983. Imbricate structure in the Ettrick area, Southern Uplands. Scott. J. Geol., Edinburgh, 19: 387-400. Williams, S. H. 1982. The Late Ordovician graptolite fauna of the Anceps Bands at Dob’s Linn, southern Scotland. Geologica Palaeont., Marburg, 16: 29-56, 4 pls. 1982a. Upper Ordovician graptolites from the top Lower Hartfell Shale (D. clingani and P. linearis zones) near Moffat, southern Scotland. Trans. R. Soc. Edinb. (Earth Sci.) 72: 229-255. 1983. The Ordovician-Silurian boundary graptolite fauna of Dob’s Linn, southern Scotland. Palae- ontology, London, 26: 605-639. 1987. Upper Ordovician graptolites from the D. complanatus Zone of the Moffat and Girvan districts and their significance for correlation. Scott. J. Geol., Edinburgh, 23: 65-92. —— & Lockley, M. G. 1983. Ordovician inarticulate brachiopods from graptolitic shales at Dob’s Linn, Scotland; their morphology and significance. J. Paleont., Tulsa, 57: 391400. —— & Rickards, R. B. 1984. Palaeoecology of graptolitic black shales. In D. L. Bruton (ed.), Aspects of the Ordovician System: 159-166. Universitetsforlaget, Oslo. Conodonts from the Ordovician—Silurian Boundary Stratotype, Dob’s Linn, Scotland C. R. Barnes’ and S. H. Williams? ‘Geological Survey of Canada, 601 Booth St, Ottawa, Ontario K1A OE8, Canada ?Department of Earth Sciences, Memorial University of Newfoundland, St John’s, Newfoundland A1B 3X5, Canada Synopsis About one hundred poorly preserved conodonts have been collected from surfaces of shale from seven graptolite zones of the Dob’s Linn boundary stratotype section, mainly from the D. anceps Zone. Attempts to recover conodonts by dissolving siltstones and cherts from the section were unsuccessful. When preserved, the conodont phosphatic material provides Colour Alteration Index values of CAI 5-7, indicating burial temperatures in excess of 300°C. The sparse, low diversity faunas assist in correlating conodont and graptolite zones. Amorphognathus sp. and Scabbardella sp. cf. S. altipes were found in the G. persculptus Zone, suggesting that the conodont turnover must lie at least high within this zone. Lowest Silurian strata yielded rare, undiagnostic coniform taxa and an element referred tentatively to Oulodus? kentuckyensis. The results encourage further efforts in retrieving conodonts from graptolitic shale sequences, but the precise correlation of the conodont turnover with respect to the defined base of the Silurian remains in question. Introduction The Ordovician—Silurian boundary was finally designated in 1985 at 1:-6m above the base of the Birkhill Shale in the Linn Branch section of Dob’s Linn, southern Scotland, at the base of the Parakidograptus acuminatus Zone (Williams 1983 and this volume; Cocks 1985). Detailed work on the rich graptolite faunas has been carried out by a number of previous researchers, especially Lapworth, Elles & Wood, Toghill and Williams (see Williams 1983, this volume). The section, however, has yielded no other biostratigraphically useful fossils in abundance; there are rare inarticulate brachiopods (Williams & Lockley 1983) and a species of a blind dalmanitid trilobite. Lamont & Lindstr6m (1957) reported conodonts from cherts in the Southern Uplands of Scotland, including Dob’s Linn, but only gave identifications and details of the Arenig and Llandeilo faunas. One critical problem in the debate concerning the definition of the Ordovician—Silurian boundary and subsequent selection of a stratotype was that few candidate sections contained both graptolites and conodonts. At the level of the G. persculptus and P. acuminatus Zones (Fig. 1) in particular, there are difficulties in correlating the graptolite and conodont zones and the two respective extinction events (e.g. Barnes & Bergstr6m, this volume). It is, therefore, both encouraging and important to report in this paper the discovery of conodonts at several levels in the Dob’s Linn boundary stratotype section. While scanning shale surfaces under the microscope during the investigation of graptolites, Williams observed a number of microfossils which have since been identified by Barnes. Further collections were made by Williams in 1985; this time, in addition to the scanning of shale surfaces, samples of shales, siltstones and cherts were processed through a variety of standard chemical rock digestion techniques employed for conodonts (e.g. acetic and hydro- fluoric acids; bleach). The latter results were disappointing in that most lithologies appeared to be barren of conodonts, although this may have been due to inadequate preservation (see below). The remaining new collections revealed many additional conodont horizons, but yielded few diagnostic elements from new stratigraphical levels. This project however demon- strates that conodonts are present, and moderately abundant at some horizons, in graptolitic shales deposited in a deep oceanic environment which has been interpreted as an accretionary prism (McKerrow et al. 1979; see other recent interpretations by Needham & Knipe 1986 and Bull. Br. Mus. nat. Hist. (Geol) 43: 31-39 Issued 28 April 1988 32 BARNES & WILLIAMS STRATOT YPE TRENCH 0? waterfall key section sheep track AUT rocky bluff uy Wi \ Ys scree ~----- boundary fault Gala Greywacke 5 Sees Birkhill Shale acuminatus persculptus extraorainarnss anceps Upper Hartfell SSeS ae Shale Lower Bs tee See Hartfell Shale wilsoni___ . peltifer se NS seman e conodonts e Extraordinarius Band a Anceps Bands Cc Complanatus Bands Fig. 1 Simplified geological map and stratigraphical section of Dob’s Linn, showing position of conodont localities and horizons (after Williams 1980). CONODONTS FROM THE ORDOVICIAN-SILURIAN BOUNDARY STRATOTYPE 33 Murphy & Hutton 1986). Careful microscopical examination of similar shales in other sequences should reveal many new conodont faunas and assist integration of graptolite and conodont biostratigraphic zonation schemes. Results Following the discovery of the microfossils, re-examination of earlier material, together with the new shale collections, has involved the study of several hundred surfaces for conodonts. Conodonts and rare scolecodonts are present. The conodonts always occur as isolated ele- ments; no fused clusters or natural assemblages were discovered. The elements are poorly preserved, typically being fractured by tectonic stretching and commonly with only part of the phosphatic skeletal material preserved. This may, in part, explain the difficulty in obtaining identifiable conodonts from dissolved samples. For some, only an external mould remains, but latex casts have been successfully made which permit specific identifications (e.g. Pl. 1, fig. 10; Pl. 2, fig. 12). The conodonts provide Colour Alteration Index values of CAI 5-7. This is in agreement with the general high thermal values reported elsewhere in the Southern Uplands of Scotland by Bergstrom (1980), indicating burial temperatures exceeding 300°C. About one hundred conodont elements have been recognized, the majority of which are identifiable only to generic level. The diversity of the fauna is low, but zonal species are present. Nearly all the conodonts come from Ordovician strata, in particular the D. anceps Zone; unfortunately, conodonts are especially rare near the Ordovician-Silurian boundary. Hartfell Shale conodonts Most of the conodonts from the Dob’s Linn section come from the Hartfell Shale. They were recovered at various levels within the Dicranograptus clingani, Pleurograptus linearis, Dicello- graptus complanatus and D. anceps Zones, but principally from the latter zone. Details of stratigraphy, together with a revision of the graptolite faunas from the D. clingani, P. linearis, and D. anceps Zones, have been published by Williams (1982a, b). Conodonts from the D. clingani Zone were not identifiable; those from the P. linearis Zone included two specimens of Amorphognathus superbus (Rhodes) from 1-1—1:2 m and 0-3—0-45 m below the top of the Lower Hartfell Shale, several specimens of Amorphognathus sp. and a specimen of Walliserodus unidentifiable to species level. The precise level at which A. superbus evolved into A. ordovicicus (i.e. base of the A. ordovi- cicus Zone) in terms of graptolite zones remains to be established. This zonal boundary appears to lie within the upper Pusgillian Stage or lower Cautleyan Stage (Bergstr6m 1971, 1983; Orchard 1890; Bergstrom & Orchard 1985), although Savage & Bassett (1985) tentatively suggest a late Caradoc age. In North America, this boundary occurs in the lower Maysvillian Stage (Sweet & Bergstrom 1971). The D. clingani—P. linearis zonal boundary is approximately equivalent to, or slightly predates, the base of the earliest Ashgill Pusgillian Stage (Williams & Bruton 1983). The samples yielding A. superbus are from the top of the Lower Hartfell Shale (mid P. linearis Zone; Williams 1982a: fig. 3) which probably falls within the Pusgillian Stage. A single identifiable conodont was recovered from the D. complanatus Zone of the Upper Hartfell Shale, namely Amorphognathus ordovicicus Branson & Mehl from the lower Com- planatus Band. At Myoch Bay in the Girvan area, southern Scotland, the D. complanatus Zone of the Upper Whitehouse Group also yields shelly fossils of Pusgillian age (Ingham 1978; Harper 1979). Conodonts from these strata (Sweet & Bergstrom 1976: 135-136; Bergstrom & Orchard 1980) do not allow a zonal assignment. It must be emphasized that the material at hand comprises only a single, poorly preserved amorphognathodontiform element; this limited evidence suggests that the A. ordovicicus Zone boundary lies within the Pusgillian rather than the Cautleyan. The D. anceps Zone is recognized in the Upper Hartfell Shale by a series of thin black shales assigned to Anceps Bands A-E (e.g. Williams 1982b). These contain the most abundant cono- dont fauna from the Dob’s Linn section. No significant difference was observed in the conodont fauna of the various bands except in terms of relative abundance. Band A yielded rare speci- 34 BARNES & WILLIAMS CONODONTS FROM THE ORDOVICIAN-SILURIAN BOUNDARY STRATOTYPE 35 mens assignable to only two species: Amorphognathus ordovicicus and Protopanderodus liripipus Kennedy, Barnes & Uyeno. Band B produced conodonts referred to A. sp. cf. A. ordovicicus, Scabbardella altipes (Henningsmoen) and an oistodontiform element that probably belonged to Hamarodus europaeus (Serpagli). Band C contained only P. liripipus, and Band D yielded A. sp. cf. A. ordovicicus and S. altipes; both had only rare fragmentary conodonts. Band E contained slightly more specimens including Amorphognathus sp., P. liripipus and S. altipes. The D. anceps Zone therefore yields conodonts belonging to the A. ordovicicus Zone. Orchard (1980) reco- vered H. europaeus from only Rawthyan and Hirnantian strata although the range of this species has now been extended into the Cautleyan by Barnes & Bergstrém (this volume). No conodonts were recovered from the l-cm black shale Extraordinarius Band of the top Upper Hartfell Shale, which yields C.? extraordinarius Zone graptolites of probable mid- Hirnantian age (Williams 1983). Birkhill Shale conodonts The Birkhill Shale includes the upper part of the top Ordovician G. persculptus Zone, the basal Silurian Parakidograptus acuminatus Zone and subsequent Llandovery graptolite zones (Toghill 1968; Williams 1983). The lower few metres of the Birkhill Shale is the critical interval from which turnover conodonts need to be recovered, but unfortunately no especially diagnostic taxa were found. In strata of the G. persculptus Zone, the few specimens observed were all coniform except for one slightly crushed and distorted specimen of Amorphognathus sp. from 0-12—-0-2m above the base of the Birkhill Shale. The coniform taxa include Dapsilodus obliquicostatus (Branson & Mehl) and Scabbardella sp. cf. S. altipes. The latter occurs at 0-95m above the base of the Birkhill Shale. Neither Amorphognathus nor Scabbardella are known with certainty from Silu- rian strata. This limited evidence, based on rare, poorly preserved specimens, suggests that most of the G. persculptus Zone may lie below the main Ordovician—Silurian conodont turnover (see discussion by Barnes & Bergstrom, this volume). The P. acuminatus Zone, beginning at 1-6m above the base of the Birkhill Shale, and the overlying Cystograptus vesiculosus Zone contained a few conodonts assigned to Dapsilodus obliquicostatus and Decoriconus sp. In addition a single, small, poorly preserved ligonodiniform element was found at 1:75m above the base of the Birkhill Shale. The form of the lateral PLATE 1 Conodonts from the Lower and Upper Hartfell Shale, Dob’s Linn, Scotland. Figs 1,2 Amorphognathus superbus (Rhodes) x 35. 1, dextral blade element, upper view. HM Y155. 1-1m below top of Lower Hartfell Shale, P. linearis Zone. North Cliff. 2, dextral blade element, upper view of mould. HM Y157. 0-3—0-45 m below top of Lower Hartfell Shale, North Cliff. Fig.3 Walliserodus sp. x 70. Lateral view. HM Y201. Top of Lower Hartfell Shale, North Cliff. Figs 4, 5,13 Amorphognathus ordovicicus Branson & Mehl. x 35. Upper Hartfell Shale. 4, sinistral blade element, upper view of mould. HM Y159. Lower Complanatus Band. 5, dextral blade element, upper view of mould. HM Y107. Anceps Band A. Long Burn. 13, dextral blade element, upper view of mould. HM Y129. Anceps Band D. Main Cliff. Figs 6, 12, 16 Protopanderodus liripipus Kennedy, Barnes & Uyeno. x 55. Upper Hartfell Shale. 6, scandodontiform element. HM Y109a. Anceps Band A, Long Burn. 12, symmetrical element. HM Y121. Anceps Band C. Main Cliff. 16, scolopodontiform element. HM Y135. Anceps Band E. Linn Branch. Figs 7, 11, 14, 15,18 Scabbardella altipes Henningsmoen. Lateral views. x 55. Upper Hartfell Shale. 7, 2acodontiform element. HM Y203. Anceps Band B. Linn Branch. 11, distacodontiform element. HM Y112. Anceps Band B. Main Cliff. 14, acodontiform element. HM Y202. Anceps Band D. Linn Branch. 15, distacodontiform element. HM Y126. Anceps Band D. Long Burn. 18, dis- tacodontiform element. HM Y204. 40cm above Anceps Band E, Linn Branch. Fig. 8 Hamerodus europaeus (Serpagli). x 55. Oistodontiform element. HM Y205. Anceps Band B. Linn Branch. Figs 9, 10 Amorphognathus sp. cf. A. ordovicicus Branson & Mehl. x 35. 9, dextral blade. Upper view of mould. HM Y114b. Anceps Band B. Main Cliff. 10, latex cast of HM Y114b (Fig. 9). Fig. 17 Amorphognathus sp. x 35. Dextral blade element, upper view of mould. HM Y136. Anceps Band E. Linn Branch. BARNES & WILLIAMS 36 CONODONTS FROM THE ORDOVICIAN-SILURIAN BOUNDARY STRATOTYPE Si process extends into the shale but its shape is revealed by a latex cast. The element is assigned tentatively to Oulodus? kentuckyensis (Branson & Mehl). The latter species is known only from Silurian strata elsewhere (e.g. Anticosti Island, McCracken & Barnes 1981). Summary About 100 conodont elements have been observed on shale surfaces from the Dob’s Linn boundary stratotype section. Most are from black shales, but occasional specimens also occur within paler grey shales and siltstones. The elements are poorly preserved, fractured and commonly occur as moulds; the Colour Alteration Index values are in the range of CAI 5-7 indicating burial temperatures exceeding 300°C. Identification of most elements can be made only to generic level; a selection of the better specimens are here illustrated (Figs 2, 3) but the photography for many proved difficult and not all details of micromorphology could be repro- duced. The diversity of the faunas is low, typically 3—5 species per graptolite zone interval. This may be expected in the deep oceanic environment of the Hartfell Shale and Birkhill Shale, but is probably also related to the limited material discovered. Siltstone, shale and chert samples were also processed chemically but yielded no identifiable conodonts. Although the sparse fauna and poor preservation must be taken into account, the following biostratigraphic conclu- sions may be drawn from this study. 1. Amorphognathus superbus is present in the Pleurograptus linearis Zone near the top of the Lower Hartfell Shale (based only on amorphognathodontiform, not holodontiform elements). Amorphognathus ordovicicus occurs in the Dicellograptus complanatus Zone of the Upper Hart- fell Shale. This suggests that the A. superbus—A. ordovicicus zonal boundary is not far removed from that of the P. linearis and D. complanatus Zones and lies within the Pusgillian Stage. 2. Most of the conodonts come from the Dicellograptus anceps Zone; all the Anceps Bands A-E of the Upper Hartfell Shale yielded specimens, which are indicative of the A. ordovicicus Zone. Conodonts also occur at several grey, silty, non-graptolitic horizons during this interval. 3. No conodonts were recovered from the 1-cm black shale of the Climacograptus? extraordi- narius Zone. 4. The lower 1-6m of the Birkhill Shale, belonging to the Glyptograptus persculptus Zone, contained two poor specimens of Amorphognathus sp. and Scabbardella sp. cf. S. altipes, known only from Ordovician strata. This suggests that the major conodont turnover (Barnes & Bergstrom, this volume) occurred at a level equivalent to at least high in the G. persculptus Zone. PLATE 2 Conodonts from the Birkhill Shale, Dob’s Linn, Scotland. Figs 1-16 arranged in order of stratigraphical occurrence of specimens. G. persculptus Zone (Figs 1-9); P. acuminatus Zone (Figs 10-14); C. vesiculosus Zone (Figs 15, 16). Fig. 1 Amorphognathus sp. x 35. Upper view, distorted specimen. HM Y142. 0:12-0:2m above base of Birkhill Shale. Figs 2, 3, 5,9 Dapsilodus sp. Lateral views. 2, HM Y206. x 90. 0-55 m above base of Birkhill Shale. 3, HM Y207. x 55. 0:95m above base of Birkhill Shale. 5, HM Y208. x 80. 0-95m above base of Birkhill Shale. 9, HM Y209. x 55. 1:5m above base of Birkhill Shale. Figs 4,6 Scabbardella altipes Henningsmoen. Lateral views. x 55. 4, HM Y210. 0-95m above base of Birkhill Shale. 6, HM Y211. 0-95 m above base of Birkhill Shale. Figs 7, 15, 16 Dapsilodus obliquicostatus (Branson & Mehl). Lateral views. 7, HM Y213. x 70. 1m above base of Birkhill Shale. 15, HM Y214. x 55. 5m above base of Birkhill Shale. 16, HM Y215. x 55. 5-5m above base of Birkhill Shale. Fig. 8 Drepanoistodus sp. x 70. Lateral view. Drepanodontiform element. HM Y212. 1m above base of Birkhill Shale. Figs 10, 12, 13 Decoriconus sp. x 55. Lateral views. 10, HM Y216. 1:75m above base of Birkhill Shale. 12, HM Y217. 1-75m above base of Birkhill Shale. 13, latex cast of HM Y217 (Fig. 12). re Figs 11, 14 cf. Oulodus? kentuckyensis (Branson & Branson). x 105. Lateral views. 11, ligonodini- form element. HM Y218. 1-75m above base of Birkhill Shale. 14, latex cast of HM Y218 (Fig. 11). 38 BARNES & WILLIAMS 5. Silurian conodonts from the Parakidograptus acuminatus and Cystograptus vesiculosus Zones include mostly coniform taxa (Dapsilodus, Decoriconus) which cross the systemic bound- ary at other localities. A poor single element assigned tentatively to Oulodus? kentuckyensis, which elsewhere is known only from Silurian strata, was found in the P. acuminatus Zone. These results suggest that more attention should be made to recover conodonts from shales, particularly in graptolitic shale sequences. The above data must be used with caution until more material is discovered. However, the situation is perhaps analogous to the presence of poorly preserved, rare graptolites within the conodont-rich Anticosti Island carbonate bound- ary sequence (McCracken & Barnes 1981; Riva, this volume). It remains one of the future challenges to find a boundary sequence that yields both well preserved and abundant, bio- stratigraphically diagnostic conodonts and graptolites across the systemic boundary. Acknowledgements Ms Felicity H. C. O’Brien provided invaluable research assistance aspects and C.R.B. acknowledges financial support from the Natural Sciences and Engineering Research Council of Canada. References Barnes, C. R. & Bergstrom, S. M. 1988. Conodont biostratigraphy of the uppermost Ordovician and lowermost Silurian. Bull. Br. Mus. nat. Hist., London, (Geol.) 43: 325—343. Bergstrom, S. M. 1971. Conodont biostratigraphy of the Middle and Upper Ordovician of Europe and Eastern North America. In W. C. Sweet & S. M. Bergstrom (eds), Symposium on Conodont Strati- graphy. Mem. geol. Soc. Am., Boulder, Col., 127: 83-157, 2 pls. 1980. Conodonts as paleotemperature tools in Ordovician rocks of the Caledonides and adjacent areas in Scandanavia and the British Isles. Geol. For. Stockh. Forh. 102: 377-392. 1983. Biogeography, evolutionary relationships, and biostratigraphic significance of Ordovician platform conodonts. Fossils Strata, Oslo, 15: 35-58, 1 pl. —— & Orchard, M. J. 1985. Conodonts of the Cambrian and Ordovician systems from the British Isles. In A. C. Higgins & R. L. Austin (eds), A stratigraphical index of conodonts: 32-67, 5 pls. London. Cocks, L. R. M. 1985. The Ordovician—Silurian boundary. Episodes, Ottawa, 8: 98-100. Harper, D. A. T. 1979. The environmental significance of some faunal changes in the Upper Ardmillan succession (upper Ordovician), Girvan, Scotland. Spec. Publs geol. Soc. Lond. 8: 439-445. Ingham, J. K. 1978. Geology of a continental margin. 2: Middle and Late Ordovician transgression, Girvan. Geol. J., Liverpool (Spec. Iss.) 10: 163-176. McCracken, A. D. & Barnes, C. R. 1981. Conodont biostratigraphy and paleoecology of the Ellis Bay Formation, Anticosti Island, Quebec, with special reference to Late Ordovician—Early Silurian chroro- stratigraphy and the systemic boundary. Bull. Geol. Surv. Can., Ottawa, 329 (2): 51-134, 7 pls. McKerrow, W.S., Leggett, J. K. & Eales, M. H. 1977. Imbricate thrust model of the Southern Uplands of Scotland. Nature, Lond. 267: 237-239. Lamont, A. & Lindstrom, M. 1957. Arenigian and Llandeilian cherts identified in the Southern Uplands of Scotland by means of conodonts, etc. Trans. Edinb. geol. Soc. 17: 60-70. Murphy, F. C. & Hutton, D. H. W. 1986. Is the Southern Uplands of Scotland really an accretionary prism? Geology, Boulder, Colo., 14: 354-357. Needham, D. T. & Knipe, R. J. 1986. Accretion- and collision-related deformation in the Southern Uplands accretionary wedge, southwestern Scotland. Geology, Boulder, Colo., 14: 303-306. Orchard, M. J. 1980. Upper Ordovician conodonts from England and Wales. Geologica Palaeont., Marburg, 14: 9-44. Savage, N. M. & Bassett, M. G. 1985. Caradoc—Ashgill conodont faunas from Wales and the Welsh Borderland. Palaeontology, London, 28: 679-714. Sweet, W. C. & Bergstrom, S. M. 1971. The American Upper Ordovician Standard. XIII: A revised time-stratigraphic classification of North American Upper Middle and Upper Ordovician rocks. Bull. geol. Soc. Am., New York, 82: 613-628. 1976. Conodont biostratigraphy of the Middle and Upper Ordovician of the United States Midcontinent. In M: G. Bassett (ed.), The Ordovician System: Proceedings of a Palaeontological Associ- ation Symposium, Birmingham, September, 1974: 121-151. Cardiff, Univ. Wales Press & Natl Mus. Wales. CONODONTS FROM THE ORDOVICIAN-SILURIAN BOUNDARY STRATOTYPE 39 Toghill, P. 1968. The graptolite assemblages and zones of the Birkhill Shales (Lower Silurian) at Dobb’s Linn. Palaeontology, London, 11: 654-668. Williams, S. H. 1980. An excursion guide to Dob’s Linn. Proc. geol. Soc. Glasgow 121/122: 13-18. —— 1982a. Upper Ordovician graptolites from the top Lower Hartfell Shale Formation (D. clingani and P. linearis zones) near Moffat, southern Scotland. Trans. R. Soc. Edinb. (Earth Sci.) 72: 229-255. —— 1982b. The Late Ordovician graptolite fauna of the Anceps Bands at Dob’s Linn, southern Scotland. Geologica Palaeont., Marburg, 16: 29—S6, 4 pls. 1983. The Ordovician—Silurian boundary graptolite fauna of Dob’s Linn, southern Scotland. Palae- ontology, London, 26: 605-639. & Bruton, D. L. 1983. The Caradoc—Ashgill boundary in the central Oslo Region and associated graptolite faunas. Norsk geol. Tidsskr., Oslo, 63: 147-191. — & Lockley, M. G. 1983. Ordovician inarticulate brachiopods from graptolitic shales at Dob’s Linn, Scotland; their morphology and significance. J. Paleont., Tulsa, 57: 391—400. Preliminary acritarch and chitinozoan distributions across the Ordovician-Silurian boundary stratotype at Dob’s Linn, Scotland G. M. Whelan Department of Geology, Glasgow University, Glasgow G12 8QQ. Synopsis Palynomorph distribution has been investigated across the Ordovician-Silurian boundary at Dob’s Linn, where the Hartfell Shale and Birkhill Shale are well exposed. Samples were taken from the anceps, extraordinarius, persculptus, and acuminatus graptolite Biozones at the stratotype Linn Branch section and also the Main Cliff. Graptolite debris is the dominant component of the organic fraction, but acritarchs, chitinozoans and scolecodonts also occur in small numbers. Although it has not been possible to define the position of the Ordovician-Silurian boundary by microflora, the presence of palynomorphs indicates that detailed sampling might provide the stratigraphical resolution necessary to do this. At Dob’s Linn in the Southern Uplands of Scotland, continuous sections through the Hartfell and Birkhill Shales (Caradoc to Llandovery) bracket the Ordovician—-Silurian boundary. These shales are replaced vertically by greywackes (the Gala Greywackes) in the maximus Zone. Fault bounded tracts showing similar transitions are common in the Southern Uplands. Systematic variations in the regional timing of this transition, and the complex younging relationships between and within tracts, are thought to reflect the progressive growth of an accretionary prism (McKerrow et al. 1977) during closure of the Iapetus Ocean. The 90m of Hartfell and Birkhill Shales exposed here (Williams 1981) represent a substantially condensed sequence, as an equivalent sequence a hundred kilometres to the west, at Girvan, is over 3000 m thick. This is a preliminary report of the distribution of acritarchs and chitinozoans across the newly formalized Ordovician—Silurian boundary at Dob’s Linn. Data from the anceps, extraor- dinarius, persculptus (all Ordovician) and acuminatus (Silurian) graptolite Biozones are present- ed. Palynomorphs were recovered from hydrofluoric and hydrochloric acid-etched residues and studied using the scanning electron microscope, or transmitted light microscope. Whilst grap- tolites are common at Dob’s Linn, and other fossils, such as scolecodonts, have been found sporadically, this is the first major palynological survey that has been undertaken on the Ordovician and the basal Silurian there. The older Upper Hartfell Shale is a sequence (28 m thick) of finely bioturbated massive grey mudstones (Williams & Rickards 1984), with subordinate thin black shale bands (two com- planatus bands, five anceps bands and one extraordinarius band), and metabentonite horizons. The Birkhill Shale (48 m) comprises a laminated, pyritous, black shale with abundant graptol- ites, and representing the persculptus to maximus Zones. The systemic boundary of the Ordovician-Silurian has been fixed at the base of the acuminatus graptolite Biozone, 1:6m above the base of the Birkhill Shale (Cocks 1985). Samples have been collected from two localities spanning the boundary, the Main Cliff and the Linn Branch section (the world stratotype of the Ordovician—Silurian Boundary). At Main Cliff the wilsoni to acuminatus graptolite Zones are exposed, and although some strike slip faulting has caused repetition of the upper anceps and extraordinarius black shale bands, the beds are consistently the right way up (Williams 1980). At the Linn Branch, the anceps to maximus Zones are present, and although the beds are overturned, the stratigraphy is not complicated by repetition. To date sampling has concentrated on the extraordinarius and anceps Zones. However, work in progress aims to characterize the distribution of palyno- morphs across the boundary. Bull. Br. Mus. nat. Hist. (Geol) 43: 41-44 Issued 28 April 1988 42 G. M. WHELAN 3 4 Figs 1-4 Chitinozoans and acritarchs from Dob’s Linn. 1, Ancyrochitina ancyrea (Eisenack 1931) Eisenack 1955. SU/DL/41, acumunatus Zone, Main Cliff, x 250. 2, Cyathochitina kukersiana (Eisenack 1934) Eisenack 1965. SU/DL/9, anceps Zone, Main Cliff, x 250. 3, Solisphaeridium nanum (Deflandre 1945) Turner 1984. SU/DL/12, anceps Zone, Main Cliff, x 530. 4, Diexallophasis sp. 1. SU/DL/10, anceps Zone, Main Cliff, x 470. Both groups of palynomorphs are unevenly distributed throughout the two sections although they are generally more abundant at Main Cliff. Acritarchs appear to be more important and better preserved in the grey mudstones, while chitinozoans appear to be more common in the black shales, although this is not always the case. Palynomorph colour varies from grey to black within a single sample, and probably reflects differences in wall thickness. Acritarchs can be divided into several groups (Downie et al. 1963): (a) Sphaeromorphs which are spherical. These are of limited biostratigraphical use as can be seen in Figs | and 2, and will not be mentioned further; (b) Acanthomorphs which have spines or processes; (c) Herkomorphs which have crested ridges forming polygonal fields; (d) Polygonomorphs which have a limited number of processes, usually between three and five; and (e) Netromorphs which are generally fusiform in shape. The Dob’s Linn samples are noticeably dominated by acanthomorph acri- tarchs and only a few samples contain representatives of the other groups. Anceps Zone Six samples have been studied from Main Cliff (only one of which is a grey mudstone) and sixteen acritarch and chitinozoa taxa have been found (Fig. 5). The chitinozoans Cyathochitina campanulaeformis (Eisenack), C. kukersiana (Eisenack) and Rhabdochitina gallica Taugourdeau all suggest a Caradoc to Ashill age. Hercochitina cf. turnbulli Jenkins has previously been described from the Caradoc of Oklahoma (Jenkins 1969), but only one poorly preserved speci- men was found at Dob’s Linn. The acritarch Solisphaeridium nanum (Deflandre) Turner ranges from Arenig to Devonian and is therefore a poor biostratigraphical indicator. Of the other acritarchs recovered Stellechinatum brachyscolum Turner has been described only from the Caradoc of Shropshire (Turner 1984), and Veryhachium reductum (Deunff) Jekhowsky from the Tremadoc to the Silurian. Diexallophasis sp. 1 has also been found from the Silurian sedgwickii Zone and is probably a new species (pers. comms Molyneux 1986). Thus palynomorphs indi- cate an Upper Ordovician age for the anceps Zone, primarily on the evidence of chitinozoan distribution. Samples from the anceps Zone at the Linn Branch section have yielded no palyno- morphs and this is attributed to the extreme weathering of this part of the section. Extraordinarius Zone The chitinozoans and the acritarch Veryhachium corpulentum Colbath found in this zone (Figs 5, 6) suggest a Caradoc to Ashgill age, although the acritarchs Veryhachium lairdii and V. reductum both range from Lower Ordovician to Silurian in age. The Linn Branch section has only yielded two non-sphaeromorph acritarchs: the acanthomorphs Baltisphaeridium sp. 1 and Armoricanium sp. 2 (Fig. 6). PRELIMINARY ACRITARCH AND CHITINOZOAN DISTRIBUTIONS, DOB’S LINN 43 GRAPTOLITE SAMPLE LITH al CHITINOZOANS ZONE NUMBER = ACANTHOMORPH ACRITARCHS SPHAEROMORPH OTHER ACRITARCHS ACRITARCHS ANCYROCHITINA ANCYREA ACUMINATUS SUDL 41 ANCYROCHITINA SP 1 KALOCHITINA SP1 LEIOSPHAERIDIA SP 1 PERSCULPTUS SU DL —| RHABDOCHITINA MAGNA DICTYOTIDIUM SP 1 VERYHACHIUM LAIRDII SUDL 38 MULTIPLICISPHAERIDIUM SP.1 LP 2 V. CORPULENTUM ; _ SYNSPHAERIDIUM SP 1 Vv. REDUCTUM = esl MICRHYSTRIDIUM SP 1 VSP SUDL 17 CYATHOCHITINA HYMENOPHORA Lo sp 4 aes MICRHYSTRIDIUM SP.2 ACTINOTODISSUS SP 1 EXT RAORDINARIUS|_ | Se == | ; +— | MULTIPLICISPHAERIDIUM SP.1 MULTIPLICISPHAERIDIUM SP 1 —— Ll. sp 1 L spe2 ee SOLISPHAERIDIUM NANUM Lead STELLECHINATUM BRACHYSCOLUM MICRHYSTRIDIUM SP 1 ESESPa2: ANCEPS a a ee a= Sew noel ea Se a | L sp 1 L. SP. 2 | GONIOSPHAERIDIUM SP.1 L sp. VERYHACHIUM REDUCTUM DIEXALLOPHASIS SP1 MULTIPLICISPHAERIDIUM SP. 2 MICRHYSTRIDIUM SP1; M.SP3 L Se ECA UMESE A CYATHGCHITINA KUKERSIANA 7] Spin ©. CAMPANULAEFORMIS RHABDOCHITINA GALLICA eee HERCOCHITINA CF TURBULLI J Fig. 5 Distribution of acritarchs and chitinozoans at Main Cliff, Dob’s Linn. In column 3 (lithology), horizontal lines indicate a black shale sample, and the dots represent a grey mudstone. Persculptus Zone At Main Cliff the chitinozoan Rhabdochitina magna Eisenack and the herkomorph acritarch Dictyotidium sp. 1 have been found, while two samples from the Linn Branch section have yielded Kalochitina sp. 1 and Conochitina tormentosa Taugourdeau. This assemblage suggests a Caradoc to Ashgill age, although Rhabdochitina magna is known to range into the Llandovery. Acuminatus Zone One sample from Main Cliff has yielded 24 specimens of the important Lower Silurian form Ancyrochitina ancyrea (Eisenack) Eisenack and a single specimen of Kalochitina sp. 1. At the Linn Branch Rhabdochitina magna is found, and both this species and Kalochitina sp. 1 extend across the boundary, and are thus of little biostratigraphical use as boundary markers. Because of the long range of most species the distributions of acritarchs and chitinozoans are less refined biostratigraphical indicators than those of graptolites. The sample from the acumin- atus Zone can be dated accurately as Lower Silurian, while all other samples which yielded an unequivocal age determination are of Upper Ordovician age. It is important to note that the chitinozoans have proved most useful in this survey and that they are often very abundant in the black shales. As the boundary is within the Birkhill Shale, it is possible that bed by bed processing will yield sufficient chitinozoan taxa to determine the position of the Ordovician— Silurian boundary accurately in terms of the microflora. As palynomorphs often occur in rocks which lack datable macrofossils, even a crude biostratigraphical zonation based on chitin- ozoans would have considerable use in word-wide correlation. 44 G. M. WHELAN a eee = on] PRE - sos GRAPTOLITE SAMPLE LITH CHITINOZOANS ACANTHOMORPH SPHAEROMORPH ZONE NUMBER ACRITARCHS ACRITARCHS ——. + — — SU DL38 —— LEIOSPHAERIDIA SP 1 ———- t— SU DL 37 RHABDOCHITINA MAGNA | | — + + SU DL 36 (L, SR. 2 | = eee == = a a \ SU DL35 KALOCHITINA SP 1 7 os SP. 2 PERSCULPTUS ACUMINATUS ‘sale ] SU DL34 CONOCHITINA TORMENTOSA | L. SP. 2 | = + |, Us SP | &. Sp 2 Jt JL —_- St | aaa as BALTISPHAERIDIUM SP 1 | es SPiat SU DL32| © | AREMORICANIUM SP 2 | L. sp.2 EXTRAORDINARIUS | |} L. SP. 14 SU DL 31 | | & SR 2 ANCEPS SU DL 43 1 = : a Fig. 6 Distribution of acritarchs and chitinozoa at the Linn Branch section, Dob’s Linn. Lithology symbols as in Fig. 5. Acknowledgements I am grateful to C. J. Burton, G. B. Curry, K. J. Dorning, P. D. W. Haughton and S. G. Molyneux for their help, and I should also like to thank D. Maclean for printing the diagrams. A N.E.R.C. grant is gratefully acknowledged. References Cocks, L. R. M. 1985. The Ordovician-Silurian boundary. Episodes, Ottawa, 8: 98-100. Downie, C., Evitt, W. R. & Sarjeant, W. A. S. 1963. Dinoflagellates, hystrichospheres and the classification of the acritarchs. Stanf. Univ. Publs, Palo Alto, 17(3): 3-16. Jenkins, W. A. M. 1969. Chitinozoa from the Ordovician Viola and Fernvale Limestones of the Arbuckle Mountains, Oklahoma. Spec. Pap. Palaeont., London, 5. 44 pp., 9 pls. McKerrow, W. S., Leggett, J. K. & Eales, M. H. 1977. Imbricate thrust model of the Southern Uplands of Scotland. Nature, Lond. 267: 237-239. Turner, R. E. 1984. Acritarchs from the type area of the Ordovician Caradoc Series, Shropshire, England. Palaeontographica, Stuttgart, (B) 190: 87-157. Williams, S. H. 1980. An excursion guide to Dob’s Linn. Proc. geol. Soc. Glasgow 121/122: 13-18. (1981). The Ordovician and Lowest Silurian Graptolite Biostratigraphy in Southern Scotland. Unpublished Ph.D. Thesis, University of Glasgow. —— & Rickards, R. B. 1984. Palaeoecology of graptolitic black shales. In D. L. Bruton (ed.), Aspects of the Ordovician System: 159-166. Universitetsforlaget, Oslo. Ordovician-Silurian junctions in the Girvan district, S.W. Scotland D. A. T. Harper Department of Geology, University College, Galway, Ireland Synopsis The Ordovician—Silurian boundary at Girvan is represented by a variety of unconformable contacts; the basal Silurian rocks both overstep and overlap the upper Ordovician strata south and southwestwards. The most complete section across the junction is in a regressive shelly facies located north of the Girvan valley in the Craighead inlier. The Hirnantian High Mains Formation contains a moderately diverse Hirnantia fauna within channel fill sandstones. The overlying basal Silurian unit, the middle Rhuddanian Mulloch Hill Conglomerate, was deposited in submarine canyons at a variety of depths and contains an entrained Cryptothyrella fauna. The continuing regression evident across the junction and the facies patterns in the lowermost Silurian are related to the local emergence of fault-bounded blocks. Introduction The Ordovician and Silurian rocks of the Girvan district, S.W. Scotland contain a wide variety of siliciclastic sediments, together with locally diverse shelly and graptolite faunas; deposition occurred in a proximal fore-arc environment (Bluck 1983). In contrast to the graptolite facies of the Ordovician-Silurian boundary sections in the shale inliers of the Southern Uplands, the most stratigraphically complete junction section at Girvan is in a shelly facies. Lapworth’s detailed study of the Girvan succession (1882) was largely confirmed by the similarly substantial researches of Peach & Horne (1899). But neither study was aware of the terminal Ordovician unit, the High Mains Formation; thus the marked contrast between the faunas of the Ladyburn Mudstones of the Upper Drummuck Group and those of the Mulloch Hill Group led Lapworth (1882: 622) to consider the apparent hiatus between the top of his Ardmillan Series and the base of his Newlands Series to represent ‘the grandest palaeonto- logical break in the entire Girvan succession’. In a detailed appraisal of the Drummuck Group, Lamont (1935: 294) noted the presence of a hitherto unrecognized unit of buff-weathering sandstone overlying the Drummuck Group and containing a distinctive shelly fauna. He considered the unit, the High Mains Sandstone, to represent the base of the Mulloch Hill Group and moreover (Lamont 1935: 289) suggested a correlation with the lower Llandovery. From this unit he briefly described and figured speci- mens of his new genus Hirnantia, which he based on material of Orthis sagittifera M‘Coy from both the High Mains Sandstone and the Hirnant beds of Bala, north Wales, and noted the presence of Meristella sp. (Hindella crassa incipiens). Subsequently, Lamont (1949) described the trilobite Flexicalymene scotica from the High Mains Sandstone and modified his views on the correlation of the unit to include the possibility of a Hirnantian age. Ingham & Wright (1970) subsequently emphasized the presence of key elements of the terminal Ordovician Hirnantia fauna and concluded a correlation with the Hirnantian Stage. Harper (1979b) noted the presence of two distinct associations of the Hirnantia fauna within the High Mains Sandstone and suggested the inapplicability of the term ‘community’ to contain the marked diversity of associations within the Hirnantia fauna. The formation has been described and mapped in detail and bulk samples of the two shelly associations investi- gated (Harper 1981). The thirteen taxa of brachiopod are currently being described (Harper 1984 and in preparation), whilst Owen (1986) has completed a monographic study of the five taxa of trilobites. Bull. Br. Mus. nat. Hist. (Geol) 43: 45-52 Issued 28 April 1988 46 D. A. T. HARPER The junction sections The basal Silurian strata both overstep and overlap the upper Ordovician rocks of the district south and southwestwards (Cocks & Toghill 1973). The most stratigraphically complete bound- ary section is thus north of the Girvan valley in the Craighead inlier (Fig. 1) whilst the largest hiatus is developed in the coastal exposures south of the Girvan Valley and southwest of the main outcrop (Fig. 2). (1) Craighead inlier. The terminal Ordovician unit, the High Mains Formation, crops out in the vicinity of High Mains farmhouse (Fig. 1). The unit is poorly exposed, and the detailed outcrop pattern (Harper 1981) was investigated by trenching and mechanical digging. The formation contains two associations of the Hirnantia fauna and a Hirnantian age is indicated. The High Mains Formation is overlain by the Mulloch Hill Conglomerate (the Lady Burn Conglomerate of Cocks & Toghill, 1973) but although the junction is not exposed it is assumed to be fairly sharp with a slight angular discordance. (ti) Main Outcrop. The main outcrop of Silurian rocks in the Girvan district extends from Saugh Hill approximately northeast to Straiton (Cocks & Toghill 1973: fig. 1). The presence of major bedding-parallel structures have locally tectonized the shale units and may be responsible for the variation of thicknesses, along strike, of several of the formations. The junction of the Silurian with the underlying Ordovician is exposed on the west bank of Pen- whapple Burn (National Grid ref. NX 23279769) some 500 metres downstream from Penwhap- ple Bridge (Cocks & Toghill 1973: fig. 4). Here, the local base of the Silurian is represented by the Tralorg Formation. At the junction the succession is inverted; however, the Tralorg Forma- tion appears to overlie conformably grey micaceous sandstones and shales of the Shalloch Formation; the junction is apparently tectonized as are the shales within the underlying Shal- loch Formation. In an adjacent quarry, graptolites of the anceps Zone indicate a middle Ashgill age for this part of the Shalloch Formation. Both units dip steeply south. (111) Coastal Exposures. The two main coastal exposures of the Ordovician—Silurian junction clearly demonstrate the southward overstep and overlap of the basal Silurian units. At the northernmost of the two exposures, the Haven (Cocks & Toghill 1973: fig. 3), the Craigskelly Conglomerate overlies the Shalloch Formation unconformably. However, farther south on Woodland Point the Woodland Formation unconformably overlies lower horizons of the Shalloch Formation, although pockets of Craigskelly Conglomerate lie between the two. Faunal and facies changes at the Ordovician—Silurian junction As noted above, the most complete boundary section is near High Mains farmhouse in the central part of the Craighead inlier (Fig. 1). The highest Ordovician strata in the Girvan district, in ascending order the Shalloch Formation, the Drummuck Group and the High Mains Formation, are sporadically exposed and the latter two units are locally highly fossil- iferous (Figs 3—22). Within the Drummuck Group a variety of shelly associations dominated by brachiopods have been noted (Harper 1979b), and are currently under detailed description, together with the continuing documentation of the brachiopod taxa (Harper 1984 and in preparation). The associations are thought to have inhabited a spectrum of environments upslope and adjacent to the proximal parts of a submarine fan system. The highest strata of the group, the upper Rawtheyan South Threave Formation (Harper 1982), contain highly fossil- iferous sandstones (the Ladyburn Starfish Beds of the Farden Member) and probable mudflow units (the Cliff Member); nevertheless background sedimentation is represented by bedded green mudstones and occasional siltstones containing low diversity faunas of minute inarticu- late, enteletacean and plectambonitacean brachiopods. The boundary with the overlying Hir- nantian High Mains Formation, although not exposed, is assumed to be fairly sharp. The High Mains Formation consists of fine-medium and medium grained quartz sandstones. The beds are massive with an apparent lack of internal sedimentary structures; horizons of shelly debris 47 ORDOVICIAN-SILURIAN JUNCTIONS IN S.W. SCOTLAND ‘passnosip do19jno UBLINIIS JO svore UIUT 9014} JO suOI}Isod ayeuNIxOI1dde oy) sMoOyYs ‘1J9] WOI10q ‘JasuT YL ‘OPISUO[BUOD [ITH YSO]NJAL 9y) UIYyIIM oUOIspuPs sNoO.aJI[ISsoJ B JO SUIS BY) SYIEUI JOP YorTG oy) ISTTYM (1 R6] Jodiey jo TH pure [H sontyesoy SNOIATISSOJ) YOUST sue YSIH OY) JO uoMIsod ay} sayeorpur Ystso}se YoRIG ayy SJorur peoysiesD oy) Jo qed exueo ay) jo dew paieiog $= ‘sy uoljewso4 yooyeys uoNeWwIO4 SUJOYIpINY UOI}BWIOY jJIH JA4eNH Sainsodxe j2Ise0D dnoiy yOnwwnig uoljewso4 uwingApe doxyjno urey = fae = sequy peeysrery NVAUYID uol}euIO4 BAPOIYL YINOS uoljewsoy suleW YdIH 8PAID $O Yysty rc O}eIBWOBuOD jyH YOonp dnoJD IIH YooNW i QUO}SPURS IHIH YOOVINW i} Ty / Zi 4 4 te} (\ “A 7 Oo ° a j A Sor. o 0 MS Ye ° ef 0 os) \ QA O Dro sBe\ Om) a ho o Ys19, o QO © © [e oo @, 0 © \ 9 Ff fo o 0 © sO 0 0 8 © © © iz fl / y. o 6 ° ° ©® © ® © © 6 o qd \/ MAR Frcs (0) ° oT o © © Oo 8 @O © \ y g 1\~¥ ° ° ° ° ° ° ° ° ° g f XO ° 6! .'6, 4Onij0? Grol Ohio I || || ° zi ° ° ° °o °o ° ORO 2-3 ee an fires ites a 4 Se woos 4ATVIS 48 D. A. T. HARPER i = = — : = = = =r 5 7 GRAPT E ! STAGES CRAIGHEAD INLIER MAIN OUTCROP CORSTAL OUIEE EXPOSURES BIOZONES — 1 = Glenwells Shale Tralorg Formation | Woodland Formation ey — SS a cyphus 2 > | I} CAUTLEYAN O op) Qa Shalloch Formation See] fe) ees noes) aaa | Shalloch Formation : | === PUSGILLIAN Shalloch Formation complanatus mi : IL Fig. 2 Correlation of Ordovician—Silurian junction sections across the Girvan district with each other and the established shelly stages and graptolite biozones. are developed at various levels in the formation. The lower 2m of the formation exposed in the High Mains trench (Harper 1981: 250) comprises grey-green fine-medium grained, thinly bedded sandstones, whilst the upper 5-5 m is a hard medium-grained, thickly-bedded sandstone. Changes in grain size, bedding characteristics and faunal composition indicate a minor regres- sion within the sequence. In view of the incomplete exposure and an apparent absence of Figs 3-22 Brachiopods and trilobites from the High Mains Formation (Hirnantian), High Mains trench, Girvan. Repository: Hunterian Museum, Glasgow. Fig. 3, fossiliferous block of the High Mains sandstone dominated by internal moulds of both pedicle and brachial valves of Hindella crassa (J. de C. Sowerby) incipiens (Williams) and crinoid ossicles, x 1. Figs 4, 8, Plaesiomys aff. porcata (M‘Coy). 4, HM L12238, latex cast of internal mould of brachial valve, x 2; 8, HM L12239, latex cast of external mould of pedicle valve, x 2. Figs 5, 9, 13, Eochonetes cf. advena Reed. 5, HM L12115, internal mould of pedicle valve, x 4; 9, HM L12117, latex cast of internal mould of brachial valve, x 4; 13, HM L12118, latex cast of external mould of pedicle valve, x 3. Figs 6, 7, 10, 11, Hindella crassa (J. de C. Sowerby) incipiens (Williams). 6, 10, HM L12242, latex cast and internal mould of pedicle valve, both x 2; 7, HM L12244a, external mould of brachial valve, x 3; 11, HM L12244b, latex cast of internal mould of brachial valve, x 3. Figs 12, 14-16, Eostropheodonta aff. hirnantensis (M‘Coy). 12, HM L12105, latex cast of internal mould of brachial valve, x 2; 14, HM L12104, internal mould of pedicle valve, x 1; 15, HM L12103, latex cast of external mould of pedicle valve, x 2; 16, HM L12653, latex cast of internal mould of brachial valve, x 2. Figs 17, 18, Hemiarges extremus Owen, HM A16153, external mould and latex cast of cranidium, both x 2. Figs 19-21, Hirnantia sagittifera (M‘Coy). 19, HM L12654, latex cast of brachial valve exterior, x 2; 20, 21, HM L1986, latex cast and internal mould of brachial valve, both x 2. Fig. 22, Achatella cf. truncatocaudata (Portlock), HM A16152, internal mould of cephalon, x 2. ORDOVICIAN-SILURIAN JUNCTIONS IN S.W. SCOTLAND 49 sedimentary structures, palaeoenvironmental analysis of the High Mains Formation is equivo- cal. Nevertheless the thickness, geometry and lithology of the unit are compatible with deposi- tion within channels which developed on the deeper parts of the shelf and the upper parts of the slope. Such environments (Dott & Bird 1979) may be characterized by apparently massive and structureless sandstones comprising channel fills in the order of 25m thick. Elsewhere, various modes of channelling characterize predominantly argillaceous upper Ashgill sequences; these developed during the time of regression in response to the end Ordovician glacio-eustatic event (Brenchley & Newall 1980). At Girvan, however, a fall in sea level in excess of the 50-100m estimated (Brenchley & Newall 1980: 34) is required and thus additional tectonic controls must be invoked. Liiqoo|i2 4 Hunt Mus 50 D. A. T. HARPER To date, the High Mains Formation contains a fauna of thirteen brachiopod (Harper 1981) and five trilobite (Owen 1986) taxa. The brachiopods are characterized by a relative abundance of Hirnantia sagittifera (M‘Coy), Eostropheodonta aff. hirnantensis (M‘Coy) and Hindella crassa (J. de C. Sowerby) incipiens (Williams), important elements of the terminal Ordovician Hirnan- tia fauna, and less common Glyptorthis, Plaesiomys, Platystrophia, Eochonetes, Eopholidostro- phia, Fardenia, Rostricellula, Hypsiptycha and Eospirigerina and an indeterminate enteletacean. With the exception of Hypsiptycha, all these forms have congeners in the underlying Drum- muck Group. Moreover small individuals of H. crassa incipiens have been described previously from the Ladyburn Starfish Beds within the upper Rawtheyan South Threave Formation near the summit of the Drummuck Group (Reed 1917: 955; pl. 24, fig. 55) whilst Mitchell (1977: 54) has described and figured a species of Hirnantia from the Cautleyan Killey Bridge Formation, which is along strike in the Pomeroy inlier of the north of Ireland. The Girvan fauna is quite distinct from other Hirnantia faunas (cf. Rong 1984a); whilst the fauna is dominated by key members of the Hirnantia fauna, it is of moderate diversity and supplemented by essentially relict North American forms. It is nevertheless different from other coeval assemblages, for example the Holorhynchus and Older Edgewood faunas (Rong 1984b: 117). Similarly, the trilobite fauna is dominated by North American relicts (Ingham in Harper 1981; Owen 1986). The succeeding Mulloch Hill Conglomerate unconformably overlies the Drummuck Group. This formation is dominated by units of polymict, poorly sorted, of either clast- or matrix- supported conglomerate. The clasts range in diameter from a few centimeters up to 15cm; a variety of lithologies is represented as is a range of shapes from near rounded to angular. The conglomerate units are separated by thinner beds of coarse impure quartz sandstone which are locally fossiliferous. Cocks & Toghill (1973) considered a shallow water environment of deposi- tion likely for the unit whilst more recently Walton (1983: 133) indicated the sedimentology and fauna of the formation to be suggestive of shallow, shelf conditions. The available data however suggest an equally feasible alternative. The nature and thickness of the formation, in excess of 100m, together with an ability to cut through some 350m of strata over a distance of about five miles, suggest the Mulloch Hill Conglomerate was deposited in a channel across a gradient of depths. Clearly in the vicinity of Girvan the unconformity was not subaerial but rather resulted from downslope channelling during the earliest Silurian (see also Ingham 1978). The fauna of the Mulloch Hill Conglomerate, although locally abundant within the sand- stone units, is of low diversity. It is dominated by crinoid ossicles and the brachiopods Crypto- thyrella angustifrons (Salter) and a species of Rhynchotreta (Cocks & Togill 1973). Both species have near identical relatives in the fauna of the upper Rawtheyan Ladyburn Starfish Beds (Harper 1979a). Such associations characterize shallow water environments created during the early Llandovery global transgression (Sheehan 1977). Discussion The faunal succession across the Ordovician-Silurian junctions indicates three phases of devel- opment: (a) above the Rawtheyan—Hirnantian transition a marked decrease in diversity con- comitant with the development of a fauna comprising relict middle Ashgill elements of the North American province together with more abundant key taxa of the Hirnantia fauna, (b) during the early and middle Rhuddanian very low diversity faunas characteristic of the, then, recently colonized shallow water environments in the North American province, and (c) the arrival during the middle and late Rhuddanian of diverse, more typically Llandovery, shelly faunas. The former two events are accompanied by channel development during the regression whilst the latter is concomitant with net transgression. Similarly in the more complete and stable boundary section of the Oslo Basin relict Ordovician forms are not displaced by more typical Silurian elements until at least the middle Rhuddanian (Baarli & Harper 1986). The mutual relationships of the basal Silurian facies and their southwestward overlap and overstep have been rationalized recently by Bluck (1983: fig. 6). Such features are considered to be the result of deposition on blocks of Ordovician strata separated by high-angle listric faults ORDOVICIAN-SILURIAN JUNCTIONS IN S.W. SCOTLAND 51 with approximately east to west trends. Evidence of fault-controlled sedimentation has been documented within the middle Ordovician succession of the Girvan district in the classic study by Williams (1962), more recently refined by Ince (1984). Whilst the disposition and relative movement of such blocks can at least partly explain lower Silurian facies patterns in the Girvan district, a mechanism is available also to provide substantial and continued local regressions during the late Ordovician and early Silurian. The relative downfaulting of sequential blocks to the south, during extensional phases, may have resulted in the rotation of each block about an axis parallel to the trend of the listric faults; consequently the leading apex of each block may have become emergent. The overall effect locally is one of regression and channel development across relatively steep slopes. Both faunal and facies development thus occurred in a tectoni- cally active environment at Girvan, against a background of global regression and transgres- sion during the late Ashgill and early Llandovery respectively. Acknowledgements I thank Dr D. M. Williams for advice regarding sedimentology and for his comments on the manuscript. Dr A. W. Owen kindly provided unpublished data on the High Mains trilobites and valuable discussion. Much of the fieldwork was carried out during tenure of a N.E.R.C. research studentship at Queen’s University, Belfast. References Baarli, B. G. & Harper, D. A. T. 1986. Relict Ordovician brachiopod faunas in the Lower Silurian of Asker, Oslo Region, Norway. Norsk geol. Tidsskr., Oslo, 66: 87-98. Bluck, B. J. 1983. Role of the Midland Valley of Scotland in the Caledonian orogeny. Trans. R. Soc. Edinb. (Earth Sci.) 74: 119-136. Brenchley, P. J. & Newall, G. 1980. A facies analysis of Upper Ordovician regressive sequences in the Oslo region, Norway: a record of glacio-eustatic changes. Palaeogeogr. Palaeoclimat. Palaeoecol., Amsterdam, 31: 1-38. Cocks, L. R. M. & Toghill, P. 1973. The biostratigraphy of the Silurian rocks of the Girvan District, Scotland. Q. JI geol. Soc. Lond. 129: 209-243, pls 1-3. Dott, R. H. jr & Bird, K. J. 1979. Sand transport through channels across an Eocene shelf and slope in southwestern Oregon, U.S.A. Spec. Publs Soc. econ. Paleont. Miner., Tulsa, 27: 327-342. Harper, D. A. T. (1979a). The brachiopod faunas of the Upper Ardmillan succession (upper Ordovician), Girvan, S.W. Scotland. Unpublished Ph.D. thesis, Queen’s University, Belfast. 1979b. The environmental significance of some faunal changes in the Upper Ardmillan succession (upper Ordovician), Girvan, Scotland. Spec. Publs geol. Soc. Lond., 8: 439-445. —— 1981. The stratigraphy and faunas of the Upper Ordovician High Mains Formation of the Girvan district. Scott. J. Geol., Edinburgh, 17: 247-255. —— 1982. The stratigraphy of the Drummuck Group (Ashgill), Girvan. Geol. J., Liverpool, 17: 251-277. —— 1984. Brachiopods from the Upper Ardmillan succession (Ordovician) of the Girvan district, Scot- land. Part 1. Palaeontogr. Soc. (Monogr.), London. 78 pp., 11 pls. Ince, D. 1984. Sedimentation and tectonism in the Middle Ordovician of the Girvan district, S.W. Scotland. Trans. R. Soc. Edinb. (Earth Sci.) 75: 225—237. Ingham, J. K. 1978. Geology of a continental margin. 2: Middle and Late Ordovician transgression, Girvan. Geol. J., Liverpool, (Spec. Iss.) 10: 163-176. & Wright, A. D. 1970. A revised classification of the Ashgill Series. Lethaia, Oslo, 3: 233-242. Lamont, A. 1935. The Drummuck Group, Girvan: a stratigraphical revision with descriptions of fossils from the lower part of the group. Trans. geol. Soc. Glasg., 19: 288-334, pls 7-9. —— 1949. New species of Calymenidae from Scotland and Ireland. Geol. Mag., Hertford, 86: 313-323. Lapworth, C. 1882. The Girvan succession. Part 1. Stratigraphy. Q. JI geol. Soc. Lond. 38: 537-666, pls 24-25. Mitchell, W. I. 1977. The Ordovician Brachiopoda from Pomeroy, Co. Tyrone. 138 pp., 28 pls. Palaeon- togr. Soc. (Monogr.), London. Peach, B. N. & Horne, J. 1899. The Silurian rocks of Britain. I, Scotland. Mem. geol. Surv. U.K., London: 1-749. Owen, A. W. 1986. The uppermost Ordovician (Hirnantian) trilobites of Girvan, S.W. Scotland with a review of coeval trilobite faunas. Trans. R. Soc. Edinb. (Earth Sci.) 77 (3): 231-239. oe D. A. T. HARPER Reed, F. R. C. 1917. The Ordovician and Silurian Brachiopoda of the Girvan District. Trans. R. Soc. Edinb. 51: 795-998, pls 1-24. Rong Jia-Yu 1984a. Distribution of the Hirnantia fauna and its meaning. In D. L. Bruton (ed.), Aspects of the Ordovician System: 101-112. Universitetsforlaget, Oslo. —— 1984b. Brachiopods of latest Ordovician in the Yichang district, western Hubei, central China. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeontology of Systemic Boundaries in China: Ordovician—Silurian boundary 1: 111-190, pls 1-14. Anhui Sci. Tech. publ. House. Sheehan, P. M. 1977. Late Ordovician and earliest Silurian meristellid brachiopods in Scandinavia. J. Paleont., Tulsa, 51: 23-43, pls 1-3. Walton, E. K. 1983. Lower Palaeozoic—Stratigraphy. In G. Y. Craig (ed.), The Geology of Scotland: 105-137. Edinburgh. Williams, A. 1962. The Barr and Lower Ardmillan Series (Caradoc) of the Girvan district, south-west Ayrshire, with descriptions of the Brachiopoda. Mem. geol. Soc. Lond. 3: 1-267, pls 1-25. Base of the Silurian in the Lake District and Howsgill Fells, Northern England R. B. Rickards Department of Earth Sciences, Downing St, Cambridge CB2 3EQ Synopsis The basal Silurian in the Lake District and Howsgill Fells is divided into four slightly different deposi- tional zones, only one of which shows a provable base to the acuminatus Zone, being underlain by a persculptus fauna and overlain by an atavus fauna. Other sections have ‘Basal Beds’ which certainly represent very condensed deposition of carbonates, perhaps involving non-sequences. The varied environments are interpreted as part of an offshore fault-scarp-cum-ridge-and-hollow system paralleling the Iapetus Suture and situated upon the southern (northward-dipping) plate. There are essentially four rather different depositional environments at the Ordovician—Silurian boundary in the Lake District proper and in the Howgill Fells; and these are each different again from the facies and faunal development at Cross Fell, dealt with by Wright elsewhere in this volume. The four types are shown in Figs 1—4: although drawn diagrammatically it is important to realize that there are no exposure gaps in the region of the boundary, and that the sections in the Howsgill Fells and western Lake District (Figs 1, 4) can be confirmed in many other nearby sections. The acuminatus Zone fauna, the new basal Silurian zone, is well represented except in one small region only, namely the classic Skelgill section (Fig. 3), the type section of the Skelgill Beds black shale formation. On this section there is a thin, hard, partly calcareous and partly siliceous shelly mudstone (usually referred to in the literature as the Basal Beds). A similar bed occurs in the Howgill Fells, but the age on Skelgill could range from the persculptus Zone to the lower atavus Zone inclusive, for it is underlain by Ashgill Shales (Hirnantian; and probably of anceps Zone age) and overlain by upper atavus Zone black shales. The Basal Beds certainly represent a period of condensed deposition and possibly of non-sequence. There is no direct evidence of hardground criteria. The shelly fossils include Atrypa flexuosa and may represent relatively deep water community life with low diversity. In the Howsgill Fells and the eastern Lake District (Figs 1, 2) the acuminatus Zone is well established but its base, and hence the base of the Silurian, cannot be proved: the Basal Beds in the Howgill Fells might be of persculptus Zone age, but a possible bentonite separates those beds from the thin acuminatus Zone black shale; and at Browgill a 0:08 m rottenstone, possibly the lithological and stratigraphical equivalent of the Basal Beds, separates Hirnantian Ashgill shales from black, acuminatus Zone shales. Only in the western Lake District (Fig. 4) can the base of the Silurian be unequivocally placed, albeit on numerous sections in the region. The Yewdale Beck section is well and continuously exposed, and above 0:3m of beds with a good persculptus Zone fauna are 11m of black shales with a very rich assemblage of acuminatus Zone graptolites (Hutt 1974). The persculptus Zone also contains numerous shelly fossils of most groups, but they have not been extensively studied. The Ashgill Shales below them yield numerous brachiopods and rarer trilobites giving a Hirnantian age to the Ashgill Shales, but graptolites in these beds are rare. The acuminatus Zone black shales yield shelly fossils only very infrequently and none to date have proved to be of diagnostic value. In every other respect, however, the Yewdale Beck section provides a good confirmatory section for the base of the Silurian, especially as an almost infinite number of both natural and artificial sections are available in the general region of Coniston and on the fells and streams to the southwest of that town. Graptolites from these sections can be collected by the hundred and, as with all other acuminatus Zone faunas mentioned above, almost all the typical species of the zonal assemblage occur. Bull. Br. Mus. nat. Hist. (Geol) 43: 53-57 Issued 28 April 1988 54 Fig. 1 R. B. RICKARDS continuous exposure into Wenlock Cc ‘) 5 black shales ao] atavus Zone 3 7 = mM. ag O-Im. acuminatus Zone <— ?P bentonite black shales O:lm Waar av aaa a fe) Skelgill Beds Paleestis: | Lia P persculptus Zone in part banded grey : calcareous nodules shales & silts anceps Zone in part Hirnantian Ashgill Shales ~~ tm Cc @ E a © @ > ® a2) @ Qa. > =_— ~— wharfe conglomerate Howsgill Fells: beds about the Ordovician—Silurian boundary on Spengill, Grid Reference SD 698998. BASE OF THE SILURIAN IN THE LAKE DISTRICT 5)5) I continuous exposure into Browgill Beds (in type development) w S| , 5 a black shales 3 = atavus Zone 2 ao 7p) brown-weathering : acuminatus Zone 008m! black shales a ol “(7171 =P persculptus Zone S 3 rey shales re = (dp) ae, Cc So = =e! £ ae (7p) = § not to scale: thickness of units given in metres Fig. 2 Eastern Lake District: beds about the Ordovician—Silurian boundary on Browgill, NY 4974 0587. Rickards (1978) attempted a general interpretation of the environment of deposition of the basal Silurian strata, envisaging a west- or northwest-facing fault scarp, according to Hutt (1974) active during deposition of the early Llandovery, against which were deposited deeper offshore, black shales and upon which and behind which were deposited the Basal Beds and their equivalents. By upper atavus Zone times the scarp feature was further submerged and covered in black shale deposition. Associated with these features were a series of ridges and hollows striking ENE/WSW, that is roughly the same as the fault scarp strike. The hollows received a greater thickness of black shale in a more highly anaerobic environment (Rickards 1964). The ridge and hollow system persisted in the Howsgill Fells region, and possibly in the main Lake District outcrop, until late in the Llandovery. Thus the onset of the Silurian in the Lake District is marked by condensed deposition of shelly limestone, and possible non-sequences, in the eastern, presumed shoreward or shallower region; and by relatively thick, black shale deposition in the western Lake District. The post- glacial marine transgression is recorded in the gradual spread of black shale deposition over the whole region, the last area to succumb being the eastern Lake District area of Skelgill which is interpreted as being on the crest of an old scarp structure, itself certainly operative as far back as the Caradoc. It seems likely that the region was situated atop the northward-dipping plate, south of the Iapetus Suture. The scarp and ridge/hollow systems may be a result of the northwards subduction process, to which they are parallel, and which resulted in a combination of compressional and extensional features. t continuous exposure to crispus Zone = oo = = Cc (Ss (2) Oo] v0 Ee black shales 3 m upper atavus Zone ox — ® ©l=5 Zo =< @ n> —_— ? persculptus — lower atavus Zones grey shales P anceps Zone and silts Hirnantian w 2 tS) or op) oO ‘iS D 4 Coniston limestone not to scale: thickness of units given in metres Fig. 3 Eastern Lake District: beds about the Ordovician—Silurian boundary on Skelgill, NY 3964 0320. BASE OF THE SILURIAN IN THE LAKE DISTRICT 57 t continuous exposure into convolutus Zone black shales atavus Zone black shales acuminatus Zone Rhuddanian Skelgill Beds blue/grey shales persculptus Zone 0O-3m grey shales and siltstones + calcareous nodules Hirnantian Ashgill Shales exposure failure not to scale: thickness of units given in metres Fig. 4 Western Lake District: beds about the Ordovician-Silurian boundary at Yewdale Beck, SD 3073 9858. References Hutt, J. E. 1974. The Llandovery graptolites of the English Lake District. Part 1. 56 pp., 10 pls. Palaeon- togr. Soc. (Monogr.), London. Rickards, R. B. 1964. The graptolitic mudstone and associated facies in the Silurian strata of the Howgill Fells. Geol. Mag., Hertford, 101: 435-451. —— 1978. In J. K. Ingham et al., The Upper Ordovician and Silurian Rocks. In F. Moseley (ed.), The Geology of the Lake District. Occ. publ. Yorks. geol. Soc. 3: 121-245. The Ordovician—Silurian boundary at Keisley, Cumbria A. D. Wright Department of Geology, The Queen’s University of Belfast, Belfast BT7 1NN, Northern Ireland Synopsis At Keisley, in the Cross Fell Inlier of Cumbria, the lowest Silurian graptolite biozone recorded until recently was that of A. atavus, with the topmost of the underlying carbonates regarded as being of either Lower Llandovery or Hirnantian age. A temporary excavation has confirmed the Hirnantian age of the latter, and with the discovery in the overlying clastic sediments of the biozones of both G. persculptus and P. acuminatus, the Ordovician—Silurian boundary is now accurately located. Although Upper Ordovician and Lower Silurian rocks crop out in the Cross Fell Inlier of northern England, the area is much faulted (Shotton 1935). Moreover, where reasonably con- tinuous graptolite sequences of the Lower Silurian are exposed in Swindale Beck (Knock) and in Great Rundale Beck (Marr & Nicholson 1888: 699; Burgess & Rushton 1979: 23), the lowest biozones (below Coronograptus cyphus) are missing. Until recently, the earliest Silurian graptol- ite biozone was that recorded from the road cutting to Keisley Quarry by Marr (1906: 485) and reported by him as indicating the Dimorphograptus confertus Zone of Marr & Nicholson (1888). The lowest part of that zone has been shown by Rickards (1970) to equate with the Ata- vograptus atavus Biozone, and the presence of beds of this age was confirmed by Rickards from graptolite material excavated in 1965 from this locality by Temple (1968: 2). On the Upper Ordovician side of the boundary the stratigraphical relationships and precise age of the main unit, the Keisley Limestone, have been debated for many years. The limestone has been a source of geological interest since the last century as it contains a prolific shelly fauna, is of distinctive lithology, and has a peculiar morphological form referred to as a ‘knoll’ by Marr (1906: 485). The views on various aspects of this mudmound have been discussed by Wright (1985); only the relationships of the carbonate mudmound to the atavus Biozone graptolite shales are relevant in the present context. Marr (1906: 485) noted that the Ashgill Shales, which do occur in Swindale Beck, were not present at Keisley; and as there was insufficient room for these beds between the Silurian graptolite shales and the nearest outcrops of Keisley Limestone, he interpreted the junction as a faulted one. Burgess (1968: 343) noted that along the track leading to the quarry, the massive limestone was succeeded by calcareous mudstones and limestone nodules which were in turn overlain by the graptolite shales ‘in apparently conformable sequence’, and the presence of this apparently unfaulted and conformable relationship was subsequently reiterated by Burgess et al. (1970: 170), despite the discontinuous nature of the outcrops. An extensive brachiopod and trilobite fauna was collected by Temple (1968, 1969) from weathered limestone bands associated with unfossiliferous shales at the bend in the quarry track; this outcrop was separated by a few metres from those of both the underlying massive limestone and the overlying atavus Biozone shales, and the extensive fauna interpreted by Temple as being of Lower Llandovery age, a view supported by Burgess & Rushton (1979: 23) but not by Ingham & Wright (1972: 47), who regarded it as being of Hirnantian age. The difficulty with the Keisley locality is that the beds immediately below the established atavus Biozone graptolite shales are concealed beneath the trackway to the quarry. To over- come this a temporary trench was dug with the aid of a mechanical digger and the complete sequence exposed (Wright 1985). Fig. 1 shows the position of the trench across the trackway and Fig. 2 the lithological log obtained. Bull. Br. Mus. nat. Hist. (Geol) 43: 59-63 Issued 28 April 1988 60 A. D. WRIGHT aC Keisley a —_ Keisley New Quarry Ss ob q vd as 5 metres Fig. 1 Plan showing the position of the temporary trench excavated across the trackway at the eastern end of Keisley New Quarry to reveal the Ordovician—Silurian boundary (National Grid Ref. NY 7137 2379). The strikes in the trench were taken on the trench floor except for the two strikes at the southern end which are in the trench walls and thus well up in the atavus Biozone. Stippling in the bank to the east of the trench indicates outcrops of fossiliferous weathered lime- stones, the fauna of which was described by Temple (1968, 1969). Stylized trees (not to scale) represent two ash (light outlines), two sycamores (dark outlines) and a hawthorn (small figure). The inset figure shows the position of Keisley in the Cross Fell Inlier (shaded) in relation to north-west England (C—Carlisle; K—Kendal). The lower part of the sequence up to and including unit 8 (numbering as in Wright 1985) consists of alternations of bedded limestones or calcareous nodules with calcareous siltstones. The bedded bioclastic limestones are fresh and although pelmatozoan debris, bryozoan frag- ments and the occasional brachiopod (including Hirnantia sagittifera) are to be seen on the bed surfaces, faunal lists are scant compared with those of Temple (1968) obtained from the well weathered material above the trackway. Gastropods, ostracodes and a few trilobites have been observed in thin sections of the trench limestones in which abundant Girvanella is probably the most revealing element palaeoenvironmentally. The unit 7 siltstone, while by no means abundantly fossiliferous, does have a shelly fauna in the form of moulds, albeit in a broken and fragmented state. The diverse fauna includes the brachiopods Dolerorthis praeclara, Hindella sp., Hirnantia sagittifera, ? Oxoplecia, Paracraniops sp., Reuschella inexpectata, Skenidioides scoliodus, Sphenotreta sp. and Toxorthis proteus ORDOVICIAN-SILURIAN BOUNDARY IN CUMBRIA 20 cm — oo | Volcanic clay Black shale Blotchy siltstone Green siltstone Calcareous siltstone Rottenstone breccia Calcareous nodules Limestone HEIL IMU 61 SILURIAN ORDOVICIAN Hirnantia shelly fauna Fig. 2 The lithological log obtained for the trench cutting across the quarry track of Fig. 1, showing the position of the Ordovician—Silurian boundary. Numbers of lithological units discussed in the text are as in Wright (1985). The black spots and bars to the right of the log respectively indicate specific horizons or bulk samples yielding graptolite assemblages. 62 A. D. WRIGHT together with dalmanellid, lingulide, orthid, sowerbyellid, strophomenide and triplesiid frag- ments. In addition to pelmatozoan and bryozoan debris, trilobite, bivalve and hyolith fragments are also recorded. This is a Hirnantia shelly fauna, and differs principally from Temple’s fauna in the apparent complete absence of craniids which accounted for more than two-fifths of the entire brachiopod assemblages from the weathered limestones (Temple 1968: 9). Overlying these beds is a thin (7cm) rottenstone breccia (units 9 and 10). This is the only indication of a break in the sequence and is interpreted as the result of minor tectonic move- ment along the surface of lithological change from the underlying carbonate dominated sequence to the overlying fine-grained and non-carbonate clastics. Angular clasts of both fossiliferous shelly Hirnantian and unfossiliferous greenish siltstone (matching the unit 11 sediment) occur in the breccia. No diagnostic shelly fossils have been located in the sequence above unit 10. The first graptolites recovered by Rickards are from a horizon 2cm below the top of unit 11 and indicate the Glyptograptus persculptus Biozone. This fauna comprises Cli- macograptus cf. miserabilis, Climacograptus ? medius, Glyptograptus sp. and Glyptograptus ex gr. persculptus. Unit 12 is an 8cm unit of silt with a blotchy and mottled appearance produced by an increase in the proportion of dark muddy silt that first appears in the greenish siltstones of unit 11 (Wright 1985: 269). Despite clear evidence of bioturbation, a small graptolite fauna from a bulk sample of the unit contained specimens of Climacograptus normalis and cf. Parakido- graptus acuminatus, and indicates the presence of the Parakidograptus acuminatus Biozone. The Ordovician-Silurian boundary at Keisley is accordingly placed at the base of lithological unit 12. This seems to be the most logical horizon although, as noted previously (Wright 1985), there is clearly a little uncertainty regarding the precise appearance of the acuminatus fauna within a bulk sample taken from the 8 cm unit. Unit 13 lithologically shows a further stage in the transition from the greenish siltstones at the base of unit 11 towards the micaceous black silty shales of the overlying sequence. In this unit the dark material is dominant, although some horizons and patches of the greenish-grey siltstones still occur; concomitantly with the overall colour change, bioturbation disappears. At 2:5cm above the base of this unit, the first of a series of bentonite clays occurs. A fauna collected from a bulk sample above this clay (Fig. 2) yielded Climacograptus medius, Cli- macograptus cf. normalis and Dimorphograptus sp. This assemblage is identified by Rickards as a post-acuminatus one, 1.e. from the base of the atavus Zone. Accordingly the acuminatus—atavus boundary is placed at the thin bentonite band, which is a useful marker that may assist with correlation elsewhere, although the appearance of atavus Biozone bentonites in the Keisley trench is a major surprise in the northern England context (Wright 1985). The increasingly rich graptolite faunas from the overlying sequence in the trench all belong to the atavus Biozone. Thus although the persculptus and acuminatus Biozones occur in thin lithological units at Keisley, both do occur and accordingly enable the Ordovician—Silurian boundary to be pre- cisely located. References Burgess, I. C. 1968. p. 343 in F. W. Shotton, A. J. Wadge & I. C. Burgess, Cross Fell Area (Field Meeting). Proc. Yorks. geol. Soc., Leeds, 36: 340-344. , Rickards, R. B. & Strachan, I. 1970. The Silurian strata of the Cross Fell area. Bull. geol. Surv. Gt Br., London, 32: 167-182. & Rushton, A. W. A. 1979. Skelgill Shales. In I. C. Burgess & D. W. Holliday, Geology of the country around Brough-under-Stainmore. Mem. geol. Surv. Gt Br., London, Sheet 31. 131 pp. Ingham, J. K. & Wright, A. D. 1972. The North of England. In A. Williams et al., A correlation of Ordovician rocks in the British Isles. Spec. Rep. geol. Soc. Lond. 3: 43-49. Marr, J. E. 1906. On the stratigraphical relations of the Dufton Shales and Keisley Limestone of the Cross Fell Inlier. Geol. Mag., London, (dec. 5) 3: 481-487. & Nicholson, H. A. 1888. The Stockdale Shales. Q. JI geol. Soc. Lond. 44: 654-732. Rickards, R. B. 1970. The Llandovery (Silurian) graptolites of the Howgill Fells, Northern England. Palaeontogr. Soc. (Monogr.), London. 108 pp., 8 pls. ORDOVICIAN-SILURIAN BOUNDARY IN CUMBRIA 63 Shotton, F. W. 1935. The stratigraphy and tectonics of the Cross Fell Inlier. Q. Jl geol. Soc. Lond. 91: 639-701. Temple, J. T. 1968. The Lower Llandovery (Silurian) brachiopods from Keisley, Westmorland. Palaeon- togr. Soc. (Monogr.), London. 58 pp., 10 pls. — 1969. Lower Llandovery (Silurian) trilobites from Keisley, Westmorland. Bull. Br. Mus. nat. Hist., London, (Geol.) 18: 197—230. Wright, A. D. 1985. The Ordovician-Silurian boundary at Keisley, northern England. Geol. Mag., Cam- bridge, 122: 261-273. Ordovician-Silurian boundary strata in Wales J. T. Temple Department of Geology, Birkbeck College, Gresse Street, London W1P 1PA Synopsis Ordovician-Silurian boundary strata in Wales belong to the shelly facies in the south and east, and to the graptolitic facies in the north and west. In the graptolitic facies the zones of Dicellograptus anceps, Glyptograptus persculptus and Parakidograptus acuminatus occur, but the Climacograptus? extraordinarius Zone is not known. The anceps Zone is restricted to central and west Wales; the persculptus Zone is widespread and is preceded by a sudden lithological change; the acuminatus Zone is preceded by a more gradual lithological change. Graptolites occur sporadically in boundary strata of the shelly facies but are not abundant enough for the base of the acuminatus Zone to be recognized in this facies. Records of the Hirnantia fauna in Wales are summarized. Introduction As a result of Caledonian and Hercynian folding the Ordovician-Silurian boundary strata in Wales form a complex arcuate pattern striking approximately NE-SW through much of central Wales but becoming east-west in south-west Wales and SE-NW in north-east Wales. The length of outcrop is approximately 750 km. The outcrop is shown in Fig. 1, together with index numbers by which individual areas and the references relating to them are cited in the text. In places on the outward (S, SE or E) side of the Caledonian fold belt in Wales, as in the adjoining parts of England, the local base of the Silurian is formed by late Llandovery (post- convolutus or post-sedgwickii) or Wenlock strata transgressive onto pre-Ashgill strata. This relation is found in the southernmost outcrop (but not in the main northern outcrop) at Haverfordwest (la), near Llandeilo (2), from north of Llandovery (4) to Garth (5a, b), near Builth Wells (6), east of Abbey-Cwmhir (7), and east of Welshpool (25, 26). Flanking this marginal area of late Llandovery transgression there is an unconformity of lesser magnitude between the early Llandovery and the Ashgill (and Caradoc) near Welshpool (27) and Llan- santffraid ym Mechain (31), and although the gap continues to diminish northwards and westwards it is recorded as still present in the Meifod and Vyrnwy areas (28, 29). Elsewhere in Wales the early Llandovery is believed to follow the topmost Ordovician with no sedimentary gap. Boundary strata Ordovician-Silurian boundary strata in Wales show two facies, shelly and graptolitic. The shelly facies consists of detrital sediments, mainly of the silt and sand grades, with a fauna predominantly of brachiopods. The graptolitic facies consists of fine detrital sediments (mudstones and shales) with some coarser horizons interpreted as turbidites, and with a fauna almost exclusively of graptolites. In pre-persculptus Zone strata the dichotomy into shelly and graptolitic facies is not as clearly defined as later. The strictly graptolitic facies, as defined by the recorded presence of the Dicellograptus anceps Zone, is much more restricted in occurrence (to central and west Wales— 16, 18, 19, 20) than the persculptus Zone, and even where both zones occur in the same area the intervening strata are either unfossiliferous (16, 18, 19) or include shelly fossils (20). Along the outcrop north-west of the Towy anticline (8-14), for instance, where the persculptus Zone is graptolitic, the very thick underlying strata yield only sporadic graptolites (not diagnostic of the anceps Zone), being otherwise unfossiliferous or with a few shelly fossils. The restriction of the demonstrable anceps Zone to central and west Wales and the wider extent eastwards of the persculptus and acuminatus Zones are consistent with regression during anceps Zone time followed by transgression during the persculptus Zone. The extraordinarius Zone has not been Bull. Br. Mus. nat. Hist. (Geol) 43: 65-71 Issued 28 April 1988 66 J. T. TEMPLE 36H oH y 33 32a,b He 34,35° Llangollen Tp HBerwyn Hills 39 30a,b cs a te Welshpool 25 (>-- shelly facies er : Early C= graptolitic facies Mandowen, absent by overlap Numbers refer toindex of areas & references H Hirnantia fauna recorded of 4? Ulandovery S/3a 4 Haverfordwest meee < see 2 Se Swansea e QO 5 10 15 2O 25 3) feo, kilometres Fig. 1 Ordovician—Silurian boundary outcrop areas in Wales and the Welsh Borderland. 1, Haver- fordwest: la, Strahan et al. 1914; 1b, Cocks & Price 1975. 2, Llandeilo, Williams 1953. 3, 4, Llandovery: 3a, Jones 1925; 3b, Jones 1949; 4, Cocks et al. 1984. 5, Garth: Sa, Andrew 1925; 5b, Williams & Wright 1981. 6, Builth Wells, Jones 1947. 7, Abbey-Cwmhir, Roberts 1929. 8, 9, Rhayader: 8, Lapworth 1900; 9, Kelling & Woollands 1969. 10, Rhayader to Abergwesyn, Davies 1928. 11, Abergwesyn to Drygarn, Davies 1926. 12, Pumpsaint, Davies 1933. 13, Llansawel, Drew & Slater 1910. 14, Llangranog, Hendricks 1926. 15, Llanidloes, Jones 1945. 16, Plynlimon, Jones 1909. 17, Machynlleth, Jones & Pugh 1916. 18, Towyn and Abergynolwyn, Jehu 1926. 19, Corris, Pugh 1923. 20, Dinas Mawddwy, Pugh 1928. 21, Llanuwchllyn-Llanymawddwy, Pugh 1929. 22, Bala: 22a, Elles 1922; 22b, Bassett et al. 1966. 23, Cerrigydrudion, Marr 1880. 24, Conwy, Elles 1909. 25, Shelve area, Whittard 1932. 26, 27, Welshpool: 26, Wade 1911; 27, Cave 1965. 28, Meifod, King 1928. 29, Lake Vyrnwy, King 1923. 30, Berwyns: 30a, Wedd et al. 1929; 30b, Brenchley & Cullen 1984. 31, Llansantffraid ym Mechain, Whittington 1938. 32, Llangollen: 32a, Groom & Lake 1908; 32b, Hiller 1980. 33, Corwen, Lake & Groom 1893. 34, Llangollen, Wills & Smith 1922. 35, Llangollen, Wedd et al. 1927. 36, Mynydd Cricor, Smith 1935. 37, Criccieth, Roberts 1967. 38, Anglesey, Greenly 1919. 39, W. Berwyn, A. W. A. Rushton & J. T. Temple (unpublished). 40, Aberystwyth and Machynlleth, Cave & Hains 1986. ORDOVICIAN-SILURIAN BOUNDARY STRATA IN WALES 67 recognized in Wales, but there is ample room for it: the barren strata between the anceps and persculptus Zones in areas 16, 18, 19, 20 are respectively 730m, 1000 m, 690 m, and 180m thick. In the persculptus Zone and the succeeding early Llandovery the dichotomy into shelly and graptolitic facies is well shown. The shelly facies forms a narrow belt running through Haver- fordwest (1a, b), the Llandovery (3a, b, 4) and Garth (5a, b) areas (which form north-westward salients from the adjacent line of outcrop along which the strata are missing), and the eastern end of the Berwyn dome (27-32, 34-36). The transition from shelly to graptolitic facies of the persculptus Zone and early Llandovery takes place in south and central Wales across the Towy anticline (between for instance Llandovery [3a, b, 4] and Pumpsaint [12]), and in north-east Wales probably north-westwards across the Berwyn dome. The persculptus and acuminatus Zones are widespread, having been recorded from north-west of the Towy anticline (10-12) as well as through most of central and west Wales (14-21, 40). G. persculptus occurs on the western outcrop around the Berwyn Hills at Nant Pant-y-llidiart, north of Lake Vyrnwy (39), and there is an informal record of the species at Bwlch yr Hwch, 5km SE of Bala (Jones in Pugh, 1929: 274-S). The persculptus Zone (but not the acuminatus Zone) has also been recorded from the north end of the Towy anticline (7, 9), and G. cf. persculptus occurs at Garth (5a). The early Llandovery graptolite succession between Bala (22a, b) and Conwy (24) is still in need of reinvestigation. In the two small isolated outliers near Criccieth (37) and in Anglesey (38) the early Llandovery is in graptolite facies, but in both cases the relationship to the Ordovician is obscure and neither the persculptus nor the acuminatus Zones are recorded. A sudden and striking lithological change heralds the incoming of persculptus Zone graptol- ites in west and central Wales (14—20, 40): the underlying strata are very thick, usually unfossiliferous, often unbedded, well cleaved or doubly cleaved, and with many ‘grit’ bands; the persculptus Zone strata (the ‘Mottled Beds’) are mudstones 5—25m thick, well-bedded, often with mottled pale bands (interpreted as bioturbated—Cave & Hains 1986) and with a thin band crowded with the zone fossil about 1m above the base. The suddenness of the lithologi- cal change preceding the appearance of G. persculptus in this part of Wales betokens some physical change in the conditions of deposition, and this evidence also is consistent with a persculptus transgression following regression. A similar lithological contrast at this horizon is also found north of the Towy anticline (9-11), although not strongly marked in the south of the outcrop (12). There is also a lithological change below the acuminatus Zone in west and central Wales (15-20, 40), but it is more gradual than that below the persculptus Zone, the hard resistant mottled mudstones of the latter zone being gradually replaced by rusty red- and yellow- weathering mudstones without bioturbation (40). A similar change occurs at this horizon north of the Towy anticline (10-12). In both areas the change probably precedes the end of the persculptus Zone (40, 12). Hirnantia fauna in Wales Around the Berwyn dome and near Llangollen there are developed ‘grits’ which have been taken as either topmost Ordovician (35) or basal Silurian (28, 29): Craig-wen Sandstone (28), Meristina crassa Sandstone (29), Allt-g6ch Grit (30), Corwen Grit (33), Glyn Grit (32), Plas uchaf Grit (35). These grits have been interpreted as channel-fill deposits formed during the Hirnantian regression (30b). ‘Grits’, possibly of the same age as those around the Berwyns, also occur in the north and east of the Bala area (Calettwr Quartzite—22b) and along the little- known outcrops north of Bala, i.e. at Cerrigydrudion (23) and Conwy (Conwy Castle Grit—24). South of the Berwyns there are ‘grit’ bands near Abbey-Cwmhir (7) which are mapped as topmost Ordovician but whose relationship to the persculptus Zone strata occurring about 3km to the west needs reinvestigation. Many of the ‘grits’ in these different areas include elements of the Hirnantia fauna (Fig. 1), for which Brenchley & Cullen (1984: 122) give faunal lists at various Welsh localities. To these 68 J. T. TEMPLE authors’ list for ‘Meifod’ (i.e. Craig-wen quarry, near Meifod) may be added the record of the tretaspid indet. discovered on the Silurian Subcommission excursion in 1979, although the presence of pebbles of underlying strata in the Craig-wen Sandstone suggests the possibility of this being a derived fossil. The Hirnantia fauna also occurs in Afon Tanat on the western outcrop of the Berwyn Hills (39). The Hirnantia fauna at its type area south of Bala (22a) was considered by Pugh (1929: 273) to be pre-persculptus in age although no single section (except Jones’ record at Bwlch yr Hwch—see above—which awaits confirmation) shows the one fauna succeeding the other. Further southwestwards along the outcrop (beyond 20) in west and west-central Wales the Hirnantia fauna dies out while the persculptus Zone fauna becomes more clearly developed. South of the Towy anticline the Hirnantia fauna has been recorded from Garth (5a, b) apparently in association with G. cf. persculptus (Williams & Wright 1981: 38), and from Haverfordwest (1b) in the St Martin’s Cemetery Beds (Cocks & Price 1975: 710) whose relations to the persculptus Zone are unknown. The Hirnantia fauna has also recently been found in the Llandovery area (4) where it is considered (Cocks et al. 1984: 144) to underlie strata probably representing the persculptus Zone. At Conwy (24) the Hirnantia fauna is underlain, as in the English Lake District, by strata containing abundant Dalmanitina [Mucronaspis auctt.], and this relationship is found also at Bala (22a) and in the Llanuwchllyn-Llanymawddwy area to the south (21). The trilobite persists southwestwards along the outcrop, as the facies change and the rocks thicken, even further (20, ?19) than the Hirnantia fauna. On the other hand the Hirnantia fauna around the Berwyns (28-30), at Abbey-Cwmhir (7) and at Garth (5) is not accompanied or preceded by Dalmanitina (except for a possible record in area 28—King 1928: 687), and although the absence of the latter trilobite in the Berwyns may be due to a stratigraphical gap below the Hirnantia ‘grits’, there is no evidence for such a gap at Abbey-Cwmhir or Garth, nor indeed at Llandovery where Dalmanitina is also absent. At Haverfordwest (1b) Dalmanitina occurs as part of an unusually rich Hirnantia fauna but is not found in underlying strata. Descriptions of sections Boundary strata of four areas merit description: Plynlimon-Machynlleth (16, 17, 40), where the sequence is graptolitic throughout and where the persculptus and acuminatus Zones are well developed; Llandovery (3a, b, 4) where the base of the Llandovery was originally defined; Haverfordwest (1a, b) and Garth (5a, b), in both of which there are apparently continuous successions in strata of predominantly shelly facies. Plynlimon—Machynlleth (16, 17, 40). The succession in this area, which is wholly in the graptoli- tic facies, has recently been described in detail (Cave & Hains 1986). The best sections of the Mottled Mudstone Member are at the Cardiganshire Slate Quarry (National Grid ref. SN 6991 9595) and in a stream near Eisteddfa-Gurig (SN 7951 8409), but the faunal transition between the persculptus and acuminatus Zones has not been investigated in detail. Strata above Mottled Mudstone Member. Dark grey rusty-weathering mudstones: in middle, sand- stones and siltstones near top of acuminatus Zone. 70-145 m Cwmere 5—25m Mottled Mudstone Member. Formation Banded mudstone with pale bioturbated layers and phosphatic concretions (both disappearing in topmost 3m). The lowest beds are unfossiliferous but about 1m above the base is a thin layer (15—-30cm) with abundant G. persculptus. Pyritized G. persculptus also occur above this layer. Bryn-glas Massive mudstone with splintery, Formation phacoidal cleavage. ORDOVICIAN-SILURIAN BOUNDARY STRATA IN WALES 69 Garth (5b). The following section is obtained by mapping in strata of predominantly shelly facies near Garth, 32km NE of Llandovery, Powys (Williams & Wright 1981). 250m + Sandstones & mudstones Rhuddanian shelly fossils 77m+ Garth Bank Formation 11-S5im Cwm Clyd Formation Eostropheodonta hirnantensis 0-30m Speckly Sandstone Hirnantia fauna. (Andrew [5a] records G. cf. Member persculptus and Mesograptus cf. modestus parvulus probably from this Member) Wenallt 0-20m Ooid Member Hirnantia fauna Formation 0-65m Siltstones Brongniartella cf. robusta (high Rawtheyan) Llandovery (4). The following section (transect i, of Cocks et al. 1984) is exposed almost continuously along a forestry road in the north Llandovery area (base of section at Grid ref. SN 8467 3962). The Hirnantia fauna, however, is extrapolated from 1-3 km further west. 120m Bronydd Formation Rhuddanian shelly fossils and graptolites suggesting atavus and acinaces Zones. Near base Climacograptus normalis 70m Scrach Formation (Hirnantia fauna in west) — Tridwr Formation Rawtheyan shelly fossils and ‘uppermost Ordovician’ graptolites Haverfordwest (1b). The following section (Cocks & Price 1975) is obtained by mapping in strata of predominantly shelly facies at Haverfordwest, Dyfed, but there are continuous expo- sures in road and railway sections upwards from about the middle of the Haverford Mudstone Formation (base of road section at Grid ref. SM 9573 1547). 85m Gasworks Sandstone Formation Graptolites indicating acinaces or atavus Zones at top. Rhuddanian shelly fossils throughout 370m Haverford Mudstone Formation Rhuddanian shelly fossils. Climacograptus cf. normalis near middle. ?C. normalis at 9m above base 65m Portfield Formation Hirnantia fauna at top, including Diplograptid undescr. sp. — Slade & Redhill Mudstone Formation Rawtheyan shelly fossils Conclusions On the assumption (cf. Temple 1978) that graptolite zones are definable and recognizable entities, then because of the wide extent of the persculptus and acuminatus faunas in central Wales, the Ordovician-Silurian boundary defined beneath the acuminatus Zone is in principle widely applicable in Wales. It is not however directly applicable in the marginal belt character- ized by boundary strata of shelly facies. Even in the recently reinvestigated Llandovery area (4), where there is an intermingling of shelly fossils and graptolites, the persculptus and acuminatus Zones are not firmly enough identified for the boundary to be recognized accurately. 70 J. T. TEMPLE References Andrew, G. 1925. The Llandovery rocks of Garth (Breconshire). Q. JI geol. Soc. Lond. 81: 389-405. Bassett, D. A., Whittington, H. B. & Williams, A. 1966. The stratigraphy of the Bala district, Merioneth- shire. Q. JI geol. Soc. Lond. 122: 219-269. Brenchley, P. J. & Cullen, B. 1984. The environmental distribution of associations belonging to the Hirnantia fauna—evidence from North Wales and Norway. In D. L. Bruton (ed.), Aspects of the Ordovician System: 113-125. Universitetsforlaget, Oslo. Cave, R. 1965. The Nod Glas sediments of Caradoc age in North Wales. Geol. J., Liverpool, 4: 279-298. & Hains, B. A. 1986. The geology of the country between Aberystwyth and Machynlleth. Mem. Br. geol. Surv., Keyworth, Sheet 163. 148 pp. Cocks, L. R. M. & Price, D. 1975. The biostratigraphy of the Upper Ordovician and Lower Silurian of south-west Dyfed, with comments on the Hirnantia fauna. Palaeontology, London, 18: 703-724, pls 81-84. —., Woodcock, N. H., Rickards, R. B., Temple, J. T. & Lane, P. D. 1984. The Llandovery Series of the type area. Bull. Br. Mus. nat. Hist., London, (Geol.) 38 (3): 131-182. Davies, K. A. 1926. The geology of the country between Drygarn and Abergwesyn (Breconshire). Q. JI geol. Soc. Lond. 82: 436-463. 1928. The geology of the country between Rhayader (Radnorshire) and Abergwesyn (Breconshire). Proc. geol. Ass., London, 39: 160-168. 1933. The geology of the country between Abergwesyn (Breconshire) and Pumpsaint (Carmarthenshire). Q. JI geol. Soc. Lond. 89: 172-200. Drew, H. & Slater, I. L. 1910. Notes on the geology of the district around Llansawel (Carmarthenshire). Q. JI geol. Soc. Lond. 66: 402-418. Elles, G. L. 1909. The relations of the Ordovician and Silurian rocks of Conwy (North Wales). Q. JI geol. Soc. Lond. 65: 169-192. —— 1922. The Bala country: its structure and rock-succession. Q. JI geol. Soc. Lond. 78: 132-172. Greenly, E. 1919. The geology of Anglesey, 2. Mem. geol. Surv. U.K., London, 389-980. Groom, T. T. & Lake, P. 1908. The Bala and Llandovery rocks of Glyn Ceiriog, North Wales. Q. JI geol. Soc. Lond. 64: 546-593. Hendriks, E. M. L. 1926. The Bala—Silurian succession in the Llangranog district (South Cardiganshire). Geol. Mag., London, 63: 121-139. Hiller, N. 1980. Ashgill Brachiopoda from the Glyn Ceiriog district, north Wales. Bull. Br. Mus. nat. Hist., London, (Geol.) 34: 109-216. Jehu, R. M. 1926. The geology of the district around Towyn and Abergynolwyn (Merioneth). Q. JI geol. Soc. Lond. 82: 465-489. Jones, O. T. 1909. The Hartfell-Valentian succession in the district around Plynlimon and Pont Erwyd (North Cardiganshire). Q. JI geol. Soc. Lond. 65: 463-537, pls 1, 2. —— 1925-49. The geology of the Llandovery district: Part I—The southern area. Q. JI geol. Soc. Lond. 81: 344-388 (1925). Part II—The northern area. Loc. cit. 105: 43-63 (1949). —— 1947. The geology of the Silurian rocks west and south of the Carneddau range, Radnorshire. Q. JI geol. Soc. Lond. 103: 1—33. & Pugh, W. J. 1916. The geology of the district around Machynlleth and the Llyfnant valley. Q. J] geol. Soc. Lond. 71: 343-383. Jones, W. D. V. 1945. The Valentian succession around Llanidloes, Montgomeryshire. Q. JI geol. Soc. Lond. 100: 309-332. Kelling, G. & Woollands, M. A. 1969. The stratigraphy and sedimentation of the Llandoverian rocks of the Rhayader district. In A. Wood (ed.), The Pre-Cambrian and Lower Palaeozoic rocks of Wales: 255-282. Univ. Wales Press. King, W. B. R. 1923. The Upper Ordovician rocks of the south-western Berwyn Hills. Q. JI geol. Soc. Lond. 79: 487-507. —— 1928. The geology of the district around Meifod (Montgomeryshire). Q. J! geol. Soc. Lond. 84: 671-700. Lake, P. & Groom, T. T. 1893. On the Llandovery and associated rocks in the neighbourhood of Corwen. Q. JI geol. Soc. Lond. 49: 426-440. Lapworth, H. 1900. The Silurian sequence of Rhayader. Q. J/ geol. Soc. Lond. 56: 67-135. Marr, J. E. 1880. On the Cambrian (Sedgw.) and Silurian beds of the Dee valley, as compared with those of the Lake District. Q. Jl geol. Soc. Lond. 36: 277-284. Pugh, W. J. 1923. The geology of the district around Corris and Aberllefenni (Merionethshire). Q. JI geol. Soc. Lond. 79: 508-541. ORDOVICIAN-SILURIAN BOUNDARY STRATA IN WALES 71 —— 1928. The geology of the district around Dinas Mawddwy (Merioneth). Q. J! geol. Soc. Lond. 84: 345-379. 1929. The geology of the district between Llanymawddwy and Llanuwchllyn (Merioneth). Q. JI geol. Soc. Lond. 85: 242-305. Roberts, B. 1967. Succession and structure in the Llwyd Mawr syncline, Caernarvonshire, North Wales. Geol. J., Liverpool, 5: 369-390. Roberts, R. O. 1929. The geology of the district around Abbey-Cwmhir (Radnorshire). Q. JI geol. Soc. Lond. 85: 651-675. Smith, B. 1935. The Mynydd Cricor inlier. Proc. geol. Ass., London, 46: 187-192. Strahan, A. et al. 1914. The geology of the South Wales Coalfield. Part XI, The country around Haver- fordwest. Mem. geol. Surv. U.K., London. viii + 262 pp. Temple, J. T. 1978. Comment on stratigraphical classification and all that. Lethaia, Oslo, 11: 340. Wade, A. 1911. The Llandovery and associated rocks of north-eastern Montgomeryshire. Q. JI geol. Soc. Lond. 67: 415-457. Wedd, C. B. et al. 1927. The geology of the country around Wrexham. Part I, Lower Palaeozoic and Lower Carboniferous rocks. Mem. geol. Surv. U.K., London. 1x + 179 pp. et al. 1929. The country around Oswestry. Mem. geol. Surv. U.K., London. x + 234 pp. Whittard, W. F. 1932. The stratigraphy of the Valentian rocks of Shropshire. The Longmynd-Shelve and Breidden outcrops. Q. JI geol. Soc. Lond. 88: 859-899. Whittington, H. B. 1938. The geology of the district around Llansantffraid ym Mechain, Montgomery- shire. Q JI geol. Soc. Lond. 94: 423-455. Williams, A. 1953. The geology of the Llandeilo district, Carmarthenshire. Q. J] geol. Soc. Lond. 108: 177-205. — & Wright, A. D. 1981. The Ordovician-Silurian boundary in the Garth area of southwest Powys, Wales. Geol. J., Liverpool, 16: 1—39. Wills, L. J. & Smith, B. 1922. The Lower Palaeozoic rocks of the Llangollen district, with special reference to the tectonics. Q. JI geol. Soc. Lond. 78: 176-223. La Limite Ordovicien—Silurien en France C. Babin,! R. Feist,? M. Mélou! et F. Paris* 1 Université Claude Bernard-Lyon 1—Département des Sciences de la Terre—27—43 Boulevard du 11 Novembre—69622 VILLEURBANNE Cédex, France 2 Centre d’Etudes et de Recherches Géologiques et Hydrogéologiques— Place Eugene Bataillon—34060 MONTPELLIER Cédex, France 3 Institut de Géologie—Faculté des Sciences—Avenue du Général Leclerc—35042 RENNES Cédex—et GRECO 130007 du C.N.R.S., France Synopsis The Ordovician and Silurian systems are well represented in France, but the boundary between them remains imprecise because there is generally a gap in the lower part of the Llandovery and/or the upper part of the Ordovician. The zctual documentation for the Armorican Massif and the south-west of France is briefly revised. Les systémes ordovicien et silurien sont largement représentes dans les massifs paléozoiques francais (notamment dans le Massif armoricain et en Montagne Noire au Sud du Massif central). Pourtant, la limite entre les deux systémes n’est nulle part reconnue avec precision dans l’état actuel des investigations. Une lacune sedimentaire semble, en réalité, étre assez généralisée, au moins pour la partie inférieure du Llandovery. Elle peut résulter de l’interférence d’un ensemble de causes, climatiques et variations eustatiques induites, €pirogéniques et tecton- iques distensives (échos taconiques) et manifestations volcaniques subordonnéees. Nous préciserons briévement ces propos pat l’examen de quelques successions. Le Massif Armoricain Différents domaines peuvent y étre considérés. En Normandie, la présence d’Ashgill est attestée par des Conodontes (zone a Amorpho- gnathus ordovicicus) pour le Calcaire de Vaux (Weyant et al. 1977). Des fragments de ces calcaires sont repris dans la formation glacio-marine dite des ‘pélites a fragments’ ou Tillite de Feuguerolles qui est également rapportee a l’Ashgill supérieur grace aux Chitinozoaires qu'elle renferme (F. Paris inédit). Dans les formations sus-jacentes l’absence apparente des Graptolites du Llandovery inférieur suggére une lacune correspondant au moins a celui-ci et débutant peut-étre dans l’Hirnantien. Dans les parties centrales et orientales du Synclinorium median armoricain, la limite Ordovicien-Silurien se place entre les Formations de Saint-Germain-sur-Ille et de la Lande Murée (Fig. 1). Le passage entre ces deux formations est expose dans diverses coupes des synclinoria du Ménez-Belair et de Laval. La Formation de Saint-Germain-sur-Ille, dans sa totalité, appartient a l’'Ordovicien supé- rieur. Elle est habituellement subdivisée en deux unités lithologiques: un Membre inferieur a dominante arénacée, puissant de 200m environ, et un Membre supérieur, argileux, et nettement moins développé (quelques dizaines de métres d’épaisseur). Des interlits argileux noirs s’intercalent dans l'ensemble grésoquartziteux constituant le Membre inférieur. Déposé dans un environnement littoral, voire tidal, ces grés livrent locale- ment une abondante faune, généralement rassemblée dans des lits d’accumulation. On y reconnait notamment des Brachiopodes (Drabovinella erratica), des Trilobites (Calymenella bayani, Homalonotidae), des Bivalves, et surtout des Graptolites qui ont permis de dater une partie de ce Membre inférieur (Skevington & Paris 1975). Ces Graptolites, limités a quelques niveaux gréso-micacés, sont exclusivement représentés par des Diplograptidae (Orthograptus truncatus truncatus, O. truncatus abbreviatus, O. truncatus pauperatus ainsi que de rares spéeci- Bull. Br. Mus. nat. Hist. (Geol) 43: 73-79 Issued 28 April 1988 74 BABIN, FEIST, MELOU & PARIS WwW & 4 >) S| es —l we Z| 26 WwW = =—|=s |e uJ Jo == aes a ty aj =) foe 9) Ww 5 J|}waz|o ef @®| OS) Si as|ee Z| 5 IF = el: = oe ) fe Membre supérieur Ww ae = a ui | ® Z ao 2 eae O ot & S| 2 ie = 7 S| 2 | 26 a | & |e (= O Ww @ 10m =e Zz Os = ra Fig. 1 Colonne stratigraphique a la limite S ri Ordovicien-Silurien dans le Synclinortum du Menez-Belair. 1. Quartzites et mudstones. 2. Grés et quartzites. 3. Mudstones et silt- stones. 4. Ampeélites. =I] iB i hr] ly Ww i mens de ? Climacograptus miserabilis et de ? Diplograptus fastigatus). S'appuyant sur la fre- quence relative des diverses sous-espéces de O. truncatus, Skevington & Paris (1975) admettent que les plus anciens niveaux a Graptolites de la Formation de Saint-Germain-sur-Ille appar- tiendraient a la partie supérieure de la Zone a D. complanatus, tandis que les niveaux les plus réecents représenteraient la Zone a D. anceps. La partie supérieure du Membre inférieur de la formation a donc été attribuée a l’Ashgill. Les Trilobites étudiés par Henry (1980) et les Brachiopodes, revisés recemment par Mélou (1985), n’apportent pas de précisions strati- graphiques complémentaires. Quant aux Chitinozoaires, ils n’ont pas été observés dans les termes les plus élevés de ce Membre inférieur (Paris 1981). Le Membre supérieur marque un net changement dans la lithologie. Sa base ravine le toit du Membre inférieur et ses caractéres sedimentologiques (mudstones et siltstones noirs a ‘ball and pillow structures’) rappellent certains faciés des formations glacio-marines décrites dans Y’Ordovicien terminal armoricain (Paris 1986). Aucune macrofaune n’y est connue. En revanche les Acritarches et les Chitinozoaires y sont relativement abondants. En dépit d’un état de conservation trés médiocre, ces microfossiles évoquent des formes de I’Ashgill supérieur. Si l’on accepte un parallélisme entre ce Membre supérieur et des formations glacio-marines finiordovi- ciennes telles que la Formation des ‘Pélites a fragments’ de Normandie ou les argiles micro- conglomératiques du Nord de l’Afrique, le sommet de la Formation de Saint-Germain-sur-Ille appartiendrait a l’Ashgill supérieur et vraisemblablement a |’Hirnantien. La Formation de la Lande Murée débute par un Membre inférieur constitué de quelques metres de quartzites noirs, pyriteux, admettant des intercalations de mudstones a Graptolites, trés riches en matiére organique (ampélites) et montrant des teneurs anormalement élevées en éléments-traces (Dabard & Paris 1986). Le contact avec le Membre supérieur de la Formation LA LIMITE ORDOVICIEN—SILURIEN EN FRANCE 75 de Saint-Germain-sur-Ille, correspondant a un brusque changement lithologique (Paris 1977), est plus ou moins bien exposé dans divers affleurements des synclinoria du Ménez-Bélair et de Laval (ex. carriére des ‘Planches’, en Guitté; carriére de ‘Pont-Douve’, en Médréac; carriére ‘Pioc’, en Vieux-Vy-sur-Couesnon; carriére du Rocher a Andouille-Neuville:; tranchée de Pautoroute Laval—Le Mans, a Ouest de Saint-Jean-sur-Erve; le ‘Moulin du Few’ en Balazé). Dans le Synclinorium du Menez-Belair, les premiers niveaux a Graptolites, parfois situés a moins d’un metre au-dessus du contact entre les deux formations, appartiennent déja au Tely- chien (sommet de la Zone a turriculatus ou Zone a crispus, selon les localités) (cf. Paris et al. 1980). Dans le Synclinorium de Laval, les premiers Graptolites récoltés dans la partie inférieure de la Formation de la Lande Murée appartiennent au Wenlock (Paris & Robardet, inédit). De toute évidence, il existe une lacune sédimentaire séparant les derniers dépéts ordoviciens (sommet du Membre supérieur de la Formation de Saint-Germain-sur-Ille) des premiers sédi- ments siluriens (base de la Formation de la Lande Murée). Cette lacune est d’ampleur variable selon les localités. Dans le Synclinorium du Ménez-Bélair, elle correspond au moins au Rhud- danien et a I’Aeronien (et peut-étre au sommet de l’Ashgill). Dans le Synclinorium de Laval cette lacune parait plus importante puisqu elle implique l'ensemble du Llandovery et une partie du Wenlock. Au Sud de Rennes, dans le Synclinorium de Martigne-Ferchaud, des travaux cartographiques (Herrouin, sous presse) ont recemment permis de préciser la succession lithologique locale, au voisinage de la limite Ordovicien—Silurien. Succedant aux siltstones micacés, a lits gréseux, de la Formation de Riadan (tradit- ionnellement rapportée au Caradoc et a lAshgill pro parte), on trouve la Formation de la Chesnaie (60 a 80m de puissance). Cette unite comprend un ensemble inférieur gréso- quartziteux et une partie supérieure a4 dominante argileuse. Pour l’instant, la Formation de la Chesnaie n’a livré aucune faune exploitable. Au-dessus se placent les grés et quartzites blancs de la Formation de Poligné (60 a 70 m d’épaisseur). Le plus souvent azoique, cet ensemble arénacé contient localement quelques Graptolites (Philippot 1950) de conservation trop médiocre pour fournir une attribution stratigraphique réellement fiable. Les premiéres faunes siluriennes sig- nificatives (Philippot 1950) apparaissent dans les mudstones noirs susjacents (ampélites). I] s’agit de riches assemblages de Graptolites de la base du Telychien (Zone a turriculatus). Dans le Synclinorium de Martigné-Ferchaud, la limite Ordovicien—Silurien se place donc entre le toit de la Formation de Riadan et les ampelites de la base du Telychien. En absence de tout controle paléontologique rigoureux, la position de cette limite reste donc trés approx- imative. Une lacune d’une partie du Llandovery, quoique vraisemblable, ne peut pour l’instant étre demontrée. Dans la partie occidentale du Synclinorium median, la presquile de Crozon permet d’approcher la limite Ordovicien—Silurien dans deux contextes différents et tous deux incom- plets. La succession observée dans lunité nord de la presquwile (plage du Veryarc’h en Camaret) demeure d’interprétation difficile (Fig. 2). En concordance sur la Formation des Grés de Kermeur, datée du Caradoc dans sa partie moyenne (biozone 14 a Jenkochitina tanvillensis; Paris 1981), la Formation du Cosquer (Hamoumi 1981; Guillocheau 1983) débute par des shales noirs a lamines gréseuses, bien stratifieés, puis se caracterise par un ensemble a blocs glissés qui passent progressivement a des boules (‘ball and pillow structures’). Les quartz de cette formation ont une origine glaciaire (Hamoumi et al. 1981). Vers le sommet, glissements et déformations s’atténuent, ce qui assure le passage a4 une stratification normale de grés a minces interlits de schistes noirs (Grés de Lamm-Saoz puissants de 6 métres environ). Ces grés sont surmonteés par les ampélites de la base du Groupe de Kerguillé qui livrent des Graptolites du Wenlock (Philippot 1950). La Formation du Cosquer n’a fourni aucun fossile dans les sédiments autochtones et son 4ge demeure imprécis. Un age ashgillien a cependant été proposé (Paris et al. 1981) par comparaison notamment avec celui attribué a la formation glacio-marine de la Tillite de Feuguerolles de Normandie. Les Grés de Lamm-Saoz furent, pour des raisons de géométrie, rapportés au Valentien (Silurien inférieur) par Philippot (1950). La recente découverte par l'un de nous (F.P.) de Armor- | SILURIEN WENLOCK ASHGILL GROUPE DE COSQUER KERGUILLE (2) [oo] P 3 og Za ud oO > a [e) (S) ac (S) Zz (ie) = <@ = a [e) je, | E=) E=2 [43 SILURIEN LUDLOW GROUPE DE KERGUILLE 2 DE LLANDOVERY CALCAIRES ? HIRNANTIEN ET TUFS HIRNANTIEN DES Zz ud S 5 = iS) fe) & el he 1S) & 10m (0) S E22 Bes Le Fig. 2 Contact Ordovicien-Silurien dans la coupe du Veryarc’>h (Camaret, presquile de Crozon). 1. Grés et quartzites. 2. Shales. 3. Shales a ‘ball and _ pillow structures’. 10m 4. Megaslumps. 5. Ampelites. Fig. 3 Contact Ordovicien-Silurien le long de YAulne, a Est de Tregarvan (presqu’ile de Crozon). 1. Schistes. 2.Grés et quartzites. 3. Hyaloclastites. 4. Calcaires. # niveau a Hir- (6+ Ts nantia. LA LIMITE ORDOVICIEN—SILURIEN EN FRANCE 77 ichitina nigerica dans le dernier interlit noir de ces grés, situeé 4 30cm sous les ampélites wenlockiennes, permet désormais de proposer, par comparaison avec les pélites a fragments du Sahara, un 4ge ashgillien supérieur pour la partie sommitale des Grés de Lamm-Saoz. Ainsi se trouve confirmée l’importance de la lacune qui correspond, dans cette unité nord, a la totalité du Llandovery. Dans l’unité sud de la presqu’ile de Crozon, la Formation des Grés de Kermeur est surmon- tée par un ensemble volcano-sédimentaire désigné Formation des Tufs et calcaires de Rosan. Aucun affleurement ne permet l’observation continue de la colonne correspondante. La base est concordante sur la Formation de Kermeur (falaise de Raguenez). Les coupes de la carriére du four a chaux de Rosan et de la route contigué livrent en abondance Nicolella actoniae. La récente révision de cette espéce (Harper 1984) permet de considérer que nous sommes ici en présence de N. actoniae ramosa, sous-espéce de |’Ashgill. Ailleurs, a Lostmarc’h, les calcaires de Rosan sont également attribuables a l’Ashgill d’aprés les assemblages de Conodontes (zone a Amorphognathus ordovicicus) selon Paris et al. (1981). Un affleurement isolé le long de I’Aulne, a Coat-Garrec, a livre des Echinodermes (Chauvel & Le Menn 1972) qui ont confirmé l’age ashgillien proposé pour cet affleurement par Melou (1971) d’aprés la faune de Leptestiina. Enfin, il semble que la partie la plus élevée de cette formation soit représentée a l'Est de Trégarvan, le long de la riviére Aulne. La sedimentation carbonatée y régresse au profit des dépots arénacés (Fig. 3). L’un de nous a réecemment découvert dans cette coupe (Mélou 1987) un niveau a Hirnantia sagittifera au-dessus duquel 90 métres de grés et de hyaloclastites avec quelques bancs carbonatés n’ont jusqu’a present fourni aucun fossile. Cette partie sommitale de la formation peut donc encore correspondre a l’Hirnantien ou représenter déja la base du Llandovery. La pile est tronquée par une faille importante qui la met en contact avec une partie élevée (Ludlow probablement) du Groupe de Kerguillé. Notons que ces observations nouvelles en presquile de Crozon tendent a réhabiliter un certain synchronisme des Formations du Cosquer et de Rosan qui avait été mis en doute recemment dans divers schémas (Paris et al. 1981; Guillocheau 1983). Dans le Sud-Ouest du Massif armoricain, les données relatives a unite vendéenne demeurent fragmentaires (Ters 1979). Les schistes et grés schisteux rapportés a l’Ordovicien supérieur comme les schistes et phtanites a Radiolaires attribués au Llandovery n’ont pas livré de fossiles déterminants. En conclusion, la présence de l’Ashgill, longtemps méconnue dans le Massif armoricain, y est désormais attestée dans plusieurs domaines et son extension inclut l’Hirnantien. Le Silurien, par contre, parait en general amputé de sa partie basale au niveau d’une lacune qui peut, suivant les régions, intéresser Rhuddanien et Aeronien (Synclinorium de Martigné-Ferchaud, Synclinorium du Menez Belair) ou affecter ensemble du Llandovery (presquile de Crozon). Le Massif armoricain ne permet donc aucune observation de la limite Ordovicien-Silurien. Le Sud-ouest de la France En Aquitaine, ’étude récente de sondages dans le socle paléozoique sous la couverture mé€socénozoique, a permis a l’un de nous (F.P.) de constater, d’aprés les Chitinozoaires, la presence d’Ashgill terminal directement surmonté par des niveaux assez élevés du Llandovery. Une lacune du Silurien basal parait donc également reconnaissable dans cette région. En Montagne Noire, la succession de l’Ordovicien et du Silurien est observable en deux endroits connus depuis Chaubet (1937): au-dessus de la “‘Tranchée noire’ pres de la Grange du Pin et au Petit Glauzy. De fagon générale, la succession ordovicienne se termine par des alternances calcaréo-argileuses dites ‘calcaires 4 Cystoides’ et reputées d’age ashgillien depuis Dreyfus (1948). Dans une récente révision des Brachiopodes de ces niveaux, Havli¢ek (1981) remet en cause cet age et estime que les associations décrites indiqueraient plutdot le Caradoc supérieur. L’Age ashgillien demeure néanmoins plausible et si la faune a Hirnantia n’a pas été reconnue, les calcaires, quoique trés pauvres en Conodontes, ont livré a l'un de nous (R.F.) quelques restes d’Amorphognathus ordovicicus. Ces niveaux terminaux, assez détritiques, n’ont fourni aucun Graptolite. Ceux-ci n’apparaissent que quelques métres plus haut dans les argilites 78 BABIN, FEIST, MELOU & PARIS carburées. Dans la partie basale de ces schistes noirs, a la Grange du Pin, des Conodontes, extraits des nodules calcaires, indiquent selon Centene & Sentou (1975), la zone a celloni (€quivalente des zones 20 a 23 des Graptolites, Llandovery moyen). Les mémes niveaux livrent au Petit-Glauzy, selon ces auteurs, Monograptus sedgwickii, M. uncinatus, M. nudus du Llando- very moyen également (zone 21). On constate ainsi que la limite Ordovicien—Silurien ne peut étre reconnue avec précision en Montagne Noire dans I’état actuel de la documentation. Faute de fossiles dans les niveaux qui assurent le passage entre les derniers carbonates a Cystoides et les premiéres ampélites a septaria, il demeure impossible de conclure a la continuité ou a l'existence de lacunes. References Centene, A. & Sentou, G. (1975). Graptolites et Conodontes du Silurien des Massifs du Midi mediterranéen. Thése 3e cycle. Université des Sciences et Techniques du Languedoc, Montpellier (inédit.). 176 pp., 13 pls. Chaubet, M. C. 1937. Contribution a étude du Gothlandien du versant méridional de la Montagne Noire. Mem. Trav. Lab. Geol. Univ. Montpellier 1: 1-223, pls 1-7. Chauvel, J. & Le Menn, J. 1973. Echinodermes de lOrdovicien supérieur de Coat-Carrec, Argol (Finistére). Bull. Soc. géol. miner. Bretagne, Rennes, 4 (1): 39-61, pls 1—3. Dabard, M.-P. & Paris, F. 1986. Palaeontological and geochemical characteristics of Silurian black shale formations from the Central Brittany Domain of the Armorican Massif (Northwest France). Chem. geol., Amsterdam, 55 (1/2): 17-29. Dreyfus, M. 1948. Contribution a etude geologique et paléontologique de ’Ordovicien supérieur de la Montagne Noire. Mem. Soc. geol. Fr., Paris, 58: 1-63. Guillocheau, F. 1983. La sedimentation paleozoique ouest-armoricaine. Histoire sédimentaire; relations tectonique-sédimentation. Bull. Soc. geol. miner. Bretagne, Rennes, (C) 14 (2): 45-62. Hamoumi, N. 1981. Comparaison des coupes du Veryarc’h et de l’Aber-Kerglintin. In: Analyse sedimentol- ogique des formations de |’Ordovicien superieur en presqu ile de Crozon (Massif Amoricain). , Le Ribault, L. & Pelhate, A. 1981. Les Schistes du Cosquer (Ordovicien supérieur, Massif Armor- icain occidental): une formation glacio-marine a la peripherie d’un islandsis ordovicien. Bull. Soc. géol. France, Paris, (7) 23: 279-286. Harper, D. A. T. 1984. Brachiopods from the Upper Ardmillan succession (Ordovician) of the Girvan district, Scotland. Part 1. Palaeontogr. Soc. (Monogr.), London. 78 pp., 11 pls. Havlicek, V. 1981. Upper Ordovician Brachiopods from the Montagne Noire. Palaeontographica, Stutt- gart, (A) 176: 1—34, pls 1-9. Henry, J. L. 1980. Trilobites ordoviciens du Massif Armoricain. Mem. Soc. géol. miner. Bretagne, Rennes, 22: 1-250, pls 1-48. Herrouin, Y. (1987). Carte geologique de la France 1/50000e. Feuille de Bain-de-Bretagne 388. B.R.G.M. éd. Melou, M. 1971. Nouvelle espéce de Leptestiina dans ’Ordovicien supérieur de l’Aulne (Finistére). Mem. Bur. Rech. geol. minier., Paris, 73: 93-105, pls 1, 2. 1985. Révision d*Orthis’ berthoisi ROUAULT, 1849, Orthida (Brachiopoda) de l’Ordovicien du Massif Armoricain. Geobios, Lyon, 18: 595-603, pls 1, 2. —— 1987. Découverte de Hirnantia sagittifera (M’Coy 1851) (Orthida Brachiopoda) dans l’Ordovicien supérieur (Ashgillien) de l’extrémité occidentale du Massif Armoricain. Géobios, Lyon, 20: 679-686, pl. 1. Paris, F. 1977. Les formations siluriennes du Synclinorium du Ménez-Bélair; comparaisons avec d’autres formations siluriennes du Massif armoricain. Bull. Bur. Rech. geol. Min., Paris, (2e sér., 1) 2: 75-87. 1981. Les Chitinozoaires dans le Paléozoique du Sud-Ouest de ’Europe. Mem. Soc. géol. miner. Bretagne, Rennes, 26: 1-412, pls 1-41. 1986. Les formations paléozoiques et leur structuration. In: Notice géologique de la feuille Combourg a 1/50 000eme. B.R.G.M. (doct provisoire). ——, Pelhate, A. & Weyant, M. 1981. Conodontes ashgilliens dans la Formation de Rosan, coupe de Lostmarc’h (Finistére, Massif Armoricain); conséquences paléogéographiques. Bull. Soc. geol. miner. Bretagne, Rennes, (C) 13 (2): 15-35. , Rickards, R. B. & Skevington, D. 1980. Les assemblages de Graptolites du Llandovery dans le Synclinorium du Ménez-Bélair (Massif Armoricain). Geobios, Lyon, 13: 153-171, pls 1, 2. LA LIMITE ORDOVICIEN—SILURIEN EN FRANCE 1S) Philippot, A. 1950. Les Graptolites du Massif Armoricain, étude stratigraphique et paléontologique. Mem. Soc. geol. miner. Bretagne, Rennes, 8: 1—295. Skevington, D. & Paris, F. 1975. Les Graptolites de la Formation de Saint-Germain-sur-Ille (Ordovicien supérieur du Massif Armoricain). Bull. Soc. geol. Fr., Paris, (7) 17: 260-265. Ters, M. 1979. Les synclinoriums paléozoiques et le Précambrien sur la fagade occidentale du Massif vendéen. Stratigraphie et structure. Bull. Bur. Rech. geol. Min., Paris, (2e sér., 1) 4: 293-302. Weyant, M., Dore, F., Le Gall, J. & Poncet, J. 1977. Un episode calcaire ashgillien dans Est du Massif Armoricain, incidence sur l’age des depots glacio-marins finiordoviciens. C.r. hebd. Seanc. Acad. Sci., Paris, (D) 284: 1147-1149. The Ordovician—Silurian boundary in the Oslo region, Norway L. R. M. Cocks Department of Palaeontology, British Museum (Natural History), Cromwell Road, London SW7 SBD Synopsis The Ordovician-Silurian boundary is exposed sporadically throughout the southern and central parts of the Oslo region; to the north there is an unconformity. In the central Oslo—Asker districts a well- developed Hirnantia fauna underlies beds with acuminatus Zone graptolites; other beds yield Holo- rhynchus faunas in the late Ordovician and early members of the Stricklandia lineage in the overlying Llandovery. Some early Silurian conodonts and acritarchs are recorded. Lower Palaeozoic rocks outcrop in the Oslo region within a 230km by 50km area which is separated from the Precambrian to the east by the faults of a Permian graben. Within this broad region, most recent work in the late Ordovician and early Silurian has been achieved in the Oslo—Asker district, which lies in the approximate centre of the region, and also in the Hadeland district, some 50 km to the north of Oslo. These and other districts will be reviewed in turn. The Ordovician and Silurian beds in the area have been known since the early work of Murchison, Kjerulf, Broegger and others, and were the subject of a monumental study near the turn of the century by Kjaer (e.g. 1908). During the past fifteen years much new work has been done, for example Worsley et al. (1983) proposed a modern system of stratigraphical nomencla- ture for the Silurian rocks of the region. Oslo—Asker District. The formation names for the late Ordovician stratigraphy (Fig. 1) were erected by Brenchley & Newall (1975), and its biostratigraphy and ecology elucidated by Brenchley & Cocks (1982), its trilobites described by Owen (1980, 1981) and its brachiopods by Cocks (1982). The topmost few metres of the Husberg¢ya Shale carries the trilobite Tretaspis sortita broeggeri, which Owen (1980) regarded as indicative of the uppermost Rawtheyan Stage. A Hirnantia fauna is known from horizons near the base of the Langgyene Sandstone Forma- tion and within the Langara Limestone-Shale Formation (Brenchley & Cocks 1982: 796), and includes common Dalmanella testudinaria, Hirnantia sagittifera, Cliftonia aff. psittacina, Hindella cassidea, Eostropheodonta hirnantensis, Mucronaspis mucronata kjaeri, bryozoans and cricco- conariids, and less common Acanthocrania, Glyptorthis, Lingula, Leptaena, Orbiculoidea, Oxo- plecia, Philhedra, Calyptaulax, Illaenus, Platycoryphe, Toxochasmops, molluscs, crinoids and carpoids. Elements of the Hirnantia fauna persist above this horizon in Hindella—Cliftonia and Dalmanella associations higher in the Lang¢yene Sandstone and there are also other faunas there such as one dominated by Trematis norvegica and modiolopsid bivalves. Above these, in the west of the area in Asker there occur thick beds largely composed of Holorhynchus gigan- teus, but with 13 other brachiopods and 17 other animals also recorded from them (Brenchley & Cocks 1982: 802), whilst in the east of the area, in the Oslo District, only trace fossils occur in rocks believed to represent a shore-face environment. At the top of the Langgyene Sandstone there occur shallow-water channel sequences which in some cases bear high abundance, low diversity faunas dominated by brachiopods such as Brevilamnulella kjerulfi and Thebesia scopu- losa. This total sequence represents a regression since at least mid-Rawtheyan times, but above the channel-fill beds there occurs a metre-thick couplet of sandstone and limestone over the whole district which carries faunas, which are not age-diagnostic, of small shells such as Onniella, Eoplectodonta, Leangella, Paucicrura and Dolerorthis, as well as crinoids and bryozoa (and 16 other rarer forms). This couplet is lithologically included within the Langgyene Forma- tion, but in fact marks the start of the ‘early Silurian’ transgression in the area. It is conform- Bull. Br. Mus. nat. Hist. (Geol) 43: 81-84 Issued 28 April 1988 82 L. R. M. COCKS ASKER OSLO RINGERIKE HADELAND E E e ra © 5 z = cc “s $ 8 2 3 ° E e é “s = QO)|< 5 3 0 re (ae || ee > rs © = O|# = 3 = oO : s £ g 3 oO < a = : [ag =) x Fig. 1 Latest Ordovician and early Silurian stratigraphy in the Asker, Oslo, Ringerike and Hadeland districts of the Oslo Region. ably followed by the basal organic-rich shales of the Solvik Formation in the Oslo District, which carries no shelly fauna, but from which Howe (1982) has identified Climacograptus transgrediens Waern from an horizon 11m above the base of the formation at Ormd¢ya, which he attributes to an horizon low in the acuminatus Zone (or perhaps high in the persculptus Zone). In the west of the area, in the Asker District, there was no break in the deposition of shelly faunas, and brachiopods are recorded from all three members of the Solvik Formation there, in a similar way to the higher parts of the formation in the Oslo District (Baarli 1985; Baarli & Harper 1986). The first occurrence of Stricklandia lens prima is at 95m above the base of the Solvik Formation (Myren Member) and the transition from S. lens prima to S. lens lens occurs between 122 and 130m above the base (Baarli 1986). Conodonts of the Icriodella discreta—I. deflecta Zone are known from 8m above the base of the Solvik Formation at Konglungen, Asker (Aldridge & Mohamed 1982). Above the three members of the Solvik Formation, the Rytteraker Formation yields pentamerides and conodonts of Aeronian age: Nakrem (1986) has identified the Distomodus kentuckyensis—D. staurognathoides conodont zonal boundary as occurring at about the boundary of the Solvik and Rytteraker Formations. The Ringerike area. The latest Ordovician of the Ringerike area remains unrevised, and thus the old stage terminology of Kiaer (1897, 1908) is employed—tt carries a rich brachiopod fauna, but one not identical to that from the Oslo—Asker region and no Hirnantia fauna is known from the area; it also differs in the presence of bioherms and patch reefs within Stage 5b, the most notable of which is at Ullerntangen. The relationships between the Ordovician and Silurian beds are obscure and a local unconformity is postulated here (Fig. 1). The overlying beds of the Saelabonn Formation are shallow-water storm deposits with lenses of displaced ORDOVICIAN-SILURIAN BOUNDARY IN NORWAY 83 shelly faunas (Thomsen 1982); their detailed age is indeterminate, but probably includes the Lower Llandovery. The overlying Rytteraker Formation includes the Borealis-Pentamerus transition near its base (Mérk 1981), and that horizon is certainly now in the Aeronian. Smelrgr (1987) has identified the acritarch zones 1 and 2 of Hill (1974) as occurring in the Saelabonn Formation. The Hadeland area. Owen (1978) has revised the late Ordovician and early Silurian of this area and established a Rawtheyan age for the Kjérrven Formation which underlies the Kalvsj¢ Formation, which carries a sparse trilobite fauna, some brachiopods and the cystoid Hemi- cosmites and the calcareous alga Palaeoporella which indicate an Ordovician rather than a Silurian age. Above this the 120m thick Sk¢yen Sandstone Formation appears to straddle the Ordovician-Silurian boundary, since beds with Zygospiraella and other typical early Silurian brachiopods occur from about the middle of the formation. The Skdyen Sandstone is succeeded conformably by the Rytteraker Formation which yields Borealis borealis near its base. Other areas. From the Skien and Porsgrunn district near the south of the Oslo Region, for example in a section at Hergyavegen, Porsgrunn, Holorhynchus beds followed by early Silurian beds yielding Zygospiraella duboisi (Verneuil) and Eostropheodonta mullochensis (Reed) are known, but the stratigraphy is unrevised. In the Oslo region north of Hadeland there is an unconformity between the late Caradoc and early Ashgill Mjgesa Limestone and the early Silurian, for example Mller (1986) has described the succession at Brummunddal, where the Helggya Quartzite of probable Aeronian age bearing Borealis borealis rests on the Mjgesa Limestone. References Aldridge, R. J. & Mohamed, i. 1982. Conodont biostratigraphy of the early Silurian of the Oslo Region. In D. Worsley (ed.), Field meeting, Oslo region, 1982. I.U.G.S. Subcommission on Silurian Stratigraphy: 109-120, 2 pls. Universitetsforlaget, Oslo (Pal. Contr. Univ. Oslo 278). Baarli, G. 1985. The stratigraphy and sedimentology of the early Llandovery Solvik Formation, central Oslo Region, Norway. Norsk geol. Tiddsskr., Oslo, 65: 229-249. —— 1986. A biometric re-evaluation of the Silurian brachiopod lineage Stricklandia lens/S. laevis. Palae- ontology, London, 29: 187—205, pl. 21. & Harper, D. A. T. 1986. Relict Ordovician brachiopod faunas in the Lower Silurian of Asker, Oslo Region, Norway. Norsk geol. Tidsskr., Oslo, 66: 87-98. Brenchley, P. J. & Cocks, L. R. M. 1982. Ecological associations in a regressive sequence: the latest Ordovician of the Oslo—Asker district, Norway. Palaeontology, London, 25: 783-815, pls 85-86. —— & Newall, G. 1975. The stratigraphy of the upper Ordovician Stage 5 in the Oslo—Asker district, Norway. Norsk geol. Tidsskr., Oslo, 55: 243-275. Cocks, L. R. M. 1982. The commoner brachiopods of the latest Ordovician of the Oslo—Asker District, Norway. Palaeontology, London, 25: 755-781, pls 78-84. Hill, P. J. 1974. Stratigraphic palynology of acritarchs from the type area of the Llandovery and the Welsh Borderland. Rev. Palaeobot. Palynol., Amsterdam, 18: 11-23. Howe, M. P. A. 1982. The Lower Silurian graptolites of the Oslo Region. In D. Worsley (ed.), Field meeting, Oslo region, 1982. I.U.G.S. Subcommission on Silurian Stratigraphy: 21-32, 2 pls. Uni- versitetsforlaget, Oslo (Pal. Contr. Univ. Oslo 278). Kiaer, J. 1897. Faunistische Uebersicht der Etage 5 des norwegischen Silursystems. Skr. VidenskSelsk. Christiania (Math.-nat.) 1897 (3): 1-76. —— 1908. Das Obersilur im Kristianiagebiete. Skr. VidenskSelsk. Christiania (Math.-nat.) 19€6 II: 1-595, pls 1-24. Mller, N. K. 1986. Evidence of synsedimentary tectonics in the Lower Silurian (Llandovery) strata of Brumunddalen, Ringsaker, Norway. Norsk geol. Tidsskr., Oslo, 66: 1-15. Mérk, A. 1981. A reappraisal of the Lower Silurian brachiopods Borealis and Pentamerus. Palaeontology, London, 24: 537-553, pls 83-85. Nakrem, H.-A. 1986. Llandovery conodonts from the Oslo Region, Norway. Norsk geol. Tidsskr., Oslo, 66: 121-133. 84 L. R. M. COCKS Owen, A. 1978. The Ordovician and Silurian stratigraphy of Central Hadeland, south Norway. Norg. geol. Unders., Oslo, 338: 1—23, pl. 1. 1980. The trilobite Tretaspis from the upper Ordovician of the Oslo region, Norway. Palaeontology, London, 23: 715—747, pls 89-93. —— 1981. The Ashgill trilobites of the Oslo Region, Norway. Palaeontographica, Stuttgart, (A) 175: 1-88, pls 1-17. Smelrgr, M. 1987. Early Silurian acritarchs and prasinophycean algae from the Ringerike District, Oslo Region (Norway). Rev. Palaeobot. Palynol., Amsterdam, 52: 137-159, pls 1-5. Thomsen, E. 1982. Saelabonn Formationen (nedre Silur) i Ringerike, Norge. Arsskr. Dansk geol. Foren. 1981: 1-11. Worsley, D., Aarhus, N., Bassett, M. G., Howe, M. P. A., M¢@rk, A. & Olaussen, S. 1983. The Silurian succession of the Oslo Region. Norg. geol. Unders., Oslo, 384: 1-57. East Baltic Region D. Kaljo, H. Nestor and L. Polmat+ Institute of Geology, Estonian Academy of Sciences, Estonia Puistee 7, Tallinn 200101, USSR + L. Polma died in January 1988. Synopsis Five confacies belts from north to south, from Estonia through Latvia to Lithuania, are described briefly through the late Ordovician and early Silurian, with their varied facies and faunas. Despite clear breaks corresponding to the Ordovician-Silurian boundary at the edges of the depositional basin, rocks of Hirnantian age are identified from the centre of the basin, including Hirnantia and Dalmanitina faunas in the Porkuni Regional Stage and basal Silurian faunas, including some graptolites, chitinozoans, brachio- pods and conodonts, from the overlying Juuru Regional Stage. Any stratigraphical break at the boundary appears to be represented by no more than a facies change. Introduction The East Baltic area is a part of the extensive gulf-like Baltic sedimentary basin (Mannil 1966; Kaljo & Jiirgenson 1977). The uppermost Ordovician and the lowermost Silurian are mostly represented by carbonate or terrigenous-carbonate rocks with an exceptionally rich benthic shelly fauna; however, pelagic groups of fossils, especially graptolites, are of a more restricted distribution. The rocks are tectonically undisturbed, and unmetamorphozed (CAI 1-1-5), with only a little dolomitization in places, and the fossils are well preserved. The bedding is almost horizontal and dips slightly to the centre of the basin. The distribution of the Ashgill— Llandovery rocks in the East Baltic is shown in Fig. 1. The outer margin of the area is erosional and corresponds to the base of the Ashgill (Vormsi Regional Stage). The axial part of the basin with the most deep-water rocks corresponds to the Baltic Syneclise (IV belt), and along its margins there occur shallower-water sediments. Most of the area is covered by younger rocks. The outcrops of the Ordovician—Silurian boundary strata are confined to North Estonia (Belt I in Fig. 1), where only comparatively shallow-water deposits are exposed. A more complete succession of facies in the basin can be seen in borehole sections. Fig. 2 presents a cross section of Ashgill and Lower and Middle Llandovery strata along the Orjaku-Remte—Ukmerge line, which is shown in Fig. 1. The section goes across the main facies belts of the basin and shows the relations between local lithostratigraphical units and their general lithology. In the figure stratigraphical units are marked with letter-indexes: their full nomenclature is given in Fig. 3. Confacies belts In the East Baltic five confacies belts can be distinguished in Ordovician—Silurian boundary beds. Their distribution is shown in Fig. 1 and their lithological composition in Fig. 2. Type 1—the most shallow-water sections in North Estonia and Hiiumaa Island represented by aphanitic, bioclastic and biohermal limestones. In the Raikkiila Formation there occur primary argillaceous dolomites of lagoonal origin in places. Some considerable stratigraphical gaps have been established (Fig. 3). The Ordovician ends with Early Porkuni bioclastic, bio- hermal and arenaceous limestones (Arina Formation), which carry a Streptis brachiopod com- munity (Hints 1986), disconformably overlain by Juuru aphanitic (Koigi Member) and biomicritic limestones (Varbola Formation) with a Stricklandia community (Rubel 1970). Type II—sections in central Estonia and Saaremaa Island. Represented by marls, aphanitic and biomicritic nodular limestones. The sections are more complete than in Type I. A distinct hiatus has been established only in the upper part of the Porkuni Regional Stage and in the west at the top of the Raikkiila Stage. The Ordovician-Silurian boundary interval is similar to the sections of Type I, but southwards the Arina Formation and the Koigi Member thin out Bull. Br. Mus. nat. Hist. (Geol) 43: 85-91 Issued 28 April 1988 86 KALJO, NESTOR & POLMA Talsi IV 4/ { Kandava @RIGA ex QREMTE te AT Vil AUN. Sue. ihe ‘ Sturi Q * 7N \ a BEF ER ae er } se J b Siupylial gS aN a Sakyna ~~ Stacgianai Ne i iors Vio, Kreékenava © A Ukmerge ~2 aes Ye O q ae See EA INGA yy, KALININGRAD i S.S.R. ; REGION 4 VILNIUS Fig. 1 Distribution of Ordovician—Silurian boundary rocks in the East Baltic area. 1—boreholes, 2—administrative boundaries, 3—outer margin of the distribution of Ashgill and Llandovery rocks, 4—boundaries of main types of sections, marked with Roman numbers. and the boundary of the systems continues in a comparatively monotonous complex of nodular limestones and marls. In places the Porkuni Regional Stage may be missing. Type III—sections in south Estonia and north-west Latvia. Marls and argillaceous lime- stones, including their red-coloured varieties, are significant lithologies. In the Llandovery, aphanitic limestones alternate with marls. A considerable erosional gap corresponds to the upper part of the Pirgu Regional Stage, and this gap increases westwards. The uppermost ‘oueu dy} Jo (Aue Jl) JUPUOSUOD JsIY BUIMOT[OJ 9Y} PUL 19}19] ISIN IY} JO S}SISUOD UONVUIIOJ & JO XOPUT OY} :¢ “SIY 99S Soxopul [vorydeisHeNs oy) Jo Suluvow Jo (sojeys) sauoispnur onyo}deis—, ‘sjieur pue sauojsouly snosoeT[isie AoIS8—g ‘souo}soUll] Ie[NpoU dI}IOTWIOIG—¢ ‘s]IBvUI puv soUO}SOULT] snosox][Is1e poi—p ‘(soy N]IoyVo) sauo}soUNT] SUTT[eISAIN0}dAIO 0} -oUY—E ‘(SoIUEIROTeO OILIedS) soUOJSOUT] 9II[OO pu oNSEID ‘ONLIedso1q—Z ‘uIsIIO [PUOOSR] JO SYOOI 9}BUOGILO-sNOsdeTIZIe poppoq-uIYyI—] “| “BIJ Ul UMOYS dUT] 9Y}) BuOTe [eAIO}UI AIepUNO sy} JO UONSeS sso [RoIydeIsyeNS 7 ‘SI 6-/dfLS 99001 IVITRAMS _ ppgfs900 92 -WAVONVH alee IL68 BL89— ee ee Vs Jl 089 = oe a, |p| 9699| 2 ie 1S Be 15 x oy) $9101. Oy In 2 Lia Sorel, @ y = ayy | #001 W 906 5 ‘xipeell eee Be Og a "ISL = > = = = o = 2 IWNTIVLS : i aS OL ~ S | ia —~ 0 | | 45-V¥TOM ie an W | | marae S | A I jo i Pom cnctliabs x 88 KALJO, NESTOR & POLMA Ordovician is represented by marls and argillaceous limestones with the Dalmanitina Fauna (Kuldiga Formation). Above this occur biosparitic, oolitic and arenaceous limestones of the Saldus Formation. The Silurian begins with marls and argillaceous limestones of the Ohne Formation with the Clorinda community (Rubel 1970). Type IV—sections in southeast Estonia, considerable part of Latvia, west Lithuania and the Kaliningrad Region. The studied stratigraphical interval begins and ends with dark graptolitic mudstones with the assemblage of the Pleurograptus linearis Zone in the Ordovician part (Fjacka Formation) and of the Coronograptus cyphus—Monograptus sedgwickii Zones in the Silurian (Dobele Formation). Between these key beds there occur red and grey calcareous mudstones, marls and aphanitic limestones. The uppermost Ordovician is analogous to the sections of Type III. The Silurian begins with marls and aphanitic limestones of the Staciunai Formation which have yielded few fossils good for correlation. Type V—sections in east Lithuania and southeast Latvia with an extensive hiatus at the boundary interval. More or less continuous Upper Ordovician deposits are represented by marls and various limestones which end at the top of the Pirgu Regional Stage with the aphanitic limestones of the Taucionys Formation which yield a Holorhynchus fauna. There is a hiatus at the level of the Porkuni, Juuru and Raikkiila Regional Stages, or in places there occur thin residual tongues and lenses of the Kuldiga, Saldus and Apascia Formations, which are transgressively overlain by mudstones and marls of the late Llandovery Adavere Regional Stage. In the westernmost part of Lithuania and in the Kaliningrad District the rocks of the Ordovician-Silurian boundary interval become still more argillaceous and graptolites occur throughout the whole section, with the exception of the uppermost Ordovician which yields a shelly Hirnantia fauna. This is a transition to a different type of facies belt which is distributed in north Poland and the southern part of the present Baltic Sea. Thus analysis of the lithologies and fossils of the various sections shows that by the end of the Ordovician the Baltic basin had experienced a considerable regression which reached its maximum in the second half of the Porkuni. This is indicated both by hiatuses in the sections (Fig. 3) and by the presence of calcareous oolites and early diagenetic (or sedimentary) dolomi- REGIONAL STAGE NORTH ESTONIA WEST LATVIA, W. LITHUANIA FAST LATVIA, EAST LITHUANIA CENTRAL | coyrtH ESTONIA ESTONIA Se Mb: SLemme Mb) DOBELE Fm 1 | SAARDE Fm SILURIAN PORKUNI Fr ar Puikule Mb. : |) [psuiaus Fm, [Rigas Bp KULDIGA Fm nooner HES JELGAVA Fm. AALLIKU Fn9SVEDASA/ JONSTORP Fm %MOE fm 8 '™ VORMS/| |KORGES- ec esicene fms TUDULINNA Fm 3 FIACKA Fm 3 MEILUNAI Fm. - 7 i cha ai ee — | Fig. 3. Stratigraphical scheme of the late Ordovician and early Silurian boundary rocks in the East Baltic area. = nooner HES rna = = Lea ptinini YS] prrey Ao / Fm. SADILA Fm. mere S) of, a PAROVEJA Fm Fm. S SN Q SS SECTIONS EAST BALTIC REGION 89 tes in the Saldus Formation in the axial part of the basin. The character of the transition from the Porkuni to the Juuru Regional Stage and the lithology of the sequences indicate a rapid deepening of the basin, obviously of glacial eustatic origin (Kaljo et al., in press). Local stratigraphy Knowledge of the local stratigraphy of rocks near the Ordovician—Silurian boundary has considerably improved in the past few decades. The correlation chart presented in Fig. 3 is based on the decisions of the regional stratigraphical conferences in Vilnius in 1976 and in Tallinn in 1984 (Grigelis 1978). The chart was compiled from material in many publications (see further references in the papers by Mannil 1966, Kaljo 1970, Kaljo & Klaaman 1982, PaSkevié- ius 1979, Grigelis 1982, Ulst et al. 1982). Dynamics of the faunas From the five regional stages from Vormsi to Raikkiila which correspond to the Ashgill and lower and middle Llandovery, extremely rich fossil faunas have been collected. The present paper uses the data obtained through the study of eight groups of fossils: stromatoporoids, tabulate corals, brachiopods, trilobites, ostracodes, chitinozoans, conodonts and graptolites. In total 734 species from 313 genera and 105 families have been identified. Table 1, which is based on data by Nestor et al. (in press), shows the distribution of species and genera by stages. It shows that the associations of the Porkuni and Juuru Regional Stages are the least diverse; and also that they have almost no common species, whereas about one third of the genera occur in both stages. At the Ordovician—Silurian boundary, besides intensive extinction of the Ordovi- cian fauna, the rate of the appearance of new fauna also rose. In Porkuni times extinction prevailed and Juuru times were characterized by the appearance of new faunas. Table 1 Numbers of species and genera of eight fossil groups recorded from the Vormsi to Raikila Regional Stages. Regional Stage Vormsi Pirgu Porkuni Juuru Raikkula Species, total number 195 252 154 a 221 transitional from the underlying 43 38 17 4 22 stage, % Genera, total number 150 175 125 109 130 transitional from the underlying 57 69 49 32 57 stage, % The dynamics of the fossil groups varied according to their ecology. For example, the shallowing of the basin in the Late Ordovician led to the radiation of the shallow-water stromatoporoids and corals, whereas the graptolites emigrated completely from the East Baltic area at the same time as the general crisis of graptolites noted by Rickards (1978) became apparent. Shallowing was also of great influence on the benthic trilobites and ostracodes, which usually inhabited deeper shelf areas and a remarkable decrease in their diversity took place in Pirgu and Porkuni times. The reverse tendency can be seen during the rapid deepening of the basin at the beginning of Juuru times; however, at that time shallow-water groups, particularly stromatoporoids and corals, were chiefly affected. Biostratigraphy and correlation Space does not allow a more detailed analysis here of the diverse biota from the boundary beds, and so only selected lists of species for each stage are presented, those which are most 90 KALJO, NESTOR & POLMA valuable for correlation (in brackets the index of the formation is shown where the species has been found). Vormsi Regional Stage Catenipora wrighti Klaamann (Kr), Plaesiomys solaris Buch (Kr), Kullervo complectens (Wiman) (Td), Acanthochitina barbata Eisenack (Td, Fj, Ml), Tretaspis seticornis (Hisinger) (Fj), Orthograptus quadrimucronatus (Hall) (Fj), Climacograptus styloides Lapworth (Fj), Hamarodus estonicus Viira (Fj), Belodina compressa (Branson & Mehl) (M1). The above species enable a clear determination of the position of the Stage at the level of the graptolite Pleurograptus linearis Zone. Pirgu Regional Stage In the lower part: Eospirigerina sulevi (Alichova) (Mo, Jn, Sv), Foramenella parkis (Neckaja) (Mo, Jn, Sv, Ad, Uk), Amorphognathus ordovicicus Branson & Mehl (Mo, Jl), Dicellograptus cf. complanatus Lapworth (Mo), Rectograptus gracilis (Roemer) (HI, Jn), Panderia megalophthalma (Linnarsson) (Jn), Tretaspis latilimba (Linnarsson) (Jn, Jl, K1). In the middle part: Clathrodictyon microundulatum Nestor (Ad), Catenipora tapaensis (Sokolov) (Ad), Esthonia asterisca Roemer (Ad, Uk), Maclurites neritoides (Eichwald) (Ad), Belodina compressa (Branson & Mehl) (Ad). In the topmost part: Conochitina taugourdeaui Eisenack (Kb), Climacograptus supernus Elles & Wood (Kb), Holorhynchus giganteus Kiaer (TC). The graptolites shown above enable a correlation of the stage with the zones of Dicello- graptus complanatus and D. anceps. Porkuni Regional Stage Paleofavosites rugosus Sokolov (Ar), Rhabdotetradium frutex Klaamann (Ar), Streptis undifera (Schmidt) (Ar), I/laenus angustifrons depressa Holm (Ar), Apatochilina falocata Sarv (Ar), Dalma- nella testudinaria (Dalman) (Kl), Hirnantia sagittifera (M‘Coy) (Kl), Eostropheodonta hirnan- tensis (M‘Coy) (KI), Dalmanitina (Mucronaspis) mucronata (Brongniart) (KI, Sl), Brongniartella platynota (Dalman) (K1), Pseudulrichia norvegica Henningsmoen (K1), Conochitina postrobusta subsp. A (Nolvak, Ms). The representatives of the Hirnantia and Dalmanitina communities enable correlation with the Hirnantian Stage at the level of the zones of Climacograptus extraordinarius and Glyp- tograptus persculptus. Juuru Regional Stage Clathrodictyon boreale Riabinin (Vr, Tm), Paleofavosites paulus Sokolov (Vr, Tm, Oh), Strick- landia lens prima Williams (Vr, lower pt), S. lens lens Williams (Vr, upper pt), Borealis borealis (Eichwald) (Tm), Calymene ansensis Mannil (Vr, Tm), Acernaspis estonica Mannil (Oh), Aitilia senecta Sarv (Vr), Steusloffina eris Neckaja (Vr, Tm, Oh), Ozarkodina ex gr. oldhamensis (Rexroad) (Oh, lower pt), Distomodus cf. kentuckyensis Branson & Branson (Oh), Ancyrochitina laevaensis Nestor (Oh, lower pt), Conochitina postrobusta Nestor (Oh), Dimorphograptus con- fertus (Nicholson) on upper pt), Pribylograptus incommodus Te etch (on top). The top of the Juuru Regional Stage is well defined by graptolites, suggesting that this level approximately coincides with the boundary of the Dimorphograptus confertus (equivalent to the Orthograptus vesiculosus) and Coronograptus cyphus Zones (Kaljo et al. 1984). The age of the lower limit of the stage can be established by Stricklandia lens prima (according to Cocks, 1971, it equates to the level of the Parakidograptus acuminatus Zone) and by the listed chitinozoans and conodonts, indicating that there was no substantial regional hiatus at the base of the Silurian in the East Baltic. However, distinct breaks occur at the margins of the basin, particu- larly to the southeast. The correlation of the Raikkiila Regional Stage is clearly defined by graptolites within the Coronograptus cyphus and Demirastrites convolutus Zones (Kaljo 1967; Kaljo 1970; Kaljo et al. 1984). Detailed correlations in Estonia were considerably improved by the study of chitin- ozoans (Nestor 1976). EAST BALTIC REGION 91 The present data from graptolites and other evidence permit only general correlation of the East Baltic section with the Dob’s Linn section, but finds of Climacograptus supernus at the top of the Pirgu and D. confertus at the top of the Juuru Regional Stage do not contradict the placing of the Ordovician-Silurian boundary (the base of the P. acuminatus Zone) at the top of the Porkuni Regional Stage. Correlation with the Anticosti section is possible by means of chitinozoans and conodonts. In this section (Achab 1981; McCracken & Barnes 1981) Member 5 of the Ellis Bay Formation is characterized by the presence of Conochitina taugourdeaui, C. micracantha and C. gama- chiana. J. Nolvak has found the first two and a form similar to the third species at the top of the Pirgu Regional Stage. At the base of Member 6 in Anticosti Ozarkodina oldhamensis appears, and somewhat higher Distomodus kentuckyensis and above bioherms Ancyrochitina spongiosa are recorded. P. Mannik, V. Nestor and V. Viira have found all these species or closely related forms in the lower part of the Juuru Regional Stage. Thus, in the Anticosti section we do not see equivalents of the Porkuni Regional Stage (at least of its upper part) which is characterized by Conochitina postrobusta subsp. A. References Achab, A. 1981. Biostratigraphie par les Chitinozoaires de POrdovicien Supérieur—Silurien Inférieur de I’Ile d’Anticosti. Résultats préliminaires. In P. J. Lespérance, (ed.), Field Meeting, Anticosti-Gaspe, Quebec, 1981 2 (Stratigraphy and paleontology): 143-157. Montreal (I{UGS Subcommission on Silurian Stratigraphy Ordovician-Silurian Boundary Working Group). Cocks, L. R. M. 1971. Facies relationships in the European Lower Silurian. Mem. Bur. Rech. geol. minier., Paris, 73: 223-227. Grigelis, A. A. (ed.) 1978. Decisions of the East Baltic regional stratigraphical conference (1976). 88 pp. and correlation charts. Leningrad, Interdep. Strat. Comm. USSR. [In Russian]. —— (ed.) 1982. Geology of the Soviet Baltic republics. 304 pp. Leningrad, Nedra [In Russian]. Hints, L. 1986. Genus Streptis (Triplesiidae, Brachiopoda) from the Ordovician and Silurian of Estonia. Proc. Acad. Sci. Estonian SSR, Tallinn, (Geology) 35: 20-26 [Eng]. summ. ]. Kaljo, D. 1967. On the age of lowermost Silurian of Estonia. Eesti NSV Tead. Akad. Toim., Yallinn, (Keem. Geol.) 16: 62-68 [Eng]. summ. ]. — (ed.) 1970. Silurian of Estonia. 343 pp. Tallinn, Valgus. [Engl. summ. ]. —— & Jiirgenson, E. 1977. Sedimentary facies of the East Baltic Silurian. In: Facies and fauna of the Baltic Silurian: 122-148. Tallinn, Acad. Sci. [Eng]. summ. ]. & Klaamann, E. (eds) 1982. Ecostratigraphy of the East Baltic Silurian. 112 pp. Tallinn, Valgus. ——, Nestor, H., Polma, L. & Einasto, R. 1988 (in press). Late Ordovician glaciation and its influence on the ecology in the Baltic cratonic basin. In: Essential biotic events in the earth history. Tallinn, Acad. Sci. [In Russian]. ——, Paskevitius, I. & Ulst, R. 1984. Graptolite zones of the East Baltic Silurian. In: Stratigraphy of the East Baltic Early Palaeozoic: 94-118. Tallinn, Acad. Sci. [Eng]. summ.]. McCracken, A. D. & Barnes, C. R. 1981. Conodont biostratigraphy across the Ordovician—Silurian boundary, Ellis Bay Formation, Anticosti Island, Québec. In P. J. Lespérance (ed.), Field Meeting, Anticosti-Gaspe, Quebec, 1981 2 (Stratigraphy and paleontology): 61-69. Montréal (IUGS Subcommis- sion on Silurian Stratigraphy Ordovician—Silurian Boundary Working Group). Mannil, R. 1966. Evolution of the Baltic basin during the Ordovician. 200 pp. Tallinn, Valgus. [Engl. summ. ]. Nestor, H., Klaamann, E., Meidla, T., Mannik, P., Mannil, R., Nestor, V., Nolvak, J., Rubel, M., Sarv, L. & Hints, L. 1988 (in press). Faunal dynamics in the East Baltic basin at the Ordovician and Silurian boundary. In: Essential biotic events in the earth history. Tallinn, Acad. Sci. [In Russian]. Nestor, V. 1976. A microplankton correlation of boring sections of the Raikkiila Stage, Estonia. Eesti NSV Tead. Akad. Toim., Tallinn, (Keem. Geol.) 25: 319-324 [In Russian with Eng]. summ. ]. Paskevitius, J. 1979. Biostratigraphy and graptolites of the Lithuanian Silurian. 268 pp. Vilnius, Mokslas. (Engl. summ. ]. Rickards, R. B. 1978. Major aspects of evolution in the graptolites. Acta palaeont. pol., Warsaw, 23: 585-594. Rubel, M. 1970. On the distribution of brachiopods in the lowermost Llandovery of Estonia. Eesti NSV Tead. Akad. Toim., Tallinn, (Keem. Geol.) 19: 69-79. Ulst, R., Gailite, L. & Jakoyleva, V. 1982. Ordovician of Latvia. 294 pp. Riga [In Russian]. The Ordovician—Silurian boundary in Poland L. Teller Zaklad Paleobiologii PAN, Newelska 6, Warsaw 01-447, Poland Synopsis Outcrops in the Holy Cross Mountains and Sudetes, as well as boreholes in the Polish lowlands, show Ordovician-Silurian boundary sediments to be variably developed or sometimes absent. The Hirnantia fauna is developed, but most other rocks are in graptolitic facies. Ordovician-Silurian boundary beds have been recognized in Poland both in outcrops and in boreholes. However, despite abundant documentation obtained from both types of sections as well as intensive investigations carried out, the boundary in Poland is still inadequately known. This is mainly because of the presence of many sedimentary gaps in the known sections, which are a result of the Taconic orogenic phase, and also because of the lack of good index fossils. In consequence, this boundary is not sharply defined in the Polish profiles, which makes good correlation with the adjacent regions difficult (Teller 1969). The Ordovician-Silurian boundary beds outcrop in Poland only in the Holy Cross Moun- tains and in the Sudetes. In the Bardo Range of the Sudetes (Teller 1962) there are no fossils known from near the junction, so the boundary has been arbitrarily designated by the presence of Lower Llandovery graptolites in black siliceous shale among the liddites The upper Ordovi- cian sediments appear to be represented in this area by alternating beds of sandstone and shale without fossils which underlie the Silurian liddites. The Ordovician—Silurian boundary has been put at the contact of these two formations, but it is not known for certain whether or not the clastic Ordovician corresponds to the uppermost Ashgill. In the Holy Cross Mountains, the boundary beds are known to occur in the Zalesie profile (Kielan 1956, 1957; Temple 1965), in the southern limb of the Bardo syncline in the Kielce region. The uppermost Ashgill silty beds contain a Hirnantia fauna with Mucronaspis mucro- nata Brongniart), M. olini Temple, Dalmanella testudinaria (Dalman), Hirnantia sagittifera (M‘Coy) and Eostropheodonta hirnantensis (M‘Coy) amongst others, and are covered by black shales with Akidograptus acuminatus at their base, accompanied by Climacograptus scalaris normalis and A. ascensus, indicating the acuminatus Zone. Thus the boundary separates the Upper Ashgill siltstone formation, containing a Hirnantia fauna, from the Lower Llandovery black shale formation with graptolites. This rapid change in facies suggests a lack of sedimentary continuity particularly since there are no graptolites in the uppermost Ashgill. In profiles in other parts of the world, the Hirnautia fauna (Cocks 1985) is generally older, or is to be found below the Ordovician Glyptograptus persculptus Biozone, the top of which is now taken as the boundary between the Ordovician and the Silurian. In many other sections in the Holy Cross Mountains (Tomezyk 1962; Bednarczyk 1973) a sedimentary gap is noted at this boundary. This gap embraces the entire Upper and partly the top of the Lower Ashgill as well as the lowermost Llandovery, and appears to be a result of the Taconic phase of orogeny. In the Polish Lowlands, the Ordovician-Silurian boundary beds show great facies variability (Modlinski 1973). In many boreholes, sedimentary gaps embrace various time spans and a change of facies toward a marly-arenaceous one is noted, which appears to indicate a gradual regression. Graptolites have only been found in the deeper parts of the platform slope clayey facies, including the Upper Ashgill Biozone of Glyptograptus persculptus and the Lower Llan- dovery A. acuminatus Zone, for example in the Lebork borehole (Tomezyk 1965). Bull. Br. Mus. nat. Hist. (Geol) 43: 93-94 Issued 28 April 1988 94 L. TELLER References Bednarezyk, W. 1971. Stratigraphy and paleogeography of the Ordovician in the Holy Cross Mountains. Acta geol. Pol., Warsaw, 21 (4): 573-616, pls 1—4. Cocks, L. R. M. 1985. The Ordovician—Silurian boundary. Episodes, Ottawa, 8: 98-100. Kielan, Z. 1956. Stratygrafia gornego ordowiku w Gorach Swieto krzyskich. Acta geol. Pol., Warsaw, 6: 253-272, pls 1-4. [Engl. summ. ]. — 1960. Upper Ordovician trilobites from Poland and some related forms from Bohemia and Scandin- avia. Palaeont. Pol., Warsaw, (for 1959) 11. 198 pp., 36 pls. Modlinski, Z. 1973. Stratigraphy and development of the Ordovician in North-Eastern Poland. Pr. Inst. geol., Warsaw, 72: 1—74, pls 1-5. Teller, L. 1962. Zagadnienie granicy Ordowik-Sylur w Gorach Bardzkich. In E. Passendorfer (ed.), Ksiega pamiatkowa ku czci prof. Jana Samsonowicza: 171-186. Warsaw. Akademia Nauk. [Engl. summ.]. 1969. The Silurian biostratigraphy of Poland based on graptolites. Acta geol. Pol., Warsaw, 19: 393-501. Temple, J. T. 1965. Upper Ordovician brachiopods from Poland and Britain. Acta palaeont. Pol., Warsaw, 10: 379-427, pls 1-21. Tomezyk, H. 1962. Problem stratygrafii ordowiku 1 syluru w Polsce w Swietle ostatnich badan. Pr. Inst. geol., Warsaw, 35: 1-134, pls 1-4. [Eng]. summ. ]. The Ordovician-Silurian boundary in the Prague Basin, Bohemia P. Storch Geological Survey, P.O. Box 85, Prague 011, 118 21 Czechoslovakia Synopsis The Ordovician-Silurian boundary in the Prague Basin is marked by an abrupt change in facies develop- ment and faunal assemblages, without significant breaks in purely marine sedimentation. Shallow marine sandstones and petromict conglomerates of the upper Kosov (Hirnantian) are followed by bioturbated mudstones due to the initial phase of a new transgression, with an abundant Hirnantia fauna in the uppermost Kosov. The mudstones are followed by dark graptolitic shales at the base of the Silurian (in the Prague Basin at the base of the Akidograptus ascensus Subzone). During the Parakidograptus acumin- atus Subzone another change of sedimentation appeared as a transition from silty-clay shales to sandy- micaceous laminites. This change corresponds to a local break in sedimentation in the north limb of the Prague Basin and in the Pankrac area, where the break continued to the Monoclimacis griestoniensis Zone. The sequence and the succession of faunal assemblages indicate an accelerated rate of transgression just below and above the Ordovician-Silurian boundary. Analysis of the faunal assemblages allows a detailed stratigraphical subdivision of the boundary beds in the Prague Basin and wide international correlation. Introduction In Bohemia, the Ordovician-Silurian boundary is well developed in the Prague Basin (Barrandian area). The Prague Basin is a tectonically predisposed linear sedimentary depression in which the sedimentation continued from the lowermost Ordovician up to the Middle Devo- nian without substantial interruptions (Havlicek 1981, 1982). In the Prague Basin, the Ordovician-Silurian boundary coincides with the boundary between the Kosov and Zelkovice Formations. Perner & Kodym (1919) supposed that there was a stratigraphical gap at the base of the Silurian in Bohemia caused by the emersion phase of the Taconic orogeny. Later, the lowermost Silurian graptolite zones, including the Parakidograptus acuminatus Zone, were documented in the Barrandian area by Marek (1951) and by Bouéek (1953) in an isolated outcrop near Béchovice. These authors denied the existance of the boundary gap east of Prague at Béchovice, but they admitted its presence in the rest of the Prague Basin. Horny (1956, 1960) found the earlier A. ascensus Zone along the whole southern limb of the Basin. He recorded that the rocks of the basal Silurian graptolite zones were only absent locally due to minor erosion caused by epeirogenetic movements that represented the aftermath of tectonic activity during the deposition of the Kosov Formation. Havliéek (1981, 1982) explained both the flysch-like Kosov Formation and the change in lithologic development at the Ordovician— Silurian boundary by invoking synsedimentary tectonic movements in the basin. More recently, basal Silurian graptolite zones have also been discovered in the northern limb of the Prague Basin and the boundary hiatus was verified only in a restricted part of the basin (Storch 1982, 1986). Investigation of the early Kosov (Storch & Mergl, in press) has shown the sequence in Bohemia to be very similar to that explained by glacio-eustatic environmental changes (Brenchley & Cocks 1982; Brenchley & Cullen 1984; Brenchley & Newall 1984). The glacio-eustatic conception of the late Ordovician to early Silurian facies and faunal changes (Brenchley 1984; Brenchley & Newall 1984) also appears to explain the Ordovician-Silurian boundary sequence in the Prague Basin. Sequence of the latest Ordovician Considerable changes preceding the Ordovician—Silurian boundary event were recorded at the top of the Kraliv Dviur Series in the Prague Basin (Storch & Mergl, in press). The deep water Bull. Br. Mus. nat. Hist. (Geol) 43: 95-100 Issued 28 April 1988 96 p. STORCH mudstones of the Kraluv Dvur Formation, with deep water faunal assemblages, were followed by coarse grained subgraywackes and silty shales at the base of the Kosov Formation. The high-diversity Proboscisambon Community of the uppermost Kraliv Dvur Formation was replaced by the low-diversity and short-lived Mucronaspis Community (Storch & Mergl, in press), the last record of which (bivalves and trilobite fragments) occurs in the shale of the lowermost Kosov Formation. The basal Kosov subgraywackes and shales were succeeded by flysch-like sediments which form most of the thickness of the Kosov Formation. This regressive sequence culminated in the deposition of shallow-water sandstones and petromict conglomerates in the upper part of the formation. In the uppermost sandstone layers a monotonous assemblage of infaunal bivalves provides evidence of intertidal environments (Havlicek 1982). In the uppermost part of the Kosov Formation, the quartz sandstones with shaly intercalations are replaced by siltstones and mudstones. Pale grey, often bioturbated calcareous mudstones and claystones containing a rich Hirnantia sagittifera Community occur near the top of the formation. The Hirnantia fauna, interpreted by Havlicek (1982) as representing a subtidal environment, has been found only in the eastern part of the Prague Basin. A gradual deepening of the sea seems likely in the uppermost Kosov (Hirnantian) of the Prague Basin. The cosmopolitan Hirnantia fauna found in the uppermost part of the Kosov Formation permits a broad international correlation. In the Prague Basin it was first recorded at Bécho- vice near Prague (Marek 1963; Marek & Havlicek 1967). Later, it was found at Nova Ves, Pankrac, Repy and Reporyje (all within the Prague area) and near Tachlovice. All the fossil- iferous localities yielded faunal associations of similar taxonomic composition, but without the depth-controlled variations of the associations reported by Brenchley & Cocks (1982) and Brenchley & Cullen (1984) from the Oslo region, Norway. Lists of the Hirnantia faunas from Bohemia were published by Havliéek (1982) and Storch (1986). The graptolite Glyptograptus bohemicus (Marek) accompanies the Hirnantia sagittifera Community in Bohemia and supports the international biostratigraphic correlation of the sequence. The layer containing the Hirnan- tia fauna is separated from the first graptolitic shales by at least 0-3m thickness of mudstone, often heavily bioturbated, with frequent limonite impregnations originating from pyrite weathering (Storch 1986). Ordovician-Silurian boundary and lowermost Silurian sequence In general, sedimentation is continuous through the Ordovician—Silurian boundary in the Prague Basin, in spite of some differences between the separate sections. By using distinctive features of the boundary sequence and also the basal Silurian lithologies, the Prague Basin may be formally subdivided into five areas (Storch 1986). The quietest sedimentation, in probably the deepest parts of the basin, is limited to the sections along the whole south limb of the basin (South limb area—Zelkovice, Vseradice, Béleé, Votkov, Zadni Treban, Hlasna Tiebani, Karlik, Cernosice and Velka Chuchle). A complete succession starting with the Akidograptus ascensus Subzone has been preserved in all the localities (exemplified by the Karlik section, Fig. 1). Clayey shales with climaco- graptids and rare glyptograptids were recorded even below the first occurrences of A. ascensus at Zelkovice and Votkov and could represent the upper part of the Glyptograptus persculptus Zone. The ascensus Subzone is represented by clayey shales with subsidiary variable siltstones. Sandy-micaceous laminites start within the Parakidograptus acuminatus Subzone. The laminites disappear in the western part of the south limb at Zelkovice and VSeradice in the Cystograptus vesiculosus Zone, and towards the east in the Coronograptus cyphus Zone, and sometimes they even reach up to the Demirastrites triangulatus Zone (Storch 1986). In the same way, the onset horizon of siliceous shales migrates in the south limb on the vesiculosus Zone at Zelkovice to the Demirastrites pribyli Zone at Cernogice and Velka Chuchle. The Repy and Béchovice sections differ in having more rapid sedimentation, giving the greatest thicknesses of graptolite zones (Repy section, Fig. 1) in this part of the Prague Basin. The layer referred to the per- | SILURIAN ORDOVICIAN Fig. 1 ORDOVICIAN-SILURIAN BOUNDARY IN BOHEMIA 3m Fy Oo Pale green clayst. (bioturbated) v rs) > °o black siliceous shales 2 5 oO mw fe) o 2 dark grey laminites fo} > iS 1 lu o S Nagel dark grey or black graptolitic shales iS ai ¥ = re v4 oO fo} o 2 cS) clayey breccia o 0 a Cc 7) i= ie) 2) = a = & c = pale grey mudst. (often bioturbated) g 3 = g 3 5 > graptolites 5 } ® é Hirnantia fauna WwW 5 =) = uw oe © 5 = oS = = ap a} D : & in Ww 5 Oo Woe ©) je > 0 oe 2 = =: ap 6G) # ae oO >t) 5 ¥ Binieas at wat (= & | @ £ x [es 2 ics = = ie ads Ss a SS sg | f= & \ io) Go FE “ 2 ) = ae St | : 2 £ = me = Eu QO = uw\ c ° oe Oo 5 a ~ to} 2 = 9 S mS NE BS = © = ° = 2 ire =6 = Pe cy > 2 rT) g > > £ = >N = > o = te @ Oo ® 2g >N = & J3 Mes fe) iS 3) = u = & 3S 0 Z —_— ara S a 4 29 Jc LE 5 = 8 ie} ?|persc. 1 bohemicus Kosov Fm. Kosov Fm. Kosov Fm. Kosov Fm Kosov Fm x . Oo BECHOVICE REPY, Zz Ox PANKRAC PRAHA fe) TACHLOVICE LODENICE| lower Silurian outcrop area section KARLIK , 7h aie . Hirnantia fauna 1% HLASNA TREBAN OZADNI TREBAN BELEC (0) 10 km ZELKOVICE Lithology, stratigraphy and faunal distribution in selected Ordovician—Silurian boundary sections of the Prague Basin; location of the sections. 97 98 Pp. STORCH sculptus Zone is also developed there. Laminites appear in the acuminatus Zone and pass up into the vesiculosus Zone. Detailed studies of both biostratigraphy and lithostratigraphy (Storch 1986) revealed that the laminites represent more condensed sedimentation than the clayey and silty shales. The onset of laminite deposition in the southern limb of the basin appears to have been synchronous with the start of the break in sedimentation in the acuminatus Zone in the north limb of the basin at Sedlec and Lodénice. The longest break in sedimentation is known from the Mala Chuchle, Pankrac, Nova Ves and Tachlovice sections (Pankrac area). The topmost Ordovician mudstones are followed there by graptolitic shales of the Litohlavy Formation, with upper Llandovery graptolites of the Monoclimacis griestoniensis Zone. In this case, reworking possibly took place of previously deposited, incoherent, clayey and muddy sediments of the basal Silurian (ascensus Subzone, the lower part of the acuminatus Subzone), and perhaps also of the topmost Ordovician (several tens of centimetres in thickness). Near Stodtlky and Reporyje (Reporyje section, Fig. 1), this break in sedimentation splits into two shorter gaps. The earlier of them starts above the ascensus Subzone and thus supports the explanation of the break presented in different parts of the Prague Basin. Sedimentation and assumed bathymetric changes The Kosov Formation, which is about 100m thick, shows sedimentation which was presum- ably controlled by glacio-eustatic regression. The subsequent transgression started in the uppermost Kosov and strongly accelerated at the base of the Silurian (Brenchley & Newall 1984). Considerable transgression is also documented by a decrease of the rate of sedimentation at the Ordovician—Silurian boundary. In the Prague Basin, the rate of sedimentation in the lowermost Silurian was approximately calculated (Storch 1986) to range between 1 m and 7-5m per 10° years in contrast to nearly 100m per 10° years during the Kosov Series (Hirnantian). During the acuminatus Zone the transgression caused a further deepening of the Prague Basin and was probably the origin of a fairly intensive bottom current in the deeper central part of the linear depression of the Prague Basin. This current is considered to have caused local breaks in sedimentation, in places perhaps accompanied by mild subaquatic erosion (Storch 1986). In the sites where this current had less erosive power, condensed sedimentation of laminites occurred, and in the quietest parts of the basin floor there were deposited siliceous shales and silty silicites ((phtanites’) which first appeared in the vesiculosus Zone. Stratigraphy The Hirnantia fauna occurs in the upper part of the Kosov Series well above the disappearance of the Mucronaspis Community in the basal part of the Series. The Hirnantia fauna, which is accompanied by Glyptograptus bohemicus, can be referred to the upper Hirnantian, namely to the upper part of the Climacograptus extraordinarius Zone or the lower part of the persculptus Zone. In Bohemia, the base of the Silurian System coincides with the base of the ascensus Subzone, which is defined by the first appearance of Akidograptus ascensus Davies (usually accompanied by Diplograptus modestus Lapworth). When compared with the British Isles, the base of the subzone in Bohemia is comparable to the base of the acuminatus Zone at the type section Dob’s Linn (Williams 1983). In the Prague Basin, the base of the ascensus Subzone mostly corre- sponds to a sudden change in both the colour and the composition of the sediments, in which the pale grey bioturbated mudstones are replaced by dark grey clayey graptolitic shales. However, a low-diversity climacograptid—glyptograptid assemblage has been recorded from several localities at the base of the graptolitic shales just below the ascensus Subzone, which is separated by an unfossiliferous bioturbated mudstone from the bohemicus Zone beneath. The first assemblage of graptolitic shales below the ascensus Subzone is referred to the upper part of SILURIAN CHRONO- STRATIGRAPHY LITHOSTRATIGRAPHY BIOSTRATIGRAPHY Llandovery Rhuddanian Aeronian Zelkovice Fm. ? persculptus — ascensus acuminatus vesiculosus triangulatus Ww) >} S E o fe fo} a cyphus Glyptograptus bohemicus Marek Climacograptus aff. miserabilis Elles & Wood Climacograptus normalis Lapworth Glyptograptus sp. (ex gr. persculptus) Glyptograptus sp. (aff. avitus) Diplograptus modestus Lapworth Akidograptus ascensus Davies Diplograptus elongatus Churkin & Carter Diplograptus aff. parvulus (Lapworth) Diplograptus parajanus Storch Cystograptus ancestralis Storch Climacograptus aff. premedius Waern Climacograptus trifilis Manck Parakidograptus acuminatus (Nicholson) Climacograptus longifilis Manck Diplograptus diminutus apographon Storch Cystograptus vesiculosus (Nicholson) Climacograptus aff. rectangularis McCoy Glyptograptus ex gr. tamariscus Atavograptus atavus (Jones) Dimorphograptus confertus (Nicholson) Orthograptus obuti Rickards & Koren Rhaphidograptus toernquisti (Elles s Wood) Lagarograptus aff. acinaces (Térnquist) Pribylograptus argutus (Lapworth) Monograptus austerus austerus Tornquist Monograptus cf. sudburiae Hutt Limpidograptus cf. posohovae Chaletzkaja Diplograptus cf. thuringiacus Eisel Diplograptus fezzanensis Desio Coronograptus cyphus cyphus (Lapworth) Orthograptus cyperoides (Tornquist) Monograptus austerus vulgaris Hutt Monograptus difformis Tornquist Petalograptus ovatoelongatus (Kurck) Rastrites longispinus Perner Demirastrites triangulatus (Harkness) Coronograptus gregarius gregarius (Lapworth) Fig. 2. Chronostratigraphy, lithostratigraphy, biostratigraphy and graptolite species ranges through the Ordovician-Silurian boundary interval in the Prague Basin. 100 p. STORCH the persculptus Zone, in spite of the fact that true Glyptograptus persculptus has not yet been found there. The ranges of graptolites up to the base of the triangulatus Zone are shown in Fig. 2. The rich graptolite assemblages of the Prague Basin were briefly described by Boucek (1953), and more recently they have been described by Storch (1986). Acknowledgements I would like to thank V. Havli¢éek and J. Kfiz for critically reading the manuscript. References Boucéek, B. 1953. Biostratigrafie, vyvoj a korrelace zelkovickych a motolskych vrstev ¢eského siluru. (Biostratigraphy, Development and Correlation of the Zelkovice and Motol Beds of the Silurian of Bohemia). Sb. ustred. Ust. geol., Prague, 20: 421-484. Brenchley, P. J. 1984. Late Ordovician Extinctions and their Relationship to the Gondwana Glaciation. In P. J. Brenchley (ed.), Fossils and Climate: 291-315. London. & Cocks, L. R. M. 1982. Ecological associations in a regressive sequence: the latest Ordovician of the Oslo—Asker District, Norway. Palaeontology, London, 25: 783-815, pls 85-86. —— & Cullen, B. 1984. The environmental distribution of associations belonging to the Hirnantia fauna—evidence from North Wales and Norway. In D. L. Bruton (ed.), Aspects of the Ordovician System: 113-125. Universitetsforlaget, Oslo. (Pal. Contr. Univ. Oslo 295). —— & Newall, G. 1984. Late Ordovician environmental changes and their effect on faunas. In D. L. Bruton (ed.), Aspects of the Ordovician System: 65—79. Universitatsforlaget, Oslo. (Pal. Contr. Univ. Oslo 295). Havlicek, V. 1981. Development of a linear sedimentary depression exemplified by the Prague Basin (Ordovician—Middle Devonian; Barrandian area—central Bohemia). Sb. geol. Véd., Prague, (Geol.) 35: 7-48. —— 1982. Ordovician in Bohemia: development of the Prague Basin and its benthic communities. Sb. geol. Véd., Prague, (Geol.) 37: 103-136. Horny, R. 1956. Zona Akidograptus ascensus v jiznim kfidle barrandienského siluru. Vést. ustred. Ust. geol., Prague, 31: 62-69. 1960. Stratigrafie a taktonika zapadnich uzavéru silurodevonskeho synklinoria v Barrandienu. Sb. ustred. Ust. geol., Prague, 26: 495-530. Marek, L. 1951. The find of Akidograptus acuminatus (Nicholson) in the Silurian of Bohemia. Vést. ustred. Ust. geol., Prague, 24: 382-384. 1963. Zprava o vyzkumu fauny vrstev kosovskych ¢eskeho ordoviku. Zpravy o geol. vyzkumech 1962: 103-104. —— & Havli¢ek, V. 1967. The articulate brachiopods of the Kosov Formation (Upper Ashgillian). Vést. ustred. Ust. geol., Prague, 42 (4): 275-284, pls 1-4. Perner, J. & Kodym, O. 1919. O rozéleneni svrchniho siluru v Cechach. Cas. Mus. Kral. éesk., Prague, 93: 6-24. Storch, P. 1982. Ordovician-Silurian boundary in the northernmost part of the Prague Basin (Barrandian, Bohemia). Vést. ustred. Ust. geol., Prague, 57 (4): 231-236. — 1986. Ordovician-Silurian boundary in the Prague Basin (Barrandian area, Bohemia). Sb. geol. Véd., Prague, (Geol.) 41: 69-99, 8 pls. —— & Mergl, M. (in press). Kralovdvor-—Kosov boundary and the late Ordovician environmental changes in the Prague Basin {Barrandian area, Bohemia). Sb. geol. Véd., Prague, (Geol.) 44. Williams, S. H. 1983. The Ordovician—Silurian boundary graptolite fauna of Dob’s Linn, southern Scot- land. Palaeontology, London, 26: 605-639. The Ordovician—Silurian boundary in the Saxothuringian Zone of the Variscan Orogen H. Jaeger Museum fur Naturkunde, Palaontologisches Museum, Invalidenstrasse 43, 104 Berlin, DDR. Synopsis In the Saxothuringian Zone of the Variscan Orogen in Thuringia, Saxonia and north Bavaria the poorly fossiliferous, thick arenaceous-argillaceous Ordovician rocks are abruptly but conformably succeeded by the very condensed sequence of Silurian—Early Devonian graptolitic alum shales and lydites beginning in both major facies with the Zone of Akidograptus ascensus. Below it, shaly interbeds in the uppermost Ordovician Dobra Sandstone yielded chiefly non-zonal graptolites, and in one section Diplograptus bohe- micus about | m below the lithological boundary. Introduction The Saxothuringian and Lugian (= West Sudetic) Zones form the middle of the three major depositional and tectonic belts of the Variscan Orogen in central Europe. They constitute the metamorphic zones that are situated between the internal Moldanubian Zone (internids) and the external Rhenohercynian Zone (externids). The latter is exemplified by the Rheinisches Schiefergebirge and the Harz Mountains, in both of which the nature of the Ordovician— Silurian junction is unknown. In this paper only the type area of the Saxothuringian Zone is considered; it lies west of the River Elbe in Saxonia, Thuringia, north Bavaria and north Bohemia. Together with the Lugian Zone (situated east of the Elbe), it forms the northern part of the Bohemian Massif and is the largest outcropping fragment of the broken Variscan orogen in central Europe. In a wider palaeogeographical and geotectonical context, the Saxothuringian—Lugian Zones are part of the Mediterranean province, and of the Palaeotethys geosyncline and sea, that is the Tethys of the early and middle Palaeozoic. In the whole of the Palaeotethys area, the Ordovician-Silurian transition is marked by a drastic change in the depositional regime. In the Saxothuringian Zone the typically 2000m thick Ordovician, consisting of poorly fossiliferous, arenaceous-argillaceous rocks with some sedimentary iron ore bodies, is rapidly replaced by 50m thick Silurian, which is made up almost entirely of interbedded euxinic lydites and alum shales rich in graptolites. From the middle Ludlow to the Pridoli, the graptolitic shales are interrupted by a peculiar limestone (Ockerkalk) or grey-green clay shales, both of which are poorly aerated deposits. Sedimentation of the alum shales, and regionally also of the lydites, recurred in the uppermost Silurian, and lasted well into the Lower Devonian (Lochkov). The Silurian (and Devonian) graptolitic shales of the Thuringian type, that is alum shales and black lydites, contain large quantities of pyrite, phosphorite (in nodules and layers) and carbon (in beds, laminae and lenses). These rocks cover vast areas in the deeper parts of the Palaeotethys sea between Thuringia and north Africa. They are the result of one of the largest oceanic anoxic events in the history of the earth, both areally and temporally. Thuringian and Bavarian Facies In the geosynclinal Palaeozoic of the Saxothuringian Zone two major facies (or rather series of facies—Faziesreihen) are distinguished, at least in the rock-sequences from the Ordovician to the Lower Carboniferous. These are known as the ‘Thuringian’ and ‘Bavarian Facies’, but it is beyond the scope of this paper to outline their features in detail. The following points may however be made. Bull. Br. Mus. nat. Hist. (Geol) 43: 101-106 Issued 28 April 1988 ‘yorqivis = 19 ‘sioquoyuel reou yorq[ynullagO Ul SiJoqINfIg = Ig ‘Usqns[usyOH = H PIUOXeS Ul aBgasyNUPIH = H jlopusqiq = 1g ‘odIIgaB1ojaryos[eiq(y = Sql ‘gS1IqoszIq = q (URLINIIg = Yor]q) URIUOAD 0} UBIIqUIR) *¢ ‘pasoydiourejour A[YeVoM 0} posoydiowejowuN ‘d10Z019}01g ‘fp ‘(es11gesueing = Os) Kilo oimog pur (4) sioquoyuely (AA) S[OJUSPTEAA (IA) Stoqyounyy Jo (sureyunoy 1xIMjOg) QBIIQIBUdYISIMZ, ULOSIIVA 9Y} JO ([euOzZR}ey AyyeordA}) syoos our[eishiD “¢ ‘adv UvOSTIBA 07 UBLIQUIBOAI, JO (soquads pur sayoIpouRss ‘soyuRs3) suoynyd 10feyy “7 ‘a8e UPOSIIVA O} UBLIQUIBIOIg jo (saytf[Ayd ‘systyos ‘sassiou3) syoo1 pesoydiowvjaur A[Su0NS “| :pussoT Jisse] URTMOYOY 94} Jo JfBY Woy yIOU SY} JO YO}9¥S jeoisojoan | “sly Jezel) I WEES i. WO 0 GiE:=s] VQ c& H. JAEGER PJesIODS mY Y/ 4 \y P, \ + FS E Y¥d9 102 ORDOVICIAN-SILURIAN BOUNDARY IN VARISCAN OROGEN 103 The Thuringian Facies represents a monotonous basin facies that exhibits only moderate lateral changes, if any. By contrast, the Bavarian Facies is complex. In the simplest model (Jaeger 1977: text-fig. 3) its site is depicted as a swell flanked on either side by deep furrows (deeper than the Thuringian basin). The central swell of the Bavarian Facies region is charac- terized by intermittent carbonate sedimentation that lasted demonstrably from the Silurian into the Carboniferous. On the swell the nature of the Ordovician-Silurian boundary is unknown. In most or all of the Saxothuringian Zone the swell-limestones are known from allochthonous blocks (olistholites) or even only from boulders, for example, the Middle to Upper Devonian stromatoporoid-coral reef-limestones at Frankenberg. The flanking depressions received non- carbonate sediments throughout their history. Typical of this Bavarian basin facies is the continuous sequence of cherts, siliceous shales and clay shales (Kieselschiefer-Fazies) spanning the long interval from the base of the Silurian to the top of the Devonian. In the Silurian interbedded graptolitic black lydites and alum shales are the typical rocks, as in the Thuringian Facies, whereas throughout most of the Devonian conodont-bearing brighter grey-green and even red cherts, siliceous shales and clay shales occur. The region of the Bavarian Facies was, at least in its Bavarian type area, the site of large- scale basic vulcanism which lasted intermittently from the earliest Ordovician to the Carbon- iferous, whereas in the Thuringian Facies the geosynclinal basic vulcanism was virtually confined to a brief phase of violent eruptions and intrusions at the beginning of the Upper Devonian. Rocks of the Thuringian Facies cover large areas in the Saxothuringian Zone. Minor occurrences are known from the southern margin of the Lugian Zone in Czechoslovakia. The Bavarian Facies rocks form a discontinuous belt that runs along the strike near the middle of the Saxothuringian Zone. They are confined to narrow strips (at the most several kilometres broad) on either side of the so-called Zwischengebirge (Betwixt Mountains) of Miinchberg, Wildenfels and Frankenberg. East of the Elbe, the Bavarian Facies reappears at the Eichberg near Weissig immediately north of the plutons that build up the area between Dresden and Gorlitz. From the Eichberg the Bavarian Facies can be traced through all of the Lugian Zone as far as the southern end of the Sowie Gory (Eulen-Gneis), where it is particularly well developed. Outside its main belt, the Bavarian Facies is typified by the Palaeozoic of the Elbtalschiefergebirge southeast of Dresden. The palaeogeography of the area of the Bavarian Facies may be envisaged as an island arc (the use of which term does not necessarily denote the implications of the theory of plate tectonics). Ordovician-Silurian Boundary At the Ordovician-Silurian boundary the distinctness of the two contrasting regional facies is particularly pronounced. In the Thuringian Facies the uppermost Ordovician is represented by the peculiar Lederschiefer, a monotonous, almost black, buff-weathering, non-bedded silty shale with high content of mica. Predominantly arenaceous rock-detritus and isolated sandstone boulders up to 30cm across (some attaining even several metres) occur in varying quantities throughout the 250m thick formation, for which it is noted. Whether the boulders represent glacial drop-stones or whether they originated from slumping are much debated questions. While the matrix of the Lederschiefer is barren, many boulders contain brachiopods, bryo- zoans, various trilobites and echinoderms, particularly loose cystoids. Most of these exotic fossils await modern expert study. Strata that compare closely lithologically with the Leder- schiefer are of wide distribution in the Mediterranean province, for example in the Orea Shale in Spain. In the uppermost two to three metres of many Thuringian sections it can be seen that the sand grains and mica flakes disappear, while many pyrite nodules appear in the shales, herald- ing the change to the otherwise abrupt transition to the Silurian euxinic graptolitic rocks. By contrast, the occurrence of sandstone beds in the uppermost Lederschiefer has been reported (Troeger 1959, 1960; Freyer 1959) from eastern sections (near Oelsnitz) that lie near the Bavar- ian Facies belt. 104 H. JAEGER In view of the intense folding, sections that exhibit a tectonically undisturbed transition from the Ordovician to the Silurian are hardly to be expected between rocks with such different mechanical properties as the Lederschiefer (below) and the lydites/alum shales (above). Never- theless, a century ago Akidograptus acuminatus was recovered from the basal graptolite shales at Ronneburg and Oelsnitz by Eisel. Recently Alder (1963) and Schauer (1971) found 4A. ascensus in the basal 4m of interbedded alum shales and lydites below the acuminatus fauna at the Weinberg near Hohenleuben in what would appear to be the most intact boundary sec- tions. The zone fossil is associated with Diplograptus modestus and several forms of Cli- macograptus (C. medius, C. rectangularis, C. scalaris normalis and C. miserabilis); there also occur unnamed climacograptids that have branched virgellae or virgellae with a distal vesicular appendage (Schauer 1971). In the succeeding half metre, Akidograptus acuminatus occurs together with all the species that are already present in the ascensus Zone, but in addition, the highly characteristic Cli- macograptus trifilis Manck and C. longifilis Manck make their first appearance. In the Bavarian basin facies the uppermost Ordovician is represented by the Dobra Sand- stone. This is an almost black, fine-grained, often quartzitic sandstone with subordinate shaly interbeds, with a maximum thickness in excess of 40m. Some sandstone beds exhibit magnificently-developed sole markings (load casts), others roll- and ball-structures. Greiling (1966: 12) interprets the Dobra Sandstone essentially as a turbidite. This peculiar rock is a characteristic formation of the Bavarian Facies, and is of wide distribution. It can be traced intermittently throughout the Saxothuringian and Lugian Zones for a total length of 400km and it has a far greater linear extent in central Europe than the coeval Lederschiefer. Lithologically virtually identical (Carnic Alps) or dissimilar (Kosov Quartzite in the Barrandian) sandstones occur in the same or analogous stratigraphical position in many areas of the Mediterranean province. In some regions they may range considerably higher, through much of the Llandovery, and not start until the base of the Silurian. The Dobra Sandstone is practically unfossiliferous, except for the uppermost two metres which yielded graptolites in shaly interbeds. Stein (1965: 119; text-figs 5, 20 and others) described Climacograptus medius, C. scalaris normalis, Diplograptus modestus, and a single thabdosome of D. cf. persculptus (Salter) from 1:90 m below its top at Dobra. At the Silurberg locality in Obermthlbach near Frankenberg Diplograptus bohemicus (Marek) was described by Jaeger (1977) from the uppermost Dobra Sandstone. This species occurs there abundantly, but to the exclusion of other graptolites, in a layer just a few mm thick in the middle of a 0-70-0-75m thick bed of homogeneous grey-black clay shale that underlies a prominent 30cm thick quartzite. The latter is overlain by 4m of platy sandstone and shale showing slickensiding, which is succeeded by 1m of broken and mylonitized alum shales and lydites indicating a major fault that throws Ludlovian (colonus and chimaera Zone) graptolite shales against the Ordovician Dobra Sandstone. The same sequence, particularly the 0:70-0:75 m thick bed of shale and the overlying compact 30cm sandstone bed, have been traced to the northeast as far as Starbach. This sequence is therefore shown as the typical one in Fig. 2 (right column). In the apparently undisturbed boundary section at Starbach the 30cm thick compact sandstone bed is immediately overlain by 40cm of weathered clay shales and siliceous shales, which in turn are succeeded by typical alum shales and lydites. Graptolites were not found in the Dobra Sandstone at other localities, nor was the occurrence of the basal Silurian graptolite zones established in this northeastern part of the Saxothuringian Zone. The basal Silurian graptolite zones were recovered in the lowermost alum shales and lydites of the type area of the Bavarian Facies along the northwest side of the Mtinchberg gneis at Dobra, Fortschenbach, Ober-Brumberg and Rauhenberg (Greiling 1957, 1966; Stein 1965). Though these workers did not formally distinguish between the Zones of A. ascensus and acuminatus 1t would appear evident from Stein’s precise documentation that the two can be differentiated. The thicknesses are approximately the same as in the Thuringian Facies, or slightly less. The associations are also the same, though the number of listed forms is somewhat smaller. Climacograptus trifilis and C. longifilis occur as frequently as in the Thuringian Facies. ORDOVICIAN-SILURIAN BOUNDARY IN VARISCAN OROGEN 105 Thuringian Basin Facies Bavarian Basin Facies Fs Upper Graptolitic === Upper Graptolitic Shales 15m —— Shales 10m = M.hercynicus - =e | Mieneymicus— _—— M.transgrediens |a=s—ae M.uniformis alee] SS Ockerkalk == /Grey-Green Shales 10 -20m — mn M.chimaera F M.chimaera 35/34 — Zam Solee == — —— Lower == Lower == Graptolitic Shales ame Craptolitic Shales ——= 35-40m M.gregarius 19 mee 8=M.gregarius s=S== Mees 18 =—= M.cyphus == peeom === S©vesiculosus == C.vesiculosus os 4 ——— 04-13m 17 —— nd) Pee A.acuminatus See § A.acuminatus et O04 = 05 m ; 6 a 02 = 05 mM ——— A.ascensus == A.ascensus 16 a 04-0,5m Sateen 0Q2-05m .=.=-. | sandstone & shale : 5m Lederschiefer pce | quartzite bed Q3m 250m = grey shale 07m Uppermost =s2a02 || Dipl. bohemicus Ordovician =~ | Débra Sandstone >40m et =? (SSI Fig. 2 Composite sections across the Ordovician-Silurian boundary in the Thuringian (left) and Bavarian basin facies (right). Legend: 1. Lydites (black layered cherts). 2. Alum shales. 3. Grey to green argillaceous shales. 4. Homogeneous non-bedded silty shales. 5. Arenaceous rocks. 6. Lime- stones. 7. Phosphoritic nodules. 106 H. JAEGER Two points of general interest may be made. Firstly, in the Saxothuringian Zone, the change from the Ordovician Lederschiefer and Dobra Sandstone, respectively, to the Silurian grapto- litic rocks takes place at the base of the Zone of A. ascensus and above beds with D. bohemicus which have only been found in one section of the Bavarian Facies. Secondly, in the Saxo- thuringian region, A. ascensus and A. acuminatus indicate two successive graptolite zones, as in the Barrandian area and southern Spain (Jaeger & Robardet 1979: 693, section 4), although 4. ascensus ranges into the acuminatus Zone, and in Sardinia even into the next higher Zone of Cystograptus vesiculosus (Jaeger 1976: pl. 3, fig. 7). References Alder, F. (1963). Biostratigraphie und Taxionomie der Graptolithen des Weinberges bei Hohenleuben. 95 pp., pls 1-47, text-figs 1-19. Diplomarbeit, Bergakademie Freiberg (unpublished). Freyer, G. 1959. Die Ausbildung der Grenze Ordovicium/Silur im Bereich der Vogtlandischen Haupt- mulde. Beitr. Geol., Berlin, 1: S—12, 2 text-figs. Greiling, L. 1957. Das Gotlandium des Frankenwaldes (Bayerische Entwicklung). Geol. Jb., Hannover, 73: 301-356. —— 1966. Sedimentation und Tektonik im Palaozoikum des Frankenwaldes. Erlanger geol. Abh., 63: 1-60, pls 1-2. Jaeger, H. 1976. Das Silur und Unterdevon vom thtringischen Typ in Sardinien und seine regional- geologische Bedeutung. Nova Acta Leopoldina, Halle a.S., 45 (224): 263-299, pls 1-3. — 1977. Das Silur/Lochkovy-Profil im Frankenberger Zwischengebirge (Sachsen). Freiberger ForschHft., Berlin, (C) 326: 45—S9, pl. 1. — & Robardet, M. 1979. Le Silurien et le Dévonien basal dans le Nord de la Province de Seville (Espagne). Géobios, Lyon, 12: 687-714, pls 1-2. Schauer, M. 1971. Biostratigraphie und Taxionomie der Graptolithen des tieferen Silurs unter besonderer Berticksichtigung der tektonischen Deformation. Freiberger ForschHft., Berlin, (C) 273: 1-185, pls 1-45. Stein, V. 1965. Stratigraphische und palaontologische Untersuchungen im Silur des Frankenwaldes. N. Jb. Geol. Palaont. Abh., Stuttgart, 121: 111—200, pls 1-2. Troeger, K. A. 1959. Kaledonische und frtihvariscische Phasen im Vogtland und den angrenzenden Gebieten. Freiberger ForschHft., Berlin, (C) 73: 1-152. 1960. Das untere Silur im Vogtland. In J. Svoboda (ed.), Prager Arbeitstagung uber die Stratigraphie des Silurs und des Devons (1958): 315-325, text-figs 1-2. Prague. Wiefel, H. 1974. Ordovizium. In W. Hoppe & G. Seidel (eds), Geologie von Thiiringen: 165-194. Gotha/ Leipzig. Zitzmann, A. 1966. Neue Conodontenfunde in der devonischen Kieselschiefer-Serie der bayerischen Fazies des Frankenwaldes. Geol. Bl. Nordost-Bayern 16 (1): 1-39. —— 1968. Das Palaozoikum im Grenzbereich zwischen Bayerischer und Thtringischer Faziesreihe des Frankenwaldes. Geol. Jb., Hannover, 86: 579-654, pls 1-3. The Ordovician—Silurian boundary in the Carnic Alps of Austria H. P. Schonlaub Geologische Bundesanstalt, PO Box 154, Rasumofskygasse 23, A 1031 Vienna, Austria Synopsis Although the Ordovician-Silurian boundary is represented in some places by a considerable uncon- formity in the Carnic Alps, in other sections a Hirnantia fauna in the Plocken Formation and possibly persculptus Zone graptolites are succeeded by the Bischofalm facies which in places has yielded graptolites of the acuminatus Zone. The shallow-water facies and unconformities at and near the boundary were partly caused by the global eustatic fall and rise in sea level and partly by Caledonian tectonic activity. Introduction The long geological history of the Carnic Alps of Austria and northern Italy lasts from the late Ordovician to middle Triassic times. For many years in this region several sections which cross the Ordovician-Silurian boundary and represent different environmental settings have been well known. Based on earlier studies by Gaertner (1931), Walliser (1964), Fliigel (1965), Serpagli (1967), Schonlaub (1969, 1971), Vai (1971), and Jaeger et al. (1975), a brief summary of knowl- edge of this interval up to the year 1975 was submitted and published in an earlier circular of the Ordovician-Silurian Boundary Working Group. Based on the final decision of the Commission on Stratigraphy that the base of the Silurian System shall be at the base of the A. acuminatus Biozone, the present paper revises the strati- graphy of the boundary beds in the Carnic Alps. In addition new field data are presented and summarized in this updated version of previous reports. I acknowledge the help of H. Jaeger, Berlin, and R. Schallreuter, Hamburg, who kindly provided unpublished data on graptolites and ostracods. Upper Ordovician sediments and stratigraphy All known late Ordovician and early Silurian boundary sequences show clear evidence of a regressive-transgressive relationship. Except for one section representing the deep water ‘Bi- schofalm graptolite facies’, for which, however, biostratigraphical data are missing for the late Ordovician, the lithology and faunal composition in the upper Ordovician reflect a stable environment of shallow to moderate depths with a considerable clastic influx in the Caradoc Stage. During this time the fossiliferous Uggwa Shales, up to 100m thick, were deposited. They comprise sandy shales and pass laterally into greenish and brownish mudstones and siltstones, the latter being widely distributed in the Central Carnic Alps in the surroundings of Plock- enpass and Lake Wolayer. In contrast to the typical Uggwa Shales, in these beds only very few fossils occur. This shale and siltstone sequence grades laterally and in part also vertically into 40-60 m of thick well-bedded and locally cross-bedded sandstones also known as the Himmel- berg Sandstone. Fossils, if any, are extremely rare except for the under- and overlying strata which suggest a late Caradoc age for this unit. Hence, this sandstone is in part equivalent to the Uggwa Shale, which is also supported by field observations. The fauna of the clastic upper Ordovician sequence is dominated by bryozoans and less frequently brachiopods, trilobites, gastropods and cystoids occur. According to Vai (1971) and Havlicek et al. (1987), this fauna suggests a close relationship to middle Caradoc sequences of Sardinia and other regions of southern Europe as well as to Bohemia. The Caradoc Uggwa Shale and its equivalent, the Himmelberg Sandstone, are overlain by distinctive limestones of Ashgill age. Two lithological types are developed in the Carnic Alps, Bull. Br. Mus. nat. Hist. (Geol) 43: 107-115 Issued 28 April 1988 H. P. SCHONLAUB 108 “AyeI] YOU pue eIysNy YINOs Ul vole sdjy os1uIeED ayQ Jo dep | “SIy S58 278 -7fef- 7 “JUSWISSDG) dUll|D}SAYD ULIM Ka||DA |IDD SUL JO ULJAON Sdjy U4eULNOS PUD Sdjvy |DI|IDH Sus JO DISSDI4L Aes 2 oy ——— (o} J19/60/N, iP ulajspjousy . A JasoyBouy > ee ¢ : 2440M — “ . , [asOy . -9aSsau|jo : 2 HOVTIIA 5 3 ae ens ve jayaiMYoons aa 4 OZ - SOF) 5 ee ee SCG ~ gene i deuuneyy . —/S4IOYDS4OH SESE suoljoes AuDPUNOG UDIAN|IS/UDIDIANOPAO Bisers MOL] OUI OMS 1O Selly MVS SUL Ui SYSor DOZO@DIDY 4O UOILNGI4LSID |Do4y ORDOVICIAN-SILURIAN BOUNDARY IN AUSTRIA 109 the first being the nodular Uggwa Limestone and the other its lateral equivalent, the coarse- grained biodetrital Wolayer Limestone. The Uggwa Limestone represents a quiet water shelf environment and contains relatively abundant microfossils, for example conodonts, ostracods and foraminiferans, but also a few trilobites, bryozoans, brachiopods and cephalopods. Yet age assignments within the Ashgill are not precisely known except for its upper part, in which the Hirnantia fauna is found. The second type, the Wolayer Limestone, comprises biodetrital cystoid-bearing light grey limestones which may be up to 18m thick, three times as much as the Uggwa Limestone. Its palaeogeographic setting suggests carbonate mud mounds on the outer shelf surrounded by rather uniform and more widely distributed shelf carbonates of the Uggwa Limestone. There is no indication of close proximity to a land area for either type. In the Carnic Alps lateral changes between the two limestone types can occur over a few km in the same tectonic unit. In other places they are tectonically separated. As shown in the diagrams (Figs 1, 2) the individual boundary sections exhibit significant differences in thickness and lithology, as far as the latest Ordovician is concerned. The Boundary Beds At the top of the Ordovician sequence in the Carnic Alps a widespread sandy facies occurs, the so-called Plocken Formation. In the old literature this horizon was termed ‘Untere Schichten’. It succeeds the Uggwa Limestone but is missing at the top of the coeval Wolayer Limestone (see below). Reinvestigation of the Plocken Formation indicates that it represents a regressive sequence starting with offshore shaly mud intercalations in the uppermost Uggwa Limestone and above, and developing into shoreface calcareous sands. In these beds contorted deforma- tion structures are very common. In the lower parts they are associated with channel fillings of coarse bioclastic material. The Hirnantian fauna which first occurs in laminated greenish-greyish mudstones overlying the Uggwa Limestone at Cellon shows evidence of transportation. The same is true for the Hoher Trieb section east of Cellon (Figs 4E, 4F). The poorly sorted, mostly disarticulated fossil debris occurs in several layers. They are characterized by internal erosional surfaces, small-scale channelling, reworking of sediment, bioturbation with subsequent infilling of fossils, and pro- nounced load deformation structures. Higher up in the sequence channelling and reworking of the sediment increase, although laminated mudstones are here less abundant. Usually channels are connected with contorted beds the thickness of which is usually between 10 and 20cm but which may reach 60cm. The channel filling consists of coarse-grained bioclastic limestones which cut into the under- lying mudstones and shales. Fossils include representatives of the Hirnantia fauna (mainly brachiopods and trilobites), pyritized ostracods and spicules. According to Jaeger et al. (1975) and Schonlaub (1980: fig. 27 and 1985: fig. 25a) the following taxa have been found in the latest Ordovician Hirnantian Stage: Kinnella kielanae (Temple) Dalmanitina mucronata (Brongniart) Rafinesquina sp. Icriodella sp. Clarkeia sp. Quadrijugator harparum (Troedsson) Hirnantia sagittifera (M‘Coy) Scanipisthia rectangularis (Troedsson) Dalmanella testudinaria (Dalman) Eocytherella troedssonia Bonnema Cryptothyrella sp. Dornbuschia ostseensis Schallreuter At Cellon (Fig. 3) and Hoher Trieb (Figs 4E, 4F) the channels are connected or grade into contorted beds composed of less pure limestones. They are irregularly coloured brownish and greyish marls with floating brachiopod valves and loosely packed matrix-supported subangular clasts of different rock types including carbonates of different size up to 20-30cm in diameter, sandstone pebbles, shales or small black phosphorite nodules. At the Nolblinggraben section at the base of the Plécken Formation there is even a layer with clasts of granitic composition (Schonlaub & Daurer 1977). 110 H. P. SCHONLAUB ORDOVICIAN/ZSILURIAN BOUNDARY SOU : | 1 2 3 FEISTRITZ— STEINWEN NOLBLING— GRABEN DER - HUTTE GRABEN WASSERFALL | (S | crenulatus oO | eet bas >, - 4 turriculotus 9 a | q aioe Do a AERONIAN Sea Res oo aa @ | ANIAN Bcuminentan c a E Bs if gi a6 | ORDOVICIAN Fig. 2 Comparative sections through the Carnic Alps near the Ordovician-Silurian boundary. The Plécken Formation has a thickness of between 1-5 and 9m, the latter occurring on the southern slope of Mount Rauchkofel. Based on the occurrences of the Hirnantia faunal assemblage at the Cellon, Hoher Trieb and Uggwa sections, a late Ashgill age, i.e. the Hirnan- tian Stage, is deduced for the Plécken Formation. This is in agreement with earlier reports of Glyptograptus cf. persculptus (Salter) from the ‘Feistritzgraben’ section in the Western Kara- | wanken Alps (Jaeger et al. 1975). We correlate this level with the basal Plocken Formation in | the Carnic Alps, although the lithologies are slightly different. t The Base of the Silurian On the carbonate shelf which was already shallow in pre-Hirnantian times the shallow water carbonate facies was re-established in the Silurian. However, in this facies disconformities with distinct karst surfaces are widely developed and depositional hiatuses are well known. The relief may be several cm or more. In particular this phenomenon can be seen on top of the carbonate mounds of the Wolayer Limestone which apparently became subaerially exposed from the ORDOVICIAN-SILURIAN BOUNDARY IN AUSTRIA HU siaimN ALPS 4 5 6 7 HOHER TRIEB CELLON RAUCHKOFEL RAUCHKOFEL SUD BODEN Kok-Fm Kok - Fm ao S0gitta - Zone Kok- Fm conodonts 2 77 Telychian (Sheinwoodian ) conodonts Fm. en Hirnantia Fauna MivAreitcic Uggwa Wolayer latest Ordovician to the middle or even upper Silurian (see Fig. 2, section no. 7). In other sequences stratigraphical gaps are of shorter duration. In any case there is an abrupt upward transition from the Hirnantian Plocken Formation to either cephalopod limestones of the Kok Formation or to the uniform dark grey graptolitic shales of the basal Silurian Bischofalm facies. According to unpublished new data of H. Jaeger (cited by Schonlaub, 1985: 78) in the Carnic Alps the graptolite facies starts in the A. acuminatus Biozone. At the ‘Steinwenderhiitte- Wasserfall’ locality the graptolitic shales succeed the greyish Bischofalm Quartzite. At other places, for example at Nolblinggraben, D. vesiculosus, the index graptolite of the lower Silurian graptolite zone 17, has been reported overlying an almost 2m thick quartzitic rock. Due to the lack of fossils the stratigraphical relationship between the two quarzitic members is yet poorly understood. They may represent fan deposits of different ages, the lower one being deposited in basin areas of the Hirnantian low sea level stand and the latter at or near the beginning of the transgressive graptolite sequence at the presumed base of the Rhuddanian Stage. In either case, in this part of the Carnic Alps an almost complete succession of strata across the Ordovician— Silurian boundary can be assumed. 112 H. P. SCHONLAUB ORDOVICIAN-SILURIAN BOUNDARY IN AUSTRIA 1113} Conclusion The Ordovician-Silurian boundary beds in the Carnic Alps reflect a regressive-transgressive cycle. Alongside probably continuous sedimentation across the systemic boundary in sections representing deeper environments, in the shallow carbonate shelf areas stratigraphical gaps are very common. This relation is in accordance with data from other regions in the world. However, this event was not solely caused by worldwide eustatic changes of sea level attributed to the famous glacial event in the southern hemisphere. Vertical block movements of Caledon- lan age also affected the Carnic Alps in the late Ordovician and, consequently, were also responsible for differences in thickness of closely-related sections as well as for greatly differing facies that developed in the Silurian after a less pronounced facies pattern in the Ordovician. Fig. 3 Ordovician—Silurian boundary beds at the Cellon section in the central Carnic Alps of Austria. A: Cellon section, lower part showing Uggwa Limestone in the lower portion and Plocken Formation above. Indicated is a coarse grained channel filling limestone bed at the base of the Plocken Formation. B: Detail from A in the upper portion of the Plocken Formation showing a multilayered fold. C: Detail from A. Coarse-grained limestone bed at the base of the Plocken Formation (no. 6 is a reference point of O. H. Walliser’s conodont-based collection). D: Internal erosional surface in the uppermost Uggwa Limestone Formation at level no. 5 of Walliser (1964). Length of the cut approx. 4cm. E: Reworked limestone clast at the same horizon as Fig. D. Long axis approx. 3-5cm. F: Fossil debris representing components of the Hirnantia fauna in the uppermost Uggwa Limestone Formation at horizon no. 5 of Walliser (1964). Width of the brachio- pod valve is 3cm. G: Same horizon as Figs D—F showing bioturbation and infilling at an internal erosional surface in mudstones. Length of the cut approx. 4cm. 114 H. P. SCHONLAUB ORDOVICIAN-SILURIAN BOUNDARY IN AUSTRIA 115 References Fligel, H. 1965. Vorbericht tiber mikrofazielle Untersuchung des Silurs des Cellon-Lawinenrisses (Karnische Alpen). Anz. 6st. Akad. Wiss. mat.-nat. Kl., Wien, 1965: 289-297. Gaertner, H. R. von 1931. Geologie der Zentralkarnischen Alpen. Denkschr. Akad. Wiss. Wien 102: 113-199. Havlicek, V., Kriz, J. & Serpagli, E. 1987. Upper Ordovician Brachiopod assemblages of the Carnic Alps, Middle Carinthia and Sardinia. Boll. Soc. paleont. ital., Modena, 25: 277-311, 9 pls. Jaeger, H., Havlicek, V. & Schonlaub, H. P. 1975. Biostratigraphie der Ordovizium/Silur-Grenze in den Stidalpen—Ein Beitrag zur Diskussion um die Hirnantia-Fauna. Verh. geol. Bundesanst., Wien 1975: 271-289. Schonlaub, H. P. 1969. Das Palaozoikum zwischen Bischofalm und Hohem Trieb (Zentrale Karnische Alpen). Jb. geol. Bundesanst. Wien 112: 265-320. —— 1971. Palaeo-environmental studies at the Ordovician/Silurian boundary in the Carnic Alps. Mem. Bur. Rech. géol. minier., Paris, 73: 367-376. —— 1980. Field Trip A: Carnic Alps. In H. P. Schonlaub (ed.), Guidebook, Abstracts. Second European conodont symposium. Abh. geol. Bundesanst., Wien, 35: 5—60, 10 pls. —— 1985. Das Palaozoikum der Karnischen Alpen. Exkursion Wolayersee. Arbeitstag. geol. Bundesanst., Wien, 1985: 34-69. — & Daurer, A. 1977. Ein auffallender Gerollhorizont an der Basis des Silures im Nolblinggraben (Karnische Alpen). Verh. geol. Bundesanst., Wien 1970: 361-365. Serpagli, E. 1967. I conodonti dell’Ordoviciano Superiore (Ashgilliano) delle Alpi Carniche. Boll. Soc. paleont. ital., Modena, 6: 30-111, 25 pls. Vai, G. B. 1971. Ordovicien des Alpes Carniques. Mem. Bur. Rech. geol. minier., Paris, 73: 437-450, 4 pls. Walliser, O. H. 1964. Conodonten des Silurs. Abh. hess. Landesamt. Bodenforsch., Wiesbaden, 41: 1-106, 32 pls. Fig. 4 Ordovician-Silurian boundary sections at Rauchkofel-Boden, Rauchkofel-Sud and Hoher Trieb in the central Carnic Alps. A: Rauchkofel-Boden section, disconformity between the Ashgill Wolayer Limestone (left) and the darker cephalopod-bearing Kok Formation (right). At the base of the latter sagitta-Zone conodonts of middle Wenlock age occur. B: Rauchkofel-Siid section showing contact between the nodular Uggwa Limestone (left) and the overlying Plocken Forma- tion (right). C, D: Reworked limestone clasts containing an Amorphognathus ordovicicus conodont fauna in the lower part of the Plocken Formation at the Rauchkofel-Stid section. E, F: Hoher Trieb section. Uggwa Limestone (left) and basal part of the Plocken Formation (right). Note channel filling coarse-grained bioclastic bed near the base of the Plécken Formation. This bed contains representatives of the Hirnantia fauna (Hirnantia sagittifera, Dalmanella testudinaria, Kin- nella kielanae, Cryptothyrella sp. and also Clarkeia sp.). The Ordovician—Silurian boundary in China Mu En-zhit Nanjing Institute of Geology and Palaeontology, Academia Sinica, Chi-Ming-Ssu, Nanjing, China. + Professor Mu died in April 1987. Synopsis After a general account of the Chinese graptolite zones about the boundary, a précis is given of the Chinese type section for the boundary, at Wangjiawan, which includes the faunal characteristics. It is followed by similar details for nine other major Chinese sections and a synthesis of the biofacial types. After a discussion of correlation problems about the boundary, it is concluded that the ascensus Zone of some European sections is equivalent to the Chinese persculptus Zone, and that the base of the Silurian is best taken above the bohemicus Zone and its correlatives, the Hirnantia—Dalmanitina fauna. Introduction Ordovician and Silurian strata are well developed in China. Many Ordovician-Silurian bound- ary sections have been defined in the Yangtze Region (or the Central China region) where the Ordovician and Silurian consist of platform deposits. These sections are small in thickness and rich in fossils, mainly graptolites, known as the Ashgill Wufeng Formation and the early Llandovery Lungmachi Formation. Between these two formations there is usually a thin bed of shelly facies, namely the Hirnantia—Dalmanitina bed (HD) or the Kuanyinchiao bed. The grap- tolite sequences of the Wufeng Formation and the Lungmachi Formation are quite complete, and thirteen graptolite zones have been established in descending order as follows: Lungmachian: L, Monograptus sedgwickii Zone L, Demirastrites convolutus Zone L,; Demirastrites triangulatus Zone L, Pristiograptus cyphus Zone L, Orthograptus vesiculosus Zone L, Parakidograptus acuminatus Zone L, Glyptograptus persculptus Zone Wufengian: W, Diplograptus bohemicus Zone s Paraorthograptus uniformis Zone W,, Diceratograptus mirus Zone W, Tangyagraptus typicus Zone W,. Dicellograptus szechuanensis Zone W, Amplexograptus disjunctus yangtzeensis Zone or Pleurograptus lui Zone The establishment of the Wufengian and Lungmachian graptolite zones is of great impor- tance in stratigraphical correlation and in the determination of the exact position of the Hirnantia—Dalmanitina bed (HD). The HD bed is underlain by beds of varying age from the Tangyagraptus typicus Zone (W3) to the lower part of the Diplograptus bohemicus Zone (W6) in different localities. By comparison, the earliest Silurian shelly facies, known as the ‘Eospiri- gerina bed or the Wulipo bed, has a less wide distribution and its upper limit varies in different places and may reach as high as the Pristiograptus cyphus Zone (L4). The relationship between the Ordovician-Silurian boundary graptolite zones and the shelly beds may be shown in Table 1. As shown in the table, the Ordovician-Silurian boundary should be drawn between the Diplograptus bohemicus Zone (W,)/Hirnantia—Dalmanitina bed and the Glyptograptus per- sculptus Zone (L,)/‘Eospirigerina’ bed. The striking faunal changes from the topmost Ordovi- cian (W,) and the lowermost of the Silurian (L,) support this assertion. Therefore, nearly all Bull. Br. Mus. nat. Hist. (Geol) 43: 117-131 Issued 28 April 1988 118 MU EN-ZHI Table 1 A correlation between the graptolite and shelly sequences across the Ordovician—Silurian boundary. L, Pristiograptus cyphus L, Orthograptus vesiculosus 3 Lis Parakidograptus acuminatus | a s EF Glyptograptus persculptus ‘Eospirigerina fauna = Hirnantia—Dalmanitina Ws Diplograptus bohemicus (Wo) fauna (HD) lower bed Paraorthograptus uniformis Kuanyinchiao Diceratograptus mirus Ww, Tangyagraptus typicus geologists and palaeontologists in China agree that the Ordovician—Silurian boundary should be placed between the D. bohemicus Zone (W,) (or the Hirnantia—Dalmanitina bed (HD)) and the G. persculptus Zone (L,). Description of the Ordovician—Silurian boundary sections In 1983 the writer reviewed sixteen Ordovician—Silurian boundary sections distributed in four stratigraphical regions and described nine sections in the Yangtze Region in detail. In recent years, some sections have been revised and some new sections recognized. There are 33 well defined Ordovician—Silurian boundary sections distributed in four regions of China. Among them, 26 are in the Yangtze Region, three in the Xizang (Tibet}}W. Yunnan Region, two in the Zhujiang Region (S. China Region) and one in the Northwest Region, as shown in the map (Fig. 1). In the northernmost region, the Ordovician—Silurian strata are very thick, complicated in structure and fossils are rare, and thus no ideal Ordovician—Silurian boundary section has been found in this region. There are no Silurian deposits in the Huanghe Region (N. China Region). In the present paper, the type section, the Wangjiawan section of Yichang, W. Hubei, and nine selected sections are described as follows. 1. The Wangjiawan Ordovician-Silurian Boundary section is the type section in China. In 1982, this section was restudied by Mu En-zhi, Zhu Zhao-ling, Lin Yao-kun, Zou Xi-ping, Wu Hong-ji, Chen Ting-en, Geng Liang-yu and Dong Xi-ping. The section is as follows (after Mu et al. 1984). Lower Silurian Lungmachi Formation (basal part): 15. Black argillaceous shale weathered greyish black, yielding (ACC768) Orthograptus vesiculosus (Nicholson), Climacograptus normalis Lapworth and C. cf. medius Tornquist more than 1:0m 14. Brownish-grey siliceous shale intercalated with black shale, with 7 siliceous beds in a distance of 20cm, yielding (ACC767) Parakidograptus acuminatus (Nicholson), Climacograptus normalis Lapworth, C. sinitzini (Chaletzkaya), Glyptograptus tamariscus magnus Churkin & Carter and Paraorthograptus sp. 0:60 m 13. Black shale with (ACC766) Parakidograptus acuminatus (Nicholson), Climacograptus bicaudatus Chen & Lin, C. normalis Lapworth, C. angustus Perner and C. sinitzini (Chaletzkaya). 0:35m 12. Black shale with sandy shale (0:15m thick) in the upper part, weathered greyish black, containing (ACC765) Akidograptus ascensus Davies, Glyptograptus sinuatus (Nicholson), G. tamariscus magnus 119 ORDOVICIAN-SILURIAN BOUNDARY IN CHINA ‘uruUNX ‘A ‘UBYysoRg ‘sueysoyeys cE ‘ueuUNA “M ‘Ixny] ‘nilsuey TE “Qeqry) Suezry ‘ezurex O¢ ‘nsueH ‘surfed ‘zSusyorys 67 ‘Ixduely MN “BuluNnM “Buyreyury gz ‘suerloyZ “‘M ‘uerbny ‘erfsuey 1/7 ‘inyquy ‘s ‘uerixsurp ‘Suosieg 97 ‘nsBuvrp ‘Surluen Jesu ueyssue yl ¢z ‘1xueeys ‘s ‘Suedrz ‘noyorleg pz ‘Ixueeys ‘s ‘equoyZ €7 :Ixueeys “§ ‘SURIXIX ZZ ‘IxueRYs ‘§ ‘SuoyzueN ‘ueyssuerT [Zz ‘uenyoIs ‘MM ‘eASuOY ‘ueyssuenyory Qz ‘ueuunx AN ‘uenseq 6] ‘ueuUn, AN ‘ullue A gy] -ueNYoIg MS ‘sulusueY_D ‘oysuenNys /] ‘ueNyoIS ‘sg ‘Bueiltd ‘ovrbutAueny oJ ‘1z3uoy ‘uerpuerluey cy] ‘izduo] ‘eASugsure7q py ‘noyziny ‘N ‘Izsuoy ‘uenkenysueny ¢] ‘noyzing MN ‘ailig ‘noyizuex Z] ‘noyzinyH ‘NI ‘enyuoy [] ‘noyziny ‘Ny ‘IAunzZ ‘isuossu0q Q] ‘noyziny _qN ‘oyue x “IXUPD 6 :NOYZIND AN ‘OvIZuOS g ‘ueNyoIS qs ‘UBYsNTX / ‘IoqnH ‘AA ‘INSIZ ‘URUTX 9 ‘laqny ‘MA ‘Buvyorg ‘Buidsuerpepy ¢ ‘Bueyory ‘eASue yp p -sueyory ‘uemerfsuem ¢ ‘SuRYyoIX “Suvixus,y 7 Sloqny “A ‘BueyorA ‘Bueyoenysueny | ‘euryo ut sdosoyno Aepunog ueLnyig—ueIoIAOpIQ | “BI Das DUIYD Y4NOS 120 MU EN-ZHI Churkin & Carter, G. tamariscus linearis Perner, G. ex gr. tamariscus Nicholson, Climacograptus angustus Perner, C. bicaudatus Chen & Lin and C. normalis Lapworth 0:20m (ACC764a) Glyptograptus sinuatus (Nicholson), G. tamariscus linearis Perner, Climacograptus angustus Perner, C. wangjiawanensis Mu & Lin, Diplograptus modestus Lapworth and Rhaphidograptus minutus Chen & Lin 0:04m 11. Black argillaceous shale weathered brownish grey in colour, rich in graptolites including (ACC763d) Glyptograptus persculptus (Salter), G. sinuatus (Nicholson), G. ex gr. tamariscus Nicholson, G. tamariscus linearis Perner, Diplograptus modestus Lapworth, Orthograptus guizhouensis Chen & Lin, Paraorthog- raptus innotatus (Nicholson), Climacograptus angustus Perner, C. normalis Lapworth, C. wangjiawanensis Mu & Lin and Rhaphidograptus minutus Chen & Lin 0-:16m (ACC763c) Glyptograptus sinuatus (Nicholson), G. lunmaensis Sun, G. tamariscus linearis Perner, G. tamariscus magnus Churkin & Carter, Diplograptus cf. coremus Chen & Lin, Orthograptus angustifolius Chen & Lin, O. guizhouensis Chen & Lin, O. bellulus Tornquist, Climacograptus angustus Perner and C. wangjiawanensis Mu & Lin 0-08 m (ACC763b) Glyptograptus sinuatus (Nicholson), G. lunmaensis Sun, G. ex gr. tamariscus Nicholson, G. tamariscus linearis Perner, G. tamariscus magnus Churkin & Carter, Diplograptus modestus Lapworth, Orthograptus angustifolius Chen & Lin, Paraorthograptus innotatus (Nicholson), P. sp., Climacograptus angustus Perner and C. normalis Lapworth 0:06 m (ACC763a) Glyptograptus persculptus (Salter), G. sinuatus (Nicholson), G. lungmaensis Sun, G. tamariscus linearis Perner, G. tamariscus magnus Churkin & Carter, Diplograptus modestus Lapworth, Climacograptus angustus Perner and C. normalis Lapworth 0:06 m Upper Ordovician Wufeng Formation: 10. Bluish grey argillaceous calcareous silicolites weathered whitish-yellow and greyish-yellow, yielding abundant brachiopods and trilobites: (ACC762) Leptaenopoma trifidum Marek & Havliéek, Kinnella kielanae (Temple), Dalmanella testudinaria (Dalman), “Paracraniops’ patillis Rong, Cliftonia cf. oxople- cioides Wright, Hirnantia sagittifera (M‘Coy), Draborthis cf. caelebs Marek & Havli¢ek, Aphanomena ultrix (Marek & Havlicek), Aegiromena cf. ultima Marek & Havlicek and Dalmanitina yichangensis Lin, D. sp. 0-33m 9. Black argillaceous shale and mudstone, yielding (ACC761) Diplograptus bohemicus (Marek) and Paraothograptus typicus Mu with a few brachiopods and cephalopods 0:26m 8. Black shale intercalated with a few siliceous shale beds of the same colour, yielding: (ACC760) Diplo- graptus bohemicus (Marek), D. sp., Glyptograptus sp., Climacograptus supernus Elles & Wood and Paraorthograptus sp. 0:23 m 7. Black argillaceous shale with siliceous shale intercalation, yielding in the upper part (ACC759) Dicel- lograptus ornatus Elles & Wood, Climacograptus supernus Elles & Wood, C. longicaudatus Geh, C. sp., Glyptograptus sp., Orthograptus truncatus Lapworth and Paraorthograptus uniformis Mu & Li 0:-42m Middle part (ACC758) Tangyagraptus typicus Mu, Climacograptus supernus Elles & Wood, C. venustus Hsu, Amplexograptus suni (Mu) and Paraplegmatograptus sp. 0:70m Lower part (ACC758a) Dicellograptus szechuanensis Mu, D. ornatus Elles & Wood, Climacograptus supernus Elles & Wood, C. sp., Orthograptus truncatus Lapworth, Orthograptus maximus Mu and Amplex- ograptus suni (Mu) 1:73m 6. Black carbonaceous siliceous shale, yielding (ACC757) Dicellograptus szechuanensis Mu, Amplexo- graptus disjunctus yangtzensis Mu & Lin, Pseudoclimacograptus sp., Orthograptus abbreviatus Elles & Wood and Parareteograptus sinensis Mu 0-40 m 5. Black carbonaceous shale, yielding abundant graptolites: (ACC756) Amplexograptus disjunctus yang- zeensis Mu & Lin, A. suni (Mu), Orthograptus cf. pauperatus Elles & Wood and Parareteograptus sp. 0-43 m 4. Black carbonaceous shale intercalated with a few siliceous beds, yielding abundant graptolites (ACC755) Leptograptus extremus modestus Chen, Dicellograptus sp., Climacograptus chiai Mu, Pseudocli- macograptus spp., Amplexograptus disjunctus yangtzeensis Mu & Lin, Orthograptus cf. maximus Mu, O. truncatus Lapworth, O. cf. pauperatus Elles & Wood and O. sp. and inarticulate brachiopods 0-20m 3. Dark grey to greyish green mudstone 0:12m Linhsiang Formation: 2. Dark yellow mudstone 0-:05m 1. Yellowish green to green argillaceous nodular limestone, yielding the trilobites (ACC754) Ham- matocnemis sp. and Microparia sp. about 2:00m 2. ‘Baoshan’ (the ‘Treasure Hill’) section, Huanghuachang, Yichang, W. Hubei (after Mu et al. 1984). ORDOVICIAN-SILURIAN BOUNDARY IN CHINA 121 Lower Silurian Lungmachi Formation (basal part): 9. Black siliceous rock weathered greyish-yellow, yielding: (ACC744) Parakidograptus acuminatus (Nicholson), Climacograptus normalis Lapworth, C. sinitzini (Chaletzkaya) 0-:10m 8. Black carbonaceous shale, black siliceous shale weathered blackish grey, containing: (ACC743) Glyp- tograptus persculptus (Salter), G. sinuatus (Nicholson), Climacograptus sp. (cf. normalis Lapworth) 0-45m Upper Ordovician Wufeng Formation: 7. Black calcareous argillaceous siliceous mudstone weathered greyish-white to greyish-yellow, yielding abundant brachiopods, trilobites and other fossils, including (ACC742) Hirnantia sagittifera (M‘Coy), Kinnella kielanae (Temple), Aphanomena ultrix (Marek & Havlicek), Cliftonia cf. psittacina (Wahlenberg), Triplesia sp., Dalmanella testudinaria (Dalman), Aegiromena cf. ultima (Marek & Havliéek), Meristina crassa incipiens (Williams) and Dalmanitina yichangensis Lin 0:10m 5—6. Black argillaceous siliceous shale, weathered dark grey, yielding (ACC741) Diplograptus bohemicus (Marek) and a few brachiopods in the upper part 0:45 m 3-4. Black siliceous shale intercalated with argillaceous shale, containing (ACC740) Dicellograptus ornatus Elles & Wood, D. sp., Glyptograptus sp., Climacograptus supernus Elles & Wood, C. hastatus Hall, C. sp. and Paraorthograptus uniformis Mu & Li 0:-51m 2. Black shale intercalated with black siliceous shale, yielding (ACC739) Diceratograptus mirus Mu, D. ornatus brevispinus Chen, Glyptograptus sp., Climacograptus hastatus Hall 0:20m 1. Black shale with a few siliceous shale intercalations, rich in graptolites including (ACC737) Tangya- graptus uniformis Mu, Dicellograptus ornatus Elles & Wood, D. ornatus brevispinus Chen, Glyptograptus sp., Climacograptus supernus Elles & Wood, C. supernus longus Geh, C. tumidus Geh, Amplexograptus suni (Mu), Orthograptus abbreviatus Elles & Wood, Yinograptus disjunctus (Yin & Mu), Y. brevispinus Mu, Paraplegmatograptus connectus Mu 0-15m Black shale with siliceous shale intercalation, yielding abundant graptolites, including (ACC737a) Tangy- agraptus typicus Mu, T. uniformis Mu, T. sp., Climacograptus supernus Elles & Wood, C. supernus longus Geh, Orthograptus truncatus Lapworth, Glyptograptus sp., Amplexograptus suni (Mu), Yinograptus dis- Junctus (Yin & Mu), Y. grandis Mu, Paraplegmatograptus sp. Ld 3. Renhuai section (after Geng Liang-yu et al. 1984). Lower Silurian Lungmachi Formation (basal part): Greyish-black silty, carbonaceous shale (0-05 m thick in single bed), cream-coloured sandy shale (in basal part), yielding an abundant graptolite fauna of Glyptograptus kaochiapienensis Hsu, G. cf. lungmaensis Sun and Orthograptus sp. etc. associated with some brachiopods 1-8m Upper Ordovician Wufeng Formation: 2. Kuanginchiao bed, including the following units: c. dark grey thick-bedded bioclastic limestone in upper part (ADR557-3) with numerous solitary corals such as Brachylasma sp., Crassilasma sp. and Dansiphyllum? sp. 1:14m b. Dark greyish thin-bedded bioclastic limestone in the middle part (ADR557-2) including Hirnantia sagittifera (M‘Coy), Dalmanella testudinaria (Dalman), Aphanomena ultrix Marek & Havli¢ek, Dalmanitina sp., Modiolopsis sp., rugose corals, and the chitinozoan Conochitina cf. sp. A of Achab 0:29m a. Dark greyish medium-bedded limestone in lower part (ADR557-1) with the monotomous chitin- ozoan Conochitina cf. sp. A of Achab 0-67 m 1. Greyish-black carbonaceous shale with a minor quantity of clayey shale in the upper part, dark greyish dolomitic limestone in the lower part and 4cm greyish black carbonaceous shale in basal part, yielding abundant graptolites such as Climacograptus hastatus Hall, C. sp., Paraorthograptus typicus Mu, P. sp., Dicellograptus ornatus Elles & Wood, D. tenuisculus Mu et al., D. szechuanensis Mu and Pleurograptus lui Mu 41m 4. The Nanzheng Formation of Liangshan, Nanzheng county, S. Shaanxi, was considered to be basal Silurian for a long time. However Zhu et al. (1986) have revised this to a late Ordovician age. According to their detailed work, the Nanzheng Formation is the equivalent of the Wufeng Formation and indicates a mixed biofacies. The Liangshan Ordovician—Silurian boundary section, Nanzheng, measured by them may be summarized as follows: Lower Silurian Lungmachi Formation (basal part): 11. Brownish grey shales with Climacograptus angustus (Perner), Diplograptus uniformis Li, Glyptograptus lungmaensis Sun, G. tamariscus distans Packham, G. tamariscus linearis Perner 0-Sm 1 MU EN-ZHI 10. Brownish grey and pinkish shale with a few cephalopods and brachiopods (NZ10) and Climacograptus normalis Lapworth, C. miserabilis Elles & Wood, C. angustus (Perner), Diplograptus ex gr. modestus Lapworth, D. uniformis Li, Glyptograptus lungmaensis Sun 0:27-0:32 m Upper Ordovician Nanzheng Formation: 9. Brownish-yellow calcareous shale rich in (NZ9) Climacograptus angustus (Perner), Orthograptus sp., Glyptograptus sp., Platycoryphe sinensis (Lu), Dalmanitina sp.; the bivalve Deceptrix sp. and some com- pressed cephalopods 0:17-0:22m 8. Brownish-grey medium-bedded argillaceous limestone with (NZ8) Diplograptus cf. bohemicus (Marek), Orthograptus sp., Climacograptus sp., Pleurorthoceras shanchongense Zou, P. jingxianense Zou, P. slender- tubulatum Zou, P. cf. clarksvillense (Foerste), Michelinoceras sp., Aegiria? sp., Platycoryphe sinensis (Lu) and Dalmanitina nanchengensis Lu 0:74m 7. Brownish argillaceous limestone, containing (NZ7) Dalmanitina nanchenensis Lu, Platycoryphe sinensis (Lu), the gastropod Rhaphistomina? sp., and brachiopod fragments 0:-46m 6. Brownish to light grey, coarse quartzitic sandstone 0:83 m 5. Light brown shale intercalated with sandstone containing bivalve fragments in the top part (NZ6) 2:30m 4. Greyish shale containing a few graptolites (NZ5) including Climacograptus sp. 0:25m 3. Grey clayey and aluminal shale rich in fossils (NZ4) with Orthograptus maximus Mu, O. cf. abbreviatus Elles & Wood, Climacograptus normalis Lapworth, Diplograptus sp., Parareteograptus sp., Dictyonema sp., Orbiculoidea, Euklesdenella, the bryozoans Stictopora, Hallopora and Escharopora; Conularia and Meto- conularia (?) proteica (Barrande) 0:28 m 2. Light grey siliceous shale containing (NZ2) Orthograptus maximus Mu, Climacograptus angustus (Perner) in the lower part 0:15m 1. Light grey and brownish siltstone and shale 0:-5m Linhsiang Formation: Light green and brownish argillaceous limestone, with Nankinolithus sp. and Protopanderodus insculptus (Branson & Mehl) in the upper (NZ2) and Paraceraurus cf. longisulcatus Lu in the lower (NZ1) 1:10m 5. Gaojiawan section, Xixiang, S. Shaanxi. A most detailed Ordovician-Silurian section was measured by Yu et al. (1986) as follows: Lower Silurian Lungmachi Formation: 10. Black siliceous and carbonaceous shale containing (XF 162-155) Orthograptus vesiculosus (Nicholson), Climacograptus transgrediens Waern and C. medius Tornquist. 2:77m 9. Black siliceous shale interbedded with carbonaceous shale rich in graptolites (KF 154-135) with Paraki- dograptus acuminatus (Nicholson), Akidograptus ascensus Davies, A. xixiangensis Yu, Fang & Zhang, A. parallelus Li & Jiao, Climacograptus sinitzini (Chaletzkaya) and Orthograptus lonchoformis Chen & Lin 4-63m 8. Black siliceous shale intercalated with black carbonaceous shale rich in graptolites (KF134—125) with Glyptograptus persculptus Salter, G. persculptus—sinuatus transient, G. tamariscus (Nicholson), G. lung- maensis Sun, G. zhui Yang, Climacograptus normalis Lapworth, Orthograptus lonchoformis Chen & Lin, Akidograptus ascensus Davies and A. xixiangensis Yu, Fang & Zhang 0-89 m Upper Ordovician Wufeng Formation: 7. Black siliceous shale weathered purplish brown in colour, containing (XF124—-118) Diplograptus bohe- micus (Marek), D. orientalis Mu, Climacograptus normalis Lapworth, Glyptograptus sp. 0:64m 6. Greyish to pale siltstone and quartzitic sandstone containing (XF117-115) Dalmanitina wuningensis Liu, Leonaspis (Eoleonaspis) olinini (Troedsson), Hirnantia sagittifera (M‘Coy), Kinnella kielanae (Temple) 0:22m 5. Black siliceous and carbonaceous shale rich in graptolites (XF 114-112) with Paraorthograptus uniformis Mu & Li, Orthograptus truncatus Lapworth, Climacograptus hastatus Hall, Paraplegmatograptus sp. and Dicellograptus sp. 0:26m 4. Black carbonaceous shale and siliceous shale containing graptolites (KF111—110) Paraorthograptus typicus Mu, Climacograptus supernus Elles & Wood, C. hastatus Hall, Paraplegmatograptus sp., Dicello- graptus graciliramosus Yin & Mu 0:17m 3. Black shale weathered brown, containing (KXF109-107) Tangyagraptus typicus Mu, Paraorthograptus typicus Mu, Climacograptus hastatus Hall, C. venustus Hsu, Amplexograptus suni (Mu), Dicellograptus ornatus Elles & Wood, Yinograptus disjunctus (Yin & Mu), Parareteograptus sp. 0-33m ORDOVICIAN-SILURIAN BOUNDARY IN CHINA 123 2. Dark grey shale with (KF 106-104) Dicellograptus szechuanensis Mu, D. excavatus Mu, Pleurograptus lui Mu, Climacograptus supernus Elles & Wood, Parareteograptus sinensis Mu, Orthoreteograptus denticulatus Mu 0:-42m 1. Dark grey to black shale, containing (XF 103-101) Pleurograptus lui Mu, Dicellograptus elegans Carru- thers, Climacograptus supernus Elles & Wood, Pseudoclimacograptus sp., Glyptograptus sp., Parareteo- graptus sinensis Mu, Orthoreteograptus denticulatus Mu 0:-44m Jiancaogou Formation: Grey and yellowish green mudstone with Nankinolithus, etc. In the section listed above, unit 1 is the Pleurograptus lui Zone which is equivalent to the Amplexograptus disjunctus yangtzensis Zone (W,). Unit 2 is the Dicellograptus szechuangensis Zone (W,) and unit 3 is the Tangyagraptus typicus Zone (W;). Unit 4 is the equivalent of the Diceratograptus mirus Zone (W,) but D. mirus itself has not been found. Unit 5 is the Paraor- thograptus uniformis Zone (W.), unit 6 is the Hirnantia—Dalmanitina bed (HD) and unit 7 is the Diplograptus bohemicus Zone (W,). Unit 8 is the Glyptograptus persculptus Zone (L,) character- ized by the occurrence of G. persculptus, G. persculptus—sinuatus transient, G. zhui and G. lungmaensis. It is noteworthy that Akidograptus ascensus first appears in the lower part of this zone and A. xixiangensis appears in the upper part. Unit 9 is the Parakidograptus acuminatus Zone (L,) characterized by the incoming of P. acuminatus and Climacograptus sinitzini in association with A. ascensus and A. xixiangensis. Unit 10 is the Orthograptus vesiculosus Zone (L;) characterized by the incoming of O. vesiculosus. 6. Bajaokou Ordovician-Silurian boundary section, Ziyang county, S. Shaanxi. The Lower Silurian Banjuguan Formation and the Upper Ordovician Bajaokou Formation are all in graptolite facies, without shelly beds. They are composed of dark grey to black carbonaceous and siliceous slate and rich in graptolites, which were deposited in deep water on the south slope of the East Qinling trough and on the north margin of the Yangtze platform. The thickness of the basal Silurian is much greater than that of the uppermost Ordovician. The section measured by Fu and others may be outlined as follows. Lower Silurian Banjiuguan Formation (basal part). Black carbonaceous and siliceous slate: L, Orthograptus vesiculosus Zone with O. vesiculosus, Neodicellograptus, Rhaphidograptus, and Atavo- graptus 274m L, Parakidograptus acuminatus Zone with P. acuminatus and Climacograptus sinitzini (F 14) 20:8 m L, Glyptograptus persculptus—sinuatus transient zone 10:-5m 4. G. persculptus—sinuatus transient, and G. tamariscus (F 13) 3. Akidograptus ascensus, Climacograptus miserabilis, Orthograptus, and Atavograptus (F 12) 2. Glyptograptus cf. persculptus, Orthograptus lonchoformis and Diplograptus cf. modestus (F 11) 1. G. cf. persculptus, G. sinuatus, G. gracilis, Diplograptus modestus, Climacograptus normalis, and C. miserabilis (F10) Upper Ordovician Bajaokou Formation (upper part). Dark grey to black carbonaceous and siliceous slate: W2 Diplograptus spp., Climacograptus sp., Orthograptus sp. (F9, F8) 2m Wi Climacograptus extraordinarius, Diplograptus spp. (F7, F6) 1-Sm W, Paraorthograptus uniformis (F 4) 1:-2m W, Diceratograptus mirus (F3) 0-6m 7. Tangshan Ordovician—Silurian boundary section near Nanjing (Jiao & Zhang 1984). Lower Silurian Kaochiapien Formation (basal part): 10. Greyish and yellowish shale with chert (ND8), containing Glyptograptus caudatus Ge, Climacograptus normalis Lapworth, and Orthograptus sp. 0.30m 9. Variegated siliceous shale with (ND7) Glyptograptus lungmaensis Sun, Orthograptus sp. and Akido- graptus? sp. 0-40 m 8. Purple siliceous shale rich in graptolites (ND6) with Diplograptus sp., Glyptograptus sp. and Cli- macograptus sp. 0:02 m 124 MU EN-ZHI Upper Ordovician Wufeng Formation: 7. Kuanyinchiao bed: greyish siliceous mudstone rich in shelly fossils (ND5) with Dalmanitina yichangensis Lin, Leonaspis sinensis Chang, Platycoryphe sp., Paromalomena polonica (Temple), Aegiro- mena ultima Marek & Havli¢ek, Triplesia? sp., Holopea? sp., Loxonema sp., Nuculoidea sp. and Hyolithes? 0:19m 6. Black sandy shale (ND4), containing Diplograptus cf. bohemicus (Marek) and Climacograptus extraordi- narius (So6) 0-28 m 5. Variegated calcareous mudstone 0:09 m 4. Purple greyish siliceous shale with graptolites (ND3) Diplograptus sp. and Climacograptus sp. 0:09 m 3. Brownish yellow shale (ND2) with the brachiopod Manosia sp., the gastropod Planetochidea and trilobite and crinoid fragments. 0:30m 2. Grey siliceous pale-weathered shale 0:-45m 1. Black siliceous shale with (ND1) Dicellograptus sp. and Climacograptus supernus Elles & Wood 0-83m 8. Xainze area, Northern Xizang (Tibet) (after Mu & Ni, 1983). Lower Silurian Dewukaxia Formation (basal part): Black graptolitic shale with Climacograptus normalis Lapworth, C. miserabilis Elles & Wood, C. xain- zaensis Mu & Ni, Glyptograptus elegantulus Mu & Ni, G. nanus Mu & Ni, G. asthenus Mu & Ni, Diplograptus lacertosus Mu & Ni, D. spanis Mu & Ni and D. temalaensis (Jones). Upper Ordovician Xainza Formation: Grey argillaceous limestone with Hirnantia, Kinnella, Cliftonia, Paromalomena, Hindella, Aphanomena and dalmanitid trilobite 8-82m Greyish-yellow shale with Glyptograptus asthenus Mu & Ni, G. daedalus Mu & Ni, G. elegantulus Mu & Ni, G. nanus Mu & Ni, Diplograptus bohemicus (Marek), D. charis Mu & Ni, D. flustrianus Mu & Ni, D. maturatus Mu & Ni, D. ojsuensis (Koren & Mikhaylova), D. orientalis Mu et al., D. spanis Mu & Ni, D. viriosus Mu & Ni, Climacograptus cf. extraordinarius (Sobolevskaya), C. miserabilis Elles & Wood, C. normalis Lapworth, C. xainzaensis Mu & Ni, C. xizangensis Mu & Ni and Orthograptus sp. 5:27m Upper Ordovician Gangmusang Formation: Limestone with shelly fauna. 9. Mangjiu section of Luxi (after Ni et al., 1983). Lower Silurian Lower Jenhochiao Formation (basal part): 4. Black shale with Climacograptus normalis Lapworth, C. miserabilis Elles & Wood, C. trifilis lubricus Chen & Lin, Akidograptus ascensus Davies, Orthograptus guizhouensis Chen & Lin, Diplograptus bifurcus Mv et al., etc. 41m 3. Sandy mudstone with Climacograptus normalis Lapworth and C. sp. c.0-Sm Upper Ordovician Wanyaoshu Formation (top part): 2. Greyish-white mudstone with Hirnantia sagittifera (M‘Coy), Hindella crassa incipiens (Williams), Cool- inia cf. dalmani Bergstrom, Plectothyrella cf. crassicosta (Dalman), Paromalomena polonica (Temple), Aph- anomena ultrix Marek & Havli¢éek and Dalmanitina sp. c.2m 1. Black shale, containing Climacograptus latus Elles & Wood, C. angustus Perner and Orthograptus maximus Mu. 10. The Ordovician—Silurian boundary strata are well developed at the locality of Shahechang, about 15km NW of Baoshan, Yunnan, where a number of graptolites were collected from the uppermost Ordovician by Ni Yu-nan, Cai Cong-yang, Chen Ting-en, Li Guo-hua, and Wang Ju-de. The stratigraphical sequence is as follows (in descending order): Lower Silurian Lower Jenhochiao Formation (basal part): 3. Upper part: Black siliceous shale with Pristiograptus sp. and Climacograptus sp. Lower part: Greyish white sandy shale with Climacograptus normalis Lapworth, C. xainzaensis Mu & Ni and Glyptograptus sp. (ex gr. persculptus) in the basal 2m. Upper Ordovician: 2. Greyish black sandy shale, rich in graptolites, the top part with Diplograptus bohemicus (Marek), Diplograptus ojsuensis (Koren & Mikhaylova), Climacograptus normalis Lapworth (ACJ196), Cli- ORDOVICIAN-SILURIAN BOUNDARY IN CHINA 125 macograptus cf. normalis Lapworth, C. xainzaensis Mu & Ni, C. extraordinarius (Sobolevskaya), Diplo- graptus cf. orientalis Mu et al., D. yunnanensis Ni (ACJ195). The middle part yields Glyptograptus daedalus Mu & Ni and Climacograptus extraordinaris (Sobolevskaya) (ACJ194); and the basal part Glyptograptus cf. elegantulus Mu & Ni, G. daedalus Mu & Ni, Diplograptus maturatus Mu & Ni, D. ojsuensis (Koren & Mikhailova) and D. temalaensis (Jones) (ACJ193). 1. Yellow argillaceous limestone with Nankinolithus? sp., Cyclopyge sp., etc. Analysis of the boundary sections The strata across the Ordovician-Silurian boundary in China fall into different biofacies types as follows. 1. Where the graptolitic Glyptograptus persculptus Zone (L,) lies upon the graptolitic Diplo- graptus bohemicus Zone (W,) without intervening shelly beds, as in the Bajaokou section, Ziyang, S. Shaanxi. 2. Where the graptolitic Glyptograptus persculptus Zone or its equivalents (L,) lies upon the graptolitic Diplograptus bohemicus Zone (W,) with a shelly bed below, as in the Xixiang section, Xixiang, S Shaanxi; the Ganxi section, Yanhe, NE Guizhou; and the Shahechang section, Baoshan, W Yunnan. 3. Where the graptolitic facies with the Glyptograptus persculptus Zone or its equivalents (L,) lies upon shelly Hirnantia—Dalmanitina beds (HD) with a graptolitic facies below, as at the Wangjiawan, Huanghuachang, Fenxiang and Tangya Sections, all in Yichang, W Hubei; the Sintan section, Zigui, W Hubei; the Shuanghezhen section, Changning, SW Sichuan; the Guanyiqiao section, Quyiang, S Sichuan; the Xiushan section, SE Sichuan; the Songtao section, NE Guizhou; the Hanjiadian and Liangfengya sections, Tongzi, N Guizhou; the Renhuai and Bijie sections, NW Guizhou; the Yanjin and Daguan sections, NE Yunnan; the Luxi section, W Yunnan; and the Xainza sections of Xizang (Tibet). 4. Where the graptolitic facies with Glyptograptus persculptus or its equivalents (L,) lies upon a mixed facies with graptolitic facies below, such as in the Honghuayuan section, Tongsi, N Guizhou; the Liangshan section, Nanzheng, S Shaanxi; the Xinkailing section, Wuning, NW Jiangxi; the Shanchong section, Jingxian, S Anhui; and the Tangjia section, Yuqiau, W Zhejiang. 5. Where the shelly Wulipo bed with an ‘Eospirigerina fauna lies upon the shelly Hirnantia— Dalmanitina bed with graptolitic facies below, as at Donggongsi, Zunyi, in N Guizhou. Strata of the first type are only known in the transitional belt between the Yangtze basin and the East Qinling trough to the north, whereas the last type is only known in the southern marginal belt of the Yangtze basin. The Ordovician—Silurian boundary sections of the second and fourth types are important for the correlation of the Diplograptus bohemicus Zone (W.,) and the Hirnantia—Dalmanitina fauna (HD). The Ordovician-Silurian boundary sections of the third type are most common and widespread in the Yangtze region. The Wufengian (Ashgill) Yangtze sea was bounded by surrounding lands and swells and became a semi-enclosed sea under aerobic conditions, but the surface water above the anoxic layer was oxygenated. The strata of the third type are rich in organic matter and graptolites flourished. The diversity of the Wufeng graptolitic fauna increases upwards stratigraphically from the Amplexograptus disjunctus yangtzeensis Zone (W,) to the Tangyagraptus typicus Zone (W3). More than twenty genera occur in the Dicellograptus szechuanensis Zone (W), apart from the dendroids. The decline of graptolite diversity took place from the Diceratograptus mirus Zone (W,) to the Diplograptus bohemicus Zone (W,) (Table 2). At the end of the Ordovician, all the axonolipous graptoloids were nearly extinct except for a few Dicellograptus which remained in China. In contrast, the Wufengian benthic shelly fauna increased in diversity. The well-known, cosmopolitan Hirnantia fauna first appeared in the equivalents of the Diceratograptus mirus Zone (W,) with 7 genera, and increased gradually to 23 genera in the uppermost Ordovician Hirnantia—Dalmanitina bed (Table 3). The sea level was lowered in late Ordovician due to the formation of the ice cap in North Africa. In the late Wufengian W,—W,g, a shallow and better aerated environment occurred due to ventilation of sea waters. The maximum glaciation was 126 MU EN-ZHI Table 2 Stratigraphical range of graptolite genera in the Wufeng Formation = fe We ic Leptograptus Pleurograptus Dicellograptus Diceratograptus ar = + Dicranograptus Tang yagraptus Glyptograptus Amplexograptus Climacograptus Pseudoclimacograptus Diplograptus Orthograptus Paraorthograptus Parareteograptus Orthoreteograptus Sinoreteograptus Neurograptus Nymphograptus Arachniograptus Phormograptus — Plegmatograptus + Paraplegmatograptus = + Yinograptus = af =F Y angzigraptus = + a ar ar || JP ap ar ap ar ar || = ++++ +++ +++ +++ qe fe te ob te de apap || te ++++4 +++ +++ }++++4+]4+)++++4 | ++4++ ++ + + + + — — Table 3 Stratigraphical range of brachiopod genera in the Upper Wufeng Formation Wie Wow we Paracraniops Dalmanella Paromalomena Leptaena Aphanomena Coolinia Hindella Trematis Hirnantia — Cliftonia ~- Plectothyrella -- Dorytreta = Philhedra — Philhedrella —_— Acanthocrania — — Kinnella — — Draborthis — — Mirorthis — a= Aegiromena — — Leptaenopoma = = Toxorthis Dysprosorthis Trucizetina Onychoplecia iste etrect eet cect | Sb GP Ab sip ae ae Ge ie ae ab IP ae ++etettetet+e | ++tetest ++tetettet+ettete | t+tett+ ORDOVICIAN-SILURIAN BOUNDARY IN CHINA 127 reached at the end of the Ordovician (W,) and the whole Yangtze basin became a nearly normal shallow sea in which the Hirnantia—Dalmanitina fauna flourished. At the beginning of the Silurian a new graptolite fauna occurred, notably with monograptids and typical Silurian diplograptids such as the Diplograptus cf. modestus and Glyptograptus cf. tamariscus groups during the Glyptograptus persculptus Zone (L,) time interval. A new brachio- pod fauna, known as the ‘Eospirigerina fauna, appeared above the Hirnantia fauna in the nearshore region. The rapid change in biofacies and faunal composition is due to the rising of sea level caused by rapid melting of the ice cap. Correlation of the Ordovician—Silurian boundary sections All the Ordovician—Silurian boundary sections may be easily correlated in China by the stan- dard of the Wufengian—Lungmachian graptolite zones and the Hirnantia—Dalmanitina bed. In order to define the Ordovician-Silurian boundary throughout the world, a precise correlation of the Diplograptus bohemicus, Glyptograptus persculptus and Parakidograptus acuminatus Zones with shelly faunas is necessary. Thus, the subdivision and correlation of the Diplograptus bohemicus Zone with the Hirnantia—Dalmanitina bed is of great importance. In the Yichang sections, Western Hubei, the uppermost Hirnantia—Dalmanitina bed is under- lain by the Diplograptus bohemicus Zone (W,) and overlain by the Glyptograptus persculptus Zone (L,), whereas in the Xixiang section, S. Shaanxi, the Hirnantia—Dalmanitina bed is under- lain by the Paraorthograptus uniformis Zone (W) and overlain by the Diplograptus bohemicus Zone (W,), which is succeeded by the Glyptograptus persculptus Zone (L,). Therefore, the D. bohemicus Zone of Yichang is equivalent to the lower part of the D. bohemicus Zone (Wé), and the D. bohemicus Zone of Xixiang is equivalent to the upper part of the D. bohemicus Zone (W2). Thus the Hirnantia—Dalmanitina bed of Yichang is the equivalent of the upper part of the D. bohemicus Zone (W2), and that of Xixiang is the equivalent of the lower part of the D. bohemicus Zone (W¢). Climacograptus extraordinarius and Diplograptus orientalis usually occur in the lower part of the D. bohemicus Zone (W6). The Glyptograptus persculptus Zone (L,) is marked by the incoming of Glyptograptus per- sculptus, G. sinuatus, G. lungmaensis, G. gracilis, Diplograptus modestus, Akidograptus ascensus and monograptids. It represents the beginning of a new developmental stage of graptolite faunas, the fifth (or monograptid) fauna as defined by the writer (Mu 1984). Thus the base of the G. persculptus Zone should be considered an important stratigraphical boundary, that between the Ordovician and Silurian. It is noteworthy that the Akidograptus ascensus Zone, directly overlying the Hirnantia— Dalmanitina beds of Europe, is usually regarded as the equivalent of Parakidograptus acumin- atus by some foreign colleagues. For defining the Ordovician—Silurian boundary the correlation of the Akidograptus ascensus Zone with the Glyptograptus persculptus Zone and the boundary between the Glyptograptus persculptus Zone and the Parakidograptus acuminatus Zone must be clarified. The Parakidograptus acuminatus Zone (L,) is marked by the incoming of P. acuminatus in association with Climocograptus sinitzini which also characterizes the P. acuminatus Zone. Akidograptus ascensus itself first appeared in the persculptus Zone (L,), much earlier than P. acuminatus, although the two forms may be present together in the P. acuminatus Zone (L,), whereas P. acuminatus is confined to the P. acuminatus Zone. Yu and his colleagues are of the opinion that Parakidograptus acuminatus is directly derived from Akidograptus ascensus and a transitional form Akidograptus xixiangensis Yu et al. was described and illustrated from the basal Lungmachi formation of Xixiang, S. Shaanxi. A. xixiangensis appears higher than A. ascensus and lower than P. acuminatus. It posseses akidograptid thecae in the proximal portion of the rhabdosome and parakidograptid thecae in the distal portion. A similar form Akido- graptus giganteus was described by Yang (1964) from the basal Silurian of W. Zhejiang. Li & Ge (1981) and Fu (1983) tried to propose a new genus for these transitional forms between Akidograptus and Parakidograptus. 128 MU EN-ZHI It is clear that the Akidograptus ascensus Zone of Europe may be correlated with the Glyptograptus persculptus Zone in China. This view was confirmed by the works of Nilsson (1984) in Sweden, and Storch (1982) in Bohemia. The same is true, in my view, for the Mirny Creek section, northeast USSR, studied by Koren et al. (1983). The Mirny Creek Ordovician— Silurian boundary section of mixed biofacies measured by Koren and her colleagues may be outlined mainly by graptolites as follows: Members 65 and 66 Paraorthograptus pacificus Zone Members 67 and 68 Climacograptus extraordinarius Zone with Hirnantia—Dalmanitina fauna Members 69 to 72 Diplograptus bohemicus Zone (=‘persculptus’ Zone) with Hirnantia—Dalmanitina fauna Members 73 and 74 Akidograptus ascensus Zone, incoming of Diplograptus of modestus group, Glyp- tograptus of the tamariscus group and Akidograptus ascensus. Members 75 to basal part of member 78 Parakidograptus acuminatus Zone, incoming of P. acuminatus and Climacograptus sinitzini.. Member 78 Orthograptus vesiculosus Zone, incoming of Orthograptus vesiculosus. It is obvious that the Paraorthograptus pacificus Zone (65—66) corresponds to the Paraortho- graptus uniformis Zone (W;), that the Climacograptus extraordinarius Zone (67-68) corresponds to the lower part of the Diplograptus bohemicus Zone (W3), and the Diplograptus bohemicus Zone (=‘persculptus’ Zone, 69-72) corresponds to the upper part of the Diplograptus bohemicus Zone (W2). The lower part of the ‘acuminatus—ascensus Zone’ (members 73-74) of Koren and others is equivalent to the Akidograptus ascensus Zone of Europe, and corresponds to the Glyptograptus persculptus Zone (L,) of China, whereas the upper part of the ‘acuminatus— ascensus Zone’ (75—basal 78) is the Parakidograptus acuminatus Zone, corresponding to the Parakidograptus acuminatus Zone (L,) of China and Europe. I am convinced that the Akidograptus ascensus Zone of the European continent is equivalent to the Glyptograptus persculptus Zone of Britain and Denmark. The Parakidograptus acumin- atus Zone and the Glyptograptus persculptus Zone of the Dob’s Linn section of Britain corre- spond to the P. acuminatus Zone (L,) and G. persculptus Zone (L,) of China respectively. The C. extraordinarius band of the Dob’s Linn section falls within the lower part of the Diplograptus bohemicus Zone (Wé), and the blind dalmanitid band of Dob’s Linn possibly falls within the upper part of the D. bohemicus Zone (W2). It seems to me that the G. persculptus Zone of Dob’s Linn as well as elsewhere represents the beginning of the Silurian transgression due to the rapid melting of the ice-cap in North Africa. Conclusions 1. The Ordovician—-Silurian boundary sections are widely distributed in China. Many Ordovician-Silurian boundary sections have been defined in the Yangtze platform of the Central China Region. 2. The graptolite sequence of the upper Ordovician (Wufengian W,—W,) and the Lower Silurian (Lungmachian L,—L,) affords a valuable standard for correlation. The position of the Hirnantia—Dalmanitina bed is confined to W,-W,. The Diplograptus bohemicus Zone (W,) is the highest level reached by the well-known and cosmopolitan Hirnantia fauna. 3. By this standard all the Ordovician—Silurian boundary sections may be easily correlated in China and even outside China. 4. The acuminatus Zone is marked by the incoming of Parakidograptus acuminatus. The underlying Akidograptus ascensus Zone of Europe is equivalent to the Glyptograptus persculptus Zone, which is the beginning of the Silurian transgression due to the rapid melting of the ice-cap in north Africa. The G. persculptus Zone was also the beginning of the monograptid fauna stage in the history of the development of the graptolite faunas. It is reasonable to place the Ordovician—Silurian boundary between the G. persculptus Zone (L,) or ‘“Eospirigerina’ bed and the D. bohemicus Zone (W,) or the Hirnantia—Dalmanitina bed (HD). 5. The C. extraordinarius Zone of the north-east USSR or the C. extraordinarius band of Dob’s Linn, Scotland, correspond to the lower part of the D. bohemicus Zone (W}). The ‘G. ORDOVICIAN-SILURIAN BOUNDARY IN CHINA 129 persculptus (= D. bohemicus) Zone of the north-east USSR corresponds to the upper part of the D. bohemicus Zone (W2) of China. 6. Many kinds of fossils have been found in the Ordovician-Silurian boundary sections such as graptolites, brachiopods, trilobites, ostracods, corals, bivalves, cephalopods, gastropods, bryozoa, crinoids, conularia, conodonts, chitinozoa, and so on. The increasing number of finds of conodonts is of great importance for correlation with the Anticosti section of Canada. At present, the correlation with Anticosti is difficult. Unfortunately there are many weak points in the Dob’s Linn section, and it is difficult to use as an international Ordovician—Silurian boundary stratotype. References Apollonoy, M. K., Bandaletov, S. M. & Nikitin, I. F. 1980. [The Ordovician-Silurian boundary in Kazakhstan]. 300 pp. Alma Ata [In Russian]. Baillie, P. W., Banks, M. R. & Rickards, R. B. 1978. Early Silurian graptolites from Tasmania and their significance. Search, Sydney, 9 (1—2): 46-47. Berry, W. B. N. & Boucot, A. J. 1973. Glacio-eustatic control of late Ordovician—early Silurian platform sedimentation and faunal changes. Bull. geol. Soc. Am., New York, 84: 275-284. —— & Wilde, P. 1978. Progressive ventilation of the oceans—an explanation for the distribution of the Lower Paleozoic black shales. Am. J. Sci., New Haven, 278: 257-275. Bjerreskoy, M. 1975. Llandoverian and Wenlockian graptolites from Bornholm. Fossils Strata, Oslo, 8: 1-94, pls 1-13. Chen Xu 1983. Influence of the Late Ordovician glaciation on basin configuration of the Yangtze platform in China. Lethaia, Oslo, 17: 51-59. — 1984. Silurian graptolites from southern Shaanxi and northern Sichuan with special reference to classification of Monograptidae. Palaeont. Sin., Peking, (N.S., B) 20: 1-102, 19 pls. [In Chinese, English summary ]. —— 1984a. The Silurian graptolite zonation of China. Can. J. Earth Sci., Ottawa, 21: 241-257. — & Lin Yao-kun 1978. Lower Silurian graptolites from Tongzi, northern Guizhou. Mem. Nanjing Inst. Geol. Palaeont. Acad. Sin. 12: 1-76. 1984. On the material of Glyptograptus persculptus (Salter) from the Yangtze Gorges, China. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and palaeontology of systemic boundaries in China. Ordovician—Silurian boundary 1: 191—202. Anhui Sci. Tech. Publ. House. Davies, K. A. 1929. Notes on the graptolite faunas of the Upper Ordovician and Lower Silurian. Geol. Mag., London, 66: 1—27. Fu Li-pu 1983. Silurian stratigraphy in Bajiaokou, Ziyang, Shaanxi. Bull. Xian Inst. Geol. Min. Res. 6: 1-18. Ge Mei-yu 1984. The graptolite fauna of the Ordovician—Silurian boundary section in Yuqian, Zhejiang. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and palaeontology of systemic boundaries in China. Ordovician—Silurian boundary 1: 389-454. Anhui Sci. Tech. Publ. House. Geng Liang-yu, Hu Zhao-xun & Gong Lian-zan 1984. Upper Ashgillian Chitinozoa from Renhuai, North- ern Guizhou. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeontology of systemic boundaries in China. Ordovician—Silurian boundary 1: 259-266. Anhui Sci. Tech. Publ. House. Ho Xin-yi 1980. Tetracoral fauna of the Late Ordovician Guanyinqiao Formation, Bijie, Guizhou prov- ince. Prof. Pap. Stratigr. Palaeont., Peking, 6: 1-45. Huang Zhi-gao 1982. Latest Ordovician and Earliest Silurian graptolite assemblages of Xainza district, Xizang (Tibet) and Ordovician-Silurian boundary. Geol. Papers of Qinghai-Xizang Plateau 7: 27-—S2. — & Lu Yan-min 1983. Lower Silurian graptolites from Xainza. Geol. Papers of Qinghai-Xizang plateau 2: 121-168. Jiao Shi-ding 1984. Discovery of Dalmanitina—Hirnantia fauna in Tangshan region near Nanjing, Jiangsu province and comment on Ordovician-Silurian boundary. Bull. Nanjing Inst. Geol. M.R., Chinese Acad. Geol. Sci. 5 (4): 66-77. Jin Chun-tai, Ye Shao-hua, He Yuan-xiang, Wan Zheng-quan, Wang Shu-bei, Zhao Yu-ting, Li Shau-ji, Xu Xing-qi & Zhang Zhong-qui 1982. The Silurian stratigraphy and palaeontology in Guanyingiao, Qijiang, Sichuan. People’s Publishing House of Sichuan, Chengdu, China. Koren, T. N., Oradovskaya, M. M., Pylma, L. J., Sobolevskaya, R. F. & Chugaeva, M. N. 1983. The Ordovician and Silurian boundary in the Northeast of the USSR. 208 pp., 48 pls. Leningrad, Nauka [In Russian ]. 130 MU EN-ZHI —— & Sobolevskaya, R. F. 1977. A new standard succession of graptolite assemblages of the Ordovician— Silurian boundary (Northeastern USSR). Dokl. Akad. Nauk SSSR 236 (4): 950-955. ,» ——, Mikhailova, N. F. & Tzai, D. T. 1979. New evidence on graptolite succession across the Ordovician-Silurian boundary in the Asian part of the USSR. Acta palaeont. pol., Warsaw, 24: 125-136. Lai Cai-gen 1979. New advances on the lower and upper boundary of Ordovician system in China. Acta geol. sin., Peking, 53 (4): 180-187. Lesperance, P. J. 1985. Faunal distributions across the Ordovician—Silurian boundary, Anticosti Island and Percé, Québec, Canada. Can. J. Earth Sci., Ottawa, 22: 838-849. Li Ji-jin 1984. Graptolites across the Ordovician—Silurian boundary from Jingxian, South Anhui. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeontology of systemic boundaries in China. Ordovician-—Silurian boundary 1: 309-388. Anhui Sci. Tech. Publ. House. & Ge Mei-yu 1981. Development and systematic position of Akidograptus. Acta palaeont. sin., Peking, 20 (1): 225-234. , Qian Yi-yuan & Zhang Jun-ming 1984. Ordovician—Silurian boundary section from Jingxian, South Anhui. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeon- tology of systemic boundaries in China. Ordovician—Silurian boundary 1: 287-308. Anhui Sci. Tech. Publ. House. Lin Bao-yu 1979. The Silurian system of China. Acta geol. sin., Peking, 53 (3): 179-181. Lin Yao-kun & Chen Xu 1984. Glyptograptus persculptus Zone—the earliest Silurian Graptolite Zone from Yangtse Gorges, China. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Strati- graphy and Palaeontology of systemic boundaries in China. Ordovician—Silurian boundary 1: 203-232. Anhui Sci. Tech. Publ. House. Mu En-zhi 1974. Evolution, classification and distribution of Graptoloidea and Graptodendroids. Scientia sin., Peking, 17 (2): 227-238. 1983. Ordovician-Silurian boundary in China. In: Studies on Stratigraphic boundaries in China: 108-154. Beijing, Sci. Press. —— 1984. Research work on Ordovician-—Silurian boundary in China. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeontology of systemic boundaries in China. Ordovician—Silurian boundary 1: 1-13. Anhui Sci. Tech. Publ. House. 1984a. Developmental stages of graptolites and stratigraphic boundaries. In: Developments in Geosci- ence, Contribution to 27th Intern. Geol. Congr. Beijing, Sci. Press. — & Lin Yao-kun 1984. Graptolites from the Ordovician-Silurian boundary section of Yichang area, W. Hubei. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeontology of systemic boundaries in China. Ordovician—Silurian boundary 1: 45-82. Anhui Sci. Tech. Publ. House. & Ni Yu-nan 1983. Uppermost Ordovician and Lower Silurian graptolites from the Xainza area of Xizang (Tibet) with discussion on the Ordovician—Silurian boundary. Palaeontologia Cathayana 1: 155-180. — & Rong Jia-yu 1983. On the international Ordovician and Silurian boundary. J. Stratigr., Nanking, 7 (2): 81-91. —, Zhu Zhao-ling, Lin Yao-kun & Wu Hong-yi 1984. The Ordovician-Silurian boundary in Yichang, Hubei. Jn Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeon- tology of systemic boundaries in China. Ordovician—Silurian boundary 1: 15—41. Anhui Sci. Tech. Publ. House. Ni Yu-nan 1978. Lower Silurian graptolites from Yichang, western Hubei. Acta palaeont. sin., Peking, 17 (4): 387-416. — 1984. Upper Ordovician Graptolites from Baoshan, western Yunnan. Acta palaeont. sin., Peking, 23 (3): 320-327. ——., Chen Ting-en, Cai Chong-yang, Li Guo-hua, Dun Yan-xue & Wang Jue-de 1982. The Silurian rocks in western Yunnan. Acta palaeont. sin., Peking, 21 (1): 119-132. ——, ——, Zhou Zhi-yi, Lin Huan-ling, Geng Liang-yu & Mu Dao-chen 1983. Boundary of Ordovician and Silurian in Western Yunnan. In: Special Papers in Scientific Expedition of the Heng Duan Mountain 1: 43-47. —, Xu Han-kui & Chen Ting-en 1981. Boundary of Ordovician and Silurian in the Xainza area, Xizang. J. Stratigr., Nanking, 5 (2): 146-147. [In Chinese]. Rickards, R. B. & Hutt, J. E. 1970. The earliest monograptid. Proc. geol. Soc., London, 1663: 115-119. Rong Jia-yu 1979. The Hirnantia fauna of China, with comments on the Ordovician-Silurian boundary. Acta Strat. Sin. 3 (1): 1-28. 1984. Brachiopods of Latest Ordovician in the Yichang district, western Hubei, central China. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeontology of ORDOVICIAN-SILURIAN BOUNDARY IN CHINA 131 systemic boundaries in China. Ordovician—Silurian boundary 1: 111-190, pls 1-14. Anhui Sci. Tech. Publ. House. Storch, P. 1982. Ordovician-Silurian boundary in the northernmost part of the Prague basin (Barrandian, Bohemia). Vést. ust’. Ust. geol., Prague, 57 (4): 231-236. Wang Xiao-feng, Zeng Quing-luan, Zhou Tian-mei, Sun Quan-ying, Li Zhi-hong, Xiang Li-wen & Lai Cai-gen 1983. Latest Ordovician and Earliest Silurian faunas from the eastern Yangtze Gorges with comments on Ordovician-Silurian boundary. Bull. Yichang Inst. Geol. Min. Res. 6: 57-163. Williams, S. H. 1983. The Ordovician—Silurian boundary graptolite fauna of Dob’s Linn, Southern Scot- land. Palaeontology, London, 26 (3): 605—639. Wu Hong-ji 1984. A species of Dalmanitina (trilobite) from Deqing and Yuqian counties, western Zhejiang. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeontology of systemic boundaries in China. Ordovician—Silurian boundary 1: 455—466. Anhui Sci. Tech. Publ. House. Yang Da-quan 1964. Some Lower Silurian graptolites from Anji, northeastern Zhejiang (Chekiang). Acta palaeont. sin., Peking, 12 (4): 628-636. 1983. Latest Ordovician graptolites from Northwestern Zhejiang. Acta palaeont. sin., Peking, 22 (6): 595-605. Yu Jian-hua, Fang Yi-ting, Liang Shi-jing & Liu Hua-bao i984. On the Ordovician—Silurian boundary in Wuning county, Jiangxi Province. J. Nanjing Univ., Nat. Sci. Edn 3: 533-542. ——, & Zhang Da-liang 1986. The Ordovician—Silurian boundary in Xixiang, S. Shaanxi. J. Nanjing Univ., Nat. Sci. Edn. Zhu Zhao-ling, Lin Yao-kun, Chen Ting-en, Zhang Sen-gui & Yu Chang-min 1986. Review on the age of “Nanzheng Shale’. J. Stratigr., Nanking, 10 (2): 98-107. —— & Wu Hongji 1984. The Dalmanitina fauna (Trilobite) from Huanghuachang and Wangjiawan, Yichang county, Hubei province. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeontology of systemic boundaries in China. Ordovician—Silurian boundary 1: 83-110. Anhui Sci. Tech. Publ. House. ae Aes ye a aay: i ewe rs | : : ay aes wie (optt oat j jdehia tb = > af <- 7 e ious s +a é * er ‘ 7 an! nq ites be J ae i ae i4 ’ = =) Po > ao” a S - = ¢ Y aie tC a | ee et or a" oe may : 7 d n= ¢ J ja Sy : aa | SP We ay pee » . 2 - Qan = PR 7 . a bing ~~ A _ ¥ ] _s f w ‘ = & 5 _ 2 << " : ——? Jo a ee a t ® fy ct f ~ =a . | ‘ P +: we | - | é vi ry oe oe PEALE, oo) er a ee se Se Se « | * : eo a a penne eal - er o' The Ordovician—Silurian boundary beds of the north-east USSR T. N. Koren’, M. M. Oradovskaya’ and R. F. Sobolevskaya°* 'VSEGEI, Srednii prospekt 74, 199026 Leningrad, USSR 2PGO ‘Sevvostokgeologia’, 44 Proletarskaya, 685000 Magadan, USSR 3VNIIOkeangeologia, 120 Moika, 190121 Leningrad, USSR Synopsis Graptolites of the supernus, extraordinarius, persculptus, acuminatus and ascensus Zones are present in sections in the north-east USSR, with the best section at Mirny Creek. Brachiopod and coral faunas also occur with the Tcherskidium and Holorhynchus beds in the supernus Zone and the Hirnantia? beds present in the persculptus Zone, both within the Tirekhtyakh Horizon. The succeeding acuminatus and ascensus Zone graptolites are developed in the Chalmak Horizon, which also bears a sparse shelly fauna. Introduction The late Ordovician and early Silurian boundary beds in the north-east USSR crop out on the Omulev Uplift in the upper Kolyma Basin. They are built up by terrigenous-carbonate and terrigenous deposits which are variable in composition and contain a mixed shelly-graptolite fauna. The rocks are exposed on limbs of extensive anticlines and show either a monoclinal succession, such as at Mirny Creek, Neznakomka River and Drevnyaya River, or represent large fragments of sections among complex faulted sequences, such as at the Ina River. The Upper Ashgill and Lower Llandovery deposits include the supernus, extraordinarius, per- sculptus, acuminatus and ascensus graptolite Zones and have a total thickness of about 300m (Fig. 1). This part of the section is designated the Tirekhtyakh and Chalmak horizons. The lower part of the Tirekhtyakh horizon (the supernus Zone) (Fig. 2) shows a diversity of facies from deep water shales yielding graptolites, for example at Khekandya River and Lukavy Creek, to biohermal and biogenic—detrital carbonates with mixed brachiopod—coral-graptolite faunas as at Mirny Creek and the Ina and Neznakomka rivers. The upper part of the Tirekhtyakh horizon (the extraordinarius and persculptus Zones) and the lower part of the Chalmak horizon (the acuminatus and ascensus Zones) are represented by sequences more Fig. 1 Distribution of Ordovician—Silurian boundary beds on the Omulev Uplift. I, Mirny Creek; II, Ina River; III, Neznakomka River Basin; IV, Tirekhtyakh River Basin; V, Mount Kharkindzha; VI, Levaya Khekandya River; VII, Drevnyaya River; VIII, Lukavy Creek. Seimchan © Bull. Br. Mus. nat. Hist. (Geol) 43: 133-138 Issued 28 April 1988 134 KOREN, ORADOVSKAYA & SOBOLEVSKAYA ‘g[BYS SNOoIeOTeO ‘Z] {[IVUWI ‘] | [VU OIWO[OP ‘O] :9U0}S}]IS snOaTvOTeO ‘6 {SUIIOYOrG ‘g {2UO}SIUMT] PorPID9EIq ‘/ !9UOISOUMT] SNOOIIS “9 SoUOysoUUT] AWTS “¢ :2UOJsoUNT] AARIO “p :9UO}SOUUI] ONSEIOIG *¢ -oUO SOUT] Ie[NGe) stydiowoyyad ‘Z souojsoumy aatsseu stydsowoyyad “7 yydq, AgjnwuO ey) Uo spaq A1epuNog UPLINIIS—URISIAOPIO 94} JO gyoid soiovjouuy 7 ‘sIy By u ¢ 20 70 we a a a SES Ds Se Bee a ne te ge \S _ SE 07 0W Of WOU Y= S— U0. = 7 ws j= 0 ‘LY DY ZPUIYLOYY ‘y ofpupywayy ‘df DNWIOVOUZA/Y wg Aus as MN ORDOVICIAN-SILURIAN BOUNDARY IN NORTH-EAST USSR 135 diverse in composition. The Upper Tirekhyakh deposits consist mainly of dolomites, marls, and siltstones representing the termination of the late Ordovician regressive cycle and the Chalmak dark carbonate clay sequences mark the beginning of the Llandovery transgression. Of greatest interest is the key section at Mirny Creek, which has the best exposed Ordovician-Silurian boundary deposits. This has been studied in detail, and forms a type section for such regional units as formations and horizons. The Tirekhtyakh horizon At the Mirny Creek and Ina River sections the horizon is 250m thick and represented by a formation of the same name (upper unit M to unit Q) which is composed of bedded and massive limestones with tabulate corals, brachiopods, ostracodes and gastropods. The lime- stones are interbedded with siltstones yielding graptolites. In the Neznakomka River the forma- tion is 315m thick and represented mainly by biohermal and biogenic-clastic limestones interbedded with siltstones. The rocks contain chiefly brachiopods but the siltstones yield rare graptolites. In the south-eastern Omulev Mountains (Khekandya River, Yasachnaya Basin, Lukavy Creek and Drevnyaya River) the Tirekhtykh horizon exhibits changes in composition. Its lower part consists of the Iryudi Formation (500-600 m) and Lukavaya sequence (100m). The Iryudi Formation is composed of clay and pelitomorphic, unevenly bedded limestones with abundant corals and brachiopods. The Lukavaya unit is represented by dark platey limestones inter- calated with calcareous shales containing abundant graptolites and rare brachiopods. As at Mirny Creek, the upper part of the horizon includes siltstones. The Terekhtyakh horizon has been subdivided by means of graptolites in sections at Mirny and Lukavy creeks, the Khekandya and Drevnyaya rivers, and at Mount Kharkindza, and by means of brachiopods mainly in Mirny Creek and the Neznakomka River (Fig. 3). The lower part of the horizon is equated with the Climacograptus longispinus supernus Zone and the Tcherskidium unicum beds. The supernus Zone is subdivided into two subzones, the lower Climacograptus longispinus longispinus Subzone and the upper Paraorthograptus pacificus Subzone. The lower subzone contains Climacograptus longispinus longispinus Hall, C. |. supernus Elles & Wood, C. hastatus Hall, C. trifidus spectabilis Koren & Sobolevskaya, and Dicello- graptus complanatus Lapworth, whose appearance marks its lower boundary. The pacificus Subzone is recognized as a taxon biozone and, along with Dicellograptus ornatus ornatus Elles & Wood and subspecies of Climacograptus longispinus, contains Climacograptus latus hekan- daensis Koren & Sobolevskaya and C. pogrebovi Koren & Sobolevskaya, while the upper part yields Glyptograptus? ojsuensis Koren & Mikhailova, Climacograptus angustus (Perner), C. normalis Lapworth and others. The supernus Zone is equated with the Tcherskidium unicum beds which also contain Ptychoglyptus bellarugosus Cooper, Holorhynchus ex gr. giganteus Kiaer and Eostropheodonta hirnantensis lucavica Oradovskaya. There are also abundant corals of the genera Agetolites, Heliolites, Propora, Calapoecia, Coxia and others (Preobrazhensky 1966). The brachiopod-coral assemblage allows the lower Tirekhtyakh horizon to be correlated with the Sb beds of Norway. On Mirny Creek the deposits also contain trilobites, gastropods, ostracodes and other fossils (Sokolov et al. 1983). The upper Tirekhtyakh horizon corresponds to the Climacograptus? extraordinarius and Glyptograptus? persculptus Zones. The extraordinarius Zone, which was first established on Mirny Creek (Koren & Sobolevskaya 1979), corresponds to the index-species range. Apart from the latter, it contains Climacograptus? ex gr. extraordinarius (Sobolevskaya), C. angustus (Perner), C. normalis Lapworth and C. mirnyensis (Obut & Sobolevskaya). Climacograptus aff. medius Tornquist and scarce Glyptograptus sp. appear in the upper part of the zone. The persculptus Zone was recognized as equal to the full range of the index-species and the zonal assemblage also contains Climacograptus angustus (Perner), C. normalis Lapworth, C. mirnyensis (Obut & Sobolevskaya) and C. torosus Koren & Sobolevskaya. 4 i) S ik wu 6 Aeron(an canella C MTG trlangulatus u Borealis -convolutus LC OR LOMErY UAAANCARN VeESLCL- LOSUS Chalma@k horizon SKenidiotdes R Hirnantia horizon oo eon perscutptus acumtnatus Torekhtyakh Ask gi Lie unicum Tcherskidium GS wp CGF fH kUS Neznakomka iver Basin » |e er M rg mem TRCCk, —————— a 10 |70 Mirny Creek BR iS ios ad N is} 8 member fo To [> [7 earelsr PF[5,3] [Tv la [v7 a ine? 54/27 [cee] =r —1— 7 i Ths g | 7Aack , Ina River eye Lithology “ r Nats 0 |e err Sol F) Ihe = IMT AT 7 TAT 30 SSS =) 55] (aos PA A lee) Pa = ara, |3 24, aaa [~a TT) Fig. 3. Correlation chart of the Ordovician-Silurian boundary beds of Mirny Creek, Ina River and the Neznakomka River Basin. For legend see Fig. 2. ORDOVICIAN-SILURIAN BOUNDARY IN NORTH-EAST USSR 37) At Mirny Creek, the Khekanda and Neznakomka rivers and Mount Kharkindzha, this zone is equated with the Hirnantia? beds (Oradovskaya 1977). Amongst the brachiopods the most common are Dolerorthis? savagei Amsden, Brevilamunella thebesensis (Savage), Rafinesquina? latisculptilis (Dalman) and Giraldibella bella (Bergstrom), and the trilobites Bumastus (Bumastus) commodus Apollonov and Mucronaspis kolymica Chugaeva. Dalmanitina olini Temple occurs near the top of the zone. The Chalmak horizon In the Omulev Mountains the lower Chalmak horizon includes the Maut Formation, and the main Chalmak Formation corresponds to the horizon in the Yasachnaya Basin. On the Omulev Uplift, the Maut Formation consists of dark calcareous shales, shales and cherts containing graptolites which are interbedded with detrital and conglomerate-like limestones with a scarce neritic fauna. Coarse clastic rocks dominate the coeval deposits further south-east. The lower part of the horizon corresponds to the Parakidograptus acuminatus and Akido- graptus ascensus Zones recognized in Mirny Creek, the Ina and Khekanda rivers, and Mount Kharkindzha. The most complete graptolite assemblage was reported from Mirny Creek (Obut et al. 1967). As well as P. acuminatus and A. ascensus, the assemblage includes Climacograptus rectangularis (M‘Coy), C. transgrediens Waern, Paraclimacograptus sinitzini Chalatskaya, Diplo- graptus ex gr. modestus Lapworth and Glyptograptus ex gr. tamariscus (Nicholson). The bound- ary of the zone is drawn by the appearance and disappearance of the diagnostic species. The acuminatus and ascensus Zone corresponds to the lower Skenidioides beds containing Skenidioides cf. scolioides Temple, Leptaena aff. aequalis Amsden, Eospirigerina putilla Oradovskaya, Zygospiraella sp. and Protatrypa sp. The assemblage is similar to the brachiopod fauna from the lower Llandovery of the Northern Appalachians (Ayrton et al. 1969). The beds also contain trilobites such as Acernaspis sp., Tropidocoryphinae gen. et sp. indet. and the corals Palaeofavosites balticus Rukhin, and Propora conferta Edwards & Haime, among others. The systemic boundary The most complete and well known section of the Tirekhtyakh and Chalmak horizons is exposed along the Mirny Creek. A point 2.5km from its mouth was chosen as a regional type section for the Ordovician—Silurian boundary in the north-east USSR. The systemic boundary is drawn at the base of unit 73 which is 1-5m thick and coincides with the base of the Maut Formation (Figs 2, 3). This level corresponds to the base of the acuminatus and ascensus Zone which in the section studied is substantiated by the appearance of representatives of such typically Silurian groups as Diplograptus modestus Lapworth and Glyptograptus tamariscus (Nicholson) (unit 73). The index-species Akidograptus ascensus Davies is known from the base of unit 74, 1-5m above the boundary, and Parakidograptus acuminatus (Nicholson) occurs in the lower part of unit 75, 11m above the boundary. Their absence from the basal layer can be attributed to the difficulty in searching for graptolites in the beds. In the section at Mount Kharkindzha, akidograptids are known from the basal beds of the Maut Formation associated with other typical diplograptids. The principal criteria for establishing the boundary on a regional scale are distinct changes in the lithological composition of the deposits as well as the change in the assemblages of graptol- ites (persculptus/acuminatus and ascensus), brachiopods (Hirnantia?/Skenidioides) and trilobites. Graptolites allow interregional and global correlations of the level. The section at Mirny Creek is well exposed and shows a continuous succession of uniformly dipping deposits containing diverse fossils. Its major advantage is bed-by-bed graptolite control within the range of the Dalmanitina—Hirnantia assemblage and the presence of shelly fauna (the Skenidiodes beds) from the base of the acuminatus and ascensus Zones. Abundant graptolites, brachiopods and corals and rare trilobites, ostracodes and conodonts are known from the Ordovician-Silurian boundary beds in the north-east USSR. All faunal groups except ostracodes and conodonts have been monographically described in different 138 KOREN, ORADOVSKAYA & SOBOLEVSKAYA publications (Sokolov et al., 1983; Nikolaev et al., 1977; Nikolaev & Sapelnikov 1969; Obut et al. 1967; Opornii razrez (Anon.) 1974; Oradovskaya 1963; Polevoi atlas (Anon.) 1968; Polevoi atlas (Anon.) 1975; Preobrazhensky 1966 and Sobolevskaya 1970, 1974). References (Anon.) 1974. Opornii razrez verkhnego ordovika na Severo-Vostoke SSSR [The Upper Ordovician key section in the north-east USSR]. In: Opornye razrezy paleozoya Severo-Vostoka SSSR: 3-136. Magadan. — 1968. Polevoi atlas ordovikskoi fauny Severo-Vostoka SSSR [Field atlas of the Ordovician fauna in the north-east USSR]. 286 pp. Magadan. 1975. Polevoi atlas siluriiskoi fauny Severo-Vostoka SSSR [Field atlas of the Silurian fauna in the North-East USSR]. 382 pp. Magadan. Ayrton, W. G., Berry, W. B. N., Boucot, A. J., Lajoie, J., Lesperance, P. J., Pavlides, L. & Skidmore, W. B. 1969. Lower Llandovery of the northern Appalachians and adjacent regions. Bull. geol. Soc. Am., New York, 80: 459-484. Koren, T. N. & Sobolevskaya, R. F. 1979. A graptolite zonation of the Ordovician-Silurian boundary deposits of the Omulev Mountains. In M. M. Oradovskaya & R. F. Sobolevskaya (eds), Guidebook to field excursion to the Omulev Mountains. Pacific Sci. Ass. 14 Pacific Sci. Cong., Magadan: 91—92. Nikolaev, A. A., Oradovskaya, M. M. & Sapelnikov, V. P. 1977. [Biostratigraphical review of the Ordovi- cian and Silurian pentamerids of the north-east USSR]. Trudy Inst. Geol. Geokhim. Akad. Nauk SSSR ural. nauch. Tsentr, Sverdlovsk, 126: 32-67, 11 pls. Obut, A. M., Sobolevskaya, R. F. & Nikolaev, A. A. 1967. Graptolity i stratigrafia nizhnego silura okrainnykh podnyatii Kolymskogo massiva (Severo-V ostok SSSR) [Graptolites and the stratigraphy of the lower Silurian of marginal uplifts of the Kolyma Massif, the north-east USSR]. 162 pp. Moscow. Oradoyskaya, M. M. 1963. Ordovikskie otlozhenia khrebta Cherskogo [The Ordovician deposits of the Chersky Ridge]. In: Materialy po geologii i poleznym iskopaemym Severo-Vostoka SSSR 16: 140-162. Magadan. — 1977. Verkhnyaya granitsa ordovika na Severo-Vostoke SSSR [The upper boundary of the Ordovi- cian in the north-east USSR]. Dokl. Akad. Nauk SSSR, Leningrad, 236 (4): 954-956. Preobrazhensky, B. V. (1966). Biostratigraficheskoe obosnovanie granitsy mezhdu ordovikom i silurom Severo-Vostoka SSSR po tabulyatomorfnym korallam [Biostratigraphic substantiation of the Ordovician-Silurian in the north-east USSR based on tabulate corals]. Avtoref. dis. kand. geol. min. nauk., Novosibirsk. 20 pp. Sobolevskaya, R. F. (1970). Biostratigrafia srednego i verknego ordovika okrainnykh podnyatii Kolyma Massif po graptolitam [Middle and Upper Ordovician biostratigraphy of the Kolyma uplifts based on graptolites]. Avtoref. dis. kand. geol. min. nauk., Novosibirsk. 26 pp. 1974. Novye Ashgillskie graptolity v basseine srednego techenia r. Kolymy [ New Ashgillian graptol- ites from the middle Kolyma River basin]. In A. M. Obut (ed.), Graptolity SSSR, Trudy I Vses. Kollokviuma: 63-71. Novosibirsk. Sokolov, B. S., Koren, T. N. & Nikitin I. F. (eds) 1983. Granitsa Ordovika i Silura na Severo-Vostoke SSSR [The Ordovician-—Silurian boundary in the north-east USSR]. 205 pp. Leningrad. The Ordovician—Silurian boundary in the Altai Mountains, USSR E. A. Yolkin, A. M. Obut and N. V. Sennikov Institute of Geology and Geophysics, Siberian Branch, Academy of Science, 630090 Novosibirsk, USSR Synopsis The Ordovician-Silurian boundary is repeatedly exposed in the Altai-Sayan fold belt, with the best- studied outcrops in the Charysh—Inya structural zone near Ust’-Chagyrka and Chineta, where the per- sculptus and acuminatus zones are both known in association with shelly faunas. Ordovician and Silurian deposits in the western part of the Altai-Sayan fold-belt are not only widely distributed in the Altai, but in the Kuznetsk Alatau, Salair and Shoria Mountains as well. The boundary interval, however, is known only from the Altai Mountains in two structural-formational zones, the Anui-Chuya and Charysh—Inya zones. In the first zone there are several sections where it is possible to see a normal stratigraphical succession from Ordovi- cian to Silurian. However, most of them are not well characterized palaeontologically, espe- cially the boundary beds (Yolkin et al. 1978; Sennikov & Sennikov 1982). Because of this, the boundary interval is shown as a biostratigraphical break in the stratigraphical correlation charts for this zone (Khomentovskiy & Tesakov 1983). The Ordovician-Silurian boundary interval is better known in the Charysh—Inya Zone. Here, in different areas, there are now more than ten known sections. In each such area there are usually several sections with transitional continuity between the two systems, though there are some differences in the faunas from area to area. The best of these sections occur near Ust’- Chagyrka and Chineta villages (Yolkin & Zheltonogova 1974; Sennikov et al. 1979, 1982, 1984). The faunal assemblages in these sections in the two areas include graptolites, conodonts, trilobites, gastropods, orthoconic cephalopods, brachiopods, ostracodes, corals, chitinozoans and polychaetes, part of which have been monographed (Sennikov 1976, 1978; Moskalenko 1977; Severgina 1978, 1984; Yolkin 1983). The most important fossils for the subdivision and correlation of these sections are the graptolites. They are the predominant group numerically and have by far the best international distribution stratigraphically. It is important to draw attention to the association, in the boundary beds, of graptolites, conodonts and trilobites, especially Dalmanitina. This indicates the possibility for future work directed towards clarifying and refining the correlation of Ordovician—Silurian boundary beds in the Altai, but perhaps also globally. The best boundary in the Altai, as in China (Chen Xu 1984) would be somewhat below the acuminatus Zone decided by the Ordovician-Silurian Boundary Working Group (Holland 1985). The beds with persculptus correspond to the onset of a wide transgression. Bull. Br. Mus. nat. Hist. (Geol) 43: 139-143 Issued 28 April 1988 OBUT & SENNIKOV YOLKIN, 140 “SPOIL UOI}DAS JO s[IeJap—ae_e ‘deul yo}ays—eE : YOdID BVYIABEYD Jo yuRQ Ia] ay} Jo deur [eaZojoaH ¢€ ‘Sly ‘OSEIIIA BIQUIYD vou YooID eyUVA[QoIng Jo yuRQ ya] ay} Jo deus [eorsojoaH 7 “SIY “SUIEJUNOJ Ie} Y 94} UL Spaq AJepUNO ULLINIIS PUL ULIDIAOPIGC JO SUOTIOIS DOUDIOJOIJO UOT}LOOT «| “BY OC — ne \areta-2| EE — — 2-783||trianguiatua-z° _ —|— — ||ewpras-Z —— -— > a |= —|fextensQeeS oinoh ten a Bi aa ee 1 Ge spe -77714 ind Be % = s a (oe 785 ZA aaa Ayo LLP ES: DO P-786 Ze 2g Wi : kG eta ‘ « C com ae — = = See, tha ~k a Loe LS 02-788 Se ~ Oo ee oa, 00-7766 ~D P-789 ia oe = Bs ee oD > ass z ~ a 4 ~~ = ~~ SS. ; le tee es CoO Ov70_ = ~~ > SS oe ot > pik a O3 4 A MY Ze va fc 3 Ms Ff 2 A? fs aa ee O77 \ as a= OP-78/14 ° Wa ax ‘ Yo = - Yi 4 aS: ¢ s ” “ Of-78e72 C-7718 __—_ i) oe 0 ee cm, «6 ; © LOM MUG OGY SE °F 2-3 Fig. 4 Profile on line A-B-C of Fig. 3b. Fig.5 Details of graptolite zonation on line A-B-C of Fig. 3b. Legend for Figs 2-7. 1—conglomerates, 2—sandstones, 3—silty sandstones, 4—siltstones, 5—cherty rocks, 6—limestones, 7—sandy limestones, 8—argillaceous limestones, 9—boundaries, 10—faults, 11— line of sections, 12—outcrops and artificial exposures (fauna collection points). YOLKIN, OBUT & SENNIKOV 142 é e Cseatqottay eqeuoronu’ Teq lo (eee eoTeqTe* rem REGIS Es Se4TQOTTID pavateo* reli ee ds olnyetehe Sosy [spooezaso eutdstq‘dg Oo mal SUMED O86 o_o —e “ds snqgderSorpey Co eeccese eves | ds sngderSoipey e——_« |e STIETEOS *pey 0 ca ‘ Sielyriniayetar tet TE) e ‘ds snqderZoptyudemuy O gy Siaeyehynalersinerel IZ Xe°TD oni: A snyeutmmoe "reg 000 6008 eee a) ae sngaderS04dhT9 o—_______ o ‘*ds snqdezZotdtq| ¥ e |?! pods snqgders04d4T4 o—e¢—_e +e Pleds snqderZ0,dfTy| e Io! STITFTSUOT*TO oeee eee call SNOTMSYOQ*’ TD] fa ar ar an a a a aa aan ° : jee OS BERG e eRe EO eave (SEES | Sngdtnoszed*TD! 4 GSISSERARS Se Gansemeg lFdeR S| SPTeIM en te Odi) oe Dip es Toate ete cA BIN aN =a Mine 4 Tseusny’sg| > KAKO MATSTNOA 1 1/1 Jon Nw ol ; op ROO NAMM AS OST donne et Se o® xf190°4¥ ss) PIS YY YY YR WAS nt ile ini cate SNTIEUTPIOVL4XE TOL] H qa ROU ° io} rs ALENON [OM cle ee eee ee AAA AAA SSNS VOnnOON NKrow 'a ® stsueseqeyu4y*gfe*ig] 2% ©) GD! GD GOO CO Gd cd 29 00 00 GD ad cd | eOIES SO’ e ‘ds erseqdo =) Tlie tonne on SUSE SS. SES Na ee ee OOOODODOOOOOOOOO DIANA Tada 1 el seodoruoeug sdsiC ese) 3 Sica utct, “olin commie S CARARA SA NRARAAAS SARA RRALSFER | ee So ee os OPE | rane om aaa Guonan ft GD Oj O MOM OM cOOdMOaON) \e | STEIOD SPUPUTOM* FeO) Bo ii titatt tit 1c na CSA tr Ty Pet : Fl le hy ye ef cd 3 76 & ma16 6 STAT aM i Ta A a inal na N T |S z-snuzedns UE AT Te cee i ESA Te a) eS HM it ; =| Nn oN 1s al us 4 g ie Jal. fe ie Aq 4 8&8 toa ot W dia: = oa = 5, o 2 da oO 30 0 Sj cae ie & 1 Bl a3 os Mple= a Atal a 3 yee Py ae le Sake 3 ' Be 5 alc ae 3 |e al 3 a i ey. A oO 8 s A SUES anolta a [os], 8 1 6 ala a Ss & p eee ee, wees Wl VUES 2 o ‘ a, ENN ac w [SeqtqoTtay,:ds snueerttr e fe aos ss a = Los, Or snqdqtnosut*yg Oe Oo | 2 ® we tase eee: sntpemzequt*pueg @| Odio = M NRsseuiat Squopouog StTde[Ttuts*oy ®@ A © ou Listas p> ls SNOTOTAOpIO*uy © oN a al ‘ds snqaderS0qdsTy e+—___+—_@ YS O|O OOO ee "ds snqderSoyq.zoereg e ob “all eeptoptoadedy e|/O0 al ee Radpeneerieetoreeae 29 71/0) ® x Ql stsuensfo°yo°1ej], @—® 5 snuiedns*duoT*TO @-@ al STSusUuTeIIOT*TN! @ a sngeasey sngegseuetTo e+———e ral ‘ds snaderSoadhkT4 ® 4 STUIOOTQ*IS xe°*TO @ Bl ‘ds snaderSou129 @ 2) sngepneo*is xe*TO e deds snaderSooeutjoopnesd ee sngeuoronutipenb*i3 xe°I09 oe a °ds snqaderSo0emtTo @-@ ~ couTu Ssnjeuro*y~ © . snuzedns *SuoT* 19 ee L sTie[nger*x e;,——® o) [SeatqoTtay, e_euoronu’Teq |e Fig. 7 Distribution of faunas in sections near Chineta village. ORDOVICIAN-SILURIAN BOUNDARY IN THE ALTAI MOUNTAINS 143 References Chen Xu 1984. The Silurian graptolite zonation of China. Can. J. Earth Sci., Ottawa, 21: 241-257. Holland, C. H. 1985. Series and Stages of the Silurian System. Episodes, Ottawa, 8: 101—103. Khomentovskiy, V. V. & Tesakov, Y. I. (eds) 1983. Resheniya Vsesoyuznogo stratigraficheskogo sovesh- chaniya po dokembriyu, paleozoyu i chetvertichnoy sisteme Sredney Sibiri, Novosibirsk, 1979. Ch. 1: Verkhniy proterozoi i nizhniy paleozoi. In: Mezhvedomstvenniy Stratigraficheskiy Komitet USSR. 215 pp. Novosibirsk. Moskalenko, T. A. 1977. Ashgill’skiye konodonty na Gornom Altaye. In A. B. Kanygin, et al., Problemy stratigrafii ordovika i silura Sibiri. Trudy Inst. Geol. Geofiz. Sib. Otdel., Novosibirsk, 372: 74-83. Sennikoy, N. V. 1976. Graptolity i stratigrafiya nizhnego silura Gornogo Altaya. Trudy Inst. Geol. Geofiz. Sib. Otdel., Moscow, 304. 270 pp., 17 pls. 1978. O nakhodke graptolitov zony persculptus na Gornom Altaye. In Novoe v stratigrafii i paleon- tologii nizhnego paleozoya Sredney Sibiri. Trudy Inst. Geol. Geofiz. Sib. Otdel., Novosibirsk, 141-144. ——,, Petrunina, Z. E., Gladkikh, L. A., Ermikoy, V. D., Zinov’eva, T. V., Mamlin, A. N. & Shokal’sky, S. P. 1984. Novye pogranichnye Ordoviksko-Siluriyskie razrezy na Gornom Altaye. Geol. Geofiz. 1984 (7): 23-27. ——, Puzyrev, A. A. & Russkikh, V. G. 1979. Ordovik i nizhniy silur rayona s.Ust’-Chagyrka (Gorniy Altai). In P. P. Kuznetsov (ed.), Problemy stratigrafti i tektoniki Sibiri: 30-45. Novosibirsk (Akad. Nauk SSSR, Sib. Otdel. Inst. Geol. Geofiz.). —— & Russkikh, V. G. 1982. Etalon llandoveriyskikh graptolitovykh zon na Gornom Altaye. Geol. Geofiz. 1982 (2): 28-35. Sennikoy, V. M. & Sennikoy, N. V. 1982. Stratigrafiya ordovika Anuysko-Chuyskogo sinklinoriya (Gorniy Altai). Geol. Geofiz. 1982 (6): 17-25. Severgina, L. G. 1978. Brakhiopody i stratigrafiya verkhnego ordovika Gornogo Altaya, Salaira 1 Gornoy Short. In J. I. Tesakov & N. P. Kulkov (eds), Fauna i biostratigrafiya verkhnego ordovika i silura Altaye-Sayanskoy oblasti. Trudy Inst. Geol. Geofiz. Sib. Otdel., Moscow, 495: 3-41, pls 1-6. —— 1984. Nekotoriye verkhneordovikskiye (Ashgillskiye) brakhiopody Gornogo Altaya. In Paleon- tologiya 1 biostratigrafiya paleozoya Sibiri. Trudy Inst. Geol. Geofiz. Sib. Otdel., Novosibirsk, 584: 39-48, pls 3, 4. Yolkin, E. A. 1983. Zakonomernosti evolutsii dekhenellid 1 biokhronologiya silura i devona. Trudy Inst. Geol. Geofiz. Sib. Otdel., Moscow, 571. 116 pp., 16 pls. ——, Obut, A. M. & Sennikov, N. V. 1978. O granitse ordovika i silura v Gornom Altaye. In B. S. Sokolov & E. A. Yolkin (eds), Pogranichniye slo ordovika 1 silura Altaye-Sayanskoy oblasti 1 Tyen- Shanya. Trudy Inst. Geol. Goefiz. Sib. Otdel., Moscow, 397: 5-14. —— & Zheltonogova, V. A. 1974. Drevneyshiye dekhenellidy (trilobity) 1 stratigrafiya silura Gornogo Altaya. Trudy Inst. Geofiz. Sib. Otdel., Novosibirsk, 130. 96 pp., 13 pls. it s) fe rh Nature of the Ordovician—Silurian boundary in south Kazakhstan, USSR M. K. Apollonov', T. N. Koren’, I. F. Nikitin’, L. M. Paletz! and D. T. Tzai‘ 1 Institute of Geology, Academy of Sciences of Kazakhstan SSR, Kalinina 69A, Alma-Ata 480100, USSR ? All-Union Geological Research Institute (VSEGEI), Sredni Prospekt 74, Leningrad 199026, USSR Synopsis Kazakhstan was the region where the coeval nature of the Dalmanitina mucronata—Hirnantia faunas with the persculptus Zone faunas was first established. The best sections are in the Chu-Ili Mountains of South Kazakhstan, the Ashchisu River and the Zhideli and Karasay sequences. A summary is given of the upper Ashgill and lower Llandovery biostratigraphy and the position of the systemic boundary. The litho- stratigraphy is also outlined. To have the Ordovician-Silurian boundary at the base of the acuminatus Zone was first advanced by Kazakhstan geologists (Rukavishnikova et al. 1968; Mikhailova 1970; Nikitin 1972; Apollonov et al. 1973; Apollonov 1974; Poltavtseva & Rukavishnikova 1972) after the discovery of Glyptograptus persculptus in association with Dalmanitina mucronata and Hirnantia in the Chu-Ili Mountains. This showed that the persculptus Zone did not succeed the Dalmani- tina beds, as was previously thought in western Europe, and that, on the contrary, it was partly coeval with the Dalmanitina mucronata—Hirnantia beds which have always been assigned to the Ordovician. Thus it became clear that tracing the persculptus boundary in the neritic facies was impossible. This new evidence has been widely discussed in the literature (Williams et al. 1972; Bergstrom et al. 1973; Lespérance 1974; Rozman 1976; Rickards 1976). The Kazakhstan Ordovician—Silurian boundary deposits are best studied in the Chu-Ili Mountains in south Kazakhstan, in the upper reaches of the Ashchisu River (Durben and Ojsu wells), as well as along the Zhideli and Karasay dry channels (Apollonov et al. 1980; Nikitin et al. 1980: textfigs 1-6). This paper is a summary of the upper Ashgill and lower Llandovery biostratigraphy and describes the position of the system boundary established in Kazakhstan on the basis of continuous sections. The succession is divided into three conformable lithostratigraphic units: the Chokpar, Zhalair and Salamat Formations. The latter is overlain by the Betkainar Formation (Figs 1—6). The Chokpar Formation consists of dark-grey and greenish-grey regularly bedded mudstones and siltstones yielding abundant graptolites characteristic of the supernus Zone (Apollonov et al. 1980). A more detailed zonation can now be suggested. The lowermost part of the Chokpar Formation contains Dicellograptus ornatus minor Toghill, Climacograptus longispinus supernus Elles & Wood, Amplexograptus inuiti (Cox) and Orthograptus amplexicaulis (Hall) and com- prises the inuiti Zone. The graptolites present above this, and in most of the Chokpar Forma- tion, are characteristic of the pacificus Zone and include Dicellograptus ornatus Elles & Wood, Climacograptus manitoulinensis Caley, Orthograptus socialis (Lapworth), Paraorthograptus paci- ficus (Ruedemann) (rare) and Nymphograptus velatus Elles & Wood. The uppermost Chokpar Formation locally contains limestone beds which are best developed in the Osju section where they are placed in a local stratigraphic unit—the Osju Limestones. The unit consists of dark- grey argillaceous limestones interbedded with aphanitic sandy limestones in which terrigenous clastics account for 15 to 20%. The Osju Limestones yields abundant brachiopods and tri- lobites including Giraldiella bella Bergstrom, Streptis altosinuata (Holtedahl), Leptaena rugosa Dalman, Cryptothyrella sp., Tscherskidium cf. ulkuntasensis Sapelnikov & Rukavishnikova, Prostricklandia prisca Rukavishnikova & Sapelnikov, Platycoryphe sinensis sinensis (Lu), Bull. Br. Mus. nat. Hist. (Geol) 43: 145-154 Issued 28 April 1988 146 APOLLONOV, KOREN, NIKITIN, PALETZ & TZAI e2 e3 Semipalatinsk Karaganda CENTRAL °6 KAZAKHSTAN Fig. 1 Localities of the Ordovician—Silurian boundary deposits in Central and South Kazakhstan. 1, Sarysu-Teniz watershed and Zhaksykon River; 2, Northeast of Central Kazakhstan—Kombabasor lake; 3, Akjar— Zhartas watershed; 4-6, Chingiz Range and Pre-Chingiz Range: 4, Mount Otyzbes; 5, Mount Mizek: 6, Mount Akdombak; 7-12, Chu-Ili Mountains: 7, Karasay River; 8, Zhideli River; 9, Anzhar River; 10, Ojsu well; 11, Durben well; 12, Mount Dulankara. @ALMA~ ATA Bumastus commodus Apollonov, Decoroproetus artus Apollonov, D. cf. evexus Owens, Otarion curvulum Apollonoyv, O. gibberum Apollonov, Dicranogmus confinis Apollonov, and Leonaspis sp. There also occur conodonts, bivalves, gastropods and cepalopods, among them Acodus similaris Rhodes, Eobelodina fornicala Stauffer, Icriodella sp., Tshuiliceras lobatum Barskov, Michelinoceras procurens Barskov and Geisonoceras fustis Barskov. The numerous graptolites that are characteristic of the pacificus Zone occur in argilliceous limestone layers. Present are Climacograptus longispinus supernus Elles & Wood, C. cf. normalis Lapworth, C. tatianae Keller, Glyptograptus posterus Koren & Tzai, G.? ojsuensis Koren & Mikhailova, Paraorthograptus pacificus (Ruedemann), Orthograptus amplexicaulis (Hall) and Plegmatograptus nebula lautus Koren & Tzai. Rare tabulate corals, radiolarians and algae are also known (Apollonov et al. 1980). The uppermost Chokpar Formation in other sections (as at the Anzhar River) is represented by massive biogenic-detrital limestones (the so-called Ulkuntas Limestones) overcrowded with tabulate corals and heliolitids. The assemblage includes Agetolites cf. mirabilis Sokolov, Hemi- agetolites insignis Poltavceva, Catenipora inordinata Kovalevsky, Plasmoporella papillatiformis Kovalevsky, Propora cancellatiformis Sokolov and Heliolites parvulus Kovalevsky. Some penta- merids such as Holorhynchus giganteus Kiaer, Proconchidium tchuilensis Rukavishnikova & Sapelnikov and Tcherskidium? ulkuntasense Rukavishnikova & Sapelnikov have been found. There occur the trilobites Holotrachellus punctillosus Tornquist, Amphylicas sp. and Sphaerex- ochus sp., which are characteristic of biohermal environments. The thickness of the Ojsu and Ulkuntas Limestones varies from 14 to 55m and the Chokpar Formation totals 140 to 180m. The Zhalair Formation rests conformably on the Chokpar deposits and is exposed in all sections studied. The section at Durben may serve as a stratotype (Figs 2, 3). The formation is composed of tobacco-green and greenish-grey siltstones interbedded locally with grey and reddish-brown fine-grained poorly sorted sandstones, the latter being of carbonate and quartz- feldspathic composition. Locally, sandstones form a separate unit more than 80m thick, for example at the Ojsu section. The lowermost Zhalair Formation includes the Durben Limestone which is 9 to 40m thick, and is easily discernible in many of the sections studied (Fig. 4). It consists of well-bedded dark grey pelitomorphic limestones. The upper part of the Zhalair Formation contains local beds of dark grey and green silty tuffites. The lower Zhalair Formation (the Durban horizon) contains graptolites of the extraordi- narius and persculptus Zones (Koren & Nikitin 1983). The former zone yields Climacograptus angustus (Perner), C. normalis Lapworth, C.? extraordinarius (Sobolevskaya) (=Glyptograptus? persculptus forma A and G. aff. persculptus of Apollonov et al. 1980) and Pseudoclimacograptus ORDOVICIAN-SILURIAN BOUNDARY IN SOUTH KAZAKHSTAN 147 Fig. 2 Schematic geological map of the Durben well area. 1, 2, Kysylsai Formation (?): 1, black siltstones and sandstones; 2, yellow sandstones; 3, Chokpar Formation black mudstones and siltstones; 4, 5, Zhalair Formation: 4, dark fine-crystalline and fine-clastic limestones; 5, green siltstones and fine-grained sandstones; 6, 7, Betkainar Formation: 6, basal conglomerate and sandstones; 7, grey sandstones; 8, red sandstones; 9, Koichin Formation: red sandstones and siltstones; 10, faults; 11, localities of fauna; 12, strike and dip. sp. The latter zone may be distinguished by the occurences of Glyptograptus persculptus (Salter) (=G. persculptus forma B of Apollonov et al. 1980), Glyptograptus sp. and Climacograptus angustus (Perner). A shelly fauna was found in limestones and siltstones within both graptolite zones, namely a typical Dalmanitina—Hirnantia assemblage including Platycoryphe sinensis (Lu), Dalmanitina mucronata (Brongniart), Dalmanitina olini Temple, Leonaspis olini Troedsson, Dic- ranopeltis sp., Dalmanella testudinaria (Dalman), Hirnantia sagittifera (M‘Coy), Anisopleurella novemcostata Nikitin, Aegiromena durbenensis Nikitin, Aphanomena ultrix (Marek & Havlicek), A. aff. urbicola (Marek & Havli¢ek), Bracteoleptaena polonica Temple, Eostropheodonta bublits- chenki Nikitin and Coolinia iliensis Nikitin. Fig. 3 A—Section on the north-east limb of the Ashchysu anticline near the Durben well. (a) the Chokpar Formation, (b—d) the Zhalair Formation: (b) limestones, (c) carbonaceous sandstone, (d) limestones, (e) siltstones, (f) Betkainar Formation; 354, f-287—localities of fauna. In the back- ground to the right are hills composed of coarse-grained sandstones of Betkainar Formation on the south-western limb of the anticline. B—enlarged part of the same section. C—section near the Ojsu well. (a) Ojsu Limestones of the uppermost part of the Chokpar Forma- tion; (b) limestones with Dalmanitina assemblage; (c) siltstone of the basal Silurian. In the fore- ground an exposure of the Ojsu Limestones is seen. D—transgressive onlapping of the basal conglomerate of the Betkainar Formation (b) on siltstones of the middle Zhalair Formation (a) in the Durben well area. Photographs I. F. Nikitin. ORDOVICIAN-SILURIAN BOUNDARY IN SOUTH KAZAKHSTAN 149 The thickness of the lower Zhalair Formation (the extraordinarius and persculptus Zones) varies from 122 to 127m in the southeastern Chu-Ili Mountains (the Durben and Osju wells), to 55m in the Zhideli River and to half a metre in the Karasay River in the northwestern Chu-Ili Mountains. The upper Zhalair Formation (the Alpeis horizon) yields early Silurian graptolites. The acuminatus Zone is well defined in the strata overlying the persculptus Zone in sections in the Karasay, Zhideli, and Ashchysu Rivers. The zonal assemblage includes abundant graptolites, namely Climacograptus acceptus Koren & Mikhailova, C.? jidelensis Koren & Mikhailova, C. mirnyensis (Obut & Sobolevskaya), C. ex gr. normalis Lapworth, Pseudoclimacograptus (Metaclimacograptus) fidus Koren & Mikhailova, P. (M.) pictus Koren & Mikhailova, Diplo- graptus modestus primus Mikhailova, G. madernii Koren & Tzai, Akidograptus cf. ascensus Davies, A. ascensus cultus Mikhailova, Parakidograptus cf. acuminatus (Nicholson) and Ortho- graptus illustris Koren & Mikhailova. The younger beds of the Zhalair Formation are eroded over most of the area studied (Fig. 4) and they are exposed only in the lower Karasay River. There, in beds overlying the acuminatus Zone, the graptolites Climacograptus miserabilis Elles & Wood, Glyptograptus sp. and abundant Priblylograptus sp. and Atavograptus sp., characteristic of the vesiculosus Zone, were found. The section is capped by strata yielding Climacograptus angustus (Perner), C. mirnyensis (Obut & Sobolevskaya), C. normalis Lapworth, Pseudoclimacograptus (Metaclimacograptus) hughesi (Nicholson), Coronograptus cyphus (Lapworth), C. gregarius (Lapworth), Monograptus revolutus praecursor Elles & Wood, Atavograptus sp. and Dimorphograptus dessicatus Elles & Wood. Shelly fauna is scarce in the Silurian part of the Zhalair Formation. In the acuminatus Zone only a single trilobite of the family Odontopleuridae occurs (exposure 280). The Zhalair Forma- tion is 51 to 133 m thick. The Salamat Formation consists of green sandstones and siltstones with abundant graptolites of the gregarius Zone. The overlying Betkainar Formation, with basal conglomerate beds, transgresses deposits of different ages, including in places the Dalmanitina mucronata beds of the Durben horizon (Figs 2, 4). The Ordovician-Silurian boundary in the Chu-Ili Mountains is drawn at the base of the acuminatus Zone, which is marked by the appearance of Akidograptus ascensus Davies, Glyp- tograptus madernii Koren & Tzai, Orthograptus illustris Koren & Mikhailova and Diplograptus modestus primus Mikhailova. The Chokpar and Zhalair Formations reflect a distinct regressive-transgressive cycle (Fig. 5). Dark pelitomorphic deposits of the Chokpar Formation (the supernus Zone) are comparatively deep-water and might have accumulated in an extensive, open, flat-bottomed sea with a remote source of terrigenous sediments. That sea was inhabited by diverse graptolites (more than 15 species). Towards the end of Chokpar time, the sea bed was elevated and a number of bio- hermal chains were developed. Each bioherm had a trail of clastic carbonate material (the Ulkuntas Limestones). In early Durben time (the extraordinarius Zone), the areas of continuously growing elevation were surrounded by thick beds chiefly consisting of limey coarse-grained sands (Fig. 6), and a broad band of the fine dark Durben Limestones accumulated which were 40m thick near the elevations and 0-5m thick further away. The areas of limestone sedimentation were inhabited by a trilobite assemblage including Dalmanitina mucronata, D. olini and Platycoryphe sinensis. In the deep-water limestones near the village of Karasay a single species of blind Dalmanitina was found. Brachiopods are commonly represented by the single species Bracteoleptaena polon- ica. The graptolite assemblage consists of 2 to 4 species, and all the fossils are large-sized, numerous but taxonomically restricted. Late Durben time (the persculptus Zone) saw the depo- sition of green fine-grained sandstones and cross-bedded siltstones with traces of turbidity and slumping. The benthic fauna shows a greater diversity (the Hirnantia—Dalmanitina assemblage) but the graptolites are limited to two to three species. An abrupt increase in the supply of tuffaceous material coincided with the beginning of the acuminatus Zone. A new and diverse (up to 15 species) graptolite assemblage appeared; however, benthic faunas are almost unknown from this level. The cosmopolitan distribution of APOLLONOV, KOREN, NIKITIN, PALETZ & TZAI 150 E B (ASHCHISU ANT SWISDU/PSODLXI } J © SECTION SVJOW10U JP 2D * sapsaduo 79° DUR NORTH Sppouiwnzp Jay * Sa]N0 snsuasD y—* 7 S720 (wd © A ° Sajid sijnovixaydwo 7) snpid (W) d —* * S/jnoo/xaj0Wod SipnoaixazOW/o 2 r : S/Svaljapi/ < jj 5 be a - ’ ro Sisuahunus 79 « snuseans snuros/Bu0p 79 — ED 22° SUENII QO ® nog SstlReny, snu/ds1u0} V7 pre, ¢ Sajsnavo 77° S710 Sdo 9° pjovosmnu/ (a snaidtiood snaitiaod, / C—* snqdowb G| sajsniuo dso }7* yusapow g»\ SmuBuipsonuzxa ]) ° snpousd 7 oxajOwy r 7 2 3 S iS waters Ce 3 © 2 “i a 8 a S y os _ eZ 6 6 8 8 8 f = 4 uy Of E aw f — > < =e 7 N i} x \ SAJDI/SSID Cs 3 | © 573.07 720001) W a Sdoydouwige (eR lx OS 10ND] p — 5 ) = = : SOMERS + ds 1dophgidg ‘ 10 os ouvoUzOg IP y ost 6 — Sq/J0021G 9 —=/bojap)g¢ FS, 5720016020 U//) 5 0 sisvansl ¢'gN | nana sy dho /00U0J0) « SISUAISD Dopiyy Day), 42/00 S/yn00/xa70wo 0 || “e iseyiny g< as sdoz0uz0pnasd « s sauseans suidsiduo} 77 a S C ~ snjsnbuo y-* —— sypulsou }7 —* a Jioz0h]9« * snoidioad snaisiood ¢ | | as \ jigosasid 19° sagsnbun jp © Snjou 1) 0U10 7 | 5 TP tess ~ = = = I-f] | oo oe ae qv Se ac => I 2 I °° c= om = = 2 Pe = 7 snoyiod | Int YuVdyWoOuHd) Nn a YuUVd WOH) UW Stet) hl SAY v MA MeL EY AN Te) tet MTA at NW IC td) 0108 sailuas WILSAS Fig. 4 Chart showing a correlation of the Ordovician—Silurian boundary deposits in the Chu-Ili Mountains. 1, black mudstones, siltstone and silty mudstones; 2, dark grey to black mudstone and siltstone; 3, grey tuffaceous pelite, tuffaceous mudstone; 4, grey siltstone, mudstone, fine-grained detrital 6, sandstone; 5, dark grey fine-crystalline evenly bedded limestone, sometimes clayey; 151 ORDOVICIAN-SILURIAN BOUNDARY IN SOUTH KAZAKHSTAN WeEtLL ICLINE ) ANZHAR RIVER SECTION SOUTH SISUIISD 19 -y © SI} /JojaTo ida * 03)2)/jojaDy SNSODN) J * S, n8090}0 ¢ . snaquooia SnazUoOW fy 0 ISUIZINGI} sq * Sn)1d S218 ONS HOU & WUsepou 09 6 * 110. SPUIM Saysepgu qe 00001) Sash oe - a sop thyg SISUAUIS. SISUBUIS sysuens/o SiSuauis a ae SISOU/DISODIS2~F TI . \ SNIINIOT SNIIIOC SIJDUOU LO Xa 79 = = OH 6 Sypousou S777 as -1037,-103 ®-1039 =] P-1039°q-295 . ad ad Ts A puie a6 Ue | 5 NX DIIUojod ig ningun ‘ta Moy * \ origin ye ELEY a aoj4n9 ° Sisueuaginp yy * c Daa opjeUnujD, | M9 1 p ee SISUAUIS J © wey a popwoujog C S — _- —y Ee WHE MND 06 D}0U0IINU «—* as Jopiyp + siynoovraiap 5/70" Wsepou 9 S11 DUI0I00.1 9X8 77 —-+ smu saans sn sijousou ii xa 77 * sisuahusius 4979 ° *-1dsiBu0} S7]00090 Jp * DELETE De as, 10oj0h 9" SSOLJOAY S: snjsnluo 77 —= = ~*-10s/dU0; a + ‘| Sisvauegnp yy — THUAYISINGNG 7. I p2jvojod sg 2J0919I0 * Dsadiz3/0DS fy 1unja’7 SISUAUIS SiSUaUIS “J — ° c . - | 10 SA ZOQIISIO ¢ ae 1 Davo S1SOUIPSODSJXI JP « | sysvahuu 4979 * | snsnduo 77 * limestone; 7, bioherm limestone; 8, middle and coarse-grained polymictic sandstone; 9, conglomer- ates and coarse-grained polymictic sandstones; 10, tuffaceous sandstone; 11, fine-clastic acid tuff and tuffite; 12, diorite sill; 13, non-deposition; 14, fossils: (a) graptolites, (b) trilobites, (c) brachio- pods. VN NO APOLLONOV, KOREN, NIKITIN, PALETZ & TZAI ONS | n I. Omen PAL. url iru OERRES N z = || ©) ro N |KARASAY ZHIDEL] DURBEN ORIRSRU ANZHAR pa RIVER RIVER WHE Ec SPRING RIVER _ ge ll| nN IT eo aT a eS = = @\— Bs — a < SS/ia Te eat al se re) Saar re S Sd z ies lu aX ros) ~icghk oa > $3} aims e oy LY « fy 3| fj e | ee ea |Qs}-—-— -—- — — — | —— ee ee ee Te e ee ee (=) Ht =x Se Be ee eee ee eee (=) (ok Be DE: Oe Es ES) ae Ep [IT vatoasness Fig. 5 Chart showing the lateral variation of different lithogenetic types within the Ordovician— Silurian boundary interval in South Kazakhstan. D, Durben Limestones; O, Ojsu Limestones; U, Ulkuntas Limestones. 1, black mudstones; 2, green sandstone; 3, detrital thin-bedded micro- crystalline limestone; 6, sandstone; 7, conglomerate and gritstone; 8, tuffite; 9, unconformity; 10, non-deposition; 11, fossils: (a) trilobites, (b) graptolites, (c) brachiopods, (d) corals, (e) other fauna groups. the acuminatus graptolite assemblage may be due to the widespread early Llandovery transgres- sion. A great crisis in graptolite evolution within the extraordinarius and persculptus Zones took place at the end of the Ordovician regressive cycle. The basal lower Silurian deposits (the acuminatus Zone) outside the Chu-Ili Mountains are established in eastern Central Kazakhstan in the Otyzbes Mountains, near the Kombabasor Lake east of the town of Bajanaul and at the watershed of the Akzhar—Zhartas Rivers north- east of Karaganda (Bandaletov 1969; Apollonov et al. 1980; Fig. 1 herein). The uppermost Ashgill deposits (the Dalmanitina mucronata beds of the Durben horizon) are known from the Zhaksykon River basin at the Sarysu-Teniz watershed in the Chingiz Range (near the town of Akdombak) and south-western Chingiz area (Nikitin 1972; Nikitin et al. 1980). The systemic boundary in the regions within the neritic development is defined by the appearance of the diagnostic brachiopods Eospirifer cinghizicus and Holorhynchus cinghizicus and tabulate corals (Borisyak et al. 1969; Nikitin 1972). However, direct correlation between the graptolite and shelly sequences within the Silurian basal beds is still not fully established, and the problem of the identification of shelly faunas diagnostic of the acuminatus Zone remains open in Kazakhstan as elsewhere. ORDOVICIAN-SILURIAN BOUNDARY IN SOUTH KAZAKHSTAN 153 Fig. 6 Schematic depositional patterns in the South Kazakhstan Palaeo-basin in the uppermost Ordovician. A (supernus Zone): a, coarse sandstones; b, biohermal (Ulkuntas) limestones; c, detrital (Ojsu) limestone; d, microitic (Ojsu) limestone; e, black (Chokpar) mudstones. B (extraordinarius and persculptus Zones): a, coarse sands; b, biohermal limestones; c, thin-bedded micritic (Durben) limestones with Dalmanitina association; d, fine sandstones with Dalmanitina— Hirnantia association. Arrows indicate the direction of transport of the clastic material. 154 APOLLONOV, KOREN, NIKITIN, PALETZ & TZAI References Apollonoy, M. K. 1974. Ashgillskie trilobity Kazakhstana [ Ashgill trilobites in Kazakhstan]. 136 pp. Alma- Ata. ——, Bandaletoy, S. M. & Nikitin, I. F. (eds) 1980. [ The Ordovician—Silurian boundary in Kazakhstan]. 300 pp. Alma Ata. [In Russian]. ——, ——,, ——.,, Paletz, L. M. & Tzai, D. T. 1973. K probleme granitzy ordovika i silura v Chu-Iliyjskikh gorakh Cans Kazakhstan) [On the Ordovician-Silurian boundary in Chu-Ili Mountains, South Kazakhstan]. In: Informatsionny sbornik nauchno-issledovatel’skikh rabot [for 1972]: 23-25. Alma-Ata, Nauka. Borisyak, M. A., Kovaleyski, O. P. & Nikolaeva, T. V. 1961. K stratigrafii silura khr. Chingiz [On the Silurian stratigraphy in the Chingiz Range]. Informatsionny sb. VSEGEI 2: 61-69. Keller, B. M. 1956. Obschij obzor stratigrafii ordovica Chu-Ilijskikh gor [General review of the Ordovi- cian stratigraphy in the Chu-Ili Mountains]. In: Ordovik Kazakhstana: 5—49. Izdatel’stvo Akad. Nauk SSSR. Koren, T. N., Sobolevskaya, R. F., Mikhailova, N. F. & Tzai, D. T. 1979. New evidence on graptolite succession across the Ordovician—Silurian boundary in the Asian part of the USSR. Acta palaeont. pol., Warsaw, 24: 125-136. —— & Nikitin, I. F. 1983. Comments on the definition of the Ordovician-Silurian boundary. Eesti NSV Tead. Akad. Toim., Tallinn, (Geol.) 32 (3): 96-100. Lesperance, P. J. 1974. The Hirnantian fauna of the Percé area (Québec) and the Ordovician—Silurian boundary. Am. J. Sci., New Haven, 274: 10-30. Mikhailova, N. F. 1970. O nakhodke Glyptograptus persculptus (Salter) vy dal’manitinovykh sloyakh Kazakhstana [On the occurrence of Glyptograptus persculptus (Salter) in the Dalmanitina beds of Kazakhstan]. Eesti NSV Tead. Akad. Toim., Tallinn, (Khim. Geol.) 19: 177-178 [In Russian with Engl. summ. ]. Nikitin, I. F. 1971. The Ordovician system in Kazakhstan. Mem. Bur. Rech. geéol. miniere., Paris, 73: 337-343. —— 1976. Ordovician-Silurian deposits in the Chu-Ili mountains (Kazakhstan) and the problem of the Ordovician-Silurian boundary. In M. G. Bassett (ed.), The Ordovician System: 292-300. Cardiff. ——,, Apollonoy, M. K., Tzai, D. T. & Rukavishnikova, T. B. 1980. Ordovikskaja sistema [The Ordovician system]. In: Chu-Ilijskii rudnyi poyas. Geologia Chu-Ilijskogo regiona: 44-78. Alma-Ata. Poltavtseva, N. V. & Rukavishnikiva, T. B. 1973. Granitsa ordovikskoj i silurijskoj sistem vy Chu-Ilijskikh gorakh [The Ordovician-Silurian boundary in the Chu-Ili Mountains]. In: Materialy po geologii i poleznym iskopaemym Yuznogo Kazakstana: 28-38. Alma-Ata. Rickards, R. B. 1976. The base of the Silurian System in the British Isles. In: Graptolity i stratigrafia: 152-153. Tallinn, Valgus. Rozman, K. S. 1976. Granitsa ordovika 1 silura [The Ordovician-Silurian boundary]. In: Granitsy geo- logicheskikh sistem: 72-93. Moscow, Nauka. Rukavishnikova, T. B., Tokmacheva, S. G. & Salin, B. A. 1968. Novye dannye po stratigrafii otlozhenii verkhnego ordovika i nizhnego silura Chu-Ilijskikh gor [New evidence on the upper Ordovician—lower Silurian stratigraphy in Chu-Ili Mountains]. Dokl. Akad. Nauk SSSR, Leningrad, 183: 420-423. Williams, A., Strachan, I., Bassett, D. A., Dean, W. T., Ingham, J. K., Wright, A. D. & Whittington, H. B. 1972. A correlation of Ordovician rocks in the British Isles. Spec. Rep. geol. Soc. Lond. 3. 74 pp. The Ordovician—Silurian boundary in Saudi Arabia H. A. McClure Arabian American Oil Company, Box 2376, Dhahran, Saudi Arabia Synopsis An account is given of the environments of deposition across the Ordovician—Silurian boundary which occurs within the Tabuk Formation, Saudi Arabia. The results of much recent work are appraised and earlier conclusions are reassessed with respect to it. The late Ordovician glaciation is considered to have been a prime factor influencing sedimentation, for example by restricting land derived clastic input. There appears to be no regionally significant depositional hiatus, and the beds about the boundary are best regarded as conformable. The general environment of deposition was of prograding sandstones, tidal flats, delta cycles and intermittent marine incursions on a tectonically stable structural platform. A basically normal graptolite sequence is deduced across the boundary region, and a précis is given of the relative dating achieved by these and other fossil groups. Introduction Early Palaeozoic rocks of the Arabian Peninsula are almost exclusively siliciclastics whose primary source was erosion from the western part of the Precambrian Arabian Shield. These rocks were successively deposited to the east along a regressive shoreline in fluvio-deltaic and shallow water marine environments. The Ordovician—Silurian boundary in Saudi Arabia occurs within the Tabuk Formation of this suite of rocks. The Tabuk Formation was originally designated by R. A. Bramkamp in 1954 in an unpub- lished report of the Arabian American Oil Company. His definition in amended form was presented on U. S. Geological Survey Miscellaneous Geologic Investigations Map I-270A (1963). The formation was formally defined by Powers et al. (1966). A summary of details of the formation is given by Powers (1968). The type section, in the Tabuk area of northwest Saudi Arabia (Fig. 1), consists of 1071 m of shale, siltstone and sandstone, deposited in shallow water in a complex of fluviatile, littoral beach, deltaic, and tidal flat sediments. Holomarine shale members, recording marine transgres- sive phases, occur at the base, near the middle and near the top. These are designated, respec- tively, the Hanadir, the Ra‘an, and the Qusaiba shales. However, in the vicinity of the type section, only the basal member, the Hanadir, shows easily mappable lateral continuity. Powers (1966) designated a reference section of 677-2m thickness in the Qasim (Qusaiba) area (Fig. 1) which is a composite section from several excellent exposures in the vicinities of Jebal Hanadir, Khashm Ra‘an, and in the Qusaiba depression. For the local definition of the Ordovician-Silurian boundary this section is best, with the advantages that (1) all three holo- marine shale members are well developed and well exposed, (2) all three shale members are graptolite-bearing, (3) additional fossil collections, including graptolites and trilobites, have been made in more recent years and serve to refine previous age assignments and strati- graphical relationships within the formation on outcrop as well as in subsurface areas several hundred kilometres to the east, and (4) glacial beds recording an ‘end-of-the-Ordovician’ glaci- ation event and stratigraphical and sedimentary details have been recently studied in the area. Fig. 2 shows a generalized stratigraphical section in the Qasim area. Stratigraphy and sedimentation Rocks of the Tabuk Formation were deposited in shallow water on a very broad and extensive, gently sloping epicontinental shelf, reflecting the underlying basement structure of a nearly flat, gently dipping, stable homoclinal platform (Powers et al. 1966). Present dips on outcrop average less than 2° eastward and have been little disturbed since deposition. Graptolitic shales Bull. Br. Mus. nat. Hist. (Geol) 43: 155-163 Issued 28 April 1988 156 H. A. MCCLURE Tabuk Fm. Outcrop SAUDI ARABIA 4 Arabian Sea Fig. 1 Outcrop map of the Tabuk Formation in Saudi Arabia. Equivalent rocks on the surface and in the subsurface have been found as far east as Oman. and sands with other macrofossils and trace fossils as well as a palynomorph suite of chitin- ozoans, acritarchs and plant spores occur on surface outcrop as well as in the deeper subsurface section of the eastern part of Saudi Arabia and Oman. The Tabuk Formation gradually thickens basinward to the east, where it is extensive in the subsurface, but, except that the three marine shales tend to be less distinct as discrete units, no gross changes in facies or depositional environment are apparent. Carbonate beds are known only as rare thin lenses at outcrop. Lithologies of the Tabuk Formation comprise three basic types: (1) medium-grained, partly cross-bedded, partly massive-bedded, channelled sandstone, (2) fine-grained, laminated and ripple-marked, micaceous sandstones and siltstones, and (3) laminated and micaceous, fossil- iferous shales, the Hanadir, the Ra‘an and the Qusaiba. The shales grade upwards through siltstone interbeds at the top into type 2 lithology. Tabuk lithologies are thus arranged in three generally coarsening-upward cycles that, together with the regional sedimentological and struc- tural framework, suggest deposition in a prograding deltaic environment dominated by fluvial sediment input. Lithology type 1 probably represents material derived from a fluvially-fed delta plain and deposited in the distributary system of a delta front; type 2, fine sandstone and siltstone, may have been deposited in intermediate mouth bars; and type 3 is considered to represent a mud-dominated platform in the pro-delta, off-shore area, where holomarine condi- tions prevailed. Each of the three cycles from bottom to top probably represents sand and silt ORDOVICIAN-SILURIAN BOUNDARY IN SAUDI ARABIA Sy) facies of a delta plain and delta front prograding during periods of eustatic stand-stills over pro-delta muds, which were the product of intermittent, possibly eustatically controlled, marine incursions. Intertidal deposition as part of the delta plain appears to have been widespread, Skolithos beds and tidalite sands being prominent towards the top of the Ordovician part of the Tabuk Formation on outcrop as well as in the subsurface. A non-barred, tidally dominated, sandy coastline was probably present, where extensive fluvially-dumped sediments were contempora- neously reworked and redistributed during periods of active progradation. Graptolite zonations, documentation of the glacial event, and sedimentary observations are the principal aids available in the area to define the nature of the Ordovician-Silurian bound- ary. Analysis of the Ra‘an and Qusaiba shales and the intervening sandstone is particularly informative, since these units bracket the boundary. (The Hanadir, the basal shale member of the Tabuk, of Llanvirn age, is not discussed here, except briefly in the section on bio- stratigraphy below.) The Ra‘an is the least distinctive and persistent of the three shale members. At the type locality at Khashm Ra‘an (latitude 26° 52’ N, longitude 43° 23’ E), the lithology consists of 67m of green-grey, silty, micaceous shale and fine-grained, red-brown, ripple marked, micaceous sandstone and siltstone with trace fossils towards the top. Glacial beds are erratically associ- ated with the top of the Ra‘an in many places at outcrop. Very rare graptolites, trilobites, brachiopods and molluscs are concentrated in several thin zones at the bottom and top. The range of the graptolite Orthograptus amplexicaulis, which occurs in the lower part of the Ra‘an, is from the clingani Zone to the anceps Zone. Glyptograptus persculptus occurs at the top of the Ra‘an and, although formerly considered to represent the lowest Silurian, is now taken as uppermost Ordovician. The trilobites indicate a less precise age ranging from about the middle Caradoc to about the late Ashgill. Overall considerations indicate the Ra‘an member at outcrop is probably late Caradoc to the latest Ashgill, persculptus Zone, in age. The sandstone overlying the Ra‘an, which is similar to the sandstone underlying it, is partly cross-bedded, partly massive-bedded, medium-grained and occasionally channelled. This unit, about 240m thick in the Qasim area, is probably lower Rhuddanian in age because of its conformable position below the well-dated Aeronian Qusaiba shale and above the persculptus Zone. It is generally barren of body fossils, but poorly preserved moulds of molluscs and brachiopods (mostly lingulids) are sometimes present. Trace fossils such as Cruziana are frequent. The Qusaiba shale, like the Ra‘an, is erratically distributed along the length of the outcrop. At its best exposure and type locality in the Qusaiba Depression (26° 53’N, 43° 34’E), it consists of 57m of varicoloured, red and grey-green laminated shale with thin interbeds of yellow shale, and red, hematitic, ripple-marked, micaceous and fine-grained sandstone with trace fossils towards the top. A medium-grained, cross-bedded sandstone overlies the shaly-silty interbedded unit. The Qusaiba is especially rich in graptolites, but also contains rare trilobites, brachiopods and molluscs. Graptolites serve to date the Qusaiba as Aeronian, convolutus Zone. Nature of the Ordovician—Silurian boundary In the Arabian section, both on outcrop and in the subsurface to the east, the persculptus Zone is present near the top of the Ra‘an shale. On the surface, persculptus occurs just below the glacial beds. While this zone has not been documented above the glacial horizon on outcrop, in the subsurface it occurs just above a diamictite suspected of being of glacial origin (Fig. 2). The contact between the Ordovician and the Silurian, both at outcrop and in the subsurface further to the east, is apparently conformable. Nothing appears to have happened across the boundary that drastically altered the depositional mode characteristic throughout the Tabuk Formation of prograding sands, delta cycles, and intermittent marine incursions on a tectoni- cally stable structural platform. Within the Ra‘an, however, extreme cold water conditions were apparently manifested in an impoverished fauna, and local glacial activity took place within the top part of the Ra‘an. Fluvioglacial channels, tillite deposits, striated and faceted megaclasts, 158 H. A. MCCLURE exotic igneous boulders, pro-glacial sandstone, and other evidence of glaciation occur in this part of the section (McClure 1978; Young 1981). This event is assumed to be approximately coeval with glaciation at this time in north Africa (Beuf et al. 1969; Hambrey & Harland 1981). The glaciation in Saudi Arabia is confined within the top part of the Ra‘an, apparently within the persculptus Zone, but is ice-marginal and ice-contact and not glacio-marine. Sub- aerial exposure due to sea level drop at the maximum of glaciation during the later phase of the Ashgill probably occurred. Thus, super-imposed upon the Ra‘an is a subsidiary sequence of events composed of (1) glacio-eustatic sea level regression, during which glaciation took place, (2) deglaciation during which glaciofluvial and fluvial sands were deposited, finally followed by (3) glacio-eustatic sea level rise, during which the upper part of the persculptus Zone shale was deposited. Sea level dropped again toward the beginning of the Silurian and regressive sands were deposited in Rhuddanian time. In later Llandovery (Aeronian) time, a marine transgres- sion apparently unrelated to glacial events deposited the Qusaiba shale. The glacio-eustatically controlled regressive—transgressive sequence at the top of the Ra‘an may be synchrononous with similar world-wide events such as those documented by Brenchley & Newall (1980) at the end of the Ordovician in the Oslo region, Norway, and those proposed by Berry & Boucot (1973). The Ordovician—Silurian boundary in Saudi Arabia may thus be taken at the base of the sandstone unit between the Ra‘an and Qusaiba shales, or above the persculptus Zone. Lithofacies to the east in the deep subsurface vary little from the outcrop sequence, except that the Ra‘an as a discrete shale unit with easily determined top and base is not always present and the sandstone of presumed Rhuddanian age between the Ra‘an and the Qusaiba at outcrop is poorly developed. The contact between the Ra‘an and the Qusaiba is consequently indistinct, and the Qusaiba sequence is considerably thicker. A distinctive feature of the subsurface is a regionally persistent and prominent, thin, highly organic, pyritic euxinic black shale, often bearing common or abundant Glyptograptus persculptus with no benthic fossils and overlying a sandstone with diamictite suspected of being equivalent to the glacial tillite of outcrop. This shale may be equivalent to the post-glaciation upper part of the persculptus Zone of outcrop mentioned above and helps place the glaciation event as within the persculptus Zone. The graptolite succession of the deep subsurface requires further study, but appears similar to that of the outcrop. Several differences are that graptolites assignable to the clingani to anceps Zones as found at the base of the Ra‘an on outcrop have not been documented in the sub- surface, and a graptolite suite assignable approximately to the boundary between the magnus and leptotheca Subzones of the gregarius Zone has been recovered in one drill hole. The most perplexing anomaly, however, is that, in another representative drill hole with continuous core sequence, a convolutus Zone graptolite suite occurs within about 6m of the euxinic persculptus Zone shale. Intervening graptolite zones of the lower Llandovery therefore appear to be largely missing or drastically telescoped. (See Note, p. 163). The ‘missing’ graptolite zones are assumed to be represented on outcrop by the non- fossiliferous Rhuddanian sandstone and their apparent absence in the deeper section where this sandstone is not present and shales were continuously deposited is puzzling. However, these zones are also ‘compressed’ in some standard British successions (R. B. Rickards, personal communication) and lower Llandovery marine fossils are rare on a worldwide basis (Berry & Boucot 1973). The apparent gap in the graptolite succession of Saudi Arabia is probably not due. to events peculiar to the Arabian platform. It is tempting to consider the euxinic black shale as well as the condensing or absence of the early Llandovery graptolite zones as in some way related to the glaciation event. Cessation or drastic restriction of fluvial flow regimes and consequent constriction of clastic input due to tie-up of water in glacial ice may have resulted in stagnant, euxinic conditions in more distal intra-platform areas on what was already a cold water, carbonate-starved platform. Fig. 2 presents outcrop and subsurface correlations within the Tabuk Formation. Thus, a firmer calibration of a time scale with depositional and climatic events and faunal occurrences is essential to define more precisely the nature of the Ordovician-Silurian bound- ary on the Arabian platform and correlate it with sequences elsewhere. The evidence accumu- lated to date, however, is informative, and the following conclusions can be tentatively made. ORDOVICIAN-SILURIAN BOUNDARY IN SAUDI ARABIA 159 W E Outcrop 600km +» Subsurface Graptolite Zones , Graptolite Zones convolutus Llandovery SILURIAN — convolutus > euxinic sh. persculptus el a a ees = aoe ee aed cf, )| COCO aCO CON | glacial sediments CO=GIEGIENIEN CamEEone * fp .€ Os glacial sediments persculptus Caradoc~—Ashgill clingani-anceps Llandeilo murchisoni 2 < O > ie) fa) c e) murchisoni Llanvirn Shale Sandstone Fig. 2 Section comparing the Tabuk Formation in the outcrop of NE Saudi Arabia (left) with that in the subsurface to the east (right). The three shale horizons at outcrop are termed Hanadir (Llanvirn), Ra‘an (Caradoc to basal Llandovery) and Qusaiba (Middle Llandovery). The Idwian Stage shown is now regarded as the lower part of the Aeronian Stage. The base of the Tabuk Formation is at the base of the Llanvirn. 1. Rates of sediment deposition may have varied on the Arabian platform across the Ordovician-Silurian boundary and can probably be attributed to the effects of glaciation. Land-derived clastic input may have been drastically restricted, resulting in euxinic, starved and stagnant areas, but: 2. No regionally significant depositional hiatus is evident and the contact between the Ordo- vician and Silurian may be considered conformable. 3. Nothing except glaciation happened at the boundary to upset significantly the mode of deposition characteristic throughout the Tabuk Formation of prograding sandstone, tidal flats, delta cycles and intermittent marine incursions on a tectonically stable structural platform. 4. The graptolite succession across the boundary appears normal, the apparent gap of early Llandovery graptolite zones being probably attributable to world-wide events and not intra- platform activity. 5. The significant boundary event on the Arabian platform appears to have been the glaci- ation at the end of the Ordovician that affected sedimentation rates and faunal suites. 160 H. A. MCCLURE Biostratigraphy Early fossil collections listed by Powers et al. (1966) and Powers (1968) were sparse. Additional surface collecting in more recent years in the Qasim (Qusaiba) area has provided fossils that serve to refine previous age assignments as well as to reveal more about the palaeobiology of the Tabuk Formation and faunal events across the Ordovician-Silurian boundary. Drill hole cores available in recent years from the deep subsurface of the eastern part of Saudi Arabia, where rocks across the boundary are extensively distributed, also provide useful information. All three shale members of the Tabuk Formation, the Hanadir, the Ra‘an, and the Qusaiba, are fossiliferous holomarine shales deposited as pro-delta muds. Intervening sands are largely of tidalite and shoreface origin and are mostly unfossiliferous of body fossils, but frequently contain trace fossils including Skolithos and Cruziana. Thin siltstone beds near the tops of the shales rarely contain poorly preserved moulds of bivalves, brachiopods (lingulids) and tri- lobites. All three shale units contain graptolites, trilobites, and an assortment of benthic fauna in addition to palynomorphs (chitinozoans, acritarchs, and spores). Except for graptolites, trilobites, and palynomorphs, the fossil suite has been little studied. R. B. Rickards has been working with the graptolites in recent years; Thomas (1977), Fortey & Morris (1982) and El-Khayal & Romano (1985) have studied some of the trilobites, H. A. McClure is working on chitinozoan and acritarch suites and J. Gray, A. J. Boucot and H. A. McClure are currently investigating spores of possible land plant affinity. The following analysis should be considered preliminary. The following lists are comprehensive compilations from both outcrop sequences (Qasim area) and cored holes of the subsurface to the east. Though not strictly pertinent to the boundary problem, the fossils of the Hanadir shale at the base of the Tabuk Formation are also listed. The Hanadir at its type section (26° 27’N, 43° 27’ E) consists of 60m of varicoloured, laminated, micaceous shale, with thin, red-brown, ripple marked siltstone and fine grained sandstone at the top. Fossils of the Hanadir include: Graptolites: Didymograptus murchisoni murchisoni (Beck), D. murchisoni geminus (Hisinger), D. pakrianus Jaanusson, D. aff. D. acutus Ekstrom, Amplexograptus cf. A. coelatus (Lapworth), A. sp. Trilobites: Neseuretus tristani (Desmarest), Plaesiacomia vacuvertis Thomas, ?Marrolithus sp. Cephalopod: Orthoceras sp. Brachiopods: ?Monobolina sp. and other articulate species and unidentified lingulids. Molluscs: ?Glyptarca sp., unidentified bivalves, unidentified gastropods. Beyrichids and other unidentified ostracodes; unidentified conodonts and palynomorphs (chitinozoans, acritarchs, spores, and scolecodonts). Based mainly on the graptolites, the Hanadir is Llanvirn in age, murchisoni Zone. The Ra‘an shale contains the following fossils, derived mainly from several thin zones at the base and toward the top from outcrop and from cores of the subsurface: Graptolites: Glyp- tograptus persculptus (Salter) s.s., Orthograptus amplexicaulis Hall s.s., Orthograptus sp. nov., Diplograptus sp., Climacograptus angustus/normalis, ?Dictyonema sp., ?Climacograptus misera- bilis and ?Diplograptus modestus. Trilobites: Kloucekia sp. and Onnia sp. Brachiopods: Com- atopoma sp. or Hirnantia sp., others (mostly lingulids). Molluscs: unidentified gastropods and bivalves and the cephalopod Orthoceras sp.; unidentified conodonts and palynomorphs (chitinozoans, acritarchs, spores, and scolecodonts). The range of Orthograptus amplexicaulis is from the clingani to the anceps Zones. Glyp- tograptus persculptus places the top part of the Ra‘an in the persculptus Zone. The Qusaiba shale contains the following: Graptolites, Suite 1: Climacograptus scalaris (Hisinger), C. aff. C. rectangularis Tornquist, Glyptograptus aff. G. incertus Elles & Wood, G. tamariscus tamariscus (Nicholson), G. (Pseudoglyptograptus) sp., Lagarograptus sp., Mono- graptus capis Hutt, M. communis Lapworth, M. convolutus (Hisinger), M. decipiens Tornquist, M. gregarius gregarius Lapworth, M. lobiferus (M‘Coy), M. cf. M. delicatulus (Elles & Wood), M. ex gr. tenuis (Portlock), Orthograptus cyperoides Tornquist, Petalograptus ovatoelongatus (Kurk), P. sp., Pristiograptus regularis (T6rnquist), Pseudoclimacograptus (Clinoclimacograptus) retroversus Bulman & Rickards, P. (Pseudoclimacograptus) sp. nov., Rastrites spina Richter, Retiolites perlatus (Nicholson), Rhaphidograptus tornquisti Elles & Wood. Graptolites, Suite 2: Climacograptus tamariscus s.l., Coronograptus gregarius cf. C. minisculus Obut & Sobolovskaya, ORDOVICIAN-SILURIAN BOUNDARY IN SAUDI ARABIA 161 Climacograptus cf. C. rectangularis, Diplograptus cf. D. magnus, Monograptus lobiferus (M‘Coy), Pristiograptus ?concinnus. Trilobite: Platycoryphe dyaulax Thomas. Bivalves: Nuculites, among others. Bellerophontids, unidentified gastropods; the cephalopod Orthoceras sp.; brachiopods: ‘Camarotoechia’ and other unidentified articulates. Unidentified conodonts and palynomorphs (chitinozoans, acritarchs, spores, and scolecodonts); ostracodes, Tentaculites, ophiuroids and fish remains. On graptolite evidence of Suite 1, the outcrop of the Qusaiba is Llandovery, Aeronian Stage, convolutus Zone. A slightly older zone in the subsurface is represented by Suite 2, from the gregarius Zone, approximately on the boundary between the magnus and leptotheca Subzones, still within the Aeronian. Palaeoecology and Palaeobiogeography The fossil content of the Tabuk Formation was the product of a remarkably stable environment and static physical conditions for a considerable period of time. Two kinds of faunal association are represented in the Tabuk suite: (1) planktic, with graptolites, chitin- ozoans and acritarchs, and (2) level-bottom benthic, with an epifauna of what were probably mostly vagrant shelly benthos such as brachiopods, molluscs, trilobites and ostracodes. Other taxa such as conodonts, scolecodonts and ophiuroids are also represented. Fine layering and lamination and lack of bioturbation of the shales indicates that a significant infauna was probably not present. In general terms, population densities were high for the planktic level and low for the benthic. Diversity was moderately high for the graptolites and very high for the chitinozoans and acritarchs. Shelly benthics identified to date indicate a low diversity. Continuity of the Tabuk suite extends for hundreds of kilometres, the fossils known from cored sequences in deep drill holes in the subsurface further to the east do not differ signifi- cantly from those of outcrop. There are no obvious indications that any element of the Tabuk biota is allochthonous. In the shales of the Tabuk, graptolites are common but of low diversity in the Hanadir, rare and of low to moderate diversity in the Ra‘an, and abundant and of high diversity in the Qusaiba. Molluscs (especially bivalves), brachiopods, trilobites and ostracodes are the next most common taxa, occurring in about equal abundance and approximately equal diversities. The shelly benthos is certainly not brachiopod-dominated as in more northern biogeographic realms. Conodonts occur in all three shale members, but are very rare and to date very little is known about them. Scolecodonts occur as a minor part of the palynomorph suites. Chitin- ozoans and acritarchs are common to abundant and of high diversity in the Hanadir, compara- tively rare and of comparatively low diversity in the Ra‘an and abundant and of high diversity in the Qusaiba. Spores, including tetrahedral tetrads that possibly represent early vascular land plants, are rare to common in the palynomorph suites. Ophiuroids are very rare; only several specimens of less than 0:5cm size are recorded from the Qusaiba. Tentaculites occurs rarely in the Qusaiba. Orthoceras is rare but ubiquitous in all three shales, being more common and robust in the Qusaiba. In one limited locality, near the base of the Ra‘an, it is concentrated in small planoconvex lenses of calcareous debris associated with algal nodules. (This is the closest resemblance to the Orthoceras limestone lenses recorded as widespread in the Silurian of north Africa by Berry & Boucot, 1973. The Arabian occurrence possibly represents a storm event.) All the shelly benthic species are small, brachiopods and molluscs being rarely more than one centimetre in maximum dimension. Shelly specimens appear to have been weakly calcified or subjected to early dissolution. Most of the material is composed of moulds of the composite type on poorly defined bedding planes and laminae. As in the case of composite-type moulds (McAlister 1962), fine interior and external morphological features are often well preserved. Although taxonomic diversity is generally maximised in shallow marine environments (Boucot 1981), this is not the case with the Tabuk fauna. The condition of the shelly benthics of the Tabuk shales may indicate the influence of low salinity, but cold-water conditions (especially marked during the glaciation at the end of the Ordovician) was most likely the over-riding control. Fortey & Morris (1982) regard the trilobite genus Neseuretus, present in the Llanvirn (Hanadir) of the Tabuk, to be a reliable and sensitive indicator of inshore facies in cold water. 162 H. A. MCCLURE Planktics do not appear to have been affected by some cold, but were clearly affected by the excessive glacial cold conditions at the end of the Ordovician. An extensive platform covered with hyposaline water can exist if an adjacent continent has a river system adequate to provide a steady influx of fresh water. In such environs today, there is a low taxic diversity, and there is no reason to think that extensive river regimes of the past flowing off large land masses may not have had the potential for producing similar hyposaline environments with appropriate faunal consequences (Boucot 1981). This condition may have prevailed on the Arabian plat- form during Ordovician and Silurian times. Outcrop sequences of the Tabuk sands, silts, and shales are oxidized to shades of red, pink, yellow and green. However, subsurface equivalents invariably range from light grey to dark grey and black. Tidalite sands are especially rich in carbonaceous laminae, each lamina possibly representing nutrient material transported by a single tidal event. Tabuk shales in the subsurface are usually medium grey to dark grey and black, the extreme case being the black, highly radioactive, ‘sooty’ shale of the subsurface persculptus Zone. Some of the palynological samples yield a distinct tetrahedral tetrad type of suspected land origin. This kind of evidence for vascular land plants may be recorded as early as the Llanvirn (Hanadir shale) in the Arabian Tabuk section. Berry & Boucot (1973) suggest that a black, radioactive shale at the ‘base of the Silurian’ in north Africa could be due to blooms of plants in mud flats and lagoons at this time. An apparently synchronous event occurs across much of north Africa and Arabia. A readily accessible and presumably useable supply of nutrients might therefore be assumed for both planktic and bottom benthics. Nutrient kind and availability may have been a significant factor in the palaeoecology of chitinozoans and acritarchs espe- cially, and perhaps also graptolites. Temperature is probably the most important variable affecting both plant and animal dis- tribution of the present and continental glaciation episodes of the end of the Ordovician and Permian—Carboniferous are associated with global diversity gradients (Boucot 1981). The change in faunal composition associated with the Ordovician glaciation is now well docu- mented (Berry 1973; Berry & Boucot 1973). A Silurian warming followed the Ordovician cold in the area and this may be reflected in taxa of the Qusaiba fossil suite being relatively more diverse and populations denser, especially planktic ones. In summary, the Tabuk palaeogeography and palaeoenvironment was probably that of a broad pro-delta mud substratum on a shallow-water, clastic-fed, carbonate-starved, tectonically stable platform area, with sediment and high nutrient input derived from a low, rapidly eroding landmass, possibly with primitive plant cover, and transported via extensive fluvial, tidal and deltaic systems. Conditions of low salinity and cold-water temperatures probably controlled diversity and density of parts of the faunal community. Conditions may be considered to have been optimum for planktic taxa such as chitinozoans and acritarchs, favourable for graptolites, and generally unfavourable for benthics. Lovelock et al. (1981) record chitinozoans and acritarchs, trace fossils, ?dalmanellid brachio- pods, and the trilobite ?Neseuretus from Early and Middle Ordovician rocks of the Amdeh Formation of Oman. Rocks of southern Jordan of age equivalent to the Tabuk Formation are sandstones, shales and siltstones bearing graptolites, brachiopods, bivalves and gastropods, nautiloids, Conularia and trace fossils such as Cruziana and Skolithos (Bender 1975). Exact equivalents in these two areas to individual units of the Tabuk Formation, the precise defini- tion of the Ordovician—Silurian boundary, and the comparison with the Tabuk fauna remain yet to be worked out. Conclusions Pending further study and documentation of the palaeobiology of Arabian Ordovician-Silurian fossil suites, the following conclusions are presented: 1. The fossils of all three Tabuk shales are similar in composition, diversity, population density and abundance levels and may be taken to represent one community. 2. Two trophic levels are readily identifiable: (a) planktic—consisting of graptolites, chitin- ozoans and acritarchs, and (b) benthic—consisting largely of vagrants on a flat-bottom mud ORDOVICIAN-SILURIAN BOUNDARY IN SAUDI ARABIA 163 substratum. 3. Water temperature was probably the main environmental control on the community. 4. Salinity was possibly a secondary control on the community. 5. Nutrient material was readily available and may have played a significant role in the palaeocology. 6. An extensive pro-delta mud platform provided the main palaeogeographical control for the bulk of the Tabuk fauna; inshore sandy facies and tidal flats were less important features. 7. The main event that affected the biological community across the Ordovician—Silurian boundary was stress imposed by glaciation at the end of the Ordovician. Otherwise, the conditions that affected the community throughout deposition of the Tabuk were also oper- ative in boundary times. 8. Similarities in the palaeobiology, palaeogeography and palaeoecology occur in the plat- form rocks of the north African Silurian sections. References Bender, F. 1975. Geology of the Arabian Peninsula—Jordan. Prof. pap. U.S. geol. Surv., Washington, 560-1: 1-36. Berry, W. B. N. 1973. Silurian—Early Devonian graptolites. In A. Hallam (ed.), Atlas of Palaeobiogeog- raphy: 81-87. Elsevier Sci. Publ. Co. & Boucot, A. J. 1973. Glacio-eustatic control of Late Ordovician—Early Silurian platform sedimen- tation and faunal changes. Bull. geol. Soc. Am., New York, 84: 275-284. Beuf, S., Biju-Duval, B., Stevaux, J. & Kulbicki, G. 1969. Extent of ‘Silurian’ glaciation in the Sahara: its influences and consequences upon sedimentation. In W. H. Kanes (ed.), Geology, Archaeology and Prehistory of the southwestern Fezzan, Libya. Ann. Field Conf., Pet. Explor. Soc. Libya, 11th: 103-116. Boucot, A. J. 1981. Principles of Benthic Marine Paleocology. 463 pp. New York, Academic Press. Brenchley, P. J. & Newall, G. 1980. A facies analysis of Upper Ordovician regressive sequences in the Oslo Region, Norway: a record of glacio-eustatic changes. Palaeogeogr. Palaeoclimat. Palaeoecol., Amsterdam, 31: 1—38. Carney, R. S. 1981. Nutrients. In A. J. Boucot (ed.), Principles of Benthic Marine Paleoecology: 136-142. New York. El-Khayal, A. A. & Romano, M. 1985. Lower Ordovician trilobites from the Hanadir shale of Saudi Arabia. Palaeontology, London, 28: 401-412, pl. 47. Fortey, R. A. & Morris, S. F. 1982. The Ordovician trilobite Neseuretus from Saudi Arabia, and the palaeogeography of the Neseuretus fauna related to Gondwanaland in the earlier Ordovician. Bull. Br. Mus. nat. Hist., London, (Geol.) 36 (2): 63-75. Hambrey, M. J. & Harland, W. B. (eds) 1981. Earth’s pre-Pleistocene Glacial Record. 1004 pp. Cambridge. Lovelock, P. E. R., Potter, T. L., Walsworth-Bell, E. B. & Wiemer, W. M. 1981. Ordovician rocks in the Oman Mountains: the Amdeh Formation. Geologie Mijnb., Den Haag, 60: 487—495. McAlister, A. L. 1962. Mode of preservation in early Paleozoic pelecypods and its morphologic and ecologic significance. J. Paleont., Tulsa, Ok., 36: 69-73, pl. 16. McClure, H. A. 1978. Early Paleozoic glaciation in Arabia. Palaeogeogr. Palaeoclimat. Palaeoecol., Amsterdam, 25: 315-326. Powers, R. W. 1968. Lexique Stratigraphique International 3 Asie (10b 1: Arabie Saoudite). 177 pp. Paris, C.R.N.S. —, Ramirez, L. F., Redmond, C. D. & Elberg E. L. jr 1966. Geology of the Arabian Peninsula: Sedimentary Geology of Saudi Arabia. Prof. Pap. U.S. geol. Surv., Washington, 560-D: i—vi, DI-D147. Thomas, A. T. 1977. Classification and phylogeny of homalonotid trilobites. Palaeontology, London, 20: 159-178. Young, G. M. 1981. Early Palaeozoic tillites of the northern Arabian Peninsula. In M. J. Hambrey & W. B. Harland (eds), Earth’s pre-Pleistocene Glacial Record. 338 pp. Cambridge. Note added in page proof. The atavus Zone (Rhuddanian) has recently been documented in the Arabian Silurian section. Atavograptus atavus (Jones) is present in both the Tabuk area of outcrop and the deep subsurface of eastern Saudi Arabia. In the outcrop section, it occurs with Climacograptus normalis Lapworth. The Ordovician—Silurian boundary in Morocco J. Destombes and S. Willefert Direction de la Géologie, Ministére de l’Energie et des Mines, B.P. 6208, Rabat-Instituts, Morocco Synopsis At only one locality, Moulay bou Anane, in Jbilet, the persculptus and acuminatus Zones are both found, although the acuminatus Zone is known from many localities throughout Morocco. The early Llandovery usually consists of transgressive shales, ranging from acuminatus Zone to cyphus Zone and above in age, overlying usually unfossiliferous glacial sandstones and microconglomerates of the latest Ordovician, from one of which a Hirnantia fauna is recorded. General survey The Ordovician-Silurian boundary in Morocco is always marked by a very acute change of facies between the two systems. The glacial episode which concludes the Ordovician deposited relatively coarse material, such as sandstones, quartzites and microconglomeratic clays, which strongly contrast with the fine argillaceous or siliceous deposits which characterize the begin- ning of the Silurian. Consequently, the scenario is one of more or less important interruption in sedimentation, the development of glaciogenic sediments, and the transgressive development of a Silurian sea after the melting of the continental ice sheet. Under these conditions, the faunas of the two systems are naturally different, apart from the single exception of Jbilet, at Moulay bou Anane (Locality 1, of Fig. 1), where selected graptolites for the official boundary (Cocks 1985), Glyptograptus persculptus Salter and Akidograptus acuminatus (Nicholson), succeed each other in the same section. Elsewhere, only A. acuminatus dates the beginning of the Silurian above more or less terminal beds of the Hirnantian: (1) in the western Anti-Atlas, at Ain Oui n’Deliouine (Locality 2); (2) in the eastern Anti-Atlas, at Tizi ou Mekhazni (Tizi Ambed) (Locality 3) and at Oued Bou-Leggou (Oued bou Oubagou) (Locality 3’); (3) on the northern slope of the central High Atlas, at Ghogoult (Locality 4) and west of Tiwghaza (Locality 4’); (4) in the substratum of the Plateau des Phosphates (Locality 5); (5) in the Moroccan central massif in the Azrou area, at Bou Ourarh (Locality 6); (6) in the Palaeozoic inliers of the north of the middle Atlas at Tazekka (Locality 7) and Immouzer du Khandar (Locality 8). Some other outcrops of the transgressive Silurian are still later Rhuddanian: (a) in the central Anti-Atlas, at Rich Mel’Alg, where graptolitic beds with Cystograptus vesiculosus (Nicholson), Dimorphograptus, and Coronograptus cyphus (Lapworth) are separated by a red layer from sandstones and clays of the Deuxieme Bani (Upper Ashgill); (b) in the coastal Meseta, at Oulad Said, south of Casablanca, where Atavograptus atavus (Jones) occurs in a boring; (c) in the Qasbat-Tadla-Azrou area, at Jbel Eguer-Iguiguena, where the same association as in (a) occurs. For (b) and (c) it is not possible to determine with precision the age of the underlying beds. The very widespread Silurian in Morocco more generally begins either with Aeronian beneath a siliceous facies alternating with phthanitic ribbons, more sandy in the Anti-Atlas at the east of the meridian of Icht, or sometimes with argillaceous—siliceous Telychian, or, in rare cases, with the upper Wenlock and/or Ludlow. Bull. Br. Mus. nat. Hist. (Geol) 43: 165-170 Issued 28 April 1988 J. DESTOMBES & S. WILLEFERT 166 ‘}X9] 9Y} UI paqliosap snypununoD snidv.boplyY YUM SIITedO] Ud} PUR OODSOIOP UIIYIIOU JO SdO19jNO URIDIAOPIO B4] | “BI cet 8 Lt yvenoz 4 Za wr SPF? 2 oa 1HdI “p W Nae VIL s SA wnissis > O82 W047 wavoV s 28vn0 ce INIT UDIZIAOPIO 0 | ea 9 woz SNyDUIWNSD | | > Hi i uf | in i s lull 1D, i s Pats! iI WIA up) lili NVINVGGNHY ' ndoiboydiowig 4 === JuZDY Xap no iziJ o) NVINVYYFTLIOGIW \ visas nog pano SNyDUIWINID 'y 4 a 2U/N01/ag,u INQ uly | woz @ snydjnosiad |9 4 snyoulwinso y + y Jeddn jUaWipes a1uabiaD)9 ys LL uDijj! aubuy nog Apjnow © © UDIJgU/DIaIg PUD 2/0Z08/Dj UDIIIA0PIO ajUy uDIs/A0piQ§ EH 2/0Z0a}Dg UDIDIAOPIQ jSOd = 20Z0a/Dg }SOg P| ORDOVICIAN-SILURIAN BOUNDARY IN MOROCCO 167 Description of partial sections (1) Eastern Jbilet, Moulay bou Anane, Locality 1 (Topographical Sheet Attaouia ech Chaibia, 1:50000, at x = 322, y = 157-2) (Fig. 1). Roch (1939) described this area as forming part of the ‘Mountains to the East of Marrakech’. Huvelin (1977) emphasized the peculiar features of the Hercynian massif of Jbilet. Huvelin and others refined the section near the boundary in 1980. Roch only pointed out that “Miss G. Elles and G. Waterlot have recognised: Monograptus (sic) modestus, M. sandersoni, M. cyphus, M. revolutus, Glyptograptus incertus, G. persculptus, Cli- macograptus Tornquisti and Cl. normalis from the base of Llandovery’ (p. 113). Specimens of Glyptograptus persculptus, determined by G. Elles, were obtained from a siliceous sandstone, weathered pink-beige, but more greyish on fresh fracture, in beds on which they are nearly orientated. They are of great size, the septum always starting at the fourth or fifth theca, and they are preserved as internal moulds, in whole or half relief. Vertical section 1 summarizes more recent collections. The usual suite of terminal Hirnantian occurs over 20m and consists of microconglomeratic clays, argillites, and sandstones with orientated sedimentary features. This is followed by a layer of quartzose sandstone not much different from those of Roch, but coarser, which yields dispersed G. persculptus with a few more irregularly orientated and smaller forms with a septum beginning at a lower level (in the third theca when visible). They are always internal moulds and are apparently narrower than those identified by G. Elles, but they show more relief. The thickness of the layer is 30cm and it can be presumed that the Roch assemblage is rather nearer the top than the base. Above this coarse facies, and without transitional beds, pink and pink-beige shales with a little mica and with a very fine cleavage, contain at their base: Climacograptus normalis (Lapworth), C. miserabilis Elles & Wood, C. rectangularis (M‘Coy), Diplograptus modestus Lapworth, Akidograptus ascensus Davies and A. acuminatus (Nicholson). The thickness of this argillaceous level is 30cm and occurs below alternations between more phthanitic beds and more or less siliceous clays which terminate the Rhuddanian. The boundary is therefore very sudden and with a sharp change of facies. (2) Western Anti-Atlas, Ain oui n’Deliouine, Locality 2 (Topographical Sheet Tiglit, 1:50000, at x = 1076-4, y = 764-2). The boundary was figured in some detail in Destombes et al. (1985: 242, fig. 46). Above green microconglomeratic strata representing the glacial upper Ordovician, a red bed makes a clear transition with argillites which are very similar in colour, although a few are greener, and shows the same alteration and preservation for fossils as at Moulay bou Anane, although the cleavage is coarser. At the contact there is C. normalis and D. modestus and two metres above a single, small, aseptate specimen of G. persculptus, together with C. normalis, C. transgrediens Waern, D. modestus, and A. acuminatus. The similarities between the two areas are striking for the early Silurian beds and, from the palaeontological point of view, the abundant D. modestus shows some intraspecific variations which recall Davies’s (1929) considerations on the similarities of G. persculptus and D. modestus, and whether it is a case of convergence or of real relationship. Internal moulds in iron-oxides only emphasize, once again, all the pitfalls in determining deformed graptolites by comparison with material which has preserved its proteic skeleton. Finally, from these two localities, which appear to be the most characteristic of those actually known from Morocco, it is difficult to imagine any Ordovician-—Silurian boundary without a break. (3) Eastern Anti-Atlas, Tizi ou Mekhazni (Tizi Ambed), Locality 3 and Bou Leggou (Oued Bou Oubagou), Locality 3’. A peculiar feature of the sections near the boundary is the presence, above conglomeratic sandstones and quartzites and lenses with very coarse green and pink sandstones, of a green siltstone with a very probable hard ground between the two deposits. (a) At Tizi ou Mekhazni (Topographic Sheet Erfoud, 1:100000, at about x = 588-8, y = 73-8), Destombes et al. (1985: 257-258, figs 54 and 55) report 10m of greenish silts followed by a black marker bed about 10 m thick of very fine silicified slates with tuff layers, followed by 75m of fine silicified white, pink and reddish violet slates, the base of which includes C. normalis, C. transgrediens, D. modestus, A. ascensus?, and A. acuminatus in the first 5m. The 168 J. DESTOMBES & S. WILLEFERT Rhuddanian and the Aeronian continue up to the M. sedgwickii Zone within 125m of siliceous sandstones, sometimes in plaquettes which weather to a very dark ferrugineous colour, but lighter on splitting. (b) At Bou-Leggou (Topographical Sheet Erfoud, 1:100000, at about x = 589-2, y = 56-6), the Rhuddanian includes about 60m of green silts which contain nine levels with classic climacograptids (Cl. normalis, transgrediens, praemedius Waern, medius and rectangularis), which are sometimes crossed by small sandy nodular structures. Towards the top, at the transition with siliceous shales, Dimorphograptus confertus Lapworth and D. confertus cf. swan- stoni Elles & Wood are found, showing a difference in thickness for the first part of the Silurian between the two localities. No trace of the black marker bed can be seen at Bou Leggou. These sections give rise to a problem in the appreciation of the precise age for the base of the silts. However, given the usual conditions of sedimentation between the end of the Ordovician and the first Silurian and the fact that there is no proof of A. acuminatus at the beginning of its biozone, one can, for cartographical purposes, take the Silurian as beginning with the silts. It remains to analyze the mineralogy of the black marker beds, and perhaps also the siliceous shales, to see whether they reflect volcanic activity, even if only very distant from this district of the eastern Anti-Atlas. (4) On the northern slope of the central High Atlas, at Ghogoult (Locality 4) and east of Tiwghaza (Locality 4’). The important Hercynian tectonics which are manifest in the central High Atlas, formerly known as the ‘Mountains to the East of Marrakech’ (Roch 1939) or ‘Demnate Atlas’ (Lévéque 1961), do not enable us to establish a sure succession for the boundary in this part of Morocco. The Silurian with A. acuminatus is present in the allochthonous inliers of Ait Mallah and Ait Mdioual (geological map Azilal 1:100000, 1985) and in the autochtonous deposits to the west of Tiwghaza (boundary of topographical sheets Telouat and Skoura 1:100000). In Ait Mallah, C. normalis, D. modestus, A. acuminatus, C. vesiculosus, Monograptus revolutus s.l. (Kurck), Pribylograptus incommodus (Tornquist) and A. cf. atavus have been identified; in Ait Mdioual, only the lower third in argillaceous or argillaceous-siliceous shales, with a very thin cleavage (overlain by drier, resonant shales, sometimes with drifted micas), and higher coarser beds with C. cyphus. The relations with the Ordovician cannot be defined since the earlier Silurian ‘constitutes a level of preferential disharmony’ (Jenny & Le Marrec 1980). West of Tiwghaza, D. modestus, A. acuminatus and C. vesiculosus are recognized from the base of the first 5m of sandy, coarse, micaceous shales underlying siliceous and phthanitic ones of the Llandovery succession. Jenny & Le Marrec (1980) described the last three metres of the upper Ordovician as composed of classic ‘massive or irregular decimetrical sandstones- quartzites, sometimes with oscillation-ripples, whitish colour with dark patina and black micro- brechic or microconglomeratic sandstones or clays with round and matt quartz’. (5) In the substratum of the Plateau des Phosphates (Locality 5). An oil-boring—BJ 105—on the geological map Qasbat Tadla (1:100000, 1985, at x = 417-7, y = 216-8) terminated at a depth of 1017m in the upper Ashgill. In a fragment of core between 963 to 988-5 m, in an argillaceous, graphitic, more or less siliceous facies, the lowest associations contain: (a) more argillaceous than siliceous beds with many slip planes with C. normalis, C. rectangularis, D. modestus, A. acuminatus, followed by (b) a more siliceous layer with the same association underlying the C. vesiculosus, Dimorphograptus and C. cyphus Zones. Although information is insufficient to define the boundary, a sudden change in facies (here between 988:5 and 1017 m) is found, with the same pattern as in other areas. (6) In the Moroccan central massif, Azrou area, at Bou-Ourarh (Locality 6) (Topographical Sheet Ain Leuh, 1:50000, at about x = 503-5, y = 302-5). The Silurian here occurs as a siliceous facies alternating with real phthanites weathering light grey. It is the “Formation dite de Mokattam’ of Choubert (1956). It always lies upon ridges of sandy or even quartzitic material, which are more resistant in the landscape, and which can be assigned to the upper Ordovician without more precision in dating. Graptolites are found more or less at the contact. At one locality, there is C. normalis, C. medius, C. rectangularis, C. vesiculosus, A. acuminatus, Glyptograptus sp. ORDOVICIAN-SILURIAN BOUNDARY IN MOROCCO 169 or Orthograptus sp., P. incommodus, A. ex gr. atavus and Raphidograptus toernquisti (Elles & Wood). The beds with A. acuminatus are less compact than those with C. vesiculosus. Rhudda- nian and Aeronian rocks with the Coronograptus gregarius Zone are found down a small valley. Sandy layers occur at several levels in the Mokattam Formation and the sequence is repetitive. It is now known that this area has suffered greatly through Hercynian tectonism, so it seems that Bou-Ourarh is constructed of a number of tectonic slices in which the Silurian has often played the role of soapstones, and so it is not possible to find any Silurian beds conformably against the Ordovician sandstones. Although this district is not important for the boundary definition, it is a supplementary paleogeographical marker for the distribution of the A. acumin- atus Zone. (7) In the Palaeozoic inliers of the north middle Atlas. (a) Tazekka (Eastern Morocco) (Locality 7). The same tectonics as at Bou-Ourarh cause repetition of the upper Ordovician and lower Silurian. At Souk et Tleta des Zerarda (Topographical Sheet Ribat el Kheir, 1:50000, at x = 594-5, y = 373-7) at the top of the usual quartzites, almost vertical upper Ashgill black argillaceous-siliceous and siliceous beds contain C. normalis, C. medius, C. rectangularis, C. probably longifilis Manck, C. probably trifilis Manck, D. modestus and A. acuminatus. Silurian beds follow, but not quite in the same section. (b) Immouzer du Khandar (Locality 8). The same situation exists at the NW end of the Immouzer du Khandar inlier (Topographical Sheet Sefrou, 1:100000, at about x = 540-1, y = 353-7), where A. acuminatus, C. normalis, C. miserabilis, C. rectangularis and D. modestus are found in the argillaceous facies of the Mokattam Formation, but the locality is altered and schistosed, with bedding plane thrusts. This has contact with sandy pelites and big well- rounded quartzites of the upper glacial Ordovician, which are equivalent to the Upper Deux- i¢me Bani Formation (Upper Ashgill) of the Anti-Atlas. Conclusions The base of the Silurian is seen in many areas of Morocco, and invariably in argillaceous facies, underlying sandstone levels and never in true phthanites. It is remarkable that in these sections no Ordovician faunas have been found, except at Moulay bou Anane. However, in the central Anti-Atlas, Tagounite area, at Jbel Larjame and at Oued Mooulili, some badly preserved bra- chiopods are known from the upper part of the Upper Deuxi¢me Bani Formation; these are from a more western region (south flank of Jbel Addana, south of Akka) and consist of Hirnantia sagittifera (M‘Coy), Eostropheodonta squamosa Havli¢ek, and Plectothyrella chauveli Havlicek, from grits above microconglomeratic clays. These faunas are very important in dating the intra-Hirnantian tillite, and in these areas of the central Anti-Atlas the Silurian begins with a hardground followed by graptolites of later Llandovery age. We do not expect to find more significant faunas in the Ordovician rocks and future studies must turn to the sedimentology of the glacial phenomena and the volcanic influences in the eastern Anti-Atlas; and also to the description and illustration of the graptolites themselves. References Choubert, G. (ed.) 1956. Lexique Stratigraphique International 4 Afrique (1a: Maroc). 165 pp. Paris, C.N.R.S. Cocks, L. R. M. 1985. The Ordovician-Silurian boundary. Episodes, Ottawa, 8: 98-100. Davies, K. A. 1929. Notes on the graptolite faunas of the Upper Ordovician and Lower Silurian. Geol. Mag., London, 66: 1—27. Destombes, J. 1981. Hirnantian (Upper Ordovician) tillites on the north flank of the Tindouf basin, Anti-Atlas, Morocco. In J. Hambrey & W. B. Harland (eds), Earth’s pre-Pleistocene glacial record: 84-88. Cambridge. , Hollard, H. & Willefert, S. 1985. Lower Palaeozoic rocks of Morocco. In C. H. Holland (ed.), Lower Palaeozoic of north-western and west central Africa: 91-336. London. 170 J. DESTOMBES & S. WILLEFERT Graf, C. (1976). Synthése géologique du bassin de Kasba-Tadla, Beni-Mellal, Tanhasset (d’aprés les données géophysiques et de forages). Rapp. BRPM/DEP. 89 pp., 15 pls (unpublished). Huvelin, P. 1977. Etude géologique et gitologique du Massif hercynien des Jebilet (Maroc occidental). Notes Mem. Serv. géol. Maroc, Rabat, 232 bis: 1-307, 12 pls, 3 maps. Jenny, J. & Couyreur, G. 1985. Carte geologique du Maroc au 100000e, feuille Azilal. Notes Mem. Serv. geol. Maroc, No. 339. —— & Le Marrec, A. 1980. Mise en evidence d’une nappe a la limite méridionale du domaine hercynien dans la boutonniére d’Ait-Tamlil (Haut Atlas central, Maroc). Eclog. geol. Helv., Basel, 73: 681-696. Levéque, P. (1961). Contribution a l’etude geologique et hydrologique de |’Atlas de Demnate (Maroc). These Sci., Paris. 242 + 161 + 42 pp. (unpublished). Roch, E. 1939. Description géologique des montagnes a Est de Marrakech. Notes Mem. Serv. Mines Carte geol. Maroc, Paris, 51: 1-438, 7 pls. Verset, Y. 1985. Carte géologique du Maroc au 100 000e, feuille Qasbat-Tadla. Notes Mem. Serv. geéol. Maroc, No. 340. The Ordovician—Silurian boundary in the Algerian Sahara P. Legrand Directeur Laboratoires Exploration (Groupe) TOTAL, 218-228 Ave du Haut-Lévéque, 33605 PESSAC Cédex, France. Synopsis Two sections, at eastern Tassili-n-Ajjer and at El] Kseib, demonstrate the Ordovician—Silurian boundary, with graptolites at intervals and rare shells, however the acuminatus Zone itself is not recorded. The sections are internationally important firstly in demonstrating excellent glacial and periglacial sediments during the late Ashgill, and secondly in showing that this continental ice-mass melted and was succeeded by, but was not the origin of, the transgression during the latest Ordovician, in persculptus Zone times. Introduction Because of the uplift that probably affected most of the Algerian Sahara near the end of the Ordovician, and the circumpolar conditions which caused the development of a continental ice sheet (Debyser et al. 1965), the Algerian Sahara seemed originally an unlikely country for biostratigraphical study of the Ordovician—Silurian boundary. However, detailed observations from the boundary beds enable us to show clearly an almost continuous succession from the Ordovician to the Silurian in the eastern Tassili-n-Ajjer, whereas to the west, in the Ougarta range, there is a probable hiatus. Moreover, these observations suggest some interesting conclu- sions about the palaeogeography because this is a country where the glacial events are particu- larly striking (Fig. 1). Eastern Tassili-n-Ajjer sections of the Djanet—In Djerane Oued tray and of the In Djerane Oued Kilian (1928) drew attention to this area by pointing out the presence of a fauna of lowermost Llandovery age. Unhappily, this discovery was forgotten and it was many years later when interest was aroused again following a preliminary collection by the ‘Mission sédimentologique sur la couverture sédimentaire du Boudin sahariem’ in 1965. Two further studies were carried out in the field (1978, 1982) despite substantial logistical difficulties; but only some of the successive results have been published, others are in press. The stratigraphical succession is as follows (Fig. 2): Above the Gara Tembi sandstones with a glacial relief: (a) the Arrkine argillaceous sandy formation (about 90m) in which a new fauna with Cli- macograptus (Climacograptus) gelidus nov. sp., C. (Climacograptus) arrikini nov. sp. and C. (Climacograptus) normalis ajjeri Legrand occurs near the base. (b) The shaley formation of Oued In Djerane in which the following distinctions can be made: Lower member (80m) of silty claystones and siltstones with a few carbonate levels; the fauna is as follows: C. (Climacograptus) normalis ajjeri Legrand, C. (Climacograptus) pseudo- venustus Legrand, C. (Climacograptus) pretilokensis nov. sp., C. (Climacograptus) tilokensis Legrand and Zygospiraella sp. Middle member (about 110m) with: a lower submember of shales with C. (Climacograptus) normalis ajjeri Legrand, Diplograptus (?) kiliani Legrand; an upper submember of siltstones and silty shales with C. (Climacograptus) freuloni nov. sp., C. (Climacograptus?) incommodus nov. sp., and Glyptograptus (Glyptograptus) sahariensis nov. sp. and near the top C. (Climacograptus) imperfectus Legrand, and ?G. (Glyptograptus) aff. persculptus (Salter). Bull. Br. Mus. nat. Hist. (Geol) 43: 171-176 Issued 28 April 1988 172 P. LEGRAND aw srt” » BECHAR 4 HASSI MESSAQUD Ss Grand Erg occidental Grands =n === oriental a" Ni \Y Oued In Ay ono . YS z Tassili de vy ae A Djerane Tarit 4 are) q Oh A Wo AW AR Fig. 1 Outcrops of lower Palaeozoic in Algeria apart from the Intermediate Series (1); Intermediate Series and Cambro-Ordovician of the syneclise of Taoudeni (2); and Precambrian and Interme- diate Series (3). Upper member of sandstones with argillaceous silty intercalations. Fossils are only found near the base and include Diplograptus africanus Legrand, and G. (Glyptograptus) tariti Legrand and then, above, Diplograptus fezzanensis Desio. A lower Llandovery age was originally suggested for the whole Oued In Djerane Formation (Legrand 1976, 1981, 1985a); then an Ordovician-Silurian boundary level at the top of the Diplograptus (?) kiliani Zone was proposed (Legrand 1985b, 1986), but a further possibility, of a boundary at the top of the Middle member, must be considered. The arguments in favour of this last possibility are as follows: (i) A new subspecies very near to Diplograptus (?) kiliani is known in the Kurama Range, Usbekistan (but not in Kazakhstan) and it occurs, according to T. N. Koren, not below the Parakidograptus acuminatus Zone, as formerly believed, but below some beds where C. (Climacograptus?) extraordinarius or G. (Glyptograptus) persculptus was collected. (ii) On the other hand, C. (Climacograptus) incommodus has some affinities with C. (Climacograptus) extraordinarius and in this respect the position of Zygospiraella, a genus ORDOVICIAN-SILURIAN BOUNDARY IN THE ALGERIAN SAHARA 173 3° section Alevé: Ph.Legrand 4 m—— /) fezzanens!s Fig. 2 = 1978 - 1982 Wu < S +ilever, f= Te oles S | D. africanus a = Mission Sédimentologique re a 8 S Y sur la bordure du Hoggar is eon AIS ma © el Ne ‘ I : e 2 . = abs ES | S S| 2: hypothese = Pw YH 3S = Suomi : Ww 1s S + g Ic S Q aS G ~ 8 I S S = S le 2 Po Ra) g = tle g Les S Is is hypothése 9S = 5 S —? — . CL. freu/oni s C ~ ® = S 4s Diplograptus (?) kiliani wy =~ = rT) & x = é ] Zygospiraela sp. = s Cl pseudovenustus 8 | L CL tilokensis WW : : = ‘0 CL. pretilokensis ro | ¥ [ — =| Kw S & G SoS Na eS eS S\% S$] S]s]..| © 4) Salie afer! ayer! miserabisis / eg. Cl Oe iri 1 OP WY Cl. normalis ayer’ Cl. normalis Cl. normalis Cl aff. gelidus CL arrikins J] CG geliqus — FORMATION ARGILO-GRESEUSE DE L’IRHARRHAR ARRIKINE F.de la Gara Tembi Fig. 2 Distribution of the principal faunas in the sections of the Djanet-In Djerane Oued tray and the In Djerane Oued, Algeria. 174 P. LEGRAND only so far definitely recorded from the Silurian, would be the same as that in Kazakhstan (Oysu River section). (iii) Finally, rare specimens of ?G. (Glyptograptus) aff. persculptus have been gathered just below the top of the middle member of the Oued In Djerane Formation. The objections to the hypothesis are the following: (i) G. (Glyptograptus) sahariensis is very close to G. (Glyptograptus) tariti and has the aspect of a Silurian Glyptograptus. (ii) Diplograptus africanus seems to belong to the Coronograptus cyphus Zone (Legrand 1976), and consequently there is a very small thickness for the Parakidograptus acuminatus Zone and the Cystograptus vesiculosus Zone. The sandstones that form the top of the middle member may be thought to be the equivalent of the zone. (ii) Parakidograptus acuminatus has not yet been found; one can think of the sandstones that form the top of the middle member as the equivalent of the biozone characterized by this species. However, nor has it been found near the Libyan boundary, where the shales take the place of the sandstones owing to the later transgression there, and where the sedimentation seems to have been more continuous. (iv) Perhaps in this apparently very confined area the vertical range of species many not have been absolutely the same as in less restricted regions. To conclude, two hypotheses can be proposed for the position of the Ordovician—Silurian boundary, but the highest seems the most likely. Moreover, there is no characteristic fauna of the Ordovician in the lower part of the section and this sets problems of correlation with the standard sections (Dob’s Linn, Kolyma River, Yangtse Valley), and consequently this section in Algeria can only be a local reference. On the other hand, it has important palaeogeographical significance since it shows the beginning of the transgression onto the southeastern part of the Saharan shield before the end of the Ordovician, which must have involved the melting of the continental ice sheet, at least locally, before the beginning of the Silurian (Legrand 1985). Ougarta Range—El Kseib section In the Ougarta Range, the stratigraphical succession of the upper part of the Ordovician includes the argillaceous sandy Bou M‘haoud Formation, which is overlain by the argillaceous sandy Jebel Serraf Formation. A mappable unconformity separates these two formations (Arbey 1962; Gomes Silva et al. 1963; BRP et al. 1964; Legrand 1974). In the eponymous locality, where that formation seems the most complete, the upper part of the Bou M‘haoud N.W. S.E. Surface a modele glaciaire Faune a Airnantia Graptolites du Llandoverien moyen Membre EBACE greso-conglomeratique “SS Membre supérieur | Formation des grés du Ksar | des argiles d’Ougarta de \'Oued Ali. Membre moyen argileux d’E! Kseib Formation gréso-conglomératique du Djebel Serraf (0) is) 10 15 20 km. Ph. LEGRAND . 1967-1981 Fig.3 Section in the vicinity of the Ordovician—Silurian boundary at El Kseib, Ougarta range, Algeria. ORDOVICIAN-SILURIAN BOUNDARY IN THE ALGERIAN SAHARA 175 Formation is apparently of Lower Caradoc age, with Kloucekia (Kloucekia?) nov. sp., Calymenella sp., Drabovinella grandis Mergl, and Drabovia sp. At first this fauna was attributed to the Upper Caradoc and the beds from which it was collected were considered to belong to the lower member of the formation subjacent to the Jebel Serraf Formation. Going to the north west (in the Daoura), the succession is apparently complete up to the lower Ashgill. Above this the Jebel Serraf Formation appears to be absent or very thin in Bou M‘haoud village, with siltstones and sandstones (channel deposits), but no fossils have been found. The quality of the outcrops does not allow us to see the contact with the lowest Silurian shales. Thus, it is near Ougarta that the Ordovician—Silurian boundary must be investigated. In the classical El Kseib section discovered by Menchikoff (1930), the Bou M‘haoud Forma- tion is reduced to its lower member. Above, the Jebel Serraf Formation consists of a well- developed sandy, conglomeratic lower member, then the microconglomeratic shales of El Kseib that prove a periglacial environment; and above these, the sandstones of the “Ksar d’Ougarta’, It is at Ougarta that some brachiopods were gathered from this member by Poueyto (1950). Unhappily this fauna (which has been recollected since 1961) is poorly diversified and consists of Plectotyrella chauveli Havlitek, Hirnantia aff. sagitiffera (M‘Coy), Lingulella sp., Pseudobolus sp., Conchilolites sp. and a homalonotid pygidium. The age of this member is uppermost Ashgill (Destombes 1971; Legrand 1974, 1985a, b). Above this the Oued Ali formation is found, whose base is characterized by a ferrugineous sandstone with ferrugineous nodules and then a bed of sandstone; there follows some varicoloured shales and coarse shaly sandstones with C. (Climacograptus) sp., and the member ends with black shales with C. (Climacograptus) aff. rectangularis M‘Coy, Orthograptus aff. mutabilis Elles & Wood, ?P. (Metaclimacograptus) phry- gonius Tornquist, and Rastrites sp., indicating a Middle Llandovery age. Although this section is only interesting from a local point of view for the definition of the Ordovician—Silurian boundary, it has the wider advantage of showing that the glacial or periglacial environment ended just before the end of the Ashgill. Conclusions The Algerian Sahara is surprisingly important in increasing our knowledge of the Ordovician— Silurian boundary period. Studies in eastern Tassili-n-Ajjer show, in an almost continuous section through coastal sediments, the nature of the endemic faunal succession, which, however, has some affinities with southern Siberia. A palaeogeography can be drawn showing the area more or less neighbouring the South Pole, and the observations in the Ougarta Range strongly suggest the almost complete melting of the Upper Ordovician continental ice sheet before the Silurian transgression. This leads us to reconsider the importance of the melting in the mecha- nism of the transgression (Legrand 1985). References Arbey, F. 1962. Données nouvelles sur la sedimentation au Cambro—Ordovicien dans les monts d’Ougarta (Saoura). C.r. hebd. Seanc. Acad. Sci., Paris, 254: 3726-3728. Bureau de recherches de pétrole et al. (compagnies pétroliéres) 1964. Essai de nomenclature litho- stratigraphique du Cambro-Ordovicien Saharien (colloque). Mem. Soc. geol. Fr., Paris (h.s.) 2. 55 pp., 11 pls. Destombes, J. 1968. Sur la présence d’une discordance générale de ravinement d’age Ashgill supérieur dans l’Ordovicien terminal de l’Anti-Atlas (Maroc). C.r. hebd. Seanc. Acad. Sci., Paris, (D) 267: 565—567. Debyser, J., Charpal, de O. & Merabet, O. 1965. Sur le caractére glaciaire de la sedimentation de Unité IV au Sahara Central. C.r. hebd. Seanc. Acad. Sci., Paris, 261: 5575. Gomes Silva, M., Pacaud, M. & Wiel, F. 1963. Contribution a l’etude du Cambro-Ordovicien des Chaines d’Ougarta (Sahara algérien). Bull. Soc. géol. Fr., Paris, (7) 5: 134-141. Kilian, C. 1928. Sur la présence du Silurien a l’Est et au Sud de l’Ahaggar. C.r. hebd. Seanc. Acad. Sci., Paris, 186 (8): 508-509. 176 P. LEGRAND Legrand, P. 1970, Les couches a Diplograptus du Tassili de Tarit (Ahnet, Sahara algérien). Bull. Soc. Hist. nat. Afr. N., Algiers, 60 (3—4): 3-58. 1974. Essai sur la paleogeographie de lOrdovicien au Sahara algerien. Notes Mem. Comp. Franc. Petrol., Paris, 11: 121-138, 8 pl. 1981. Contribution a étude des graptolites du Llandovérien inférieur de POued In Djerane Tassili N’Ajjer Oriental (Sahara algerien). Bull. Soc. Hist. nat. Afr. N., Algiers, 67 (1-2): 141-196. —— 198la. Essai sur la paléogéographie du Silurien au Sahara algérien. Notes Mem. Comp. Franc. Petrol., Paris, 16: 9-24, 9 pls. 1985. Lower Palaeozoic Rocks of Algeria. In C. H. Holland (ed.), Lower Palaeozoic of North Western and West Central Africa: 6-29. London. 1985a. Réflexions sur la transgression silurienne au Sahara algérien. Act. Cong. Nat. Soc. Sav. Sect., 6: 233-244. — 1986. The lower Silurian graptolites of Oued In Djerane: a study of populations at the Ordovician— Silurian Boundary. Spec. Publs geol. Soc. Lond. 20: 145-153. Menchikoff, N. 1930. Recherches géologiques et morphologiques dans le Nord du Sahara occidental. Rev. geogr. phys. et geol. dyn. 3 (2): 103-247. Poueyto, A. 1950. Coupe stratigraphique des terrains gothlandiens a Graptolites au N d’Ougarta (Sahara occidental). C.r. somm. Seanc. Soc. geol. Fr., Paris, 1950: 44-46. The Ordovician—Silurian boundary in Mauritania S. Willefert Direction de la Géologie, Ministére de l’Energie et des Mines, B.P. 6208, RABAT-Instituts, Morocco Synopsis Three sections are described across the Ordovician-Silurian boundary in Mauritania, each bearing well- developed glacial deposits succeeded by graptolitic shales. In general, fossils of the latest Ordovician and earliest Silurian are absent, apart from the southeastern section between Aratane and Oualata, at a cliff in Hodh, where the persculptus and atavus Zones are recorded. Introduction Three areas in Mauritania (Fig. 1) shed some light on the question of the Ordovician—Silurian boundary; however, the pioneer stage of work in these large areas encourages caution. The areas are: 1 Zemmour Noir (northern Mauritania), known from the masterly contribution of Sougy (1964) and included in the northern flank of the Reguibat uplift in Deynoux et al. (1985). 2 The Mauritanian Adrar, monographed by Trompette (1973), in the western part of the Taoudeni Basin (Deynoux et al. 1985). 3 Hodh, whose Precambrian and Ordovician glacial deposits were studied by Deynoux (1980); this is in the eastern extension of Tagant, which reaches the Adrar towards the S and SE. The Hodh escarpment frames a Cambro—Ordovician-Silurian ribbon to the N of the southern margin of the Taoudeni Basin before the post-Palaeozoic oversteps it (Deynoux et al. 1985). In each area, the glacial upper Ordovician has been carefully studied and these deposits are more remarkable than those of Morocco, since they were nearer to the Lower Palaeozoic pole, and so record even more glacial activity, and, moreover, the glacial episode lasted for a longer time. The Ordovician-—Silurian relationships are very gradual at Hodh and marked by an acute change of facies at Adrar and Zemmoutr. Regional descriptions 1 Zemmour Noir (Fig. 2A, but chiefly Deynoux et al. 1985: 347, fig. 4; 354, fig. 6; and 369, fig. 7). The upper Ordovician consists of the Garat el Hamoueid Group and overlies rocks of Precambrian to Llanvirn age. Its upper boundary is correlated with the upper Ashgill by analogy with comparable deposits in Morocco and Algeria and its thickness varies between 0 and 200m. The rocks are typical glacial deposits but these characteristics become less clear to the NW in the Dhlou Chain because of tectonic complications. Some sedimentological features suggest a more periglacial regime near the top. Faunas are very rare and consist only of ‘indeterminable Camarotoechia compared by Havlicék (1971) with other brachiopods of the upper sandstones of the Deuxiéme Bani of Morocco; and of Cornulites. The base of the Silurian is marked by a very sharp discontinuity, and the system is well developed on the eastern margin of Zemmour, striking SSW—NNE. It always starts with Demirastrites triangulatus (Harkness) (determined by A. Philippot) in a facies of black, argilla- ceous, and some micaceous, shales. Its thickness seems to decrease evenly from 30m in the north to 6m in the south. Among the detailed sections of Sougy (1964), the more northern, west of Gara Bouya Ali, has its base concealed by about 27m of sandy ‘oued’: in the 3m of overlying shales there are specimens of Monograptus sedgwickii (Portlock) (determined by A. Philippot), while a 30cm bed of sandstones separates the top of the Garat el Hamoueid Group from the hidden part. Bull. Br. Mus. nat. Hist. (Geol) 43: 177-182 Issued 28 April 1988 178 Carboniferous Folded chains (caledono-hercynian) Siluro-Devonian Terminal Precambrian and Cambro-Ordovician S. WILLEFERT Tindouf e PSN esmana WG Z Le lates \Z eS eatn Se Se 4 Veta FOUG GARA wy +e ar Upper Precambrian Lower and middle Precambrian Aloun Bea el Malek 4 EI Mzere| A 24 aha + 4 a& c > 4 + Ad-Dokhla =F . | a Tenoymer{- Bi Amrapeg KATE El Mreit Nouadhibou ; VE 7 S Oar’ (| Quodone @:: — “@Herrour é # Ree Nee +f S @ Tika a —— \ ) NOUAK CHOTT | ) a0| Ss os ~ KR \) | Le A S Wek 27 hienti@ess Hon re |e \ - AOJUK ER ¢ i ) be : S | Kf oo A A és . f \ enthanet: ERD faa as = & |. 5 re

&2C!kKounou | Bere! bh ; DAKAR 5 R. Idrissi Fig. 1 Geological sketch of the western margin of the Taoudeni Basin, Mauritania, after Deynoux (1980). Elsewhere, the surface of the sandstones at the contact with the shales is sometimes covered by a yellow coating. At Gara Foug Gara there is 2m between ‘Camarotoechia’ and Demirastrites triangulatus. There is therefore not much hope of defining the boundary exactly in Zemmour Noir, unless new discoveries are made in the western tectonized part. The Silurian has been noted in the Dhlou Chain but has not been systematically studied. 2 The Mauritanian Adrar (Fig. 2B, but chiefly Deynoux et al. 1985: 371, fig. 11; 374, fig. 12; 378, table 3). This area geomorphologically consists of (roughly from NNE to SSW), the Atar plain, the cliff, the plateaus (tabular zone) and the SW margin (folded zone), overlapped by the Mauritanides chain. The Ordovician-Silurian boundary is exposed in the two last units, but the area can be treated as a whole, whilst noting that the Silurian becomes more sandy to the 179 ORDOVICIAN-SILURIAN BOUNDARY IN MAURITANIA ja xnoukaq Jaye ) (S861 12 UO!}99s YPOH 24} Ul souojspues © yY pue “Y Udamjaq sI AJepuNog sy] “BIURILINeJ] Ul SUOTIOAS AJBpUNOG UeLINIIS—URIDIAOPIQ 7 ‘BIy N WOOr | $9}1]0}dDI9 a] 3U0}SPUDS HYUN IS W @ N sl @ in © W A s| ©) | a2 Bis S9}NI1j41DUl sapodolysnig & SaNd'jIDUI sapodolysnig JINVGYOIS/G —————. —_ ,2iMaiiadns ayy. WOSCL Oi Oe ih (7) a (o) a ec Ss a e -wosS v & (7) L : a ny a] £ Wool JDIPY UDIUDJIIND;W Sev AW NY3lSIAIM dnoi9 PlanowDdY skD|2 y90)]g je 30109 he ue WOp 49 ISSIJPI “Y inowwaez 180 S. WILLEFERT WSW. The glacial formation and the Silurian have been called ‘Supergroup 3’ by Trompette (1973), subdivided into the Abteilli Group and the Oued Chig Group. (a) The Abteilli Group represents the glacial upper Ordovician whose lower boundary is difficult to establish because the glacial deposits occur in a landscape long exposed to continen- tal deposition and weathering. The only earlier marine palaeontological horizon consists of lingulids of probable Cambro-Ordovician boundary age (determined by P. Legrand). The top of the group is marked by sandy eskers which reflect the withdrawal of the land ice to the south-east. At the time of Monod’s survey (1952) in this district, some brachiopods in a sandstone from the folded zone at Ayoun Lebgar were determined by D. Le Maitre, who recognized the genera Camarotoechia, Rhynchonella (especially R. ex gr. borealis), Orthis, Dal- manella etc., but frequently with nomenclatural doubt. Monod thought that these sandstones were of Silurian age and that influenced the palaeontologist in her attribution to a high level in the Wenlock. However, these brachiopods may perhaps better be compared with those from Gara Foug Gara. J. Drot considers that in Zemmour as well as in Adrar all these fossils are indeterminable, but it is tempting to compare the total fauna directly. In the section, collected again by Trompette (1973), the usual graptolitic shales are immediately above the brachiopod- bearing lenticular sandstones, which indicate a marine incursion which might have been con- temporaneous with those of Zemmour or the upper sandstones of the Deuxiéme Bani, and so Trompette has suggested that they belong to the lower Silurian. However, prudence is neces- sary with such weak data and both possibilities remain hypotheses. (b) The base of the Oued Chig Group. In the fifteen sections and complementary support sections, Trompette (1973) was able to verify the concordance between the Abteilli Group and the Oued Chig Group and also the striking difference in sedimentation between the two groups. Their contact is rarely clear: there is often 1m or more of sandy debris masking the extreme base of the Silurian. The oldest graptolites are: Climacograptus normalis Lapworth, C. cf. rectangularis (M‘Coy), C. cf. scalaris (Hisinger), ?C. sp. or Pseudoglyptograptus sp., cf. Pseudo- climacograptus (Metaclimacograptus) hughesi (Nicholson), Diplograptus magnus Lapworth, D. modestus Lapworth or D. magnus, Pristiograptus regularis (TOrnquist), Lagarograptus tenuis (Portlock), M. sedgwickii and ?Cyclograptus sp. or Calyptograptus sp. There is no Akidograptus acuminatus (Nicholson) but a part of the Rhuddanian may be present when the lowest associ- ation contains only the first Climacograptus and Diplograptus either modestus or magnus. In Adrar it appears that the Llandovery Series begins earlier than in Zemmour because of the scarcity of monograptids at the base. 3 The Hodh (Fig. 2c, but chiefly Deynoux et al. 1985: 389, fig. 16 and unpublished determinations). The subdivisions adopted here are Tichit Sandstones for the glacial formation and Aratane Group for the sandstones and shales with graptolites. The definition of the Ordovician-Silurian boundary (Cocks 1985) may modify somewhat the Silurian attribution of some of the basal graptolitic sediments. The glacial complex rests on any formation among those defined as Cambro-Ordovician. The major erosional disconformity which opens the glacial cycle is perhaps also in places an angular unconformity, for example in Tagant (Dia et al. 1969). Deynoux (1980) has recognized a lower and an upper part in a total thickness of the order of 100-150 m. The upper part, with several members, includes sandstones and microconglomeratic clays underlying a landmark sandstone R,, followed by sandy clays (still with microconglomeratic layers) under a second sandy landmark R,, above which are the clays with graptolites of the Aratane Group. To the east there are further sandstones termed R; and R,. This group ranges from 100-130m in thickness. In the more southeastern section, about halfway between Aratane and Oualata, a bed with graptolites between R, and R, contains some diplograptids identified as amplexograptids of Ashgill type. Following the escarpment to the north and west, the sandy landmarks become less easy to correlate but the zone of Glyptograptus persculptus is well represented: (a) The more western layer, a portion of the Aratane cliff, appears to be deposited in a glacial gully under R, and contains only Climacograptus normalis and C. transgrediens Waern. ORDOVICIAN-SILURIAN BOUNDARY IN MAURITANIA 181 (b) The persculptus Zone contains: Glyptograptus persculptus (Salter), ?Acanthograptus sp. or ?Koremagraptus sp., C. normalis, C. miserabilis Elles & Wood, C. transgrediens, C. cf. praeme- dius Waern, C. medius (Tornquist), C. cf. rectangularis, C. cf. indivisus Davies, C. minutus? Elles & Wood, a more amplexograptid than climacograptid new form which recalls some figures of Comatograptus Obut & Sobolevskaya or Hedrograptus Obut, although more oval; rare frag- ments of Orthograptus ex gr. truncatus Lapworth, and ?Akidograptus sp. Some climacograptids show basal spines (Elles & Wood 1906; series of species of Manck 1924 (see Miinch 1952); reminiscent of more ancient species such as those described by Ross & Berry, 1963). The septa of G. persculptus begins at the 4th theca. These beds, except one, are in the portion of the Oualata-cliff, therefore to the NW-SE and above R, (but Deynoux cannot always decide between R, and R, towards the NW) in a facies of argillaceous shales and sandy layers and lenses, and some more micaceous beds. (c) Above in the same member and in the portion of Oualata-cliff: (i) A layer in a more sandy facies: C. normalis, C. transgrediens, C. medius, C. probably praemedius, the amplexograptid form, a proximal part of Rhaphidograptus?, a proximal part of Akidograptus? and some monograptid thecae. (ii) In the same facies as (b): C. normalis, C. miserabilis, C. minutus, amplexograptid form narrower than those above, Orthograptus truncatus abbreviatus Elles & Wood, Dimorpho- graptus sp., Pribylograptus incommodus (Tornquist) and Atavograptus ex gr. atavus (Jones). (iii) C. normalis, C. miserabilis, Pseudoclimacograptus (Metaclimacograptus) hughesi or undulatus (Kurck), Diplograptus modestus, D. diminutus Elles & Wood, and a single Peira- graptus or pathological specimen of Diplograptus sp.? (d) The landmark bed R, is above these layers, except in one section where it has not been recognized (C. normalis, P. (M.) hughesi, Dimorphograptus cf. confertus Lapworth), and the same facies as (b) begins again with C. normalis, C. rectangularis, P. (M.) hughesi or undulatus, D. modestus, Glyptograptus ex gr. tamariscus (Nicholson), G. tamariscus linearis? Perner, G. either angulatus Packham or distans Packham, ?Raphidograptus sp., A. atavus, A. strachani Hutt & Rickards, Lagarograptus acinaces? (Tornquist), and Coronograptus cyphus? (Lapworth). To the north of Aratane, beyond the post-Palaeozoic cover, towards Mejahouda and in the vicinity of Tinioulig, Sougy & Trompette (1976) have sampled the usual climacograptids, D. modestus, Cystograptus vesiculosus (Nicholson) and 4A. ex gr. atavus. All these graptolites are often irregularly flattened, preserved in iron oxides or with a fragile black pellicule. There is never an impression of fusellar tissue. Their deposit is rarely homogeneous along the rhabdo- some. Some layers contain brachiopods and numbers of other organic fragments. The Ordovician-Silurian boundary is therefore situated between the sandy landmarks R, and R, in the east of the Hodh. G. persculptus terminates the Ordovician, A. acuminatus is only suspected, and the remaining Rhuddanian is well represented. One should not forget that these collections are the first made systematically from this adverse environment, and reflect limited field-work, which was part of a large programme executed in a short time and with no possibility of immediate revision. The cliff at Hodh, in the Oualata area, if it were more accessible, would nevertheless be a first-rate place for a parastratotype, since it records the end of the African glacial phenomenon and has a good Ordovician-—Silurian transition. Recently, Legrand (1986) has described in detail (before the choice of the boundary) the lower Silurian at Oued in Djerane, Algeria, and has recognized new taxa. There is certainly some correlation between the Hoggar margin and the west of the Taoudeni Basin. However, before defining an ‘African’ fauna, it would be very useful to demonstrate with more certainty the effects of diagenesis on the preservation of graptolites, the more so because sections in proteic tissues have revealed the ability of the cortical layers to trap exogeneous particles. These extraneous particles could, of course, modify considerably any part of a rhabdosome. Conclusions From the Hodh to the Adrar, the post-glacial transgression would seem to have begun in the Ordovician and extended towards the west in the earliest Silurian, arriving later in the 182 S. WILLEFERT Zemmour. The cliff to the north-west of Oualata is the best exposure of the local Ordovician— Silurian boundary, though it is still necessary to fully describe and figure the graptolites and complementary faunas from there. References Cocks, L. R. M. 1985. The Ordovician—-Silurian boundary. Episodes, Ottawa, 8: 98-100. Deynoux, M. 1980. Les formations glaciaires du Précambrien terminal et de la fin de l’Ordovicien en Afrique de l'Ouest. Deux exemples de glaciation d’inlandsis sur une plate-forme stable. Trav. Lab. Sci. Terre St Jerome, Marseille, (B) 17: 1-315. ——., Sougy, J. & Trompette, R. 1985. Lower Palaeozoic Rocks of West Africa and the western part of Central Africa. In C. H. Holland (ed.), Lower Palaeozic of north-western and west central Africa: 337-495. London. Dia, O., Sougy, J. & Trompette, R. 1969. Discordances de ravinement et discordance angulaire dans le Cambro-Ordovicien de la region de Mejeria (Tagant occidental, Mauritanie). Bull. Soc. géol. Fr., Paris, (7) 11: 207-221. Elles, G. L. & Wood, E. M. R. 1901-18. A monograph of British Graptolites. Palaeontogr. Soc. (Monogr.), London. m + clxxi + 539 pp., 52 pls. Haylicék, V. 1971. Brachiopodes de lOrdovicien du Maroc. Notes Mém. Serv. géol. Maroc, Rabat, 230: 1-135, pls 1-26. Legrand, P. 1986. The lower Silurian graptolites of Oued In Djerane: a study of populations at the Ordovician-Silurian boundary. Spec. Publs geol. Soc. Lond. 20: 145-153. Manck, E. 1924. Grosskolonien von Climacograptus, Abdriicke von Zelltieren von Graptolithen. Natur, Leipzig, 16. Monod, T. 1952. L’Adrar mauritanien (Sahara occidental). Esquisse géologique. Bull. Dir. Mines Afr. occ. fr., Dakar, 15. Munch, A. 1952. Die graptolithen aus dem Anstehenden Gotlandium Deutschlands und der Tschechoslo- wakei. Geologica, Berl. 7: 1-157, pls 1-62. Ross, R. J. & Berry, W. B. N. 1963. Ordovician Graptolites of the Basin Ranges in California, Nevada, Utah and Idaho. Bull. U.S. geol. Surv., Washington, 1134: 1-177. Sougy, J. 1964. Les formations paléozoiques du Zemmour noir (Mauritanie septentrionale); étude strati- graphique, pétrographique et paléontologique. Annls Fac. Sci. Univ. Dakar 15: 1-695. Trompette, R. 1973. Le Précambrien supérieur et le Paléozoique inferieur de !Adrar de Mauritanie (bordure occidentale du bassin de Taoudeni, Afrique de l'Ouest). Un exemple de sédimentation de craton, étude stratigraphique et sédimentologique. Trav. Lab. Sci. Terre St Jerome, Marseille, (B) 7: 1-702. Ordovician—Silurian boundary in Victoria and New South Wales, Australia A. H. M. VandenBerg! and B. D. Webby? ‘Geological Survey Division, Department of Industry, Technology & Resources, P.O. Box 173, East Melbourne, Victoria, 3002, Australia *Department of Geology & Geophysics, University of Sydney, New South Wales, 2006, Australia Synopsis The late Ordovician and early Silurian is often represented by an unconformity or otherwise by beds bearing graptolites: no significant shelly faunas are known. In Darraweit Guim, Victoria, and in the Forbes—Parkes area of New South Wales, there may be beds spanning the Ordovician-Silurian boundary without a break, but nowhere have both the persculptus and acuminatus Zones been found in a single, structurally uncomplicated, succession. Introduction Ordovician and Silurian rocks crop out extensively in the Lachlan Fold Belt of southeastern Australia (Figs 1 and 3). A variety of facies is represented, from deep marine chert, black shale and turbidites, to shallow marine mudstone and sandstone. Carbonates and volcaniclastics occur, associated with island arc-type andesites in central New South Wales. The turbidite— black shale—chert association often contains rich and diverse graptolite assemblages and cono- donts, but virtually no shelly fossils. Mixed graptolite-shelly fossil assemblages occur in some of the volcaniclastic deposits, but the shallow marine carbonates only contain shelly fossils. Sections in central and eastern Victoria No single section spanning the Ordovician—Silurian boundary has yet been located in Victoria, although there is reasonably convincing evidence of a complete but fault-disrupted succession at Darraweit Guim, near Melbourne (Fig. 1). Poor exposure and deep weathering, and the scarcity of fossils in the Silurian rocks, are the main difficulties in locating further sections. Another limiting factor is due to the effects of the Benambran Orogeny, a major accretionary event which took place at about the Ordovician—Silurian boundary and produced the Wagga Metamorphic Belt in eastern Victoria (Cooper & Grindley 1982). The orogeny is marked by a prominent facies change, from black shale with or without turbidites, to massive mudstone or quartzite. East of the metamorphic belt, the facies change follows a break in sedimentation, which in some places was accompanied by folding. No such break in sedimentation occurs in the Melborne Trough in central Victoria, but here the lithological contrast produced by the Benambran Orogeny is such that the boundary interval became the preferred site for strike faulting during the Middle Devonian Tabberab- beran orogeny, thus causing considerable complexity in the boundary sections. Darraweit Guim The only apparently complete succession spanning the Ordovician-Silurian boundary in Victo- ria occurs at Darraweit Guim, a hamlet 46km NNW of Melbourne (Fig. 1). It is situated near the western margin of the Melbourne Trough, a basin in which there is record of continuous marine sedimentation from early Ordovician to late Early Devonian time (VandenBerg & Wilkinson, in Cooper & Grindley 1982). The boundary sequence recognized by VandenBerg et al. (1984) consists of three units, the Bolinda Shale, Darraweit Guim Mudstone and Deep Creek Siltstone (Fig. 2). Bull. Br. Mus. nat. Hist. (Geol) 43: 183-190 Issued 28 April 1988 184 A. H. M. VANDENBERG & B. D. WEBBY Ns Ww VICTORIA ORRYONG e¢ hoe ee BENDIGO ® EATHCOTE N ®& SEYMOUR Ke s DARRAWEIT eEipone \ & (1) MOUNT e7, CI BALLARAT GISBORNE | EASTON \ = WELLINGTON ( ) oF. PROVINCE © RIVER WARBURTON MELBOURNE s PROVINCE @ 2 2 BAIRNSDALE @ @ MORWELL ae SILURIAN [ised ORDOVICIAN : REGIONAL METAMORPHICS ct) 60 100 km _ SS ee | BASS STRAIT Fig. 1 Distribution of Ordovician and Silurian rocks in central and eastern Victoria. Localities mentioned in text and Fig. 2 are: 1, Darraweit Guim; 2, Mount Easton region; 3, Yalmy River; 4, Delegate (southeast N.S.W.). The Bolinda Shale is composed of 800m or more of thin-bedded coarse-grained black shale and fine sandstone with a rich Bolindian graptolite fauna, comprising mostly cosmopolitan species. The assemblage consists of very abundant Climacograptus latus, C. longispinus supernus and Orthograptus amplexicaulis (sensu lato), somewhat less abundant C. hastatus, C. cf. tubuli- ferus, Paraorthograptus pacificus pacificus and Dicellograptus ornatus, and rare specimens of Orthograptus fastigatus, Orthoretiograptus denticulatus and Pleurograptus linearis (sensu lato). This assemblage constitutes the Zone of D. ornatus and C. latus of VandenBerg (in Webby et al. 1981) and is virtually identical to that of the Paraorthograptus pacificus Subzone at Dob’s Linn (Williams 1982). The overlying Darraweit Guim Mudstone consists of. 20 to 45m of sparsely fossiliferous black calcareous mudstone and slump-folded mudstone of partly evaporitic origin, and may be the only unit in Australia to show the effects of the late Ordovician glaciation (VandenBerg, in prep.). The impoverished shelly fauna consists of small bivalves, hyolithids, straight nautiloids, and. a single trilobite, Songxites darraweitensis. More important, however, is the occurrence of Climacograptus? extraordinarius which is associated with C. angustus and C. cf. acceptus (VandenBerg et al. 1984). This assemblage represents the upper Bolindian Zone of C.? extraor- dinarius and is considered to correlate with the C.? extraordinarius Zone at Dob’s Linn (Williams 1983). Contacts between the Darraweit Guim Mudstone and the overlying Deep Creek Siltstone are usually poorly exposed and marked by bedding-parallel faults. The Deep Creek Siltstone is very thick (800-1000 m) and consists of poorly bedded, massive and bioturbated siltstone and thin rippled sandstone. Fossils are very rare. The lowest graptolite horizon occurs about 75m above the base of the formation (and about 90m above C.? extraordinarius) and contains Glyptograptus sp. (VandenBerg et al. 1984: fig. 11). A somewhat richer assemblage occurs 85m ORDOVICIAN-SILURIAN BOUNDARY IN AUSTRALIA 185 GLOBAL| GRAPTOLITE ZONAL MELBOURNE TROUGH | YALMY SERIES @| BIOSTRATIGRAPHY eects RIVER — DELEGATE STAGES DARRAWEIT | MT EASTON | MOUNT (SE NSW) | BRITISH |AUSTRALIAN| GUIM PROVINCE | TINGARINGY PZ ya 7 |S Mc ADAM a Sere elasyay 8 te th |_ magnus | SANDSTONE 5 | SS \ > we <¢ SS BS SAX SILURIAN LO ORDOVICIAN {¢) 100km laa pa Fig.3 Map showing the distribution of Ordovician and Silurian rocks in central and southern New South Wales, and the location of Ordovician-Silurian boundary sections represented in Fig. 4. 188 A. H. M. VANDENBERG & B. D. WEBBY Silurian boundary interval. The latest Ordovician deposits east of the Wagga Metamorphic Belt accumulated with associated graptolites in deeper waters as did much of the overlying Early Silurian, but many sections show physical breaks (unconformities, disconformities with associated facies changes or faults) reflecting the Benambran orogenesis or subsequent events. The few sections which appear to show conformity unfortunately have an incomplete record of Late Ordovician to Early Silurian graptolite assemblages—late Bolindian occurrences fol- lowed by a significant barren interval to the succeeding mid-Llandovery assemblages, making it impossible to position the boundary closely (Figs 3—4). In addition to the rarity of proven early Llandovery deposits, there is an even greater paucity of established late Bolindian to early Llandovery shelly faunas. Indeed the graptolites are the only group to be adequately represent- ed in the New South Wales successions. The sections with the best potential for establishing the Ordovician—Silurian boundary in New South Wales are in the Forbes area and east of Cano- windra. Two less important sections occur in the Angullong—Four Mile Creek area and east of Goulburn. 1. Forbes—Parkes. The Cotton Siltstone of the Forbes area comprises separate exposures of a lower unit of late Ordovician age and an upper unit of Early Silurian age (Sherwin 1970, 1973) with an extensive strip of ground in between, representing unexposed intervening beds. Sherwin identified two graptolite assemblages from the lower unit, fauna A characterized by Cli- macograptus supernus, C. hastatus, C. latus, Dicellograptus cf. elegans and Orthograptus trun- catus subsp., and assigned a Bolindian age; and fauna B typified by C. normalis and placed by Sherwin at or just above the Ordovician—Silurian boundary. The upper unit contains faunas C and D which are correlated with the late Llandovery (sedgwickii and turriculatus Zones); see also Sherwin (1974). C. normalis is the only determinable graptolite in fauna B and is a long-ranging species, and consequently can be of little use in establishing the position of the GLOBAL SERIES & STAGES GRAPTOLITE ZONAL BIOSTRATIGRAPHY FORBES- EAST OF (3) ANGULLONG- (4) EAST OF PARKES CANOWINDRA |'°’FOUR MILE CK.| ~ GOULBURN S >|< Ww m|=a Oa BEDS Lu a = => || ae O (shales) ZZ, = Ss COTTON DT IqRL ‘poq wW0y}eId o1foouo = ggO ‘UIaIaY [ 9IQeL Ul pu (MT g6I) SouIeY 7 UdyYOBIDOW Aq udAIs sojdues asoy}] WOY eyep UONNINsIP JUOpOoUOD ‘souleg pue ‘souleg 2 Ployjnd ‘squieg 2 UayORIQOW, Aq po}o9[[09 sofdures yUOpouod jo uontsod Surmoys (] “SI uo ATaAMoodsel gg puR “qz “WZ SertTeOOT) (YBII) JaATY UOWTeS pue (21u99) Avg sif[q Jo apis sam ‘(Jo]) asroquresyey oyUIOg ye [eAIoVUT Alepunog uvLIN[IS—URIDIAOPIO 9y} Jo AydessHeNsoyyy jo [rejeq ¢ “BY ORDOVICIAN-SILURIAN BOUNDARY IN ANTICOSTI Ail = ORDOVICIAN SILURIAN > < GAMACHIAN MENIERIAN Q S 6,7, = = — Qo? leae wo < ttt - 262m ---; 7 Ba ys w= 2 © 5 250m es a S| wz o|4 peor? ape oo wo J Lt = > ely N Se Q wo So (= oo] © ° 3 G o ro) ec Sis S = a aN 3° T=} = = Q wn N @ = is) = =|2 < nS = ze = A el e 78 a| 5 9 186m - SSee sess is} Wn) s/s} |,f=s 6 rs) SS 5 @ || S LS -— | on 3 [ 150 ° — m ol= 4 [a] s = 135m c [o) | 2 = _ = o & — Lt S o o iS Sale x4 = 2 S 2) fo) > = 8 a 2 © | <@ = 100m < 4 S i S — _ i= 3 S iS <= =| S35 (al oes Se Taco es Q sles 3 oe cars smn ag _——e Go wn aq a FE ao 3 ~ 3 i. Ww) 2 ae cc ee w/e * Sp oS Se ss Ss ee o| o | | = 2 < S 2 & & 8 o|Q 34m o- -2 ei “32 3 = Ses ly a Q Q is) c 1S) J2 | a S 2) | = iS € a a o 1 o lo So Ia i ee Ss & Bll cc Fig. 1 Graptolite ranges in the upper Vauréal, Ellis Bay and lower Becscie Formations of Anticosti Island. the Climacograptus prominens—elongatus Zone and interpreted its fauna (constituted primarily of biserial graptolites not easily related to those of other successions) as representing a level ‘intermediate between ... the youngest Ordovician and the oldest Silurian’ (1969: 551). He also referred the species used to name the zone to Climacograptus rather than Amplexograptus, as Barrass (1953) had done, because most specimens recovered from the core possessed clima- cograptid thecae with everted apertures rather than amplexograptid thecae. In 1981 he re- named the zone the Amplexograptus inuiti Zone on the recognition that A. elongatus Barrass was identical to Amplexograptus inuiti described by Cox (1933) and also its junior synonym. He also re-interpreted Amplexograptus prominens Barrass as a subspecies of A. inuiti. In 1985, I studied and sorted out the type material of Climacograptus latus Elles & Wood and came to the conclusion that this species belongs to Amplexograptus rather than Cli- macograptus, s.l., and is also identical to, and the senior synonym of, A. inuiti. A. latus was erected on flattened, fragmentary material and A. inuiti (Figs 4b—c) on excellent, isolated speci- mens from Akpatok Island in northeastern Canada. Cox (1933: 2) pointed out the similarity of A. inuiti to A. latus, but refrained from considering the two species identical because the thecal apertures of A. latus were ‘more even’ and lacked genicular flanges. In reviewing the type specimens of A. latus, | recognized apertural lappets in all specimens retained in the species and also residual genicular flanges (Figs 2a—h), but not in the specimens that I have excluded from it (Figs 2i-)), which belong to Climacograptus tubuliferus Lapworth. These features are even more pronounced in the topotype material recently identified as C. latus by Williams (1982: pl. 3, figs 12-18). The occurrence of A. latus in the A. prominens Zone of Anticosti is critical, for it allows us to correlate this zone with the Dicellograptus anceps Zone of Scotland and the Cli- GRAPTOLITES FROM ANTICOSTI ISLAND 223 macograptus pacificus Zone of the U.S.S.R. and their equivalents in China, the Yukon, and elsewhere, something which had not hitherto been possible. I have, however, refrained from naming this zone after either D. anceps or C. pacificus because neither graptolite has been recovered from Anticosti. Amplexograptus prominens itself is morphologically quite distinct from Amplexograptus latus and cannot be regarded as a mere subspecies or a morphological variant of it. The study of an original collection of A. prominens (made by Col. C. C. Grant) from the type strata at Observa- tion Cliff on the north shore of Anticosti Island fully confirms Barrass’ (1953) original diagnosis of this species. A. prominens is characterized by broad, short rhabdosomes which expand rapidly from a narrow proximal end (first pair of thecae), by prominent genicular flanges and the absence of a mesial spine on th 11. The long genicular flanges and the lack of a mesial spine on th 1’ set A. prominens well apart from all other species of Amplexograptus, although it shares with them a similar type of proximal-end development (early prosoblastic) and thecal style (amplexograptid with well-developed lappets) (Riva 1987). A. prominens is a unique species, known up to now only from the upper Vauréal Formation of Anticosti. It is the last Amplexo- graptus. It could well be the immediate ancestor to Paraclimacograptus decipiens sp. nov. which has a long range through the upper Vauréal and with which it has been confused in the past. P. decipens differs from A. prominens both in thecal form and the nature of genicular ornaments (Fig. 2s) but otherwise it shares with it the same type of proximal development and general distal rhabdosome structure (Figs 20-r). On the other hand, the isolated specimens from Manitoba referred to A. prominens by Jackson (1973: 2-4; text-figs 2B, E and F) are close to the topotypes of the older Paraclimacogratus manitoulinensis (Caley) shown here as Figs 5g, h and 1. Occasional low or incipient lappets are present both on the everted thecal apertures of the Manitoba specimens and the topotype specimens of P. manitoulinensis, and the Manitoba specimens have also a keel-like appression on outer margin of th 1’. One specimen referred to Amplexograptus inuiti by Jackson (1973: text-fig 2D) has also a mesial spine on th 1! in addition to the keel-like structure. This sort of structure has not been observed in topotype specimens of P. manitoulinensis, but a mesial spine has been reported and figured by Walters (1977) in specimens from the Lorraine Group of the St Lawrence Lowlands. The name Paraclimacograptus decipiens is proposed below for the short, stubby biserial graptolites which stratigraphically follow on A. /atus in the upper Vauréal Formation (Fig. 1). P. decipiens is morphologically close to A. prominens for which it may be easily mistaken (hence its specific name), but its thecae are of the paraclimacograptid type with clearly everted thecal apertures and reduced genicular flanges supported by two short genicular spines (Fig. 2s). The development of the proximal end is of the prosoblastic type and th 1! lacks a mesial spine, much as in A. prominens. The problem now arises as to the proper generic affiliation of the new species, which could be either in the genus Paraclimacograptus Pribyl, 1948 or Paraortho- graptus Mu, 1974. Paraclimacograptus has P. innotatus (Nicholson) as type species. P. innotatus (Figs 5l-n) is a thin, short graptolite, restricted to the lower Llandovery, with an advanced prosoblastic type of proximal-end development, thecae slightly inclined to the axis of the thabdosome with wide apertural excavations, everted thecal apertures and short genicular processes which turned out to be flanges in isolated Siberian specimens (Crowther 1981: pl. 13, fig. 4). It lacks a mesial spine on th 1'. Rickards (1970: 32) has also noted a complete median septum on deformed specimens identified as C. innotatus, but it is probably the trace of the virgula. Paraorthograptus has P. typicus Mu from the Upper Ordovician Paraorthograptus uniformis Zone of the Wufeng Shale of central China as type species. This species was described as having *... thecae of the orthograptid type with paired ventral spines ... pointed obliquely downward at the proximal end, horizontal at the distal end ... Interthecal septa straight, slightly inclined, not curved; apertural margins everted, not horizontal .... (Mu et al. 1974: 161; translated). No mention was made of the proximal end, which is not preserved in the holotype specimen (Fig. Sa); it is preserved, however, on a complete specimen on the type slab (Fig. 5b) and shows an apparently advanced type of proximal-end development, much as in Paraortho- graptus pacificus (Ruedemann) (Figs Sc-f). The type species of Paraclimacograptus and Para- orthogratus share the same basic rhabdosome morphology, i.e. a prosoblastic type of 224 J. RIVA proximal-end development, thecae inclined to the rhabdosome axis and wide thecal excavations with everted apertural margins. They differ, however, in the type and size of genicular processes which are flanges in species assigned to Paraclimacograptus (Fig. 5j) and genicular spines of various length in species included into Paraorthograptus. The latter also have a mesial spine on th 1’, a virgella and antivirgellar spines, whereas the former generally lack a mesial spine on th 1* (except in some specimens of P. manitoulinensis figured by Walters 1977) and also, appar- ently, antivirgellar spines in the younger species such as P. innotatus (see Crowther 1981: pl. 13, fig. 4). The problem is whether two genera are needed to group species on the basis of external morphology, conspicuous as it may be. Lin & Chen (1984: 216), for instance, have tried to solve this problem by simply assigning Climacograptus innotatus Nicholson to Paraorthograptus in describing Chinese specimens identified and figured as Paraorthograptus innotatus (Nicholson). However, a study of the Chinese specimens has revealed that they are fragmentary growth or juvenile stages of P. typicus. One of them, complete with mesial spine on th 1’ and long, paired genicular spines, is shown here as Fig. 5k. This deviation notwithstanding, I feel that the genus Paraclimacograptus should group species characterized by a prosoblastic proximal develop- ment (advanced as in the type species or more primitive as in P. manitoulinensis), thecae inclined to the rhabdosome axis, wide thecal excavations, everted apertures and genicular flanges. The genus Paraorthograptus should group all species which, in addition to the basic morphology of the paraclimacograptids, have genicular spines rather than flanges, a mesial spine on th 1' and antivirgellar spines. Paraclimatograptus decipiens has genicular processes consisting of reduced flanges supported by short, lateral spines (Fig. 2s). It may be regarded as a transitional form between species assigned to Paraclimacograptus and Paraorthograptus, but the fact that flanges are still present, genicular spines poorly developed and the rhabdosome lacks a mesial spine on th 1' support its inclusion in Paraclimacograptus, and it will be so described below. | The following graptolites have been identified from the P. prominens Zone from surface outcrops and the N.A.C.P. drill core (Fig. 1): Amplexograptus latus (Elles & Wood), Amplexo- graptus prominens Barrass, Paraclimacograptus decipiens n.sp., Glyptograptus cf. G. hudsoni Jackson, Peiragraptus fallax Strachan, Rectograptus abbreviatus (Elles & Wood), Orthograptus? and Desmograptus sp. In the N.A.C.P. well (Riva 1969), Amplexograptus latus has a short, 34m long range at the base of the P. prominens Zone, from the 2047 to the 1933 ft level (614-579 m), whereas P. decipiens ranges through the middle and upper part of the zone, from the 1647 to the 1376 ft level (493-412 m), for a total of at least 80m. Glyptograptus cf. G. hudsoni (Figs 2k—n) was described by Jackson (1971) from the Upper Ordovician of Southampton Island, north of Labrador and Akpatok Island; in the N.A.C.P. well it has a long range extending through both the D. complanatus and the A. prominens Zones to terminate somewhere in the upper Vauréal Formation (Fig. 1), for a total of at least 650m; P. fallax is a rare graptolite and has been recognized in only one collection from the mouth of the Patate River in association with A. latus, R. abbreviatus and G. cf. G. hudsoni (Riva & Petryk 1981); R. abbreviatus occurs sporadi- cally through both the D. complanatus and A. prominens Zones and two specimens were also collected by A. A. Petryk from member 6 of the Ellis Bay Formation, just below the Ordovician-Silurian boundary (Fig. 31). Correlation of the A. prominens Zone. A. latus is a cosmopolitan graptolite long recorded from the D. anceps Zone of southern Scotland and, especially, the D. complexus and P. pacificus Subzones (Williams 1982). This allows us definitely to correlate the A. prominens Zone of Anticosti Island with the uppermost British Ordovician. A. latus also occurs in the C. supernus Zone of Kazakhstan (Koren et al. 1980), where it has been described as Amplexograptus stukalinae, the C. pacificus Subzone of the Omulev Mountains of Siberia (Koren et al. 1983), where it is represented by A. latus hekandaensis, the Amplexograptus yangtzensis to the Diplo- graptus bohemicus Zones of the Wufeng Shale of central China (Mu & Lin 1984), where A. latus has been called A. suni and A. yangtzensis (Fig. 4a), and from the Bolindian D. ornatus and C. latus Zones of Victoria, Australia (VandenBerg 1981a). The A. prominens Zone of Anticosti is correlated with all the above-mentioned zonal levels (Fig. 1). GRAPTOLITES FROM ANTICOSTI ISLAND 225 Graptolites from the Ellis Bay and lower Becscie Formations. Graptolites are scarce above the A. prominens Zone. Few graptolites have been collected above member 2 of the Vauréal Formation besides a few specimens of G. cf. hudsoni (Figs 2l-n). Graptolites are also scarce in the Ellis Bay and Becscie Formations: the few collected are either indicative of the uppermost Ordovician or are long-ranging species that straddle the Ordovician—Silurian boundary. Members 4 and 7 of the Ellis Bay Formation have yielded fragmental climacograptids which I have assigned to Scalarigraptus angustus (Elles & Wood); one of them is shown as Fig. 31. In Scotland this graptolite ranges through the D. anceps Zone (Williams 1983: fig. 3) and may be taken to indicate that member 6 of the Ellis Bay is of uppermost Ordovician age. At Salmon River, the Becscie Formation has yielded fragments of S. angustus from its contact with the top of reef structures of the Ellis Bay upwards (Fig. 1). An excellent three-dimensional specimen of S. angustus was collected by A. A. Petryk a few metres above the base of the Becscie (Figs 3t, u); two small collections of this species were made 7 and 30m above the base of the formation at pool 9 on Salmon River (Figs 3j—s) and one specimen (Fig 3v) was collected from the Gun River Formation, well above the Ordovician—Silurian boundary. This is the longest specimen of S. angustus collected on Anticosti Island. S. angustus ranges from the Ashgill to the lower Llando- very, and it is common in the D. anceps, G. persculptus, A. acuminatus and other Zones at or above the Ordovician-—Silurian boundary and cannot be regarded as a good zonal indicator. In closing, it will be noted that a large climacograptid approaching Scalarigraptus normalis (Lapworth) in size (Fig. 3w) was collected by T. E. Bolton from the basal Becscie Formation on the east side of Ellis Bay at Cap-a-l’Aigle. S. normalis is only known from the G. persculptus Zone to the lower Llandovery. The new genus Scalarigraptus. The occurrence of graptolites of the normalis (or scalaris) group in the Ellis Bay and Becscie Formations brings again to the fore the problem of their generic affiliation which cannot any longer be the traditional polyphyletic genus Climacograptus Hall. Climacograptus was created by Hall (1865: 111-112) for ‘simple stipes with sub-parallel margins having a range of cellules (thecae) on each side’, which were to be ‘short and square’. Graptolithus bicornis was designated as the type species and the members of the G. scalaris group of Linné were ‘conceived’ as the ‘veritable species of this genus’. (The generic name Climacograptus was obtained by adopting the Greek noun klimax, equivalent to the Latin scala, ladder, of which scalaris is the adjective). Since its creation, this genus has known enormous popularity, having been used as a generic umbrella for all sorts of biserial graptolites characterized by square or climacograptid thecae, at least in the mature or distal part of the thabdosome. Elles & Wood (1906) attempted to deal with the large number of British grapto- lites assigned to Climacograptus by dividing them into five groups on the basic of thecal outline, type of apertural excavation or thecal ornaments such as spines, but did not propose new genera or subgenera. Pribyl (1947, 1948), on the other hand, went a step further and proposed the genus Pseudoclimacograptus for climacograptids characterized by a zig-zag median septum connected by transverse rods to the thecal septa and the genus Paraclimacograptus for cli- macograptids with genicular spines. The genus Pseudoclimacograptus has since been widely accepted by graptolite specialists, but the genus Paraclimacograptus has been overshadowed by the genus Paraorthograptus Mu, 1974. Riva (1974b, 1976) showed, on the basis of three- dimensional topotype material, that C. bicornis had a primitive diplograptid, or streptoblastic, type of proximal-end development and thus differs significantly from other climacograptids with a prosoblastic type of proximal-end development. The graptolites of the scalaris group, considered by Hall (1865: 112) as the ‘veritable species’ of Climacograptus, have an advanced prosoblastic type of proximal-end development and cannot be regarded as true cli- macograptids, although they share with C. bicornis a similar distal development. For this reason, I am proposing the new genus Scalarigraptus for all graptolites of the ‘scalaris’ or normalis group and for all Ordovician climacograptids with an advanced prosoblastic type of proximal-end development and with a septate or partly septate rhabdosome. C. normalis will be designated as the type species of the new genus. In 1949 Obut erected the genus Hedrograptus for early Silurian climacograptids with insig- nificant or incomplete apertural excavations on one side of the rhabdosome and complete on 226 J. RIVA the other. The figures of the type species, H. janischewskyi Obut, show that it is also character- ized by an advanced prosoblastic type of proximal-end development, just like C. normalis. In 1975, Obut extended Hedrograptus to include all climacograptids of the scalaris group. This would mean that Hedrograptus rather than the proposed Scalarigraptus is actually the genus intended for climacograptids of the scalaris group. I have, however, thanks to the cooperation of A. M. Obut, been able to examine a latex cast of the holotype of H. janischewskyi (Fig. 6a) and conclude that Hedrograptus is based on an incomplete and distorted specimen preserved in +-face view which does not allow us to ascertain whether the thecae are climacograptid or glyptograptid. Another specimen from the type locality, also preserved in }-face view (Fig. 6b), is much larger than the holotype of H. janischewskyi and probably not conspecific with it. For these reasons, I have been reluctant to adopt Hedrogratus and propose instead the genus Scalarigraptus. Systematic palaeontology Family DIPLOGRAPTIDAE Lapworth, 1873 Genus AMPLEXOGRAPTUS Elles & Wood, 1907 Amplexograptus latus (Elles & Wood) Figs 2a—h, 4 1906 Climacograptus latus Elles & Wood: 204-205; pl. 27, figs 3a—e and g, non figs 3f-h; text-figs 135a—d. 1933 Climacograptus inuiti Cox: 1-19, pls 1, 2. 1953 Amplexograptus elongatus Barrass: 62-66; figs 6-8. non 1970 Climacograptus latus Elles & Wood; Toghill: 22; pl. 15, figs 1, 2. 1974 Amplexograptus disjunctus yangtzensis Mu & Lin; Mu et al.: 162; pl. 70, fig. 6. 1980 Amplexograptus stukalinae Mikhailova; Koren et al.: 125-126; pl. 4, figs 1, 2. 1982 Climacograptus latus Elles & Wood; Williams: 39-40; pl. 3, figs 12-18. [See also for other synonyms. | 1983 Climacograptus latus hekandaensis subsp. nov.; Koren & Sobolevskaya: 116-117; pl. 30, figs 2-6; pl. 31, figs 1-3. 1983 Climacograptus latus Elles & Wood; Wang et al.: pl. 3, fig. 1. 1984 Amplexograptus suni (Mu); Mu & Lin: 56; pl. 5, figs 4-6. Fig. 2 Type specimens of Amplexograptus latus (Elles & Wood, 1906) and graptolites from the Vauréal Formation. ah, Type specimens of A. /atus from the upper Hartfell Shale, Main Cliff, Dob’s Linn; a, SM 19683b (Elles & Wood 1906: text-fig. 135a), paralectotype, x 5; b, SM A19680 (Elles & Wood 1906: pl. 27, fig. 3a), proposed lectotype, x 5; c, BU 1195 (Elles & Wood 1906: pl. 27, fig. 36), paralectotype, x 5; d, SM 19682a (Elles & Wood 1906: pl. 27, fig. 3g, text-fig. 135c), paralectotype, x 5; e, BU 1412b, unfigured paralectotype (on the same slab as BU 1412a of Fig. 2j), x 5; f, SM A19683c, unfigured growth stage, x 10; g, BU 1411a (Elles & Wood: pl. 27, fig. 3e), paralectotype, x 5; h, BU 1411b, unfigured paralectotype, x 5; 1-j, Scalarigraptus tubuliferus (Lapworth) originally included in the type material of A. latus; 1, BU 1413 (Elles & Wood: pl. 27, fig. 3h) doubtfully included, x 5; j, BU 1412a (Elles & Wood 1906: pl. 27, fig. 3f), x 5; k—n, Glyptograptus cf. G. hudsoni Jackson; k, G.S.C. 82880, from the 2739 ft (822 m) level in the N.A.C.P. core, x 5; 1, m, G.S.C. 82881, from member 2 of the Vauréal Formation at Cap Crotté, Anticosti Island (A. A. Petryk’s coll. 76 AP29-1), respectively x 10 and x 5; n, G.S.C. 82882, same locality and collection, x 5; o-s, Paraclimacograptus decipiens sp. nov.; 0, G.S.C. 82883, holotype, longest specimen recovered from the 1376ft (413m) level in the N.A.C.P. core x 5; p, GS.C. 82884, paratype, a large macerated specimen (A. A. Petryk’s coll. 83 AP6-5), from 90m above the mouth of Patate River, member 2, Vauréal Formation, x 5; q-s, G.S.C. 82885, 82886, paratypes, isolated specimens from the 1381 ft (414m) level in the N.A.C.P. core showing the development of the proximal-end thecal structure, x 15. Note the development of vertical cortex filaments in the apertural excavations of th 2? and 37. MDT GRAPTOLITES FROM ANTICOSTI ISLAND JX tA een S AP ay = SS = \= Be OO IO une y } \ ted & ) ¢ 228 J. RIVA Lectotype. SM A19680 (Fig. 2b) (Elles & Wood 1906: pl. 27, fig. 3a) from the upper Hartfell Shale, D. anceps Zone, Main Cliff, Dob’s Linn, Scotland. Herein selected. PARALECTOTYPES. SM A19683b and A19682a (Figs 2a, b), BU 1195 and 1411a (Figs 2c, g), BU 1414 and 1196 (not figured because of poor preservation) and the following specimens from the type collection, not previously figured: BU 1412b (Fig. 2e), 1411b (Fig. 2h) and a growth stage, SM A19683c (Fig. 2f). BU 1413 and 1412 (Figs 21, j) have been excluded from A. latus and assigned to C. tubuliferus. OTHER MATERIAL EXAMINED. Several topotype specimens of A. inuiti from Akpatok Island, the N.A.C.P. drill core and surface collections made by A. A. Petryk from member 2 of the Vauréal Formation, Anticosti Island. The type and topotype material of Amplexograptus stukalinae Mikhailova and of Climacograptus latus hekandaensis Koren & Sobolevskaya stored either at the VSEGEI in Leningrad or at the Institute of Geology and Palaeontology of the Akademya Nauk, Moscow, U.S.S.R.; the type or topotype material of Amplexograptus suni (Mu) and Amplexograptus disjunctus yangtzensis Mu & Lin at the Institute of Geology and Palaeontol- ogy, Academia Sinica, Nanjing, and at the Institute of Geology and Mineral Resources, Academy of Geological Sciences, Yichang, China. DESCRIPTION. Rhabdosome up to 5 to 6cm in length, gradually widening from 0-8—1-1 mm at the level of th 17 aperture to a maximum of 2:2-2:-4 (exceptionally 2:6) mm distally, attained within 2 or 3 cm. The average width, however, is less than 2mm, generally 1-6-1:83mm. A waist-like constriction may also be noted in some specimens above the the first pair of thecae. Thecae 14-12 in 10mm proximally, decreasing to 11-12 distally. Development of proximal end of prosoblastic type (Cox 1933: 6, 7; figs 1-21). The sicula is 1-5 mm long; it secretes a virgella and two antivirgellar spines. Th 1' originates low in the metasicula, grows down along the virgellar side to the sicular aperture, then turns out and upwards, secreting a mesial spine at the point of upward growth; th 1? buds off from the downward-growing portion of th 1', grows around the reverse side of the sicula to turn up at the point of issuance of the antivirgellar spines (Fig. 4). Th 2’ buds off th 1’ and th 2? from th 1? and so on alternately to the distal end of the aseptate rhabdosome. Thecae are of the amplexogratid type with apertural lappets and thecal excavations occupying about 4 of the rhabdosome width. A selvage runs around the thecal apertures and the infragenicular walls to form a short genicular flange. REMARKS. The type material of A. latus was mixed, containing two specimens herein assigned to C. tubuliferus (Figs 2i, j). Because of its world-wide distribution, this species has been identified and described as C. latus and also under a number of names such as C. inuiti and A. stukalinae Mikhailova, Climacograptus latus hekandaensis Koren & Sobolevskaya for specimens from Kazakhstan and NE Siberia, and as Amplexograptus disjunctus Mu & Zhang, Climacograptus suni (Mu) and Amplexograptus disjunctus yangtzensis Mu & Lin for specimens from the Upper Ordovician Wufeng Shale of central China. A. yangtzensis is a species in its own right and not a subspecies of A. disjunctus, a nomen nudum, the type of which could not be located in a recent study visit to Nanjing. It is based on a single three-dimensional specimen (Mu et al. 1974: pl. 70, fig. 4), here refigured as Fig. 4a, from a zone of the same name in the lower Wufeng Shale, where graptolites are generally preserved in relief in a black shale. Farther up in the Wufeng Shale, A. yangtzensis is replaced by A. suni, which differs from A. yangtzensis only in being preserved as flattened, brown, flaky films. The specimens from Dob’s Linn, Scotland, identified as C. latus by Toghill (1970) belong to either S. normalis or S. tubuliferus. STRATIGRAPHICAL AND GEOGRAPHICAL OCCURRENCE. A. latus is restricted to the D. anceps Zone of Scotland (Williams 1982) and may be considered as one of its diagnostic fossils. It is a widely distributed, cosmopolitan species, known from the C. supernus Zone of Kazakhstan and NE Siberia, the Upper Ordovician of China and correlative strata elsewhere. In SE Australia it helps name the upper Bolindian D. ornatus—C. latus Zone (VandenBerg 198 1a). GRAPTOLITES FROM ANTICOSTI ISLAND 229 Genus PARACLIMACOGRAPTUS Pribyl, 1948 TYPE SPECIES (by original designation). Climacograptus innotatus Nicholson (Nicholson 1869: 238; pl. 11, figs 16, 17). DIAGNOSIS (amended from Pribyl 1948: 40-41, 47-48, fig. 6). Rhabdsome aseptate, apparently ovoid in cross-section; thecae of the paraclimacograptid type, inclined to the axis of the thabdosome; apertural excavations wide and deep with everted thecal apertures and genicular flanges, strengthened by a selvage (list) split into two short spines at the geniculum in some species. Proximal end characterized by a prosoblastic type of development, and provided with a virgella, antivirgellar spines and, exceptionally, a mesial spine on th 1' (in older species). INCLUDED SPECIES. The following species may be included in Paraclimacograptus: Paraclimaco- graptus innotatus (Nicholson), Paraclimacograptus manitoulinensis (Caley), Paraclimacograptus decipiens sp. nov., Paraclimacograptus sp., an undescribed species from the Climacograptus wilsoni Zone of Gaspé, Canada. Climacograptus innotatus nevadensis Carter (Riva 1974a: figs 2k—m) from the late mid- Ordovician of Nevada, Texas (Marathon region), Oklahoma (unpubl. data) and Australia (VandenBerg 1981b) is close to Scalarigraptus. This species has an advanced prosoblastic proximal-end development, thecae of the climacograptid type with stiff genicular spines in the first six to twelve pairs, a long virgella accompanied by a sicular downgrowth, a long inflated virgula and a sicula lacking the prosicula. These characteristics brings it closer to the scalarigraptids of the tubuliferus group of the Upper Ordovician rather than to the para- climacograptids. Paraclimacograptus decipiens sp. nov. Figs 20-s Ho.otyPe. G.S.C. 82883 (Fig. 20), from the 1376 ft (413m) level in the N.A.C.P. core, upper Vauréal Formation, Anticosti Island. PARATYPES. G.S.C. 82884 (Fig. 2p), from 90m above the mouth of Patate River, Anticosti Island, member 2 of the Vauréal Formation; G.S.C. 82885 and 82886 (Figs 2q-s), isolated growth stages from the 1381 ft (414 m) level of the N.A.C.P. core, upper Vauréal Formation. Name. Latin decipiens, deceiving. DESCRIPTION. Rhabdosome of moderate length, usually not exceeding 2 to 3cm, maximum observed 4cm (Fig. 20), widening rapidly from 0-8-1-0mm at the level of the aperture of th 1? to 1-6—2-0mm (maximum observed 2:-4mm) at the level of the 4th to Sth pair of thecae. Thecae numbering 8 in 5mm, or 15 in 10mm, proximally, decreasing to 12-13 in 10mm distally, of the paraclimacograptid type with everted thecal apertures, except for the first two which have low lappets (faintly visible also on the second pair of thecae in Fig. 2s). Interthecal septa inclined at 20° to 40° to the rhabdosome axis; supragenicular walls parallel or slightly inclined to it. Thecal excavations wide, occupying + of the rhabdosome width, reinforced by a selvage running around the thecal aperture and the infragenicular wall and terminating as two short, stiff genicular spines supporting a reduced hood (Fig. 2s). Development of the proximal end of the prosoblastic type. Sicula about 1-Smm long, partly exposed on the obverse side of the thabdosome (Figs 20, s). Th 1! originates low in the metasicula, grows down the virgellar side to the sicular aperture before turning out and upwards to terminate about level with its point of origin. Th 1? buds off the downward-growing portion of th 1', grows diagonally around and up on the obverse side of the rhabodosome (Fig. 2r); th 2! buds off from th 17 and th 2? from th 1? and so on alternately to the distal end of the rhabdosome. A thin nema passes through the rhabdosome and extends a short distance beyond it. The rhabdosome is aseptate. REMARKS. The development of the proximal end of P. decipiens is identical to that of A. latus and A. prominens, suggesting a close genetic relationship between the three species. P. decipiens is much larger than P. innotatus which has a proximal development of the advanced prosoblas- 230 J. RIVA tic type. P. decipiens is much closer to P. manitoulinensis from the lower Upper Ordovician of NE North America (Riva 1969) (Figs 5g, h and 1), but this species is thinner, of uniform width and has genicular flanges strengthened by a thickened selvage (Fig. 5j). A mesial spine on th 1! may occur sporadically in some rhabdosomes (Walters 1977). STRATIGRAPHICAL AND GEOGRAPHICAL OCCURRENCE. P. decipiens is only known from the 4A. prominens Zone of Anticosti, where it has a stratigraphical range of at least 80m in the upper Vauréal Formation (Riva 1969). It has been found also sporadically in recent surface collections made by A. A. Petryk and in an older collection (Y.P.M. 3036/4) made by W. H. Twenhofel and stored at the Peabody Museum of Yale University (Riva & Petryk 1981: 160). Genus SCALARIGRAPTUS nov. TYPE SPECIES. Climacograptus normalis Lapworth (Lapworth 1877: 138; pl. 6, fig. 31; Elles & Wood 1906: pl. 26, fig. 2a; Williams 1983: text-fig. 4a). NAME. From the Latin scalaris, ladder-like. DIAGNOsIS. Rhabdosome septate or partly septate, ovoid to subrectangular in cross-section; thecae of the climacograptid type with definite genicula, deep horizontal apertural excavations and straight supragenicular walls, usually parallel to the axis of the rhabdosome. Proximal-end development of the advanced prosoblastic type with only th 1’ initially growing down along the sicula. The virgella is the only proximal spine. INCLUDED SPECIES. The following species, among others, fall within the limits of the diagnosis of Scalarigraptus: C. normalis, C. angustus (Perner), C. transgrediens Waern, C. medius Tornquist, C. praemedius Waern, C. rectangularis M‘Coy, C. brevis Elles & Wood, C. putillus (Hall), C. tubuliferus Lapworth, C. nevadensis Carter, C. yumenensis Mu and C. biformis (Mu & Lee). Fig. 3 Syntypes of Climacograptus miserabilis Elles & Wood, 1906 and graptolites from the Ellis Bay and the lower Becscie Formations. a—c, Syntypes of C. miserabilis; a, BU 1148b (Elles & Wood 1906: text-fig. 120b), proximal end with long virgella (freed from matrix), x 5; b, BU 1150 (Elles & Wood 1906: text-fig. 120a), typical specimen with long virgella (freed from matrix), x 5; c, BU 1146a (Elles & Wood 1906: pl. 26, fig. 3b and text-fig. 120c), distal fragment showing thread-like virgula passing through the thin rhabdosome, x 5. d-h, Scalarigraptus angustus (Perner) from the Ellis Bay Formation; d, G.S.C. 82887, obverse view of growth stage preserved in relief, showing climacograptid thecae and wavy median septum, from the oncolite platform bed, basal member 7 (A. A. Petryk’s collection 84AP8-2-1F), Pointe Laframboise, Cape Henry, x 10; e-h, G.S.C. 82888— 82891, large distorted or fragmentary rhabdosomes from upper member 4 (A. A. Petryk’s collection 81AP3-2), Baie des Navots, Ellis Bay, x 5. i, G.S.C. 82892, Rectograptus abbreviatus (Elles & Wood), macerated specimen from member 5, Ellis Bay Formation, immediately below reef bio- herms, 7 km upriver from mouth of Salmon River, right bank (A. A. Petryk’s collection 75APt3-3), x 5. jm, S. angustus (Perner) from the basal beds of the Becscie Formation; j, k, G.S.C. 82893, 82894, a growth stage and an adult individual showing a thin virgella distally prolonged (A. A. Petryk’s collection 81AP13-1-1F), from pool 9, Salmon River, 13m above the base of the forma- tion, x 5; l-m, G.S.C. 82895, 82896, from the basal Becscie at pool 9 on Salmon River (collected by J. Riva 1981), x 5. n-s, G.S.C. 82897-82902, growth series of S. angustus (A. A. Petryk’s collection 79AP48-4), 7m above base of the Becscie, base of pool 9 on Salmon River, x 5. t, u, G.S.C. 82903, observe view of S. angustus preserved in excellent relief, showing wavy median septum in proximal part of rhabdosome (A. A. Petryk’s collection 76AP22-30-6’), 2-3m above base of Becscie Forma- tion on Salmon River, respectively x 10 and x 5. v, G.S.C. 82904, longest specimen of S. angustus recovered from the mid-part of the Gun River Formation, 3-5km from mouth of Chute Creek, eastern Anticosti (A. A. Petryk’s collection 75MPt18-L8C-1F), x 5. w, G.S.C. 69157, Scalarigraptus normalis (Lapworth), collected by T. E. Bolton in 1981 from the basal Becscie Formation on the east shore of Ellis Bay near Cap-a-l’Aigle, Anticosti Island, x 5. 231 GRAPTOLITES FROM ANTICOSTI ISLAND ~ 5 . ms SO a \ CWS : Ne gS Ua SH Ween TY. 2 WOK EY SSM MENA Salsas AOD SN ‘ = = a ROE ee oe PS aS é — SR Se ; ES SSS ED ; Ep i an ie Sep A FCR = Oe 232 J. RIVA Scalarigraptus angustus (Perner, 1895) Figs 3a—u 1895 Diplograptus (Glyptograptus) euglyphus Lapworth var. angustus Perner: 48; pl. 8, figs 14a, b. 1906 Climacograptus scalaris (Hisinger) var. miserabilis Elles & Wood: 186; pl. 26, figs 3a, b, d, e, g, h, non figs 3c, f; text-figs a—c. 1951 Climacograptus angustus (Perner) Pribyl: 7; pl. 2, figs 2-9. 1975 Climacograptus angustus (Perner); Bjerreskov: 23; fig. 9A. 1980 Climacograptus angustus (Perner); Koren et al.: 131; pl. 37, figs 2-7; text-figs 34a-e. 1983 Climacograptus angustus (Perner); Koren & Sobolevskaya: 106-108; pl. 27, figs 1—5; text-fig. 34. 21983 Climacograptus mirnyensis (Obut & Sobolevskaya); Koren & Sobolevskaya: 132-133; pl. 37, figs 2-5; text-figs 47K—H. 1983 Climacograptus miserabilis Elles & Wood; Williams: 615-616; text-figs 3fH, ?j, 4f, Sa—b. [See also for a more extended pre-1983 synonymy. | Ho.otyPe. National Museum of Prague CD 1835, partly figured by Perner (1895: pl. 8, figs 14a—b) and refigured in full by Pribyl (1951: pl. 2, fig. 8). MATERIAL STUDIED. The type collection of C. miserabilis in the Lapworth collection of Birming- ham; part of the collections made by P. Toghill at Dob’s Linn; the type and topotype material of C. angustus in Prague; the collections of C. angustus and C. mirnyensis at VSEGEI, Lenin- grad, several collections made by A. A. Petryk from the Ellis Bay, Becscie and Gun River Formations of Anticosti Island. DESCRIPTION. Rhabdosome up to 2cm in length, widening imperceptibly from 0-8—0-9mm at the level of th 1? aperture to a maximum of 1-0-1-1 mm (exceptionally 1-2 mm) within one pair of thecae. Thecae of the climacograptid type, numbering 11-12 in the first 10mm, decreasing to 10-11 distally, with sharp genicula and supragenicular walls parallel to slightly inclined to the rhabdosome axis. Thecal apertures horizontal to slightly everted; thecal excavations wide and semicircular, occupying about 4 of the rhabdosome width and reinforced by a thin selvage around the aperture and the infragenicular walls, terminating as a slight genicular flange (Figs 3d, t). Development of the proximal end of the advanced prosoblastic type. Sicula from 1-2 to 1-6mm long, secreting a long virgella; it is mostly exposed on the obverse side of the rhabdosome (Williams 1983: text-fig. 3h). Th 1’ first grows down along the sicula and then turns out and upwards at the sicular aperture (Figs 3d, t); th 1? grows up from th 1! and th 2? from th 17. Th 2! is also the dycalical thecae which gives rise to two independent linear series separated by a median septum. The median septum begins on the obverse side of the rhabdo- some at about the level of th 1* aperture (its point of origin is marked by a notch in some specimens, Fig. 3t) and follows a wavy pattern through the first 5 or 6 pairs of thecae before straightening out (Figs 3d and t). A thin, thread-like nema passes through the rhabdosome and extends for some distance beyond. REMARKS. In 1951 Pribyl pointed out that C. miserabilis Elles & Wood 1906 was identical to, and the junior synonym of, C. angustus (Perner 1895). This synonymy was accepted by some workers (for instance Bjerreskov 1975: 23) but not by British workers for a number of reasons best summarized by Williams (1983: 616). Recently, I have been able to study the type material of both C. miserabilis and of S. angustus. C. miserabilis is based on seven specimens from the D. complanatus Zone and two from the D. anceps Zone of Dob’s Linn, Scotland. The two speci- mens from the D. anceps Zone do not belong to C. miserabilis: one, BU 1145b (Elles & Wood 1906: pl. 26, fig. 3c), is a distal fragment of tubuliferus, and the other, BU 1149 (Elles & Wood: pl. 26, fig. 3f), is of uncertain affiliation. The specimens from the D. complanatus Zone (three of which are shown here as Figs 3a—c) are preserved as thin, flaky, abraded films. They all belong to C. miserabilis. They are from 0-8 to 1-1mm wide and have 12-11 thecae per 10mm proxi- mally and 11 distally. The proximal end bears a long virgella, and a thin nema passes through the rhabdosome. This is all that can be learned from the type material of C. miserabilis. The type and topotype material of S. angustus is more diversified and contains several specimens in partial relief. (I was unable to draw any specimens, but was assisted in my work by Dr A. GRAPTOLITES FROM ANTICOSTI ISLAND 233 Fig. 4 a, N.I.G.P. Catalogue Number 21410, holotype of Amplexograptus yangtzensis Mu & Lin (=A. latus), x 20; b and c, SEM montages of Amplexograptus inuiti (Cox) (=A. latus) from Akpatok Island, Canada: b, SM A102524, obverse view; c, SM A102521, reverse view, both x 20 (courtesy of Peter Crowther). Pribyl). The specimens attain a width of 1:0-1-:1 mm, have 12-11 thecae per 10mm proximally and 10 distally. The thecae are all of the climacograptid type with strong genicula. The proxi- mal end bears a long virgella and a thin virgula passes through the rhabdosome. The holotype is a complete, not partial, specimen as claimed by Strachan (1971: 34); it has been refigured in full by Pribyl (1951: pl. 2, fig. 8). With the aforesaid in mind, I do not see any morphological differences between the types of C. miserabilis and S. angustus and do not hesitate to place the former in synonymy with the latter. The specimens from the basal Becscie Formation (Figs 3j—u) are all practically identical to the type of S. angustus and so are those from the Gun River Formation. The specimens from member 4 of the Ellis Bay Formation (Figs 3e—-h) are wider (from 1-1 to 1-3mm) because of poor preservation and distortion; that from the base of member 7 (Fig. 3d) has the same dimensions as the holotype in Prague. STRATIGRAPHICAL AND GEOGRAPHICAL OCCURRENCE. S. angustus is a cosmopolitan graptolite ranging through the Upper Ordovician and part of the Lower Silurian. In NE Siberia (Omulev Mountains) it is common from the base of the C. extraordinarius Zone to the top of the A. acuminatus Zone (Koren et al. 1983: figs 62, 64). On Anticosti Island it first occurs at the top of the P. manitoulinensis Zone (Riva 1969: figs 11, 13), below the base of the D. complanatus Zone, and extends all the way up into the Gun River Formation of mid-Llandovery age. ee Nu \ ill rn InnNaniy Wi Yi Y "Ty AN _Illl n 1 Fig. 5 a, b, Paraorthograptus typicus Mu; a, N.I.G.P. Cat. No. 21418a, counterpart of the holotype (better preserved than the part) from the Wufeng Shale north of Yichang, central China, showing the characteristic long, paired genicular spines of the species but with the proximal end missing (a thabdosome of Climacograptus longispinus supernus Elles & Wood lies diagonally across its proxi- mal end), x 5; b, unfigured specimen of P. typicus, with a complete proximal end, occurring on the same slab as the holotype, x 5. c-f, U.S.N.M. 415038415401, rhabdosomes of Paraorthograptus pacificus (Ruedemann) from the Phi Kappa Formation at Trail Creek, Idaho, U.S.A., near the type locality of the species, showing their characteristic short genicular spines, both paired and triple, and stubby form; note the tectonic deformation undergone by specimens of Figs Sc and d lying normal to each other, x 5. gj, G.S.C. 56899, 56895, 56900 and 56901, respectively, topotypes of Pseudoclimacograptus manitoulinensis (Caley) from the upper Whitby Formation, 5 km south of Little Current west side of Rt 68, Manitoulin Island, Ontario, Canada; g—1, growth series showing distinct fusellar rings, x 10; j, detail of thecal excavations showing everted thecal apertures and well-developed genicular lappets strengthened by a selvage, x 20. k, N.ILG.P. Cat. No. 82816, proximal end of P. typicus figured as Paraorthograptus innotatus (Nicholson) by Lin & Chen (1984: pl. 4, fig. 7), showing the spinose processes typical of the species: virgella, antivirgellar spines, mesial spine on th 1! and genicular spines, x 10. l-n, Paraclimacograptus innotatus (Nicholson), topotypes from the lower Birkhill Shale (Lower Silurian) at Dob’s Linn, southern Scotland; 1, SM A20222, specimen figured by Elles & Wood (1906: pl. 27, fig. 10a) as a ‘typical specimen’ (but not the ‘type’ of Nicholson), x 5; m, n, SM A20232 (op. cit.: pl. 27, fig. 106), specimen showing advanced prosoblastic development of proximal end and a partly uncovered sicula below th 17, x Sand x 10, respectively. GRAPTOLITES FROM ANTICOSTI ISLAND DBS a Dh a {7 a aN =le raul a } elas ley [ | ais P 4 a4 tH ac WW all iad wae | i : \ Ly } ) Sp LB iss t | 5 J ( rs fia LS | 2 Fig. 6 a, I.G:G—COAH-SSSP No. 278/5, 1945, te 2 a camera lucida drawing of latex cast of the : i 2 holotype of Hedrograptus janischewskyi Obut from the Lower Silurian (Llandovery) of the southern Ural Mountains, U.S.S.R., preserved as a 3-face view impression, x 4; b, IL.G.G— COAH-SSSP No. 278/6, 1945, a ‘topotypic specimen’ of H. janischewskyi ‘from the same locality as the holotype and the closest to the type’ (Obut, in litt. 1984), preserved as a 3-face impression in a light-grey aphanitic limestone b with most of the periderm missing, x 4. ) =e toa = Wah a —— Das == SS Re Neen Acknowledgements I thank Dr Barrie Rickards for hospitality and facilities during several visits to the Sedgwick Museum, Mr P. J. Osborne for the loan of specimens from the Lapworth Collection, Birmingham, Dr Tatyana N. Koren, VSEGEI, Leningrad, for permission to study the collections from the Omuley Mountains and Kazakhstan; Dr A. Pribyl for his help in studying graptolites at the National Museum of Prague, Dr A. A. Petryk for permission to study his Anticosti collections, Miss Claire Carter, U.S.G.S., for the loan of 236 J. RIVA topotype specimens of P. pacificus, Li Ji-jin for his assistance at the Academia Sinica in Nanjing, Wang Xiao-feng for organizing a field excursion near Yichang, Mme Aicha Achab for the use of the INRS- Géoressources photolaboratory and my daughter Patricia for translations of Russian papers. This work was supported in part by a research grant from the N.R.C. of Canada and by a sabbatical travel grant from Université Laval. References Barrass, R. 1953. Graptolites from Anticosti Island. Q. JI geol. Soc. Lond. 110: 55-75. Bjerreskov, M. 1975. Llandoverian and Wenlockian graptolites from Bornholm. Fossils Strata, Oslo, 8: 1-94, pls 1-13. Cox, I. 1933. On Climacograptus inuiti sp. nov. and its development. Geol. Mag., London, 70: 1-19. Crowther, P. R. 1981. The fine structure of the graptolite periderm. Spec. Pap. Palaeont., London, 26: 1-119. Elles, G. L. & Wood, E. M. R. 1901-18. A monograph of British Graptolites. Palaeontogr. Soc. (Monogr.), London. m + clxxi + 539 pp., 52 pls. Hall, J. 1865. Graptolites of the Quebec Group. Figures and Descriptions of Canadian organic-remains, Dec. II. 151 pp. Montreal, Canada geol. Surv. Jackson, D. E. 1973. Amplexograptus and Glyptograptus isolated from Ordovician limestones in Mani- toba. Bull. geol. Surv. Can., Ottawa, 222: 1-8. Koren, T. N., Mikhailova, N. F. & Tsai, D. T. 1980. Class Graptolithina. Graptolity. In M. K. Apollonoy, S. M. Bandaletov & I. F. Nikitin (eds), The Ordovician—Silurian boundary in Kazakhstan. 300 pp. Alma Ata, Nauka Kazakh S.S.R. Publ. Ho. ——, Oradovskaya, M. M., Pylma, L. J., Sobolevskaya, R. F. & Chugaeva, M. N. 1983. The Ordovician and Silurian boundary in the Northeast of the U.S.S.R. 208 pp., 48 pls. Leningrad, Nauka [In Russian ]. Lapworth, C. 1877. On the graptolites of County Down. Rep. Proc. Belf. Nat. Fld Club 1876-77 (Appendix): 125-144, pls 5—7. Lesperance, P. J. 1985. Faunal distributions across the Ordovician-Silurian boundary, Anticosti Island and Percé, Québec, Canada. Can. J. Earth Sci., Ottawa, 22: 838-849. Lin Yao-kun & Chen Xu 1984. Glyptograptus persculptus Zone—the earliest Silurian graptolite zone from Yangzi Gorges, China. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Strati- graphy and Palaeontology of Systemic Boundaries in China. Ordovician—Silurian Boundary 1: 199-223, pls 1—6. Anhui Sci. Tech. Publ. House. Mu En-zhi, Ge Mei-yu, Chen Xu, Ni Yu-nan & Lin Yao-kun 1974. In: A Handbook of the stratigraphy and palaeontology of Southwest China: 154-221. China Publishing House, Nanjing. & Lin Yao-kun 1984. Graptolites from the Ordovician-Silurian boundary sections of Yichang area, W. Hubei. Jn Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeontology of Systemic Boundaries in China. Ordovician—Silurian Boundary 1: 45-73. Anhui Sci. Tech. Publ. House. Nicholson, H. A. 1869. On some new Species of Graptolites. Ann. Mag. nat. Hist., London, (4) 4: 231-242. Obut, A. M. 1949. Polievoj atlas rukovodyashchikh graptolitov verkhnego silura Kirghizskoj S.S.R.: 1-57, pls 1—7. Publishing House of the Academy of Science of the U.S.S.R., Frunze. — 1975. Tip Hemichordata—Klass Graptoloidea. In A. A. Nikolaev et al. (eds), Polievoj atlas silurijskoj fauny severo-vostoka S.S.S.R.: 145-183. Magadan. Perner, J. 1895. Studie o ceskych graptolitech, cast II. Palaeontogr. Bohem., Prague, 3b: 1-52, pls 1-8. Petryk, A. A. 1979. Stratigraphie revisee de l'Ile d’Anticosti. Québec Ministére de l’Energie et des Ressourses, DPV-711: 1—24. Pribyl, A. 1947. Classification of the genus Climacograptus Hall, 1865. Bull. int. Acad. tcheque Sci., Prague, An. 48 (2): 1-12, pls 1-2. —— 1948. Some new subgenera of graptolites from the Families Dimorphograptidae and Diplograptidae. Vest. st. geol. Ust. ésl. Repub., Prague, 23: 37-48. 1951. Revision of the Diplograptidae and Glossograptidae of the Ordovician of Bohemia. Bull. int. Acad. tcheque Sci., Prague, 50 (1949): 1—S1, pls 1-5. Rickards, R. B. 1970. The Llandovery (Silurian) graptolites of the Howgill Fells, Northern England. Palaeontogr. Soc. (Monogr.), London. 108 pp., 8 pls. Riva, J. 1969. Middle and Upper Ordovician graptolite faunas of the St Lawrence Lowlands, and of Anticosti Island. Mem. Am. Ass. Petrol. Geol., Tulsa, 12: 513-556. — 1974a. Graptolites with multiple genicular spines from the Upper Ordovician of Western North America. Can. J. Earth Sci., Ottawa, 11: 1455-1460. GRAPTOLITES FROM ANTICOSTI ISLAND Way) — 1974b. A revision of some Ordovician graptolites of eastern North America. Palaeontology, London, 17: 1-40. — 1976. Climacograptus bicornis bicornis (Hall), its ancestor and likely descendants. In M. G. Bassett (ed.), The Ordovician System: Proceedings of a Palaeontological Association symposium, Birmingham, September 1974: 589-619. Cardiff, Univ. Wales Press & Natl Mus. Wales. 1987. The species Amplexograptus praetypicalis n. sp. and the origin of the typicalis group. Can. J. Earth Sci., Ottawa, 24 (5): 924-933. & Petryk, A. A. 1981. Graptolites from the Upper Ordovician and Lower Silurian of Anticosti Island and the position of the Ordovician-Silurian Boundary. In P. J. Lespérance (ed.), Field Meeting, Anticosti-Gaspe, Quebec, 1981] 2 (Stratigraphy and paleontology): 159-164. Montreal (I.U.G.S Subcom- mission on Silurian Stratigraphy Ordovician—Silurian Boundary Working Group). Strachan, I. 1954. The structure and development of Peiragraptus fallax gen. and sp. nov. Geol. Mag., Hertford, 91: 509-513. —— 1971. A synoptic supplement to ‘A Monograph of British Graptolites by Miss G. L. Elles and Miss E. M. R. Wood’. Palaeontogr. Soc. (Monogr.), London. 130 pp. Toghill, P. 1968. Graptolite assemblages and zones of the Birkhill shales (Lower Silurian) at Dobb’s Linn. Palaeontology, London, 11: 654-668. 1970. Highest Ordovician (Hartfell Shales) graptolite faunas from the Moffat area, South Scotland. Bull. Br. Mus. nat. Hist., London, (Geol.) 19: 1—26, pls 1-16. Twenhofel, W. H. 1928. Geology of Anticosti Island. Mem. geol. Surv. Brch Canada, Ottawa, 154: 1-481. VandenBerg, A. H. M. 1981a. Victorian stages and graptolite zones. In B. D. Webby (ed.), The Ordovician System in Australia, New Zealand and Antarctica: 2-6. 1.U.G.S. Publication 6. —— (1981b). A complete Late Ordovician graptolite sequence at Mountain Creek near Deddick, eastern Victoria. Unpubl. report, geol. Surv. Victoria 1981/81. Walters, M. 1977. Middle and Upper Ordovician graptolites from the St Lawrence Lowlands, Québec, Canada. Can. J. Earth Sci., Ottawa, 14: 932-952. Wang Xiao-feng 1983. Latest Ordovician and earliest Silurian faunas from the eastern Yangtze Gorges, China, with comments on Ordovician-Silurian boundary. Bull. Yichang Inst. Geol. Min. Res. 6: 129- 163. Williams, S. H. 1982. The Late Ordovician graptolite fauna of the Anceps Bands at Dob’s Linn, southern Scotland. Geologica Palaeont., Marburg, 16: 29-56, 4 pls. — 1983. The Ordovician-Silurian boundary graptolite fauna at Dob’s Linn, southern Scotland. Palae- ontology, London, 26: 605-639. >. > < ~——— : — a : J Pty = < . ,. z —i | t a9 { = “Tt an r s a 4 a ae eh - Ai - i faye Pies Wa i 7 pos —* ey = a al 2 . hud = = a " 2 —_ ? 7 ~ i a... leat = Wrasse Ww 002 eeieriigh n't te ; cy he sae ee ~ ae Aad peo aed a j, @elte4 ‘ a Bee er eee Yate ; ie: =s i, = : - @ is ae VE hive wing) UJ eas a Ph, ; A en 2 ky me J ; he 7 ‘1s Cw i — ~~ & at a i : arabs, Te. - my aa in a : . — > «a1.0e Ort. i in a oe “> a ; o . a ws vig uta a : 2 ‘e : 2 : “ i, ' ie at, baie, 4 a i. est ay rel ~2own ¥ : ge as ' - =e wader ery ot “ O O ZZ = r t= ot . , Lor. bi » | ’ ae. ! i“? ee aie 7 a - i —— = —— bs = 6 £ z- " * ; a ; Sonk rr Ser Per 5 ae iy <) a a uiniewtin acta’? i | é & one ; y a Fes bu Moe ret jim +g St afog 2 ae eee | ‘ ; NR ine The Ordovician—Silurian boundary on Manitoulin Island, Ontario, Canada C. R. Barnes and T. E. Bolton Geological Survey of Canada, 601 Booth St, Ottawa, Ontario, K1A OE8, Canada Synopsis The Ordovician-Silurian boundary in southern Ontario is reviewed. Sections on Manitoulin Island have been regarded by earlier workers as representing continuous sedimentation in a shallow carbonate plat- form environment on the north-east flank of the Michigan Basin. The best section across the boundary, exposed in the Kagawong West Quarry, is described and illustrated. Lithological studies have demon- strated a minor karst development near the systemic boundary. Conodont and macrofossil data demon- strate that the Kagawong Member, Georgian Bay Formation and the lower 15cm of the overlying Manitoulin Formation are of Richmondian age (Ordovician, Cincinnatian Series). The remainder of the Manitoulin Formation is of Rhuddanian age (Silurian, Llandovery (Anticostian) Series). A hiatus is shown to occur 15cm above the base of the Manitoulin Formation that represents the Gamachian Stage, Cincinnatian Series and possibly also the latest Richmondian Stage and earliest Rhuddanian Stage. Although the section on Manitoulin Island possesses many of the prerequisites of a boundary stratotype, the hiatus at the systemic boundary ruled it out of consideration as the formal stratotype. It 1s, however, one of many similar sections in the North American Midcontinent with a hiatus of this proportion at this level which is interpreted as reflecting the eustatic sea level drop in the latest Ordovician related to the north African continental glaciation. Regional setting In southern Ontario, undeformed, gently-dipping Ordovician and Silurian carbonates form the eastern margin of the Michigan Basin, affected slightly by the Algonquin Arch (Fig. 1). Over much of this area, the boundary between Ordovician and Silurian strata is a disconformity, but to the north, on Manitoulin Island (Fig. 1), several previous workers have considered it to be conformable with continuous sedimentation. More recent palaeontological and sedimentologi- cal work has revealed a paraconformable relationship. South of the Algonquin Arch (Fig. 1) exposures of the systemic boundary near the base of the Niagara Escarpment reveal a sharp disconformable contact between the Queenston and Whirlpool formations. The Queenston red shales have been generally regarded as continental deposits of the “Queenston Delta complex’ with their widespread distribution being attributed to lowered sea-level caused by the Late Ordovician glaciation (Dennison 1976). A few limestone interbeds low in the Queenston Formation have yielded a marine fauna, including conodonts, brachiopods, and bryozoans with at least the former indicating a littoral community (Barnes et al. 1978) and suggesting a Richmondian (Late Ordovician) age. The overlying Whirlpool For- mation is a white, cross-bedded sandstone barren of diagnostic fossils, but overlying shales within the Medina Group yield Llandovery fossils. The classic reference section for this area is that of the Niagara Falls gorge. North of the Algonquin Arch (Fig. 1), the red shales are replaced progressively by shallow water limestone with minor grey shale of the Kagawong Member (30m) of the Georgian Bay Formation (130m). On Manitoulin Island the red shales are absent and these Late Ordovician carbonates are overlain by carbonates of the Manitoulin Formation (20m), regarded as approximately equivalent to the sandstone of the Whirlpool Formation of the Niagara region. These regional stratigraphical relationships are illustrated in Fig. 1. Bull. Br. Mus. nat. Hist. (Geol) 43: 247-253 Issued 28 April 1988 248 C. R. BARNES & T. E. BOLTON MANITOULIN ALGONQUIN NIAGARA ISLAND ARCH GORGE CABOT HEAD SBINSB YY. z CABOT =a < HEAD = 2) CABOT = | @ | MANITOULIN ~ HEAD MANITOULIN ~ _—————————— WHIRLPOOL WHIREROOL Oe OD gO ee ee z > ZZ SS Ze ~ 5 < 3 QUEENSTON QUEENSTON a 6 « Cs ares bd 10 o [e) Ww Lo Joel m Collingwood LAKE ONTARIO Toronto Hamilton Niagara Gorge Kilometres Fig. 1 Map of southern Ontario showing Manitoulin Island, main tectonic elements, and generalized stratigraphical successions across the Ordovician—Silurian boundary for Manitoulin Island, Algonquin Arch, and Niagara gorge. Detailed stratigraphy On Manitoulin Island, the systemic boundary is best exposed and most accessible at the small, disused Kagawong West Quarry (Figs 2, 3) on Highway 540, 3km west of Kagawong (Alguire & Liberty 1968, Stop 2; Sanford & Mosher 1978, Stop 10; Telford et al. 1981, Stop 14; Kobluk & Brookfield 1982, Stop 6.3). The adjacent roadcut exposes additional strata of the Kagawong Member and the Manitoulin Formation. The following sequence is exposed: Manitoulin Formation: 6:5m Dolostone, massive to thick bedded at base, weathering into thin beds separated by irregular shale partings; medium to light brown with grey patches, weathering to a buff colour; medium crystalline; minor vugs in basal 15cm; abundant fossil debris, especially brachiopods and rugose corals; minor silicification. 0-15m Dolostone, thin bedded to laminated, argillaceous; medium brown, weathering to a very light brown colour; finely crystalline; beds separated by even shale partings; sharp upper and lower contacts; recessive unit. ORDOVICIAN-SILURIAN BOUNDARY ON MANITOULIN ISLAND 249 me Fig. 2 Kagawong West Quarry showing Kagawong Member, Georgian Bay Formation and Manitoulin Formation. Ordovician-Silurian boundary is drawn (black arrow) at top of 15cm recessive argillaceous dolostone unit. Georgian Bay Formation, Kagawong Member: 1-7m Dolomitic limestone, medium bedded weathering to thin beds; medium grey brown, weathering to blue grey; finely crystalline; poorly fossiliferous, bryozoans and stromatopo- roids. Liberty (1954: 13) and Bolton & Liberty (1954: 28) placed the systemic boundary at the top of the shaly recessive unit, including it within the Kagawong Member. Later Alguire & Liberty (1968: 8) included it in the Manitoulin Formation and considered this sequence to represent continuous sedimentation with no disconformity. Sanford & Mosher (1978: 13) and Sanford et al. (1978: 99) from lithological and geochemical evidence placed the systemic boundary 11cm above the top of the shaly recessive unit, the unconformity probably developing under sub- marine rather than subaerial conditions. Kobluk (1984) defined two paleokarst surfaces— erosional disconformities below the base and 10cm above the top of the recessive shaly unit. The lower paleokarst was regarded as at, or very close to, the systemic boundary. Johnson & Telford (1985), however, noted that the disconformable contact between the Manitoulin and Georgian Bay Formations is devoid of scour, rill or other features indicative of extended periods of erosion. Palaeontology Conodonts. Eight samples from this section, with particular emphasis on the Georgian Bay— Manitoulin formational contact, yielded nearly 1000 conodonts (Fig. 3). This fauna formed part of earlier studies by Tarrant (1977) and Barnes et al. (1978). The fauna of the Kagawong Member of the Georgian Bay Formation was listed by Barnes et al. (1978: fig. 3) and includes Aphelognathus grandis (Kohut & Sweet), A. pyramidalis (Branson, Mehl & Branson), Oulodus ulrichi (Stone & Furnish), Panderodus staufferi (Branson & Mehl), Pseudobelodina vulgaris Sweet, Rhipidognathus symmetricus Branson, Mehl & Branson. The last species dominates the fauna in the uppermost bed, indicating a littoral environment (e.g. Rhipiodognathus community 250 C. R. BARNES & T. E. BOLTON of Barnes & Fahraeus 1975). The progressive decrease in diversity upwards in the member also suggests upward shallowing. Most taxa are of late Maysvillian to Richmondian age. In the Composite Standard Section for the Middle and Upper Ordovician rocks of the Midcontinent Province, Sweet (1984, Appendix) reports A. pyramidalis and P. staufferi as restricted to the Richmondian interval. The Kagawong West fauna is herein assigned to the Richmondian Aphelognathus divergens Zone. Although several of the taxa range into Gamachian strata on Anticosti Island (McCracken & Barnes 1981: fig. 12), the presence on Manitoulin of Plectodina tenuis, A. grandis rather than A. sp. cf. A. grandis, P. staufferi rather than P. sp. cf. P. staufferi, and the absence of Gamachignathus spp., suggests a Richmondian rather than a Gamachian age. The fauna may be generally correlative with other Richmondian units such as the Bull Fork and Drakes formations, Cincinnati area (Sweet 1979a), the Noix Limestone, Edgewood Group of Missouri (McCracken & Barnes 1982) and the Vauréal Formation of Anticosti Island (Nowlan & Barnes 1981), but biofacies differences between these faunas make precise correla- tion difficult. The thin shaly recessive bed, at the base of the Manitoulin Formation, contains a similar fauna with Rhipidognathus (Fig. 3). Only P. gracilis and possibly O. sp. are known to range into Silurian strata elsewhere; no characteristic early Silurian taxa are present. The shaly recessive unit is therefore considered to be of Ordovician (Richmondian) age. The dolostones of the Manitoulin Formation yielded a conodont fauna (Fig. 3) that includes Icriodella discreta Pollock, Rexroad & Nicoll, Spathognathodus comptus Pollock, Rexroad & Nicoll s.f., and Ozarkodina hassi Pollock, Rexroad & Nicoll. The conodont fauna from the Lower Silurian of southern Ontario, including Manitoulin Island, and northern Michigan was described by Pollock et al. (1970), with other documentation by Barnes et al. (1978). The lower, but not lowest, part of the Manitoulin Formation thus includes forms indicative of the Icrio- dina irregularis Zone of Pollock et al. (1970), who also noted (p. 746) that in some sections ‘the oldest parts of the Manitoulin ... seems to correspond with the pre-/criodina irregularis Zone in the Midwest ... and with the lower part of Walliser’s (1964) Bereich I.’ I. discreta and O. hassi are known from earliest Silurian strata, Menierian Stage of Barnes (in press), in the Anticosti Island sections that are continuous across the Ordovician—Silurian boundary although S. comptus is absent (McCracken & Barnes 1981: fig. 12; Barnes, this volume). Herein, the Mani- toulin Formation is assigned to the Icriodella discreta—I. deflecta Zone of Aldridge (1972). In the Manitoulin section, there is therefore no evidence of the latest Ordovician conodont Fauna 13 characteristic of the Gamachian Stage as described by McCracken & Barnes (1981) from Anticosti Island. Other sections in the Midcontinent in North America also lack this interval, e.g. the Cincinnati area (Sweet 1979a; Sweet et al. 1971; Sweet 1984), the Noix Limestone and Bowling Green Dolomite of the Edgewood Group, Missouri (McCracken & Barnes 1982), and elsewhere in the western Midcontinent (Sweet 1979b), and the Hudson Bay region (LeFevre et al. 1975). McCracken & Barnes (1981) attributed this pattern to the latest Ordovician (Gamachian) regression, induced by the north African glaciation, which restricted areas of continuous sedimentation to subsiding marginal cratonic basins or non-eroding oceanic basins. Macrofossils. The general fauna of the Kagawong Member, Georgian Bay Formation, as detailed by Caley (1936), suggests the inclusion of these carbonates within the standard North American Richmondian Stage. Within the upper 5m, only Stromatocerium, Tetradium and poorly preserved undiagnostic bryozoans, bivalves and gastropods have been identified. According to Copper (1982: 680), ‘the post-Richmondian Ellis Bay Spirigerina—Hindella faunas of Anticosti Island are absent, suggesting an interval of erosion or non-deposition’. The fauna of the overlying Lower Silurian Manitoulin Formation is scattered throughout with concentrations confined to the uppermost beds (Bolton 1966, 1968). Characteristic forms include the corals Paleofavosites asper (d’Orbigny), Palaeophyllum williamsi Chadwick, cystoid Brockocystis tecumseth (Billings), brachiopods Resserella eugeniensis (Williams), Mendacella sp., ‘Orthorhynchula bidwellensis Bolton, Zygospiraella planoconvexa (Hall), Sypharatrypa (?) lati- corrugata (Foerste), Eospirigerina parksi (Williams), and Dolerorthis sp. An early Llandovery (Anticostian) pre-C, age, within the “Coelospira’ planoconvexa—Atrypa laticorrugata Zone of ORDOVICIAN-SILURIAN BOUNDARY ON MANITOULIN ISLAND 251 Ss (ep) = [= + Zz =, B aS re SECON JOB \ 6 4 @ = & e229 O L 2) BE Ss. © & oo > 5 Zz S35 2 8 oe SEO! ts ZZ < Zz ee @aass 3 © SS < =| a > Eo oot .2= a © — no 5 © 8S SESS aAOVWVSELNQ jog Oo Sees 3 ®@eosQvuT =) re) O @ ze af 8 Be®@S2A1 2 5 9 Hei e2 4 = = Q OSs = @ Os a SS o S86 8s 2 8 Ze = tis COVERED Sees es 2 |<] mG | INTERVAL Sess ue ee 5 Pez > qa 260 & ® bed tide ae ae > = "6 Za Elo |c o 5 gz (a2 a =? 4 2 ff y er) >) oe c =F Pied ’ : ye aes A = , — =" { eee a \ » i 4G rl Petr: st) Gi ea EB) Se Tye ay °c beats embatie> Me a ~ ly en ~ i —— ; or j . i cs > mir’ i % ( i Yu rut 7? Fea ‘ i ee ~ * wl ‘ re wy a J ¢ es oa nai \ : . % 4 “+ ms > - ar om 9 A ah P ' . ii ee ab = =) > r iter _ ¥ Ve _ =f 7’ ee f = p = us A abs dias te, ae u = mee a - ‘T4 q J — ra aay 7 2 ; ee. o; =f j : eve = ye a * 7 ee a Ga Ae ote? pl, a) wage stead oh call -apal Vibe «Cl ells 6 a aes < a DD, aay cer - mae 9 <2 2a Oe, Meds Ghat + haere. FF =a. PD Preliminary report on Ordovician—Silurian boundary rocks in the Interlake area, Manitoba, Canada H. R. McCabe Manitoba Mineral Resources Geological Services, 535-330 Graham Avenue, Winnipeg, Manitoba R3C 43, Canada Synopsis Both Ashgill and early Llandovery rocks are represented in both surface outcrop (Stonewall Quarry) and the subsurface of Manitoba, but there is no definite evidence of continuous sedimentation through the boundary period. The Interlake area of central Manitoba and its northwestward extension to eastern Saskatche- wan (Fig. 1) provides the only outcrop area for the Lower Palaeozoic strata of the Williston Basin, and the only Lower Palaeozoic outcrops between Hudson Bay and the western Cor- dillera. Unfortunately, outcrops are sparse and expose only limited stratigraphical intervals, so that it is not possible at present to propose a definitive locality for the Ordovician—Silurian boundary there. No single outcrop area is at present known which exposes completely the required stratigraphical interval. Nevertheless, because of the critical location of the Manitoba outcrop belt, the following will present a brief summary of data relevant to the delineation of the boundary. Stearn (1953, 1956), on the basis of detailed faunal studies, placed the Stonewall Formation in the Ordovician and placed the Ordovician—Silurian boundary at the contact between the Stonewall Formation and the overlying Fisher Branch dolomite of the Silurian Interlake Group. However, because of erosion of the uppermost beds, the type section of the Stonewall Formation at the Stonewall Quarry is incomplete. At the time of Stearn’s studies, firm correla- tion with the complete subsurface sequence had not been established. Subsequently, Porter & Fuller (1959) established a subsurface reference section for the Stonewall Formation, based on correlation of regional marker horizons (B. A. Morriseau, 3-20-90-6W; 875’-920’). Detailed correlations between the Morriseau well and the Stonewall Quarry (about 72km to the east) indicate that, at the Stonewall Quarry, the uppermost 4 to 6m of the Stonewall beds, including a prominent medial arenaceous-argillaceous marker bed (t-horizon) has been eroded. Brindle (1960), from subsurface faunal studies, suggested that the Ordovician—Silurian boundary falls within the Stonewall Formation, rather than at the top, and may be marked by the medial arenaceous bed. It must be noted that marker beds at the top, middle and bottom of the Stonewall Formation can be correlated through almost the entire Williston Basin, indicating little or no stratigraphical discontinuity at the Ordovician—Silurian boundary. Preliminary results of conodont studies (C. R. Barnes, personal communication) indicate an Ordovician—Richmondian (Ashgill) age for the Stonewall Quarry beds. Also, a possible late Lower Llandovery fauna was obtained from a core hole drilled near the outcrop belt north of Grand Rapids. Exact correlation of this core hole with the surface section is uncertain, but it appears that the sampled interval may be upper Stonewall, and the upper Stonewall beds may, at least in part, fill the apparent gap between the lower Stonewall beds of Ashgill age and the Middle Llandovery Fisher Branch beds. Recent stratigraphical core hole drilling in the Interlake outcrop belt, and mineral explora- tion drilling in the area north and west of Grand Rapids, have obtained a number of cores for the Fisher Branch—Stonewall-Stony Mountain succession, so that the complete lithological sequence through the Ordovician—Silurian boundary interval is now available. Also, recent geological mapping has outlined several new outcrops that may expose this interval. Although no systemic boundary outcrop can be defined with certainty, two newly accessible occurrences may possibly include the boundary zone, but precise faunal data for these outcrops are not yet Bull. Br. Mus. nat. Hist. (Geol) 43: 255-257 Issued 28 April 1988 256 H. R. MCCABE © Cormorant I Grand Rapid | ap Sa IB Sten i LAKE vr W/INN/PEGOS/S SILURIAN INTERLAKE GROUP B.A. Morriseau | 8-20-9-6 L AKE\, MAN/TOBA Stonewall L Fisher Br. 8-20-9-6 t-marker = SSS Stonewall Quarry STONEWALL FORMATION Williams ear 7 as ee ORDOVICIAN LEGEND Dolomite — Highly fossiliferous 10 — Argillaceous STONY MOUNTAIN FORMATION — Pebbly, calcarenitic Fig. 1 Correlation of the Stonewall Formation and adjacent rocks in the Interlake area, Manitoba, Canada (in part after Porter & Fuller 1959). Correlation with the subsurface is also shown. available. A thin sequence of dolomites, including an argillaceous marker bed believed to be the mid-Stonewall (t-horizon) marker, is exposed at the parking lot for the Manitoba Hydro powerhouse at Grand Rapids, but the remaining stratigraphical exposure is minimal. A large bedrock hill south of the village of Cormorant (approx. Sec. 14, Tp. 60, Rge. 22 WPM), on the south shore of Cormorant Lake, is traversed by a recent extension of Provincial Road 287. This hill is believed to comprise an outlier of the Stonewall Formation, although exposure is by no means complete (Fig. 1). Good exposures occur in a roadcut at the top of the hill, in a small quarry near the top, and in a number of scattered natural outcrops on the slopes of the hill. Total topographic relief (partially exposed stratigraphical section) is 33m, and the estimated Stonewall thickness is only about 10-6m. A preliminary examination shows, at the top of the hill, a 2-3m cap of massive to nodular bedded, buff mottled, variably fossiliferous dolomite with numerous corals and minor brachiopods and gastropods, but no recognizable ORDOVICIAN-SILURIAN BOUNDARY ROCKS IN MANITOBA AST Virgiana decussata (the diagnostic fossil of the Fisher Branch Formation). These beds have not yet been identified palaeontologically, but on the basis of lithology are believed to be Fisher Branch Formation (Middle Llandovery). These beds overlie sharply, and with apparent slight unconformity, a pebbly argillaceous marker bed (0:9m), which in turn is underlain by fine- grained dense conglomeratic dolomite (2:87m). This in turn overlies a 0:64m reddish grey dolomitic shale and argillaceous dolomite (possible t-marker?) which passes downward to microcrystalline dense dolomites. The conglomeratic beds are believed to be stratigraphically equivalent to similar dolomites described by Stearn for an outcrop on P.T.H. 10 near Rocky Lake, 26-7 miles (42:-6km) north of The Pas (Stearn 1956: 13). Stearn reported an Ordovician fauna from these strata, suggesting that, at this locality and at Cormorant, a portion of the Upper Stonewall may be missing because of non-deposition or pre-Fisher Branch (Middle Llandovery) erosion. Core-hole drilling and microfossil studies for the Cormorant section and for the Stonewall area, planned for 1986-87, may permit more precise determination of the Ordovician—Silurian boundary in Manitoba. It should be noted that the conglomeratic beds occurring in the Stonewall Formation in central Manitoba (e.g. the Cormorant area) are not known in southern Manitoba, where the Stonewall beds are slightly thicker and possibly comprise a more com- plete, but not completely exposed, Ordovician—Silurian boundary sequence. The summary faunal list for the Stonewall Formation is as follows: Upper Stonewall fauna (after Brindle 1960 for Saskatchewan subsurface): Above t-marker: streptelasmid, Favosites cf. favosus Goldfuss, Syringopora sp., bryozoan. Below t-marker: Halysites (Catenipora) gracilis Hall, ?Oepikina stonewallensis Stearn. Spathognathus manitoulinensis (Pollock, Rexroad & Nicoll)—C. R. Barnes (pers. comm. 1975). Lower Stonewall fauna (Stonewall Quarry section—after Stearn 1956): Kochoceras cf. pro- ductum, Antiplectoceras shammattawaense, Paleofavosites capax, P. okulitchi, Tryplasma gracilis, Angopora manitobensis, Beatricea regularis, Megamyonia nitens, ?Oepikina stone- wallensis, Ephippiorthoceras minutum, Metaspyroceras meridionale, Bickmorites insignis. (after C. R. Barnes 1975, pers. comm.): Belodina profunda (Branson & Mehl), Rhipidognathus symmetrica discreta Bergstrom & Sweet, Panderodus staufferi (Branson, Mehl & Branson). References Brindle, J. E. 1960. The faunas of the lower Palaeozoic carbonate rocks in the subsurface of Saskatche- wan. Res. Rep. Saskatchewan Dept. Min. 52: 1—45, pls 1-8. Porter, J. W. & Fuller, J. G. C. M. 1959. Lower Paleozoic rocks of the northern Williston Basin and adjacent areas. Bull. Am. Ass. Petrol. Geol., Tulsa, Ok., 43: 124-189. Stearn, C. W. 1953. Ordovician—Silurian boundary in Manitoba. Bull. geol. Soc. Am., New York, 64: 1477-1478. —— 1956. Stratigraphy and palaeontology of the Interlake Group and Stonewall Formation of southern Manitoba. Mem. geol. Surv. Brch Canada, Ottawa, 281: 1-162. r : a ' + 9 T “ t ‘ a Se a ec “0 ‘ To = oe “a j Vy ata es, : & a ‘ 4 =~ rf = cra f - Pra = ihe! oes mh ae _ i@ 7 ten? = -9 ¥ oo « tT iota” waaay a 2 a ot | = We ~ oe = ~ Ara | oe 7 ‘ q a. ae 2 ar , i. oie - Vea Be 3, feu - a Vv Lo Oe & Ph + E tise se She oe The Ordovician—Silurian boundary in the Rocky Mountains, Arctic Islands and Hudson Platform, Canada B. S. Norford Institute of Sedimentary and Petroleum Geology, Geological Survey of Canada, 3303-33rd St N.W., Calgary, Alberta T2L 2A7, Canada Synopsis The Ordovician-Silurian Boundary is developed within sequences of platform carbonates at Pedley Pass (southeastern British Columbia) and in the Kaskattama well (northeastern Manitoba). At Snowblind Creek (Arctic Islands), the boundary is documented within a transitional facies between platform carbon- ates and basinal rocks, but access to the locality is difficult and expensive. Further detailed palaeontol- ogical studies are needed to establish the precise position of the boundary at all three localities. The Rocky Mountains Silurian carbonates are widespread in parts of the Rocky Mountains (1400 km long, 50-140 km wide) and a graptolitic facies is locally present in the northwestern and west-central parts. Access is expensive except close to the very few roads. In the graptolitic facies (Road River Group), the Ordovician-Silurian boundary interval has not been studied in detail. Exposures are not good and unconformities are present within or below the Llandovery part of the sequence. The Ordovician ornatus Zone and the Silurian cyphus Zone are well documented (Cecile & Norford 1979; Jackson et al. 1965; Davies 1966) and taxa identified by Davies may indicate some of the intervening persculptus, acuminatus, atavus and acinaces Zones. The carbonate facies consists of resistant dolomites almost throughout the Rocky Moun- tains. North of Peace River an unconformity is present below Silurian dolomites of the Nonda Formation. South of the Peace, the Beaverfoot Formation appears to span latest Ordovician and most of Llandovery time. The section at Pedley Pass is typical of many in southeastern British Columbia, except that access is simple and inexpensive. The locality is within a carbonate platform, a considerable distance inboard of the platform-front. Exposure is excellent along a steep ridge above the timberline, and complete through more than 500m of Upper Ordovician and Lower Silurian limestones and dolomites. Retreat of glaciers was relatively recent and the rocks are essentially unweathered. The terrane is folded and thrust but structure is simple within the thrust plates, with moderate dip parallel to the ridge. Disconformities have not been recognized within the boundary interval, but discontinuities could be present within the sequences of shallow water carbonates. Conodont alteration indices (CAI) of 4 are known from just above the Beaverfoot Formation near Pedley Pass (Goodarzi & Norford 1985: 1091, sample D) and thus the rocks of the boundary interval have high thermal maturity and are quite unsuitable for palaeomagnetic and many geochemical studies. At Pedley Pass, 130 m of poorly fossiliferous dolomites separates an Upper Ordovician coral and brachiopod fauna (Bighornia—Thaerodonta Fauna of Ashgill age) from the lowest brachio- pods (Nondia sp.) and corals (Rhegmaphyllum sp., Streptelasma sp.) confidently dated as Silurian (Eostropheodonta Zone, part of Virgiana fauna, upper Lower to Middle Llandovery). Macro- fossils are present in the intervening rocks but are poorly preserved. Conodont studies of these beds have not been completed, but preliminary data (T. T. Uyeno in Norford 1969: 39) from a corresponding interval at Mount Sinclair, 25km north of Pedley Pass, indicate that the Ordovician—-Silurian Boundary lies somewhere within the upper 75m of the poorly fossiliferous interval at Pedley Pass. Bull. Br. Mus. nat. Hist. (Geol) 43: 259-263 Issued 28 April 1988 260 B. S. NORFORD Thus, the Beaverfoot Formation seems to show sedimentation across the Ordovician— Silurian Boundary but the problems are those of precisely locating the boundary and the high thermal maturity (CAI 4) of the rocks. The region is not suitable for a stratotype of more than local application. The Arctic Islands The Arctic Platform and the Inuitian Orogen comprise a vast region (2000 by 1000km) in which Ordovician and Silurian rocks are widely distributed, both in outcrop and subsurface. Exposures are mostly good, but logistic dependency on aircraft makes access expensive and then only possible during the short summer. A carbonate shelf is bounded to the northwest by a graptolitic facies, locally stratigraphical sections show the interfingering of the two facies in great detail, for example, along Snowblind Creek, Cornwallis Island (Thorsteinsson 1959). Broad open folds characterize the structure in most of the Arctic Platform; thermal maturities are low on Cornwallis Island (Conodont Alteration Indices 1 to 2, Uyeno 1981 and in Goodarzi & Norford 1985: 1091, sample B). Macrofossils are not common in the carbonate facies, but the graptolitic facies is very fossiliferous, locally with exquisite preservation of graptolites in full relief within limestone nodules. Palaeontological studies of both macrofossils and microfossils are only at a reconnaissance level at present, but the region has great promise for the achievement of detailed correlations of zonal schemes based on various phyla. Carbonates of the Allen Bay Formation, the Baillarge Formation and correlative rocks contain corals, cephalopods, brachiopods, gastropods, trilobites and receptaculitids. Ashgill faunas resemble those of northwestern Greenland and the Hudson Platform. Conodont faunas indicate Fauna 12 of the United States with the same fauna present in latest Caradoc rocks; Fauna 13 may also be present below conodont faunas indicative of the mid-continent Lower Silurian kentuckyensis Zone (Ryley 1984). Very early Silurian macrofaunas have not yet been collected from these formations, and, similarly, the conodont faunas are poorly known. Most probably, all of latest Ordovician and earliest Silurian time is represented within the Cape Phillips and Ibbett Bay Formations of the graptolitic facies. However, the graptolite faunas have not yet been described taxonomically and the presence of the pacificus, extraordi- narius, persculptus and acuminatus Zones have not been established. Cephalopods, radiolarians, sponge fragments, ostracodes, polychaetes and trilobites are associated with the latest Ordovi- cian graptolite faunas and allow correlation into the carbonate facies. Thus, the Late Ordovician and Early Silurian macrofaunas and microfaunas have yet to be described, but the intricate facies relations of carbonates and graptolitic rocks make it a region of international importance for the discrimination of the Ordovician—Silurian Boundary. The section at Snowblind Creek on Cornwallis Island is eminently suitable as a key section for intercontinental correlations except for its difficult access. The variety of fossil groups within the graptolite zones provides for detailed correlation of shelly benthic zones with the standard graptolite zonation. The Hudson Platform The Hudson Platform is a large remnant (1600 by 1000km) of a sequence of Palaeozoic carbonates and evaporitic rocks that once covered much of the Canadian Shield. The platform now floors Hudson Bay, but the rocks extend onshore in the Hudson Bay and James Bay Lowlands to the south and on Southampton, Coats and Mansel Islands to the north. Access to all of these areas is difficult and costly. Outcrop is very sparse in the Lowlands and limited to the major rivers and some intertidal regions; exposures are less rare in the northern islands but stratigraphical sections are few and incomplete. The rocks are essentially flat-lying with rare faults. Thermal maturities are low. A number of wells have been drilled in the Lowlands and offshore in the central regions of Hudson Bay; these provide the best stratigraphical sections and several (including Sogepet-Aquitaine Kaskattama Province No. 1) took continuous slim core through the Ordovician—Silurian Boundary. 261 ROCKY MOUNTAINS, ARCTIC ISLANDS AND HUDSON PLATFORM “eQo}URI] Ulo}SvaYIIOU ‘WIO}e[q UOSPNY ‘JOM T ‘ON SUIAOIg VUE} eYse y oUIe}INbY—-jadaBo¢g ‘> ‘salIOPIIO] ISAMYIION ‘PURIS] SI[[EMUIOD ‘ UOT}DEg Y22ID pul[qmoug “g “RIQUIN[OD YsHiig UJa]se9YyINOs Jo suIeJUNO AYDOY ‘uOTdag sseg Ao[pog ‘y ‘dey Aijeo07 Ig Is 7 *Oer O\WVLNO 2 Vi! 7 VAOLINYW | DIAIOVd bole es Pes =~ SaICLINN SL sama y \ é& Yon == d es ~~ fo) A Bi AL Ae & I ‘314 262 B. S. NORFORD Two sets of nomenclature have been used for an interval of shallow water dolomites between the Churchill River Group (Ashgill) and an unconformity at the base of the Severn River Formation (basal beds high Lower or Middle Llandovery). The Port Nelson Formation (Savage & Van Tuyl 1919; Norford 1971) is based on an outcrop on Nelson River, northern Manitoba; the Red Head Rapids Formation (Nelson 1963, 1964; Sanford 1974; Heywood & Sanford 1976) is based on two outcrops on Churchill River in the same region. The relations are uncertain between these three outcrops and an interval recognized in the subsurface between the Churchill River Group and the Severn River Formation. Nomenclatorial priority suggests the use of the term Port Nelson Formation. The interval reaches 35m in the sub- surface of northern Manitoba and is more than 60m thick on Southampton Island, where some additional younger strata may be present beneath the sub-Severn River unconformity. The Silurian Severn River Formation rests on 32m of the Port Nelson Formation in the Kaskattama well (Norford 1970). The contact is apparently conformable in the well but else- where there is evidence of an erosion surface beneath the Severn River Formation. The lower part of the Severn River Formation can be dated as late Early or Middle Llandovery, primarily on the presence of Virgiana decussata (Whiteaves). In the well, the Port Nelson Formation consists of dolomites and dolomitic limestones with mudstone partings and nodules, isolated crystals and thin beds of anhydrite and locally halite. In the Kaskattama well, corals and brachiopods in the basal 11m indicate a very late Ashgill age and correlation with the lower GRAPTOLITE |CONODONT|MACROFOSSIL} SOUTHERN ROCKIES ARCTIC ISLANDS HUDSON PLATFORM ZONES FAUNAS PEDLEY PASS SNOWBLIND CREEK KASKATTAMA WELL convolutus ? c SEVERN RIVER VIRGIA m = FORMATION (9 gregarius f | | kentuckyensis BEAVERFOOT PHILLIPS nathani FORMATION PORT NELSON Pacificus FORMATION BIGHORNIA- if Py THAERODONTA CHURCHILL RIVER m GROUP Cc Fig. 2 Correlation diagram of the three areas. <#—— ORDOVICIAN-SILURIAN BOUNDARY ——3 ROCKY MOUNTAINS, ARCTIC ISLANDS AND HUDSON PLATFORM 263 and middle part of the Stonewall Formation of southern Manitoba. The upper 21m contains only fragmentary fossils in the well, but sparse conodonts (Conodont Alteration Index | to 1-5) have been recovered from outcrop 5:5m below the top of the Port Nelson Formation in its type section. T. T. Uyeno has identified Panderodus cf. P. simplex Branson & Mehl s.f. and tentatively dates the horizon as early Llandovery, but comments that the form shows some transitional features to those of the Middle and Upper Ordovician form-species Panderodus compressus Branson & Mehl. Thus, in the Hudson Platform the Ordovician—Silurian Boundary lies either within the Port Nelson Formation or within a regional unconformity below the Severn River Formation. The sediments that formed the Port Nelson Formation were inhospitable to animal life, and although one can hope for more refined dating of the upper beds and thus more precise positioning of the Boundary, the region is not suitable for a stratotype of more than local application. References Barnes, C. R., Norford, B. S. & Skevington, D. 1981. The Ordovician System in Canada, correlation chart and explanatory text. Int. Un. geol. Sci., Paris, 8: 1-27. Cecile, M. P. & Norford, B. S. 1979. Basin to platform transition, Lower Paleozoic strata of Ware and Trutch map areas, northeastern British Columbia. Geol. Surv. Pap. Can., Ottawa, 79-1A: 219-226. Davies, E. J. L. (1966). Ordovician and Silurian of the northern Rocky Mountains between Peace and Muskwa Rivers, British Columbia. Univ. Alberta, unpubl. Ph.D. dissertation. Goodarzi, F. & Norford, B. S. 1985. Graptolites as indicators of the temperature histories of rocks. J. geol. Soc. Lond. 142: 1089-1099. Heywood, W. W. & Sanford, B. V. 1976. Geology of Southampton, Coats and Mansel Islands, District of Keewatin, Northwest Territories. Mem. geol. Surv. Can., Ottawa, 382: 1-35. Jackson, D. E., Steen, G. & Sykes, D. 1965. Stratigraphy and graptolite zonations of the Kechika and Sandpile Groups in northeastern British Columbia. Bull. Can. Petrol. Geol., Calgary, 13: 139-154. Nelson, S. J. 1963. Ordovician paleontology of the northern Hudson Bay Lowland. Mem. geol. Soc. Am., New York, 90: 1-152, pls 1-37. — 1964. Ordovician stratigraphy of northern Hudson Bay, Lowland, Manitoba. Bull. geol. Surv. Can., Ottawa, 108: 1-36. Norford, B. S. 1969. Ordovician and Silurian stratigraphy of the southern Rocky Mountains. Bull. geol. Sur. Can., Ottawa, 176: 1—90. —— 1970. Ordovician and Silurian biostratigraphy of the Sogepet—Aquitaine Kaskattama Province No. | well, northern Manitoba. Geol. Surv. Pap. Can., Ottawa, 69-8: 1—36. —— 1972. Silurian stratigraphy of northern Manitoba. Spec. Pap. geol. Ass. Can., Toronto, 9: 199-207. Ryley, C. C. (1984). Late Ordovician and Early Silurian conodont taxonomy and biostratigraphy, lower Allen Bay Formation, Cornwallis Island, NWT. Univ. Western Ontario, unpubl. B.Sc. dissertation. Sanford, B. V. 1974. Paleozoic geology of the Hudson Bay region. Geol. Surv. Pap. Can., Ottawa, 74-1B: 144-146. Thorsteinsson, R. 1959. Cornwallis and Little Cornwallis Islands, District of Franklin, Northwest Terri- tories. Mem. geol. Surv. Can., Ottawa, 294: 1-134. — 1963. Ordovician and Silurian stratigraphy. Mem. geol. Surv. Can., Ottawa, 320: 31—SO. —— & Tozer, E. T. 1970. Geology of the Arctic Archipelago. Econ. Geol. Rept. Geol. Surv. Can. 1 (5th edn): 547-590. Uyeno, T. T. 1981. Systematic study of conodonts. In Stratigraphy and conodonts of Upper Silurian and Lower Devonian rocks in the environs of the Boothia Uplift, Canadian Arctic Archipelago. Bull. geol. Surv. Can., Ottawa, 292: 1—75, pls 1-10. oe 1 2 a = = ies oe bie en ry oh . - ; Par’ me St bh a ea ” aie d a 2 ? hag” thie iy ji . iW ’ ] r a 5 oa um . J iy dncathade ae tte steely, jaye * ’ i as ne : : , ene eo eat ve ai Peis y i. » smen alps teal) y . _— eres bal a a " i." ce i aid a) ol f a i Bal exc , o>? 4 = rn (ae Mes . ' me | 4 em 4 ai <— 7 —— - @ 1. . ate 2 ‘>wts wy ah 42 =< m4 4 1 2 : " gry’ ef | a) Ge aa : rh a : « = 4 . 4 oo . FA EE asl) -4 | ai alite nt i a ee ec Ore iP ly) lea La PS a vi, oe aa 1 ; ; 7 a at ay en 7 a : wal a 1 7 : a : y ‘ se 7 ak a eee wae af ah Lies, r i a 1? ; | J " ‘ a a ee a mc et ade Ve a ms - - ’ ; ; i: 2 ° is = : i = . a > , r bd [ oie - - Ordovician—Silurian boundary, northern Yukon, Canada A. C. Lenz’ and A. D. McCracken? ‘Department of Geology, University of Western Ontario, London, Ontario, N6A 5B7, Canada ?Department of Geology, Laurentian University, Sudbury, Ontario, P3E 2C6, Canada. Synopsis The Ordovician-Silurian boundary is described from three graptolite and conodont-bearing sections of northern Yukon. Upper Ordovician graptolite biostratigraphical units comprise the Dicellograptus ornatus, Pacificograptus pacificus and tentatively the Glyptograptus persculptus zones; that of the cono- donts being the Amorphognathus ordovicicus Biozone and the North American Fauna 12. A strati- graphical hiatus between the P. pacificus Zone and the G. persculptus Zone?, and probably equivalent to the Diplograptus bohemicus and Climacograptus? extraordinarius graptolite Zones, and to North American conodont Fauna 13, appears to be present everywhere in the region. Introduction The presence of excellently exposed graptolite-bearing sequences in the Richardson and Ogilvie mountains of northern Yukon has been recognized for more than 20 years. Graptolitic strata of the Road River Formation are known to be widely distributed throughout the northern Cor- dillera of Canada and adjacent Alaska (e.g. Lenz & Perry 1972; Lenz 1972, 1982; Churkin & Brabb 1965). For the purpose of this paper, three key sections are discussed; these are Peel River, Pat Lake and Blackstone River, the first in the Richardson Mountains, the latter two in the Ogilvie Mountains (Figs 1, 2). The Peel River section is chosen because the Ordovician—Silurian boundary beds are completely exposed and are well! studied, and are defined on both grapto- lites and conodonts; Pat Lake contains a thick conodont and shelly fauna-bearing limestone of probable latest Ordovician age in an otherwise entirely graptolitic sequence; and Blackstone River is a much thicker boundary sequence containing both graptolites and conodonts. These sequences have already been discussed in Lenz & McCracken (1982). The three sections discussed are in remote and isolated areas of northern Yukon, the field season is relatively short, seldom more than two and a half months, and the cost of access is high. Access to any of the three localites is via regular scheduled aircraft service to Whitehorse in southern Yukon, and then to the villages of either Mayo or Dawson City in central Yukon (Fig. 1), and by privately chartered helicopter thereafter. Weather in the region can vary considerably, but is generally pleasant in July and early August. Stratigraphy The Road River Formation, the type of which is in the Richardson Mountains (Jackson & Lenz 1962), is a thick basinal sequence of dominantly dark grey to black shales and cherts with minor dark limestone beds and a few relatively thick-bedded debris-flow carbonates. Grapto- lites are common to abundant in the shales and conodonts occur in some of the thin, dark limestone beds. The Peel River section is a more or less typical Richardson Mountains bound- ary sequence, but is without significant carbonates. The Road River strata of the Ogilvie Mountains, of which the Pat Lake and Blackstone River sections are representative, are characterized mainly by thinly bedded dark shales and calcareous shales, much greater amounts of dark limestone beds and laminae, and much less chert. The shales and calcareous shales contain abundant graptolites, while the dark limestones Bull. Br. Mus. nat. Hist. (Geol) 43: 265-271 Issued 28 April 1988 266 A. C. LENZ & A. D. MCCRACKEN 4 poo Pa) © Je > Py 1S) of) O Zz 150 KMS. _——— DAWSON 100 MILES ——— Fig. 1 Index map of northern Yukon showing localities. 1 = Blackstone River; 2 = Pat Lake; 3 = Peel River. ORDOVICIAN-SILURIAN BOUNDARY IN N. YUKON 267 4 BLACKSTONE RIVER 3 62°56'N, 137°20'W PEEL RIVER 65°53’'N, 135°42'°W 2 : PAT LAKE 4 65°O9'N, 136°42’W ing ee A: cas ; = eee ., Eee L. acinaces Zone $ acuminatus Fane ee ' E f =~. G. persculptus a Zone a }t— 4 S Kb] & P|] << = - 22 Ne 4 9 | a o b= / “ I] Z eens | A Cc) = 7 < Sa Ui coup oe =— Ta x = pee fefa) S SESS) 5 7% er a Us Sa =. oO, of | i =a > Se] 4 i Jo oS O == == Wane) S|) (s=2/ cc eee S a= Dicellograptus es ra aa | CHERT O| EES. = [o%e | sa |—~+ z ——_]| SHALE @ | GRAPTOLITES a] =a — Je 1m (20 ft.) =e L~A™-N'7] ES) CALCAREOUS SHALE 4 iat LIMESTONE ia) CONODONTS oO CORALS Fig. 2 Correlation of graptolite zones of localites 1—3 (Fig. 1), using the base of the Silurian as a datum. may contain conodonts and, rarely, trilobites (e.g. Lenz & Churkin 1966; Ludvigsen 1981). A relatively thick sequence of light-coloured, probably shallow water, conodont and coral-bearing limestone of probable latest Ordovician age (Glyptograptus persculptus Zone?) occurs in the Pat Lake section (Fig. 1). The presence of the limestone is anomalous, and its origin may be related to the widely recognized latest Ordovician glacially induced regression (e.g. Lenz 1976, 1982; Lenz & McCracken 1982). Graptolites Ashgill graptolite faunas of the northern Cordillera are divisible into two biostratigraphical units, a lower Dicellograptus ornatus Zone and the upper Paraorthograptus pacificus Zone. The uppermost Ordovician, the G. persculptus Zone, is less well developed and is clearly absent from the Peel River section, but is tentatively recognized in the Pat Lake and Blackstone River sections. Lowest Silurian (Llandovery) strata, represented by the Parakidograptus acuminatus Zone and the overlying Atavograptus atavus and Lagarograptus acinaces Zones are widely recognized (Lenz 1982). The D. ornatus Zone is characterized by the index species, and by D. minor, Glyptograptus latus, Climacograptus longispinus, C. latus, C. hvalross, C. hastatus, C. supernus, Orthograptus abbreviatus, O. cf. fastigatus, Orthoretiograptus denticulatus, Arachniograptus laqueus and Lepto- graptus spp. Dicellograptus is common, as are most of the diplograptid species. The P. pacificus Zone is taxonomically a much more impoverished fauna and is characterized by an abundance of C. supernus and P. pacificus. Most of the species of diplograptids noted in the D. ornatus Zone are present, but in much lesser numbers, and dicellograptids are rare. In 268 A. C. LENZ & A. D. MCCRACKEN addition, the exotic Diceratograptus cf. mirus is represented by two specimens in the Peel River section (Chen & Lenz 1984). The supposed G. persculptus Zone, which was considered to be lowest Silurian in Lenz & McCracken (1982), is characterized by a fauna of low diversity, and is only tentatively recog- nized. The index species has not, to date, been recovered from the northern Canadian Cor- dillera, although it does occur in southeastern Alaska (Churkin et al. 1971). This bio- stratigraphical unit is distinguished by the relatively sudden appearance of narrow forms of Climacograptus normalis and C. miserablilis, a very spinose form of ?Paraorthograptus and Orthograptus cf. abbreviatus. Other species appearing in the interval, but not confined to it, include Diplograptus modestus, Glyptograptus tamariscus, G. gnomus, G. cf. laciniosus, and G. cf. lanpheri. Monograptids have not been recovered. The G. persculptus Zone? is absent in the Peel River section. The P. acuminatus Zone, the lowest Silurian biostratigraphical unit, is readily recognized by the appearance of the index species, as well as Climacograptus cf. trifilis, ? Akidograptus ascen- sus, Cystograptus vesiculosus and Diplograptus modestus diminutus. Monograptids have not been found. The A. atavus and L. acinaces Zones are discussed together since they witness the incoming of monograptids, particularly Atavograptus and Pribylograptus, as well as being characterized by Dimorphograptus confertus swanstoni, D. physophora (and subspecies) and common Cystograptus vesiculosus. Graptolite correlation The graptolitic sequences of the northern Cordillera are directly comparable to those in central China and the Kolyma and Kazakhstan regions of the U.S.S.R., and indirectly with that of southern Scotland (Lenz & McCracken 1982; Chen & Lenz 1984). The D. ornatus Zone is directly comparable to the C. longispinus Subzone of Koren et al. (1979), more or less compar- able with the D. szechuanensis Zone and possibly the Amplexograptus yangtzeensis Zone of central China (Chen & Lenz 1984), and probably with the D. complanatus Zone of Scotland (Williams 1982). Correlation of the P. pacificus Zone of Yukon is almost certainly directly with the P. pacificus Subzone of U.S.S.R., but comparison with the Chinese succession is more difficult. Faunally, the P. pacificus Zone is most similar to the D. szechuanensis and A. typicus Zones; however, the presence of rare Diceratograptus in the Peel River section suggests correlation with strata as high as the Paraorthograptus uniformis Zone of China. The latter correlation would appear to be even more reasonable if P. uniformis of China is, as suggested by Williams (1982), synony- mous with P. pacificus. Correlation of the G. persculptus and P. acuminatus Zones 1s relatively straightforward, and it therefore appears that strata equivalent to the Diplograptus bohemicus Zone of China, and the Climacograptus? extraordinarius Zone of U.S.S.R. and Scotland are unrepresented by grapto- lites or missing from the Yukon sections. Conodonts and conodont correlation Ashgill conodonts from the Blackstone and Peel River sections are regarded as being within the Amorphognathus ordovicicus Biozone and the North American Fauna 12. The conodont fauna at Blackstone River (Figs 1, 2) occurs 3m below the supposed G. persculptus Zone and 13:7m above the last occurrence of graptolites of the P. pacificus Zone. Significant taxa include A. ordovicicus Branson & Mehl, Belodina confiluens Sweet, Besselodus n. sp., Gamachignathus ensifer McCracken et al., Icriodella superba Rhodes?, Noixodontus girardeauensis (Satterfield), Oulodus ulrichi (Stone & Furnish), Panderodus? sibber Nowlan & Barnes, Plectodina florida Sweet, P. tenuis (Branson & Mehl), Protopanderodus sp., Scabbardella altipes (Henningsmoen) and Walliserodus amplissimus (Serpagli). Not all of these species were initially listed by Lenz & McCracken (1982) and some have since undergone taxonomic revision. ORDOVICIAN-SILURIAN BOUNDARY IN N. YUKON 269 Fig. 3 Ordovician-Silurian boundary section on Peel River. Arrow on upper photograph is Ordovician— Silurian boundary. Lower photograph is a close-up of the boundary beds; the lower arrow is the top of the P. pacificus Zone and the upper arrow is the base of the P. acuminatus Zone. 270 A. C. LENZ & A. D. MCCRACKEN One of the most noteworthy species, N. girardeauensis, was also found by McCracken & Barnes (1982) in Missouri in association with Aphelognathus grandis (Branson, Mehl & Branson) and A. ordovicicus. The recent work of Sweet (1984) established the A. grandis Chronozone; the nominal species not only occurs in the Missouri fauna, but also in the Richmondian Vauréal Formation of Anticosti Island (Nowlan & Barnes 1981). This species was not recognized in the Gamachian Fauna 13 by McCracken & Barnes (1981), but they recorded the related species A. aff. A. grandis. The range of A. grandis is reported to be from the upper Maysvillian through much of the Richmondian A. divergens Chronozone; it does not appear to range into post-Richmondian, pre-Silurian strata (Sweet 1984). The close stratigraphical proximity of the Blackstone conodont fauna to the G. persculptus Zone? graptolites does not necessarily imply that it is latest Ordovician. The rare co-occurrence on Blackstone River of G. ensifer with A. ordovicicus, B. confluens (= B. compressa of Lenz & McCracken 1982), O. ulrichi, P.? gibber, P. florida and P. tenuis is comparable to the upper Vauréal Formation fauna (late Richmondian) of Nowlan & Barnes (1981). Unless the upper limit of A. grandis is younger than is at present known, the co-occurrence of N. girardeauensis and A. grandis in Missouri may indicate the Richmondian or, possibly, the late Maysvillian (based strictly on published microfossil data). Hence, the occurrence of N. girardeauensis at Blackstone River may favour a late, rather than the latest, Ordovician age. The Lower Llando- very shale and chert from both the Blackstone and Peel River sections have not been collected for conodonts. A single f element of G. ensifer co-occurs at the Peel River section (Figs 1, 2) with some of the species listed above for the Blackstone River; I. superba?, N. girardeauensis, O. ulrichi and P. florida are absent from this fauna, whereas O. rohneri Ethington & Furnish and Pseudobelodina vulgaris vulgaris Sweet are present only at the Peel River section. Unlike the Blackstone section, where the Ordovician conodont fauna is within a thick, 16-7m interval barren of graptolites, the Peel River conodont-bearing stratum is within a thin, 2.5m interval bounded by shales containing graptolites of the P. pacificus Zone, and hence this conodont fauna is regarded as late, but not latest, Ordovician. The fauna occurs in strata 1:6-1:9m below the systemic boundary. Lenz & McCracken (1982) did not report Ashgill conodonts from the Pat Lake section Figs 1, 2). The sparse faunas there comprise poorly preserved conodonts that were originally assigned an early Silurian age on the basis of ramiform elements and on their stratigraphical SOUTHERN SCOTLAND P. acuminatus CORDILLERAN CANADA KOLYMA and KAZAKHSTAN, USSR CENTRAL CHINA P. acuminatus P. acuminatus P. acuminatus =| G. persculptus ? G. persculptus G. persculptus persculptus Climacograptus D. bohemicus C.?extraordinarius Z extraordinariuS fF D < | P. uniformis 9 O D D. mirus > P. pacificus Z P. pacificus | se ee O D> T. typicus P. pacificus S Qe D. anceps na oa O ez D. szechuanensis ? Fi = Climacogr D. complanatus D. ornatus O longispinus A. disyunctus > 2 yangtzeensis I J} aaa Fig. 4 Correlation of Ordovician—Silurian strata of Yukon with those of central China, U.S.S.R. and Scotland. ORDOVICIAN-SILURIAN BOUNDARY IN N. YUKON 271 position with respect to G. persculptus Zone? graptolites. Ozarkodina sp. A Lenz & McCracken has a definite Silurian aspect, although this does not demand an assignment to that system since the genus has elsewhere been occasionally recognized from Upper Ordovician strata. The poor preservation of the coniform elements limits their biostratigraphical value; they could be assigned to either Ordovician or Silurian taxa. Thus an age determination for these post-P. pacificus Zone and pre-persculptus Zone? conodonts may depend upon the positive identification of Ozarkodina sp. A. An unequivocal age based solely on the conodont taxa cannot be determined for the Pat Lake conodont faunas. Their occurrence below the G. persculptus Zone? suggests an Ordovician age. Llandovery and younger conodont faunas are much more diverse and better preserved than those discussed herein; study of these faunas is in progress. Acknowledgements Assistance provided in the field by the Geological Survey of Canada, and particularly by A. E. H. Pedder and D. G. F. Long, is acknowledged. Financial support for the project was provided by a National Sciences and Engineering Research Council operating grant to Lenz, and in part to McCracken by the Northern Research Group, the University of Western Ontario. References Chen Xu & Lenz, A. C. 1984. Correlation of Ashgill graptolite faunas of central China and Arctic Canada, with a description of Diceratograptus cf. mirus Mu from Canada. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeontology of Systemic Boundaries in China. Ordovician—Silurian Boundary 1: 247-258, | fig. Churkin, M., jr & Brabb, E. E. 1965. Ordovician, Silurian and Devonian biostratigraphy of east-central Alaska. Bull. Am. Ass. Petrol. Geol., Tulsa, Ok., 49: 172-185. ——,, Carter, C. & Eberlein, D. E. 1971. Graptolite succession across the Ordovician-Silurian boundary in southeastern Alaska. Q. JI geol. Soc. Lond. 126: 319-330, 1 pl. Jackson, D. E. & Lenz, A. C. 1962. Zonation of Ordovician and Silurian graptolites of northern Yukon. Bull. Am. Ass. Petrol. Geol., Tulsa, Ok., 46: 30-45. Koren, T. N., Soboleyskaya, R. F., Mikhailova, N. F. & Tsai, D. T. 1979. New evidence on graptolite succession across the Ordovician—Silurian boundary in the Asian part of the USSR. Acta palaeont. pol., Warsaw, 24: 125-136. Lenz, A. C. 1972. Ordovician to Devonian history of northern Yukon and adjacent District of Mackenzie. Bull. Can. Petrol. Geol., Calgary, 20: 321-361. 1976. Late Ordovician—Early Silurian glaciation and the Ordovician—Silurian boundary in the northern Canadian Cordillera. Geology, Boulder, Colo., 3: 313-317. —— 1982. Llandoverian graptolites of the northern Canadian Cordillera: Petalograptus, Cephalograptus, Rhaphidograptus, Dimorphograptus, Retiolitidae, and Monograptidae. Contr. Life Sci. R. Ont. Mus., Toronto, 130: 1-154. —— & McCracken, A. D. 1982. The Ordovician-Silurian boundary, northern Canadian Cordillera: graptolite and conodont correlation. Can. J. Earth Sci., Ottawa, 19: 1308-1322, 2 pls. —— & Perry, D. G. 1972. The Neruokpuk Formation of the Barn Mountains and Driftwood Hills, northern Yukon; its age and graptolite fauna. Can. J. Earth Sci., Ottawa, 9: 1129-1138. Ludyigsen, R. 1981. Biostratigraphic significance of Middle Ordovician trilobites from the Road River Formation, northern Cordillera. Prog. Abstr. geol. Assoc. Can., 6: A36. McCracken, A. D. & Barnes, C. R. 1981. Conodont biostratigraphy and paleoecology of the Ellis Bay Formation, Anticosti Island, Québec, with special reference to Late Ordovician—Early Silurian chronostratigraphy and the systemic boundary. Bull. Geol. Surv. Can., Ottawa, 329 (2): 51-134, 7 pls. —— —— 1982. Restudy of conodonts (Late Ordovician—Early Silurian) from the Edgewood Group, Clarksville, Missouri. Can. J. Earth Sci., Ottawa, 19: 1474-1485, 2 pls. Nowlan, G. S. & Barnes, C. R. 1981. Late Ordovician conodonts from the Vauréal Formation, Anticosti Island, Québec. Bull. geol. Surv. Can., Ottawa, 329 (1): 1-49, 8 pls. Sweet, W. C. 1984. Graphic correlation of upper Middle and Upper Ordovician rocks, North American Midcontinent Province, U.S.A. In D. L. Bruton (ed.), Aspects of the Ordovician System: 23-35. Uni- versitetsforlaget, Oslo. Williams, S. H. 1982. The Late Ordovician graptolite fauna of the Anceps Bands at Dob’s Linn, southern Scotland. Geologica Palaeont., Marburg, 16: 29-56, 4 pls. » —~ x ay e a -_ cones eG) & ii 4 7) low - ae 4 ‘wera ee is ‘ Sit 2 ) a 7 et ee s@ i Ag — PTaA iy ¢ > Gar oa & >) a Ss oa wiguel \ m ey “* ia ee se net) — be wy : he 7p S) a a a ae = e = : nod ag wr Stacie ee a “| = neat 1 23 ei a ~ ’ -) se why: ~( gar tr Ramat be artes -_ = 7 ) fw Ac SP al i “< = —™ a cece ii yo." ae id + 7. ae = 7?) a ts oor ie. <_< ry » y te 1 ¢ . } 3 = Woe : @ ‘Ly i : Nj oy uw =— i j of i> HN is | 9? pt Vi i > “al a : * ‘6 a Fae x - Ly) ' = 5 ‘ a; é ‘ i? y ah aS os ape> inde as Gey = kh “at is r a to te “ i : da CS, Ba The oc Re ouwgy rea pate -h = 9f 7 bee Va. ee an om j 4 eee y rRipemate cutilb: tise) op a & 6 . : i, a ihe ‘ rae Wh. a ae! ul -~ ‘ 7 ie a ° i = ~ ; : a eT ip rs Ee! ota a ater pe ibe th ile ar aT a = ; >? el 9 a Pe S 4 p> Ol an rage ae ee ’ y Coo - 2 is ‘ vst appt aginst | Tea © é 1s, ae ie pas rt 3 ge | Ja & - all - St sheet LS ee Cas. eee The Ordovician—Silurian boundary in the United States S. M. Bergstrom’ and A. J. Boucot? ' Department of Geology and Mineralogy, The Ohio State University, 125 S. Oval Mall, Columbus, OH 43210, U.S.A. ? Department of Zoology, Oregon State University, Corvallis, OR 97331, U.S.A. Synopsis Ordovician and Silurian rocks are widespread in the United States and there are numerous outcrops in many regions displaying the systemic boundary interval. However, a regional review of key sections in all the major outcrop areas shows that biostratigraphically closely controlled and stratigraphically complete or nearly complete boundary successions are quite rare. Indeed, the Esquibel Island section in south- eastern Alaska, where the systemic boundary is in a continuous graptolitiferous sequence, is not only the only known occurrence in the United States of a typical P. acuminatus Zone fauna, but also the only known place in the country where the systemic boundary can be established precisely on graptolites in a continuous succession. Elsewhere, relatively complete, if not complete, boundary successions are present in the Appalachians and in the Great Basin, as well as Alaska, but in virtually all cases the bio- stratigraphical control is not good enough to establish the boundary level with certainty. Most of the sections in these regions display a gap in the boundary interval, and this is the case also in most of the many boundary sections in the Midcontinent region. The best known, and stratigraphically most nearly complete, cratonic sections are in Arkansas, Oklahoma, Missouri, and Illinois, where strata having a taxonomically varied Hirnantia fauna are overlain, with locally only a minor, if any, stratigraphical gap, by rocks containing Llandovery fossils. No graptoloid graptolites are known from these sections, and the precise level of the systemic boundary is uncertain in some sections. It is concluded that further studies are urgently needed on fossils and rocks in the boundary interval, particularly to establish the precise age of the conodont faunal turnover as well as to clarify the mutual relations between the distribution patterns in time and space displayed by different groups, and their relations to the graptolite-based systemic boundary. Introduction Ordovician rocks are present in the subsurface over much of the United States and they are exposed in several major regions (Cook & Bally 1975). Although less widespread than those of Ordovician age, Silurian rocks are likewise distributed over major parts of the country and exposed over considerable areas. Accordingly, it is not surprising that the interval of the Ordovician-Silurian systemic boundary is available for study at a large number of localities from the Appalachians in the east to the Great Basin in the west. In many of these sections, the faunal succession is incompletely known or fossils are absent in critical intervals, which applies to the cratonic areas in the continental interior as well as to the geosynclinal areas along the continental margins. Nevertheless, because in most sections, particularly the cratonic ones, the systemic boundary is associated with a stratigraphical gap and a change in lithology, its level in those sections can be readily recognized. As is the case elsewhere in the world, nearly complete successions in continuously fossiliferous facies across the boundary interval are quite rare in the United States both in shelly and graptolitic facies. For instance, we are not aware of a single section outside Alaska where the precise level of the base of the P. acuminatus Zone, that is the internationally accepted base of the Silurian, can be recognized by means of graptol- ites or other fossils. It is quite clear that the choice of this level for the systemic boundary at the present time makes its recognition difficult, if not impossible, in stratigraphically more or less complete successions like those in the Great Basin (Ross et al. 1979; Leatham 1985; etc.) and in the Mississippi Valley region (Amsden 1986). Bull. Br. Mus. nat. Hist. (Geol) 43: 273-284 Issued 28 April 1988 274 S. M. BERGSTROM & A. J. BOUCOT Fig. 1 Index map showing areas with Ordovician and/or Silurian outcrops (black) and systemic boundary sections. 1, northern Maine; 2, eastern New York and western Vermont; 3, central Appalachians (Pennsylvania and adjacent states); 4, eastern Tennessee; 5, Alabama and Georgia (southern Appalachians); 6, the Cincinnati region and adjacent areas in Ohio, Kentucky, and Indiana; 7, the Nashville dome in central Tennessee; 8, northern Arkansas (including the Batesville district); 9, eastern Missouri and southwestern Illinois; 10, southern Oklahoma (including the Arbuckle Mountains); 11, Black Hills (South Dakota and Wyoming); 12, North Dakota; 13, Colorado; 14, west Texas; 15, New Mexico; 16, Bighorn Mountains, Wyoming; 17, Montana; 18, Idaho; 19, Nevada; 20, Utah; 21, southeastern California; 22, southeastern Alaska (inset map). The purpose of the present paper is to review briefly the biostratigraphy of the systemic boundary interval in key sections in the principal outcrop areas. Page limitations make it necessary to restrict ourselves to data essential for the understanding of the local and regional geology of this interval in the United States. For convenience, we will deal with each of the major outcrop regions separately, from the Appalachians in the east to the Great Basin in the west. For the location of these regions, see Fig. 1. Northern Appalachians In large parts of the Northern Appalachians in the United States (Maine to New York State), Silurian or younger rocks rest with a conspicuous, in many cases angular, unconformity on the Ordovician (Berry & Boucot 1970: fig. 6). This stratigraphical gap varies in magnitude both locally and regionally but includes in most cases portions of both the Ordovician and Silurian systems. Conventionally, this gap is explained as a product of the Middle to Late Ordovician Taconic orogeny, but it is evident that the apparently global drop in sea level during the latest Ordovician (Hirnantian) contributed to emergent conditions, at least locally. In this region, biostratigraphical control through the systemic boundary interval is, in general, poor. This is partly due to the fact that the rocks were largely deposited in environments with small numbers of shelly organisms, and those that became fossilized were in many cases strongly affected by the subsequent metamorphism of the host rocks. That diagnos- ORDOVICIAN-SILURIAN BOUNDARY IN UNITED STATES 275 tic fossils are present locally is shown by Neuman’s (1968) finds of shelly fossils of the Hirnantia fauna in east-central Maine, the only occurrences of this type of fauna from the northeastern United States. Another, and in terms of geology of the systemic boundary even more inter- esting, sequence is that of the Carys Mills Formation of northeastern Maine and adjacent New Brunswick. The lower part of this thick unit has yielded specimens of Glyptograptus persculptus (Rickards & Riva 1981) and Llandovery age graptolites are known from higher parts of the formation (Pavlides 1968). The Carys Mills has also produced well preserved conodonts of the Icriodella discreta—I. deflecta Zone of probable Rhuddanian (early Llandovery) age (Barnes & Bergstrom, this volume) but, unfortunately, the precise stratigraphical position of these cono- donts within the formation is uncertain because of the scattered exposures, considerable thick- ness, monotonous lithology, and structural deformation of the unit. At any rate, it appears rather likely that the Carys Mills represents a stratigraphically complete succession from the uppermost Ordovician to the lower Silurian, but further studies are needed to pinpoint the level of the systemic boundary. Central and Southern Appalachians In southern New York and parts of eastern Pennsylvania and Virginia (Fig. 1), the Ordovician— Silurian boundary is marked by an unconformity (Dennison 1976) and parts of the Ordovician, and possibly also of the lowermost Silurian, are missing. From north-central Pennsylvania to eastern Tennessee, the systemic boundary is somewhere in a succession, several hundred metres thick, of near-shore to non-marine clastic sediments lacking shelly fossils of stratigraphical utility. Although the precise level of the systemic boundary remains undetermined in these successions, it has been common practice to classify the Juniata and Sequatchie formations as Ordovician and the overlying Tuscarora and Clinch formations as Silurian. Recent work by Colbath (1986) has raised the possibility of establishing a viable palyno- morph (acritarch and chitinozoan) biostratigraphy useful for precise recognition and correla- tion of the systemic boundary in these successions. Likewise, Gray’s work (1985) on higher land plant spore tetrads permits recognition of the approximate boundary interval. Both the spore tetrads and the marine palynomorphs occur in some abundance in near-shore marine sedi- ments. The spores are also found in purely non-marine facies provided they have not been destroyed by low-temperature metamorphism of the host strata. However, palynomorph work in the systemic boundary interval in this region has not passed the pioneer stage, and much additional study is needed to assess the local and regional biostratigraphic utility of these fossils. In the southernmost Appalachians, in the Birmingham area of Alabama, the systemic bound- ary is marked by a conspicuous stratigraphical gap that includes the entire Upper Ordovician and probably the lowermost Llandovery as well (Hall, unpublished; Berry & Boucot 1970). Near the Alabama—Georgia boundary, the stratigraphical gap also includes the entire Middle Ordovician (Dennison 1976), but in northwesternmost Georgia, Chowns (1972) considered the systemic contact conformable on lithological evidence. The youngest Ordovician strata in much of Alabama, which are referred to the Sequatchie Formation (Drahovzal & Neathery 1971), are of Late Ordovician (Maysvillian and Richmondian) age. In Limestone County in northern Alabama, the Devonian Chattanooga Shale contains reworked Late Ordovician (probably Richmondian) conodonts (Bergstrom, unpublished) apparently originating from now-eroded rocks that may be younger than the biostratigraphically dated parts of the Sequatchie Forma- tion. Where dated biostratigraphically, the Sequatchie is separated from overlying rocks by a stratigraphical gap corresponding not only to the uppermost Ordovician but also some part of the post-Ordovician succession. Locally this gap is substantial and may include more than a system. Eastern North American Midcontinent We include in this area the Cincinnati Arch region in Ohio, Kentucky, and Indiana, and the Nashville Dome area in central Tennessee (Fig. 1). 276 S. M. BERGSTROM & A. J. BOUCOT The Cincinnati region contains the Reference Standard of the North American Upper Ordo- vician, the Cincinnatian Series. Both faunal and lithological evidence suggest an appreciable hiatus between the Ordovician and the Silurian over the entire outcrop area in the Cincinnati region. The stratigraphically most complete succession is apparently on the eastern side of the Cincinnati Arch in southern Ohio and adjacent Kentucky. There is no record of Hirnantian (latest Ordovician) age rocks anywhere in the Cincinnati region and the youngest Cincinnatian stage, the Richmondian, is considered to be of pre-Hirnantian age. Based on the succession of Anticosti Island, Québec, Canada, Twenhofel’s (1921) Gamachian Stage has in recent years been recognized as a post-Richmondian, pre-Silurian standard unit (Barnes & McCracken 1981). Although rocks of Gamachian age are\ not known to be represented in the Cincinnatian type area, the Gamachian is now classified as the uppermost part of the Cincinnatian Series (Ross et al. 1982). One of the best exposed and most representative sections through the Ordovician—Silurian boundary interval on the eastern flank of the Cincinnati Arch is a series of exposures along Ohio Highway 41 between West Union and Ohio Brush Creek, Adams County, Ohio (Summerson 1963; Rexroad et al. 1965; Gray & Boucot 1972; Grahn & Bergstrom 1985). In this section, the beds are horizontal, developed in fossiliferous limestone and shale, and there are no structural complications. The topmost Ordovician unit, the Drakes Formation of Rich- mondian age, is overlain conformably and without conspicuous lithological break by the Belfast Member of the Brassfield Formation (Fig. 2). This unit has produced a relatively undiagnostic conodont fauna of general early Llandovery type (Rexroad 1967) as well as chitinozoans suggesting a C. cyphus Zone age (Grahn & Bergstrom 1985). Grahn & Bergstrom (1985) interpreted the stratigraphical gap as corresponding to about four graptolite zones and it is surprising that there is no channelling, development of a conglomerate, or other lithic evidence of a sedimentary break. The major body of the Brassfield, that is, its post-Belfast part, contains a rich megafossil fauna of early to middle Llandovery age (Berry & Boucot 1970) as well as a stratigraphically diagnostic conodont fauna of the Distomodus kentuckyensis Zone (Rexroad 1967; Cooper 1975) and chitinozoans (Grahn 1985). There are no graptolites known from this succession. In many other Cincinnati region sections, especially on the west flank of the Cincinnati Arch, the stratigraphical gap associated with the systemic boundary is even greater than in the Ohio Brush Creek sections (Rexroad & Kleffner 1984). In parts of the Nashville Dome in central Tennessee, the Devonian Chattanooga Shale unconformably overlies Middle Ordovician rocks (Dennison 1976). In other parts of the Nash- ville Dome, strata dated as Richmondian are overlain unconformably by the Brassfield Lime- stone of middle Llandovery age (Wilson 1949), which indicates the presence of a stratigraphical gap of magnitude similar to that in the Cincinnati region. Central North American Midcontinent We include in this area Oklahoma and adjacent Texas Panhandle, Arkansas, Missouri, Illinois, Minnesota, and Wisconsin (Fig. 1). In a recent comprehensive study, Amsden & Barrick (1986) provided a useful summary of the geology of the Ordovician-Silurian boundary interval in this region. Of particular significance is the confirmation of the widespread occurrence of latest Ordovician strata having shelly fossils of the Hirnantia fauna and conodonts of the Noixodontus fauna. The stratigraphically most informative sections are in the Batesville district of north-central Arkansas and in eastern Missouri. Both locally and regionally, the stratigraphical succession varies a great deal, and in several cases, sections in close proximity to each other exhibit striking differences in lithological and stratigraphical development. This is well illustrated by the conditions in the Batesville district as well as in eastern Missouri. In the Batesville district two sections are of particular interest. One of these sections is in the Love Hollow Quarry (Craig 1968, 1986a, 1986b; Amsden 1968, 1986). In this large and recently active quarry, the beds are horizontal and there are no notable tectonic complications. A ORDOVICIAN-SILURIAN BOUNDARY IN UNITED STATES 277 | CONODONTS | |] CHITINOZOANS | SHELLY FOSSILS | Platymerella Zone Ancyrochitina sp iklaensis C. gregarius Z. equiv Aeronian Ozarkodina hassi Oulodus sp Conochitina sp. cf (6. Icriodella discreta Zz o a 2/0 “jo DSla a/z n|< a el z < x S) > sy Lu ie Fe < 2 fe) ea uu Aulacognathus bullatus Pterospathodus celloni Oulodus pelitus Ozarkodina sp. cf. oldhamensis Distomodus kentuckyensis Pt. amorphpg I— p staurognathoides D. kentucky. & Pt. celloni (Cason Oolite Cason Shale “Fernvale’® Ls undatus HIRNANTIAN Phragmodus Protopanderodus insculptus ORDOVICIAN ASHGILL Am. ordovicicus Noixodontus Carniodus sp Amorphognathus ordovicicus girardeauensis conodonts from this locality confirm that the Cason Oolite is of Hirnantian age and that the overlying Brassfield is coeval with the Brassfield of the Cincinnati region (Craig 1986b; Barrick 1986). The systemic boundary is placed at the base of the Brassfield and is not strongly expressed lithologically; it may be associated with a stratigraphical gap corresponding to the lowermost Llandovery, but conodonts and other fossils do not provide sufficient stratigraphical resolution to assess its magnitude precisely. The Cason Oolite equivalent in southern Oklahoma is apparently the Keel Limestone (Amsden 1986) that has yielded Hirnantian age brachiopods as well as conodonts of the Noixodontus fauna (Barrick 1986). Its topmost part has also produced stratigraphically younger conodonts of general Silurian aspect but no Silurian index species. Barrick assigned the latter fauna to the Llandovery and placed the systemic boundary within the Keel. Amsden, on the basis of his brachiopod studies, placed the entire Keel in the Ordovician (Amsden 1986: text-fig. 37) and noted that the unit is separated from the overlying Cochrane Formation by a large stratigraphical gap corresponding to the lower and middle Llandovery. In our opinion, the Silurian-type conodont fauna reported from the upper Keel by Barrick (1986) does not provide firm evidence of Silurian age because, as shown by Barnes & Bergstrom (this volume, p. 325), the turnover from an Ordovician-type to a Silurian-type conodont fauna may well have taken place in very latest Ordovician (late G. persculptus Zone) time, within a time interval older than ORDOVICIAN-SILURIAN BOUNDARY IN UNITED STATES 279 the base of the Silurian. Whether or not this alternative dating is correct can be solved only after the conodont faunal turnover has been firmly dated in terms of graptolite zones. As noted by Amsden (1986), there are two important outcrop areas of the systemic interval in the Mississippi Valley, one in west-central Illinois and northeastern Missouri, and the other in southwestern Illinois and southeastern Missouri. A considerable number of sections through the uppermost Ordovician and overlying Silurian strata have been described by Amsden (1974, 1986) and Thompson & Satterfield (1975). The former also described the brachiopod faunas and the latter reported on the conodonts (also cf. McCracken & Barnes 1982). The strati- graphically most complete systemic boundary sequences are in the former area; in the latter area, the Edgewood Group, of Hirnantian age at the top, is overlain directly and unconform- ably by the Sexton Creek Limestone that contains brachiopods suggesting a late Llandovery (late Aeronian—Telychian) age (see Fig. 4). One of the biostratigraphically most instructive sections is along the west side of Missouri Highway 79 at Clinton Springs at the south edge of Louisiana, Pike County, Missouri, where the horizontal beds are easily accessible along a major highway. Good brachiopod collections of Hirnantian age have been described from the Noix Oolite at this locality (Amsden 1974, 1986) and conodonts (of Noixodontus fauna type) studied by Thompson & Satterfield (1975). The overlying Bryant Knob Formation has yielded a few brachiopods (Amsden 1974) and conodonts interpreted as indicating early Llandovery age (Thompson & Satterfield 1975). The OKLAHOMA | _S.£. MISSOURI N.E. MISSOURI Coal Creek | Thebes N. Clarksville ] 4 mi S. of Clarksville Bowling Green Dolomite U Llandovery U Llandovery Undiagn Bowling Cochrane Fm Sexton brachs SILURIAN Green shelly & conodont Creek Fm. }| brachiopods Silurian- aspect conodonts Conodonts Dolomite P. celloni Z ?0.? nathani Z faunas of Silurian Undiagnostic conod. faunas aspect r? | LLANDOVERY a ee ee [a Bryant Knob Em e Hirnantia Noix Oolite Hirnantia Ni Hirnantia Ni Hirnantia N =] Keel Oolite $ = and RB and 3 Noix Qolite 2 and 3 Zz o S 8 ~ Pag Noixodontus 2 Noixodontus $ Noixodontus $ Neix Oolite | Nosxodontus S z 3 a) 3 S z o faunas S faunas 9° fauna v faunas x fj o/> 3 9 5 5 — a ola Ideal Quarry S |\Girardeau Ls| $ 2 3 a Mbr ° s rs < 3 i] o 2 O15 = 5 5 5 a 3 & 2 2 5 y = S S fe) <= is) is) ° 2 € iS S S St So zt Tz D Orchard ; Maquoketa Maquoketa Sylvan Sh complanatus| & Creek Sh Sh Sh Z Fig. 4 Occurrence of key fossil assemblages, and general biostratigraphy, in the systemic boundary interval at some localities in Oklahoma, southeastern Missouri, and northeastern Missouri. For the location of these sections, see Amsden & Barrick (1986), and Thompson & Satterfield (1975), and these papers provide most of the data upon which this diagram is based. As is clear from the diagram, it is a review of the general stratigraphy in each of the areas and no correlation is implied between a unit in one column and one at the same vertical position in another column. In Oklahoma and southeastern Missouri, the systemic boundary is associated with a prominent stratigraphical gap whereas in the illustrated sections from northeastern Missouri, the succession across the systemic boundary may have only a minor, if any, stratigraphical gap. 280 S. M. BERGSTROM & A. J. BOUCOT succession does not show any distinct lithic break between these units and it may be one of the stratigraphically most nearly complete boundary successions in the Midcontinent region. A stratigraphically similar section is present along Highway 79 about 6:5km south of Clarksville and about 19km southeast of the Clinton Springs locality (Fig. 4; Amsden 1974). In his recent reassessment of the data at hand, Amsden again placed the systemic boundary at the top of the Noix Oolite but indicated (1986: 42) that ‘the brachiopod biostratigraphy requires no signifi- cant interruption in the Noix-Bryant Knob sequence’. Interestingly, McCracken & Barnes (1982) reported conodonts of Silurian aspect, by them interpreted as representing either the O.? nathani Zone or the D. kentuckyensis Zone, from the lowermost 1:65m of the Bowling Green Dolomite from a locality near Clarksville, Where this unit directly overlies the Noix Oolite, which yielded a representative Noixodontus fauna. The conodont faunas from the Noix and the Bowling Green are quite different, and there is obviously a faunal turnover between these units. Unfortunately, as noted by Barnes & Bergstrom (this volume), the precise age of this faunal turnover is currently unknown in terms of the graptolite succession, but it is quite possible that it took place in the latest Ordovician. If so, it cannot be excluded that the systemic boundary is above the base of the Bowling Green. However, the fact that the latter unit is overlain by the late Llandovery Sexton Creek Limestone (Amsden 1986) makes it clear that the systemic boundary must be below the base of the latter unit. Western Midcontinent Important outcrop areas in this vast region (Fig. 1) include the Black Hills in South Dakota, the Bighorn Mountains in Wyoming, and areas in Montana, Colorado, southern New Mexico, and western Texas. Most of the Upper Ordovician in these areas consists of shallow-water carbonates with few megafossils but with taxonomically varied and biostratigraphically useful conodont faunas (Sweet 1979). The biostratigraphy of the overlying beds is less well known. No biostratigraphically well-controlled section is currently known that is stratigraphically reason- ably complete in the Ordovician—Silurian boundary interval, and the data suggest that every- where rocks of Ordovician age are separated from younger rocks by an unconformity representing a significant stratigraphical gap (Ross et al. 1982). The most nearly complete boundary section may be in the subsurface of North Dakota; however, data from the deposi- tional basin extension in adjacent Manitoba, where the succession is quite similar to that in North Dakota, suggest the absence of at least the lowermost Llandovery (Barnes & Bergstrom, this volume). Great Basin We include in this region western Utah, Nevada, Idaho, and southern California (Fig. 1). There are numerous excellent sections of Upper Ordovician and Lower Silurian rocks in carbonate facies with virtually 100% exposure in the Great Basin, and most of these sections may be reached by car and by foot under reasonable conditions. However, many localities are structur- ally complex, and widespread secondary dolomitization, particularly in the Ordovician, makes it difficult to obtain well-preserved megafossils. Furthermore, diagnostic shelly megafossils are not common and graptolites are rare. Conodonts are known from some sections and they offer great potential for detailed stratigraphical work in the widespread carbonates; unfortunately, the problem of dating the conodont faunal turnover referred to above currently restricts their use in establishing precisely the position of the Ordovician—Silurian boundary. Accordingly, it is currently impossible to recognize with certainty the exact level of the systemic boundary, or even to assess whether or not deposition was continuous, at those carbonate sections where there is not a conspicuous unconformity in the boundary interval. Much of the pertinent biostratigraphical information from megafossils was summarized by Berry & Boucot (1970). Additional data from shelly fossils have been published by, among others, Budge & Sheehan (1980a, 1980b) and Sheehan (1980, 1982). ORDOVICIAN-SILURIAN BOUNDARY IN UNITED STATES 281 Although conodont work in the systemic boundary interval is still in the pioneer stage in the Great Basin, it is apparent that conodonts offer greater potential than any other group for detailed biostratigraphical work. Two recent conodont studies deserve mention in a discussion of the systemic boundary. Ross et al. (1979) described the conodont biostratigraphy of the Hanson Creek Formation near Eureka, Nevada. They suggested that this unit represents continuous deposition from Ordovician to Silurian time. This is quite possible, but it is perhaps equally possible that all the conodont samples referred to in their study are of Ordovician age and that the systemic boundary is at a higher, as yet undetermined, level in the Hanson Creek than that advocated by Ross et al. because, as noted by Barnes & Bergstrom (this volume), none of their conodont collections contain index conodonts of definite Silurian age. In another recent study, Leatham (1985) described the conodont biostratigraphy of the Fish Haven Dolomite and immediately overlying strata in a section in northernmost Utah. He identified a prominent conodont faunal turnover and a transitional fauna interval of 5-Sm thickness in the uppermost Fish Haven. The systemic boundary was placed at the base of this transition interval, but Leatham (1985) was uncertain whether or not there was a strati- graphical gap at this level. He was also uncertain about the nature of the mixed faunal association and suggested that it might be a product of reworking or stratigraphic leak. In our view, it cannot be excluded that the interval with the mixed fauna, regardless of its nature, is of Hirnantian age, and that the systemic boundary, as it is now defined by means of graptolites, is at a somewhat higher stratigraphical level, in the lowermost part of the Laketown dolomite. Of special interest in a review of the Ordovician—Silurian boundary biostratigraphy in the Great Basin is Berry’s (1986) record of an uppermost Ordovician to lower Silurian sequence of graptolite faunas in cherts and dolomites of the upper Hanson Creek Formation in the Monitor Range, central Nevada. A quartz sand-bearing dolomite, which evidently represents a period of shallowing near the end of the Ordovician, is underlain by strata having the Dicello- graptus complanatus ornatus graptolite assemblage, and directly overlain by rocks containing the diagnostic species association of the Glyptograptus persculptus Zone. Stratigraphically higher beds contain species that may represent the P. acuminatus Zone but the zonal index has not been found. Graptolite-bearing shale sequences of Ordovician and Silurian age are widespread in the mountain ranges in the Great Basin but the studied successions appear to be stratigraphically incomplete and display a gap in the systemic boundary interval. For instance, in the carefully studied and well-known graptolite shale succession in the Trail Creek area, central Idaho, Llandovery beds older than the M. convolutus Zone are missing (Carter & Churkin 1977). Alaska With one important exception, little information is currently available concerning the geology of the Ordovician—Silurian boundary in Alaska. This exception is the Prince of Wales region in southeastern Alaska (Fig. 1) where in the long-ranging Descon Formation there is a quite condensed succession through the systemic boundary interval, which displays a complete sequence of late Ordovician—early Silurian graptolite zones. The best known succession is on Esquibel Island (Churkin & Carter 1970; Churkin et al. 1971) where a few metres thick sequence of cherty shales spans the systemic boundary without any indication of depositional breaks. A less than 3m thick interval with the G. persculptus Zone fauna is overlain by about 1-5m of strata containing graptolites characteristic of the P. acuminatus Zone, including the zonal index. The Esquibel Island graptolite species associations show close similarity to those of coeval strata in the Birkhill Shale in the Ordovician—Silurian boundary stratotype at Dob’s Linn in south Scotland, making it possible to recognize the level of the systemic boundary with considerable precision. This may be the only place in the United States where the level of the systemic boundary can be fixed conclusively on graptolite evidence in a stratigraphically con- tinuous section, and one can only regret that this key locality is located in a remote region that is likely to be visited by very few geologists. 282 S. M. BERGSTROM & A. J. BOUCOT Zz < ac =) = 2p) Zz x 2 > e) Qa oc fe) Conclusions . The Ordovician—Silurian boundary interval is well exposed at numerous localities through- out the United States from the Appalachians in the east to the Great Basin in the west. . Available biostratigraphical and/or lithostratigraphical evidence suggests that in the vast majority of these sections, there is a stratigraphical gap, of greatly different magnitude in different sections, in the boundary interval (Fig. 5). Particularly in the shallow-water cratonic successions, this gap reflects the global drop of sea-level near the end of the Ordovician, but there is evidence that local uplifts have been of importance in some areas. Currently, we are aware of only a single biostratigraphically closely controlled section in the United States, on Esquibel Island, southeastern Alaska, which displays continuous deposition throughout the boundary interval. However, such sections may exist elsewhere, particularly in the Appa- lachians and in the Great Basin. . Some of the best, and biostratigraphically most closely controlled, boundary sections are in Arkansas, Oklahoma, Missouri, and Illinois where rocks having the Hirnantia shelly fauna and the Noixodontus conodont fauna are overlain by Llandovery-age strata with, at least locally, only a minor, if any, stratigraphical gap. Regrettably, no stratigraphically diagnostic graptolites are known from these sections. . A considerable number of well-exposed, thick, and apparently stratigraphically relatively complete sections in shallow-water carbonate facies are known from the Great Basin. Dolo- mitization has seriously affected the state of preservation of the megafossils, which are rather scarce in most sections, but conodonts are moderately common and taxonomically varied. Yet, because the conodont biostratigraphy is not tied reliably to the graptolite zone suc- cession in the G. persculptus and P. acuminatus Zones, currently the conodonts cannot be used to pinpoint the level of the systemic boundary in carbonate sections without a signifi- cant stratigraphical gap. . As far as we are aware, in the United States the P. acuminatus Zone has been identified with certainty only on Esquibel Island, Alaska, and this is the only place where the level of the systemic boundary can be established precisely by means of zonal graptolites. The suc- cessions of shelly fossils, conodonts and palynomorphs are thus far calibrated only impre- N.E. MAINE} APPALACH N. ARKAN.| S.OKLAH. JE.MISSOURI} N. UTAH NEVADA jS.E. ALASKA Cochrane cs Sexton [Foul Creek Fm Brassfield Ls (Tripl. cs ? c alata beds) LLANDOVERY eps Brassfield Tuscarora and Clinch Fms Bowling Laketown Green Dol Descon Juniata and Oolite & c«s Sequatchie c c Sh Fms csp Richmond- janice Sylvan Sh peg | Maquoketa Sh Fig.5 Summary diagram showing important formations and degree of stratigraphical completeness of systemic boundary sections in nine important outcrop areas in the United States. Small letters, which indicate the presence of biostratigraphical control by means of a particular index fossil group, denote the following: c, conodonts; g, graptolites; s, shelly fossils (especially brachiopods); p, palynomorphs (especially chitinozoans). For further data on each of these successions, see the text. The section of northern Utah is that described by Leatham (1985). Vertical ruling marks proved or assumed stratigraphical gaps. Only formations near the systemic boundary are listed in the diagram. ORDOVICIAN-SILURIAN BOUNDARY IN UNITED STATES 283 cisely and broadly with the graptolite zone succession, and therefore these fossils cannot yet be used successfully to pinpoint the precise level of the Ordovician—Silurian boundary, especially in sections without a significant stratigraphical gap in the boundary interval. If the base of the P. acuminatus Zone is to be a viable and useful level for the base of the Silurian, then it is clearly necessary to determine the precisely equivalent level in the successions of shelly fossils, conodonts and palynomorphs. Because of the absence of graptolite control in the critical sections in the United States, that biostratigraphically most important correla- tion work will have to be carried out elsewhere in the world. However, the mutual strati- graphical relationships between non-graptolitic taxa are well displayed in sections in the United States. A detailed study of these relations no doubt will produce interesting and useful information of regional significance. Acknowledgements We are indebted to Dr W. B. N. Berry for valuable information and to Ms Karen Tyler for technical assistance with the manuscript. Special thanks are due to T. W. Amsden and J. E. Barrick for proof copies of their important study on Hirnantian and associated faunas in the central United States. References Amsden, T. W. 1968. Articulate brachiopods of the St Clair Limestone (Silurian), Arkansas, and the Clarita Formation (Silurian), Oklahoma. Paleont. Soc. Mem. 1 (J. Paleont., Tulsa, 42 (3 suppl.)). 117 pp., 20 pls. —— 1974. Late Ordovician and Early Silurian articulate brachiopods from Oklahoma, southwestern Illinois and eastern Missouri. Bull. Okla geol. Surv., Norman, 119: 1-154, 28 pls. 1986. Part I—Paleoenvironment of the Keel-Edgewood oolitic province and the Hirnantian strata of Europe, USSR, and China. In T. W. Amsden & J. E. Barrick, 1986 (q.v.). — & Barrick, J. E. 1986. Late Ordovician—Early Silurian strata in the central United States and the Hirnantian Stage. Bull. Okla geol. Surv., Norman, 139. 95 pp., 7 pls. Barnes, C. R. & McCracken, A. D. 1981. Early Silurian chronostratigraphy and a proposed Ordovician— Silurian boundary stratotype, Anticosti Island, Québec. In P. J. Lespérance (ed.), Field Meeting, Anticosti—Gaspe, Quebec, 1981 2 (Stratigraphy and paleontology): 71-79. Montréal (I.U.G.S. Subcom- mission on Silurian Stratigraphy Ordovician—Silurian Boundary Working Group). Barrick J. E. 1986. Part II—Conodont faunas of the Keel and Cason Formations. In T. W. Amsden & J. E. Barrick, 1986 (q.v.). Berry, W. B. N. 1986. Stratigraphic significance of Glyptograptus persculptus group graptolites in central Nevada, U.S.A. In R. B. Rickards & C. P. Hughes (eds), Palaeoecology and Biostratigraphy of graptol- ites: 135-143. Oxford (Spec. Publs geol. Soc. Lond. 20). & Boucot, A. J. (eds) 1970. Correlation of the North American Silurian rocks. Spec. Pap. geol. Soc. Am., New York, 102: 1-289. Budge, R. D. & Sheehan, P. M. 1980. The Upper Ordovician through Middle Silurian of the eastern Great Basin. Part 1. Introduction: Historical perspective and stratigraphic synthesis. Contr. Biol. Geol. Mil- waukee Public Mus. 28: 1-26 (1980a). Part 2. Lithologic descriptions. Loc. cit. 29: 1-80 (19805). Carter, C. & Churkin, M. jr 1977. Ordovician and Silurian graptolite succession in the Trail Creek area, Central Idaho—A graptolite Zone reference section. Prof. Pap. U.S. geol. Surv., Washington, 1020: 1—33, 7 pls. Chowns, T. M. 1972. Depositional environments in the Upper Ordovician of northwest Georgia and southeast Tennessee. In T. M. Chowns (ed.), Sedimentary environments in the Paleozoic rocks of northwest Georgia. Guidebk Dep. Mines Min. Geol., Ga, Atlanta, 11: 3-12. Churkin, M., jr & Carter, C. 1970. Early Silurian graptolites from southeastern Alaska and their correla- tion with graptolite sequences in North America and the Arctic. Prof. Pap. U.S. geol. Surv., Washington, 653: 1-51, 4 pls. ——, —— & Eberlein, G. D. 1971. Graptolite succession across the Ordovician—Silurian boundary in southeastern Alaska. Q. JI geol. Soc. Lond. 126: 319-330, 1 pl. Colbath, G. K. 1986. Abrupt terminal Ordovician extinction in phytoplankton associations, Southern Appalachians. Geology, Boulder, Colo., 14: 943-946. Cook, T. D. & Bally, A. W. (eds) 1975. Stratigraphic Atlas of North and Central America. 272 pp. Princeton. 284 S. M. BERGSTROM & A. J. BOUCOT Cooper, B. J. 1975. Multielement conodonts from the Brassfield Limestone (Silurian) of southern Ohio. J. Paleont., Tulsa, 49 (6): 984-1008, 3 pls. Craig, W. W. 1969. Lithic and conodont succession of Silurian strata, Batesville District, Arkansas. Bull. geol. Soc. Am., New York, 80: 1621-1628. 1986. Conodont paleontology of Middle and Upper Ordovician strata, Batesville District, Arkansas. In W. W. Craig, R. L. Ethington & J. E. Repetski, Guidebook to the conodont paleontology of uppermost Lower Ordovician through Silurian strata, northeastern Arkansas. Geol. Soc. Am., Annual Meeting (S-C & S-E sects, Memphis) Guidebook Field Trip 5: 1-19 (1986a). Conodont paleontology of Silurian strata, Batesville District, Arkansas. Loc. cit.: 21-35 (19865). Dennison, J. M. 1976. Appalachian Queenston delta related to eustatic sea-level drop accompanying Late Ordovician glaciation centred in Africa. In\M. G. Bassett (ed.), The Ordovician System: 107-120. University of Wales Press. Drahovzal, J. A. & Neathery, T. L. (eds) 1971. The Middle and Upper Ordovician of the Alabama Appalachians. Guidebook, Alabama Geol. Soc. 9th Annual Fieldtrip. 240 pp. Grahn, Y. 1985. Llandoverian and early Wenlockian Chitinozoa from southern Ohio and northern Kentucky, U.S.A. Palynology, Dallas, 9: 147-164, 2 pls. —— & Bergstrom, S. M. 1985. Chitinozoans from the Ordovician—Silurian boundary beds in the eastern Cincinnati region in Ohio and Kentucky. Ohio J. Sci., Columbus, 85 (4): 175-183, 1 pl. Gray, J. 1985. The microfossil record of early land plants: advances in understanding of early terrestriali- zation, 1970-1984. Phil. Trans. R. Soc., London, (B) 309: 167-192. —— & Boucot, A. J. 1972. Palynological evidence bearing on the Ordovician-—Silurian paraconformity in Ohio. Bull. geol. Soc. Am., New York, 83: 1299-1314. Leatham, W. B. 1985. Ages of the Fish Haven and lowermost Laketown Dolomites in the Bear River Range, Utah. Publs Utah geol. Ass., Salt Lake City, 14: 29-38. McCracken, A. D. & Barnes, C. R. 1982. Restudy of conodonts (Late Ordovician—Early Silurian) from the Edgewood Group, Clarksville, Missouri. Can. J. Earth Sci., Ottawa, 19: 1474-1485, 2 pls. Neuman, R. B. 1968. Paleogeographic implications of Ordovician shelly fossils in the Magog belt of the northern Appalachian region. In Zen E-an, W. S. White, et al. (eds), Studies of Appalachian Geology: northern and maritime: 35-48. New York. Pavlides, L. 1968. Stratigraphic and facies relationships of the Carys Mills Formation of Ordovician and Silurian age, Northwest Maine. Bull. U.S. geol. Surv., Washington, 1264: 1-44. Rexroad, C. B. 1967. Stratigraphy and conodont paleontology of the Brassfield (Silurian) in the Cincinnati Arch area. Bull. Indiana geol. Surv., Bloomington, 36: 1—64, 4 pls. ——, Branson, E. R., Smith, M. O., Summerson, C. & Boucot, A. J. 1965. The Silurian formations of east-central Kentucky and adjacent Ohio. Bull. Ky geol. Surv., Lexington, (X) 2. 34 pp. —— & Kleffner, M. A. 1984. The Silurian stratigraphy of east-central Kentucky and adjacent Ohio. Geol. Soc. Am., Annual Meeting (S-E & N-C sects) Field Trip Guides: 44-65. Rickards, R. B. & Riva, J. 1981. Glyptograptus? persculptus (Salter), its tectonic deformation, and its stratigraphic significance for the Carys Mills Formation of N.E. Maine, U.S.A. Geol. J., Liverpool, 16: 219-235. Ross, R. J. & 28 co-authors. 1982. The Ordovician System in the United States. Correlation chart and explanatory notes. Internat. Union geol. Sci., (A) 12. 73 pp. , Nolan, T. B. & Harris, A. G. 1979. The Upper Ordovician and Silurian Hanson Creek Formation of Central Nevada. Prof. Pap. U.S. geol. Surv., Washington, 1126-C: C1—C22, 4 pls. Sheehan, P. M. 1980-82. The Late Ordovician and Silurian of the Eastern Great Basin. Part 3. Brachio- pods of the Tony Grove Lake Member of the Laketown Dolomite. Contr. Biol. Geol. Milwaukee Publ. Mus. 30. 23 pp. (1980). Part 4. Late Llandovery and Wenlock brachiopods. Loc. cit. 50. 83 pp. (1982). Summerson, C. 1963. A résumé of the Silurian stratigraphy of Ohio. A. Fld Excurs. Michigan Basin geol. Soc., Lansing, 1963: 12-34. Sweet, W. C. 1979. Late Ordovician conodonts and biostratigraphy of the western Midcontinent Prov- ince. Geology Stud. Brigham Young Univ., Provo, 26 (3): 45-85, 5 pls. Thompson, T. L. & Satterfield, I. R. 1975. Stratigraphy and conodont biostratigraphy of strata contiguous to the Ordovician-Silurian boundary in eastern Missouri. Rep. Invest. Mo. geol. Surv., Rolla, 57 (2): 61-108. Twenhofel, W. H. 1921. Faunal and sediment variations in the Anticosti sequence. Mus. Bull. Can. geol. Surv., Ottawa, 33 (Geol. ser. 40): 1-14. Wilson, C. W. 1949. Pre-Chattanooga stratigraphy in central Tennessee. Bull. Tenn. Div. Geol., Nashville, 56: 1-407, pls 1-28. The Ordovician—Silurian boundary in South America A. J. Boucot Department of Zoology, Oregon State University, Corvallis, Oregon 97331, U.S.A. Synopsis In South America late Ashgill rocks followed in the same succession by the early Llandovery are known only in the Precordillera of San Juan, Argentina. Early Llandovery fossils are known from the Puna Well, Argentina, the basal Trombetas Formation of Brazil, west of Lake Titicaca in Peru, and in the Merida Andes of Venezuela. Glaciogenic deposits of presumed Ordovician-Silurian boundary age are known from Argentina, Bolivia, Brazil and Peru. Introduction There are unfossiliferous and relatively unfossiliferous strata in South America whose assign- ment to either the Ordovician or Silurian is a problem. But I am unaware of any South American area where there are fossiliferous strata involved in real Ordovician—Silurian bound- ary indecision. In South America the assignment of fossiliferous beds to either the Ordovician or Silurian has been easy because there are no areas recognized to date where fossiliferous beds of latest Ordovician and earliest Silurian age are present in conjunction with each other. Discussion of the Ordovician—Silurian boundary in South America may be broken into two parts: (1) the strata present on the shield areas; (2) the strata present in the structurally complex Andean regions bordering the shield areas to the north and west. Recognized, fossiliferous Ordovician rocks have not yet been shown to exist on the shield areas except for a few areas very close to the Andean, disturbed rocks, whereas there is widespread Ordovician scattered here and there in the Andean regions; fossiliferous Silurian rocks are widespread on the shield areas, as well as in the Andean regions. There are potentially Ordovician, unfossiliferous strata, possibly latest Ordovician, rocks on the shield areas, but until some means of dating them precisely emerges it would be futile to spend time discussing them. For example, Caputo & Crowell (1985) have described diamictites that may be tillites of Ashgill age, that occur not too far below Silurian strata containing higher land plant spores of earlier Llandovery age (Gray, unpublished data from the Amazon Basin). I will, as stated, not devote attention to such difficult and biostratigraphically ambiguous beds. In the following summary statement I will review, geographic region by geographic region, what is currently known about the lowest Silurian and highest Ordovician fossiliferous rocks of the continent. It should, however, be kept in mind that the later Ordovician and earlier Silurian of South America are very poorly known, or known only in a rough reconnaissance manner, when contrasted with rocks of similar age in Europe. Conclusions arrived at here, particularly in the many poorly understood Andean regions, will certainly be subject to serious revision during the next few decades as additional field and laboratory studies take place. The Silurian correlation chart for South America (Berry & Boucot 1972) provides a good summary of the data available up to about 1970, but can now be significantly supplemented by additional published and unpublished data. Extra new data are also published by Cuerda et al. and Baldis & Pothe de Baldis (this volume, pp. 291-295). Argentina Amos (in Berry & Boucot 1972) provided an authorative review of the Argentinian Silurian, and its relations with the underlying Ordovician where present. The Argentinian Palaeozoic may be easily divided into that associated with the Andes in the north and the west, as contrasted with that present on the shield areas to the east. Much of the shield area Palaeozoic in Argentina is present in the subsurface beneath Mesozoic and Cenozoic cover, but there are Bull. Br. Mus. nat. Hist. (Geol) 43: 285-290 Issued 28 April 1988 286 A. J. BOUCOT limited areas where high-angle faulting has brought Precambrian and Palaeozoic rocks to the surface. The shield regions in the Buenos Aires, La Pampa and Rio Negro regions (Amos in Berry & Boucot 1972: fig. 2) have not yielded any body fossils of proved Ordovician age, although some unfossiliferous units have been assigned for varied reasons to the Ordovician. Fossiliferous Silurian rocks are present in these regions, but no fossils of proved Lower Llandovery age have been demonstrated. The Silurian fauna consists of Malvinokaffric Realm brachiopods for the most part, and, as is characteristic of that cool to cold climate Realm, few taxa are present. It is presently unclear in these regions whether strata that could conceivably have crossed the Ordovician-Silurian boundary are present.‘ The prevalence of late Ordovician to earlier Silu- rian continental glaciation in the Southern Hemisphere opens up the possibility that any such beds might well be in the non-marine category that can only be dated with a certain level of uncertainty for this time interval. The presence in the Cape Mountain System (Gray et al. 1986) of nearshore marine and possibly non-marine beds of probable Lower Llandovery or Ashgill age, or both, has some bearing on the Argentinian shield type occurrences in the Sierra de la Ventana, to the southwest of Buenos Aires in the Sierras Australes, which are commonly considered to be a pre-Jurassic continuation of the Cape Mountain System by many. In any event, it is reasonable to conclude (in the total absence of any dated Ordovician or early Llandovery fossils) that non-marine, or very nearshore, relatively unfossiliferous boundary beds might have been, or still might be present in the shield portions of Argentina. More subsurface data could demonstrate this possibility, particularly through the use of palynomorphs. For purposes of considering the Ordovician—Silurian boundary, the Andean regions of Argentina should be divided into the Precordillera de San Juan, where the Cambrian and Ordovician fossils have North American platform biogeographical affinities and occur in plat- form carbonate type rocks, and the Andes proper with their Malvinokaffric Realm Ordovician and Silurian faunas occurring in siliciclastic rocks. Amos (in Berry & Boucot 1972) has provided a summary for the Silurian of the Precordillera de San Juan. Nowhere are there fossiliferous Silurian rocks suspected to be older than Upper Llandovery, and the underlying Ordovician is nowhere thought to be younger than Caradoc, i.e. the Precordillera de San Juan is not a place in which to find a close approximation to the Ordovician-Silurian boundary as far as was then known, but see Cuerda et al. and Baldis & Pothe de Baldis (this volume). The only exception to this statement about the absence of the Ashgill is in a limited area, where the Cantera Formation (Furque & Cuerda 1979: 473) has yielded Ashgill trilobites and brachiopods (Baldis & Blasco 1975; Nullo & Levy 1976; Levy & Nullo 1974), although interrupted above by ‘contacto tectonico’ with a Lower Devonian unit. Tillites are not reported from this region, which suggests that the area may not necessarily have been subjected to continental glaciation, and might have been the site of a major regression associated with the terminal Ordovician—earliest Silurian glaciation. Amos (in Berry & Boucot 1972) has summarized the Andean Silurian of northwestern Argen- tina, chiefly in the Provinces of Salta and Jujuy. The fossiliferous Silurian is no older than about Upper Llandovery based on available data, except for the single Lower Llandovery fossiliferous occurrence in the Puna well to the west of the material summarized by Amos (see Boucot et al. 1976). This fossiliferous Silurian is underlain by the tillites of the Mecoyita Formation which lack diagnostic fossils, and have been commonly considered (Laubacher et al. 1982) to be of Ashgill age (although shown by Amos, in Berry & Boucot 1972, to be well up into the Upper Llandovery). The underlying fossiliferous Ordovician is nowhere demonstrated to be of Ashgill age, although Caradoc equivalents are recognized (Amos in Berry & Boucot 1972). It is clear that there are few places anywhere in Argentina for a palaeontologically-based close approach to the Ordovician-Silurian boundary. Bolivia Fossiliferous Ordovician (Hughes 1981, summary) and Silurian (Laubacher et al. 1982) rocks are well known in the Andean portions of Bolivia. However, no proved fossiliferous Silurian of ORDOVICIAN-SILURIAN BOUNDARY IN SOUTH AMERICA 287 Lower Llandovery age is known, nor fossiliferous beds of Ashgill age. Tillite separating fossil- iferous rocks belonging to the two systems is widespread. The oldest fossiliferous Silurian at present recognized is of Upper Llandovery age (Berry & Boucot 1972) from the Pojo region, where both brachiopods and graptolites provide the date. It is likely that there is a major, glacially correlated disconformity over most of Bolivia between the two systems (Berry & Boucot 1972: fig. 2). There is no reliable palaeontological evidence for placing any of the Andean tillites above the Llandovery: Berry & Boucot (1972: 26-27) summarize the graptolitic and brachiopod evidence from the overlying Kurusillas and Llallagua Formations, which contradicts that provided by Crowell et al. (1980); Crowell et al. (1981) suggest a Wenlock or Ludlow lower limit based on palynomorphs. An Ashgill age is most consistent for these tillites, in view of the overall emphasis on a glacial peak during that interval as contrasted with earlier Ordovician and later Silurian times. Antelo (1973) described Llandovery fossils from the Canca- niri, but the fossils actually come from above the tillite horizon (Cuerda & Antelo 1973) in beds which at Pojo were assigned by Berry & Boucot (1972) to the Llallagua Formation, which overlies the tillite proper. Brazil Fossiliferous Ordovician from the shield areas is unknown, except far to the west in the Amazonian region in the subsurface close to the areas of Andean disturbance. Silurian (Lange, in Berry & Boucot 1972) has been known from the Brazilian shield areas for over a century, but the graptolitic Silurian featuring Climacograptus has been conventionally assigned to the Llan- dovery, and not the latest Llandovery, because that genus was unknown above the Llandovery in the classic European and North American areas. Since 1972 there has been an accumulation of data indicating that Climacograptus can occur as high as the Lower Devonian (Jaeger 1978) in Austria, and that the palynomorphs associated with the graptolite show that the graptolites are no older than about Ludlow, rather than being of Llandovery age as had always been assumed. The palynomorphs in the Amazon Basin, where they occur with the graptolite, include acritarchs being studied by Luis Quadros, chitinozoans being studied by Florentin Paris, and higher land plant spores being studied by Jane Gray. All three specialists concur in assessing the age of the graptolitic part of the Trombetas Formation, the unit in question, as being no older than Ludlow. There is a possible tillite beneath the Trombetas Formation (Caputo & Crowell 1985). The tillite and associated strata are unfossiliferous, but an Ashgill age has been inferred, largely because the overlying, fossiliferous Trombetas Formation was concluded earlier to have been of Lower Llandovery age; this is now known to be an error. But basal Trombetas Formation beds, strata lacking any marine megafossils or marine palyno- morphs, have yielded spore tetrads to Jane Gray which are of earlier Llandovery age and which also indicate in the absence of any marine organisms a possible non-marine environment. Similar spore tetrads of similar age have been recovered from the Brazilian Parana Basin (Gray et al. 1985) and from the Cape Mountain System of South Africa (Gray et al. 1986). Silurian strata have been reported from the Parnaiba Basin (Lange in Berry & Boucot 1972 gives a summary) based on palynomorph studies. However, there is still uncertainty about the precise parts of the Silurian present within this Basin, and no fossils of proved Ordovician age are known. Fossiliferous Silurian was unknown in the Brazilian part of the vast Parana Basin until this decade (see Gray et al. 1985, for a summary, including the initial recognition of these beds and their fossils by de Faria). Now, with the aid of both acritarchs and higher land plant spore tetrads there is no doubt about the presence of shallow water, Benthic Assemblage 1, marine earlier Llandovery on the northeastern flank of the Basin. Earlier Silurian, based on graptolites from the southwestern flank of the basin in Paraguay, has been known for some time (Harrington in Berry & Boucot 1972), but no trace of any fossiliferous Ordovician is known anywhere to be associated with the Parana Basin. In summary the Brazilian shield areas are not ones where the Ordovician—Silurian boundary may be located by means of fossils, owing to the total absence of any Ordovician fossils immediately beneath the available Lower Silurian fossils. 288 A. J. BOUCOT Chile Fossiliferous Silurian rocks are unknown in Chile. The rocks from the Salar de Atacama region in northern Chile, assigned by Cecioni & Frutos (1975) to the Lower Palaeozoic (Ordovician, Silurian and Lower Devonian) are probably of Lower Carboniferous age, due to the similarity of their brachiopods to those found nearby (Bahlburg et al. 1986) which were assigned by Boucot to the Lower Carboniferous (fossiliferous Devonian beds are known from this area, yielding Tropidoleptus and Australocoelia, but these shells are unlike those figured by Cecioni & Frutos 1975 as contrasted with the Lower Carboniferous brachiopods). Fossiliferous earlier (Arenig) Ordovician is known in the Puna de Atacama, well to the east of the Salar de Atacama, but unassociated with fossiliferous Silurian. The nearest fossiliferous Silurian consists of a single Lower Llandovery locality in the Argentinian Puna, which yielded Cryptothyrella among other things (Boucot et al. 1976), which is unassociated with any fossiliferous Ordovi- cian. The fossiliferous Devonian beds in the Salar de Atacama region are no older than about Siegenian—Emsian, and rest unconformably on an older basement complex. We do not know whether there is any possibility of finding Ordovician—Silurian boundary region strata in Chile. The older Palaeozoic rocks of Chile are almost unknown, although there are many suspect regions that warrant careful attention. Colombia Fossiliferous Silurian rocks are presently unrecognized in Colombia, while none of the known Ordovician has been shown to even reach the Caradoc, much less the Ashgill (Hughes 1981). The presence in the Perija Andes, on the Colombian—Venezuelan boundary, of Lower Devon- ian fossiliferous beds, resting unconformably on basement complex, indicates that at least in some spots one would not expect fossiliferous Silurian or Ordovician strata to be preserved. Ecuador Fossiliferous Ordovician and Silurian rocks have not yet been recognized in Ecuador, although there is no reason to doubt their potential presence in the Andean part of the country. Paraguay See discussion of the Paraguayan Lower Silurian occurring on the southwestern margins of the Parana Basin under ‘Brazil’, p. 287. Peru There is widespread fossiliferous Ordovician and Silurian in southern Peru, both to the east and west of Lake Titicaca (see discussion of the Silurian in Laubacher et al. 1982; Hughes, 1981, summarizes the Ordovician, which has reliable palaeontological evidence only up to beds of Caradoc age). Laubacher et al. (1982) recognized Early Llandovery brachiopods to the west of Lake Titicaca, in the absence of the tillite that so commonly separates fossiliferous Ordovi- cian and Silurian rocks from each other in the central Andean region. But these fossiliferous Early Llandovery fossils are removed stratigraphically some distance from the youngest Ordo- vician rocks which have yielded fossils no younger than Caradoc. In southern Peru, therefore, there is no locality known where a close approach to the Ordovician—Silurian boundary is made within fossiliferous rocks. In central and northern Peru, as well as along the coast, fossiliferous Silurian rocks are unrecognized. The lack of tillite to the west of the Titicaca region does raise the possibility that an Ordovician-—Silurian transition may eventually be discovered in southern Peru or adjacent Bolivia, since a major disconformity might be more likely in the more easterly regions characterized by tillite. ORDOVICIAN-SILURIAN BOUNDARY IN SOUTH AMERICA 289 Venezuela The Ordovician and Silurian rocks related to the Ordovician—Silurian boundary are restricted in their occurrence to the Merida Andes, well to the south of Lake Maracaibo. Hughes (1981) comments that the Ordovician faunas of the Merida Andes are of Caradoc age; they are structurally well removed by faulting from immediate contact with the Lower Llandovery faunas of the Merida Andes described by Boucot et al. (1972). The Lower Llandovery faunas of the Merida Andes are dominated by brachiopods that cannot be dated any closer than Lower Llandovery; thus we are ignorant about whether or not these faunas are actually very close to the Ordovician-Silurian boundary. Graptolites that might help to resolve the age problem are unknown from the Merida Andes Llandovery. The shallow water nature of the Merida Andes Lower Llandovery, the medium-grained sandstones of the Silurian portion of the Caparo Formation with a Benthic Assemblage 2 set of communities dominated by such genera as Mendacella, is, however, consistent with the concept that there might be a disconformity between the two systems there, related to possible glacial regression, as is the case in many other parts of the world. In any event, the recognition of a close approximation to the Ordovician—Silurian boundary in Venezuela is as yet unknown. Acknowledgments I am indebted to Chris Hughes and Barrie Rickards for their friendly advice on a draft of the manuscript, and to Jane Gray and the palaeontologists of the PETROBRAS office in Rio de Janeiro for their assistance with the Brazilian data. References Antelo, B. 1973. La fauna de la Formacion Cancaniri (Silurico) en los Andes Centrales Bolivianos. Revta Mus. La Plata, (Paleont.) 7 (45): 267-277. Bahlburg, H., Breitkreuz, C. & Zeil, W. 1986. Palaozoisches Sedimenten nordchiles. Abh. Berliner Geowiss. (A) 66: 147-168. Baldis, B. A. & Blasco, G. 1975. Primeros trilobites Ashgillianos del Ordovicico Sudamericano. Actas I Congr. argent. Paleont. Bioestratigr., Tucuman, 1: 33-48. Berry, W. B. N. & Boucot, A. J. (eds) 1972. Correlation of the South American Silurian Rocks. Spec. Pap. geol. Soc. Am., New York, 133: 1—59. Boucot, A. J., Isaacson, P. E. & Antelo, B. 1976. Implications of a Llandovery (early Silurian) brachiopod fauna from Salta Province, Argentina. J. Paleont., Tulsa, 50: 1103-1112. ——, Johnson, J. G. & Shagam, R. 1972. Braquiopodos Siluricos de los Andes Meridefios de Venezuela. Boln Geol. Minist. Minas V enez., Caracas, Publ. Esp. 5 (Mem. 4 Congr. geol. Venez. 2): 585-727, 34 pls. ——., Rohr, D. M., Gray, J., de Faria, A. & Colbath, G. K. 1986. Plectonotus and Plectonotoides, new subgenus of Plectonotus (Bellerophontacea: Gastropoda) and their biogeographic significance. N. Jb. Geol. Palaont. Abh., Stuttgart, 173: 167-180. Caputo, M. V. & Crowell, J. C. 1985. Migration of glacial centers across Gondwana during Paleozoic Era. Bull. geol. Soc. Am., New York, 96: 1020-1036. Cecioni, A. & Frutos, J. E. 1975. Primera noticia sobre el hallazgo de Paleozoico Inferior marino en la Sierra de Almeida, Norte de Chile. Actas I Congr. argent. Paleont. Bioestratigr., Tucuman, I: 191—207. Crowell, J. C., Rocha-Campos, A. C. & Suarez-Soruco, R. 1980. Silurian glaciation in central South America. In M. M. Cresswell & P. Vella (eds), Fifth International Gondwana Symposium: 105-110. Balkema. ——, Suarez-Soruco, R. & Rocha-Campos, A. C. 1981. The Silurian Cancaniri (Zapla) Formation of Bolivia, Argentina and Peru. In M. J. Hambrey & W. B. Harland (eds), Earth’s pre-Pleistocene glacial record: 902-907. Cambridge. Cuerda, A. J. & Antelo, B. 1973. El limite Silurico-—Devonico en los Andes Centrales y Orientales de Bolivia. Actas 5 Congr. geol. argent., Buenos Aires, 3: 183-196. Furque, G. & Cuerda, A. J. 1979. Precordillera de La Rioja, San Juan y Mendoza. 2 Simp. geol. regional Argentina, Cordoba, 1: 455—522. Gray, J., Colbath, G. K., de Faria, A., Boucot, A. J. & Rohr, D. M. 1985. Silurian-age fossils from the Paleozoic Parana Basin, southern Brazil. Geology, Boulder, Colo., 13: 521—S25. 290 A. J. BOUCOT ——, Theron, J. N. & Boucot, A. J. 1986. Age of the Cedarberg Formation, South Africa and early land plant evolution. Geol. Mag., Cambridge, 123: 445-454. Hughes, C. P. 1981. A brief review of the Ordovician faunas of northern South America. Actas II Congr. argent. Paleont. Bioestratigr. (y I Congr. Latinamericano Pal.), Buenos Aires, I: 11—22. Jaeger, H. 1978. Late graptolite faunas and the problem of graptoloid extinction. Acta palaeont. pol., Warsaw, 23: 497-521. Laubacher, G., Gray, J. & Boucot, A. J. 1982. Additions to the Silurian stratigraphy, lithofacies, biogeog- raphy and paleontology of Bolivia and southern Peru. J. Paleont., Tulsa, 56: 1138-1170. Levy, R. & Nullo, F. 1974. La fauna del Ordovicico (Ashgilliano) de Villicun, San Juan, Argentina. Ameghiniana, Buenos Aires, 11 (2): 173-194. Nullo, F. & Levy, R. 1976. Consideraciones faunisticas y estratigraficas del Ashgilliano de Sudamerica. Actas 6 Congr. geol. argent., Buenos Aires, 1: 413—422. The Ordovician—Silurian boundary in Bolivia and Argentina A. Cuerda’, R. B. Rickards’ and C. Cingolani* 1 La Plata Museum, Paseo del Bosque, 1900 La Plata, Argentina ? Sedgwick Museum, Department of Earth Sciences, Downing Street, Cambridge CB2 3EQ 3 Centro de Investigaciones Geologicas, Universidad Nacional de la Plata, Calle 1, no 644, 1900 La Plata, Argentina Synopsis The Ordovician-Silurian boundary level has been identified in few areas, although there is considerable potential for future work. The following sections are the best: 1 Lampaya, Bolivia; 2 the Don Braulio Valley, Argentina; 3 Talacasto, Argentina. Recent fieldwork has established that Talacasto appears the best of these, and a sequence of persculptus Zone, probably acuminatus Zone, and approximate equivalent of the atavus Zone has been established. The base of the Silurian at Talacasto is taken at 60cm above the base of the La Chilca Formation, following a persculptus Zone assemblage. Several stratigraphically important graptolites are recorded from South America for the first time. Introduction In Bolivia undoubted low Silurian rocks are exposed in the Eastern Cordillera, and in Argen- tina in the Precordillera (Fig. 1). The Cancaniri Formation is the basal unit of the Silurian in Bolivia (Castanos & Rodrigo 1978) and consists of 105m of diamictites, shales and sandstones yielding palynomorphs and, in some sections, scarce brachiopods. The Precordilleran Argentin- ian Silurian is recognized as three facies types: the Eastern Facies, some 2500-3000 m of shales, sandstones and conglomerates with associations of brachiopods, corals and graptolites; the Central Facies, 450-500 m of green shales, orthoquartzites, and fine grained limestones, with rich assemblages of brachiopods, corals, trilobites and graptolites; and the Western Facies, restricted to the Calingasta region, approximately 1000m of shales and turbidite sandstones, yielding some brachiopods. Each facies type (Cuerda, in press) is interpreted as having a different palaeoenvironment, respectively: a N—S trough between Pre-Cambrian ridges; proxi- mal to distal platform; distal platform to abyssal plain. The stratigraphically lowest formations in these facies are the La Rinconade Formation, the La Chilca Formation, and the Calingasta Formation. Bolivia The Lampaya section is located near Cochabamba. Three lithological units have been recog- nized in the Silurian, the Cancaniri Formation at the base, and above it the Kirusillas and Catavi Formations, a total of 1355m spanning the Llandovery to Ludlow. The Ashgill Series seems to be absent in Bolivia so that the Cancanfiri Formation rests upon Caradoc or earlier strata. At Lampaya the Cancafiri Formation consists of 105m of diamictites with shales and sandstones intercalated as thin layers. A Llandovery age is supported by palynomorphs refer- able to the Veryhachium rhomboidium Zone (Suarez-Riglos 1975). Macrofossils have been re- covered including trilobites, brachiopods, corals and ostracods by one of us (A.C.). The Cancaniri Formation at Lampaya rests upon the Caradoc. Argentina Villicum Hills Section. The Don Braulio Valley drains the eastern slopes of the Villicum Hills, where the Ashgill black shales and grey sandstones are topped by a ferruginous oolite. The grey Bull. Br. Mus. nat. Hist. (Geol) 43: 291-294 Issued 28 April 1988 292 CUERDA, RICKARDS & CINGOLANI e\ Tucunuco Talacasto -- ' WJ o < c ig =F a (e) a = < - o San Juan f ar lsazic \ & deca ge Tontal s EC Eastern Cordillera Vary - Pate CF Frontal Cordillera R.d. | Sierra del |P Precordillera SP Pampean Ranges Fig. 1 Distribution of Silurian facies in the Precordillera of San Juan, Argentina. The western facies is shown around Calingasta, the central facies in the close stipple and the eastern facies in open stipple to the right. sandstones have yielded the trilobites Calymenella (Eohomalonotus) villicumensis Baldis & Blasco and Dalmanitina sudamericana Baldis & Blasco (Baldis & Blasco 1974) and the brachio- pods Fascifera punctata, Arenorthis cuyana, Villiscundella muozetici, Bagnorthis garrigoui and Kjaerina (Neokjaerina) florentina (all Levy & Nullo 1977). The Silurian commences with argillaceous sandstones and has a palynomorph assemblage referable to the Llandovery, which Volkheimer et al. (1980) list as Ancyrochitina sp., A. cf. ORDOVICIAN-SILURIAN BOUNDARY IN BOLIVIA AND ARGENTINA 293 ancyrea (Eisenack), Conochitina cf. chydaea Jenkins, Desmochitina sp., Cyathochitina cf. cam- panulaeformis Eisenack, Euconochitina cf. filifera Tangourdeau, Rhabdochitina sp. ‘A’, Spathochi- tina cf. clarindoi de Costa and Sphaerochitina sp. Above the argillaceous sandstones the beds grade into medium and coarse sandstones of Wenlock and Ludlow age (Magotes Negros Formation). Baldis & Pothe de Baldis (1988, this volume) have reviewed and revised this section. The Talacasto section (Figs 1, 2) is located some 16km WNW of Talacasto railway station and has been studied by Cuerda et al. (1982). Recent collecting by the authors yielded several hundred graptolites throughout the whole of the 3-65m of the La Chilca Shale Formation. Collecting was done every few centimetres, as closely as the friability of the shale would allow. Several confirmatory collections were made nearby. Glyptograptus persculptus occurs com- monly, both flattened and in three dimensions, in association with equally common specimens of Climacograptus angustus Perner and in addition Pseudoclimacograptus sp. nov., Glyp- tograptus sp. (an undescribed form commonly seen in the persculptus Zone in other parts of the world), Climacograptus cf. medius Tornquist, and Climacograptus normalis Lapworth. This assemblage is taken to indicate the latest Ordovician G. persculptus Zone. At 55cm above the base of the formation G. persculptus s.s. disappears, but the remainder of the fauna continues. Rhaphidograptus sp. at 90 cm, and G. ex gr. persculptus (late forms, smaller, and with a delayed median septum) also occur between 1-1 m and 1-38 m, where Pseudoclima- cograptus sp. nov. is also especially abundant and dominates the fauna. The Pseudoclimaco- graptus sp. nov. is close to P. fidus and P. pictus described from the acuminatus Zone of Kazakhstan by Koren & Mikhailova (1980). From 60 cm to 1:7 m we have recorded specimens Arenig limestones (San Juan Formation), above thrust se CAAA atavus z zone AZ BA La Chilca Shale Formation (3°65m) BB mcrae basal Silurian conglomerate - ao > °o ao) i= o =!) as a no) c =) o = n = Fig. 1 B. A. BALDIS & E. D. POTHE DE BALDIS € ire ° 3 ° = DO o ° ct) wn 4 ° (=) Lower 10m 31°15! Ff. C.G.B. J 2384 56 7 8 Y | Dalmanitina 2 Calymenella 3 Arenorthis 4 Bagnorthis 5 Fascifera 6 Hirnantia 7 Dalmanella 31915! JEL SALADO OLA LAJA AS PALOMITAS Om 10 Il 12 13 14 8 Modiolopsis 9 Nuculopsis 10 Palaeoneilo Il Undet grapt. 12 Cl. hughesi I3Lw Sil Acritarchs l4Lw Sil Chitinozoa The location of the type locality of the Don Braulio Formation (above), and a section through the formation showing the distribution of the fossils mentioned in the text (below). ORDOVICIAN-SILURIAN BOUNDARY IN ARGENTINA 297 was recorded for the first time from the Lower Silurian (formerly only described from the Lower Devonian). Other forms such as Veryhachium tetraedron Deunff, Marrocanium simplex Cramer et al., Multisphaeridium alloiteaui Deunff and M. cf. remotum (Deunff) are typical of Ordovician sediments. The age of the association is based on the presence of Tunisphaeridium tentaculaferum (Martin) and Domasia limaciforme (Stockman & Williére) whose first appearance is in the Lower Llandovery of England and Belgium. The graptolites appearing in all parts of the Upper Section of the Don Braulio Formation were determined by Peralta (in press) as the typical Lower Silurian assemblage of Cli- macograptus aff. C. hughesi (Nicholson), Monograptus sp., Glyptograptus sp. and Rastrites sp. The Don Braulio Formation has 40 m thickness in the type locality. A brief description of the section is as follows (also see Fig. 1): A. Hematitic Member: 9 Pale green-greyish shales (strongly deformed), with spots of iron oxides and scattered ramous CALDION IES oc nda qooe so da re RO ORER CEE ROH oad tEaae reader osese eaten cece con OONConC Ee aeee eee aes 2m 8 Oolitic sandstones weathering dark red; red brownish to green in fresh fracture. Acritarchs GING! GNINCOAOCEIE SeacadoanposasmondddcoososccogsvoucUdmoanenentod loa menESdae Sooo Basen cr acO Cone: 0:50m 7 Pale green-greyish shales, intercalated with thin hard siltstones and fine-grained sandstones of 2 Gin WMS SNES nouboreute eenuusAeueubobon cs doobU OURO ROaUDHIORE OAR OE Hen tanud toad acne eee oares 2m 6 Oolitic sandstone similar to Bed 8 but with fewer oolites, with indeterminable monograptid BUDS. coosvodprlb smactidhindguadaeseads leaaboaases eRe Ee OT CeCe eC nO ONU TS RTOT ROR cer ar eee a aiars aera 1m 5 Green-greyish hard siltstone with some shaley levels and Climacograptus hughesi .............. 3m elematiticisiltstones (poorly bedded) oe ca..cececcswe nes ce cee eh asienearees occ aee sreeu cise esnes seeds 1m B. Silty and Shaly Member: 3 Dark green shales and siltstones (highly deformed) alternating with pale green-greyish clay, with fragments of unidentifiable graptolites (Monograptidae?) ................cec cece eee ee eee eees 10m C. Conglomerate and Sandstone Member (Lower Section): 2 Dark green to green-greyish fine-grained sandstones (poorly bedded) with the trilobites Caly- menella (Eohomalonatus) villicumensis, Dalmanitina sudamericana and the brachiopods Fas- cifera punctata, Hirnantia cf. sagittifera, Dalmanella cf. testudinaria ............0.0.000cceeeeeeeees 12m 1 Basal oligomictic conglomerate with intercalations of green-greyish lenses ...................... 6m Unconformity ~ ~ ~ ~ ~ ~ La Cantera Formation (Llanvirn to Caradoc). From the above we may conclude that the Ashgill is well dated in the whole sequence with trilobites, brachiopods and Hirnantia cf. sagittifera at its top. Ten metres of barren shales follows this section, followed by a Llandovery graptolite fauna and acritarchs and chitinozoans of Lower to Middle Llandovery age. References Baldis, B. A. & Blasco, G. 1975. Primeros trilobites Ashgillianos del Ordovicico sudamericano. Actas I Congr. argent. Paleont. Bioestratigr., Tucuman, 1: 33-48. Benedetto, J. L. 1985. El hallazgo de la tipica fauna de Hirnantia en el Ashgilliano tardio de la Sierra de Villicum. Asoc. Paleont. Argent., Reun. Com. en San Juan, 1: 56-57. San Juan. Levy, R. & Nullo, F. 1974. La fauna del Ordovicico (Ashgilliano) de Villicum, San Juan, Argentina. Ameghiniana, Buenos Aires, 11 (2): 173-194. Peralta, S. 1988. Graptolitos del Llandoveriano Inferior en el Paleozoico Inferior clastico del pie oriental de la Sierra de Villicum. Act. J Jorn. Geol. Precord., San Juan (in press). Pothe de Baldis, E. D. 1980. Lower Silurian acritarchs from Villicum, Province of San Juan, Argentine Precordillera, Argentina. Abstr. 5th Int. Conf. Palynology, Cambridge: 315. Sanchez, T. M. 1985. El Género Mediolopsis en el Ashgilliano de la Sierra de Villicum y la comunidad Hirnantia—M odiolopsis. Asoc. Paleont. Argent., Reun. Com. en San Juan, 1: 58-59. San Juan. Volkheimer, W., Pothe, D. & Baldis, B. 1980. Quitinozoos de la base del Silurico de la Sierra de Villicum (Provincia de San Juan, Republica Argentina). Revta Mus. argent. Cienc. nat. Bernardino Rivadavia, Buenos Aires, (Paleont.) 2 (6): 121-135. Late Ordovician and Early Silurian Acritarchs F. Martin Département de Paléontologie, Institut Royal des Sciences Naturelles de Belgique, Rue Vautier 29, B-1040 Bruxelles, Belgium Synopsis The principal stratigraphical data for late Ordovician and early Silurian acritarchs are reviewed; at present they do not justify any formal zonation on a broad geographic scale. The systemic basal boundary stratotype at Dob’s Linn, southern Scotland, has not yielded index acritarchs. A preliminary selection of taxa from correlative strata on Anticosti Island, Québec, eastern Canada, indicates that the area has the most continuous palynological record from at least the Ashgill to the late Llandovery, with the best potential for establishing detailed acritarch systematics and interregional correlation. Introduction In general, the biostratigraphical tool provided by the acritarchs is still only partly exploited for interregional correlation, for the following reasons: (1) sufficiently detailed systematic descrip- tions have become available only during the last fifteen years or so, through the use of SEM, and a coherent taxonomic framework is still lacking; (ii) precisely defined taxa are most often reported only from regions where their total stratigraphical range is not established; (iii) a large number of data relate to dispersed samples, for which there is no macrofossil age control. In particular, acritarchs of latest Ordovician and earliest Silurian age have received little docu- mentation. This scarcity of data reflects the lack of palynological investigations rather than of suitable marine deposits, for these probably planktonic, organic-walled microfossils appear to be relatively weakly facies-controlled when compared with macrofossils. Nevertheless, the Ashgill extinction that affected numerous other fossil groups also involved the acritarchs. Differences in composition of assemblages between the end of the Ordovician and the begin- ning of the Silurian are indicated in the following areas: Anticosti Island, eastern Canada; southern Appalachians, U.S.A.; Belgium; and the Algerian Sahara. These differences are ampli- fied by the absence of Hirnantian or Gamachian strata, except on Anticosti, where, on the basis of preliminary data (Duffield & Legault 1981, and author’s personal observations), the disap- pearance of numerous Ordovician taxa seems to occur in the Gamachian. A marked change between acritarch associations from the late Ashgill and the Llandovery is mentioned briefly (Le Heérissé 1984) for the subsurface rocks in southern Gotland. Colbath (1986) has reviewed different hypothetical causes for these acritarch extinctions, ranging from the effects of sea-level and climatic changes associated with glaciation to a bolide impact model. Review of data The map (Fig. 1) shows the distribution of late Ordovician and early Silurian acritarchs and indicates detailed references. Numbers (see explanation of Fig. 1) refer generally to the most recent publication that indicates previous data; exceptions are Anticosti and Great Britain, for which further references are given. Anticosti and southern Scotland provided the two final candidate sections for the Ordovician-Silurian boundary stratotype considered by the Subcom- mission on Silurian Stratigraphy (Holland 1984). Since then the International Commission on Stratigraphy (Bassett 1985) has chosen to fix the base of the Llandovery Series, together with that of its lowest stage, the Rhuddanian, at Dob’s Linn, southern Scotland; the boundary stratotypes for the two other Llandovery stages, Aeronian and Telychian, are located in the type area of the Llandovery in Wales (Cocks 1985). Areas from which no index acritarchs are known (for example, the Ashgill of southwest France, Rauscher 1974) are omitted. Owing to the lack of agreement on precise correlation between the North American and British upper Ordovician standard successions (Barnes et al. Bull. Br. Mus. nat. Hist. (Geol) 43: 299-309 Issued 28 April 1988 300 F. MARTIN Fig. 1 Generalized world map showing late Ordovician and early Silurian acritarch localities. The following abbreviations indicate the information included in publications 1—37 listed below: (CA), undifferentiated late Caradoc and Ashgill; A, Ashgill; P, Pusgillian; R, Rawtheyan; H, Hirnantian; G, Gamachian; L, undifferentiated Llandovery, possibly including Rhuddanian; Rh, Rhuddanian; Ae, Aeronian; T, Telychian; p.d., palynological dating only. Chronostratigraphic units groups within parentheses are not differentiated from each other. 1, Hill 1974, Rh-T: 2, Aldridge et al. 1979, Rh-T: 3, Hill & Dorning in Cocks et al. 1984, Rh-T: 4, Downie 1984, Rh-T: 5, Eisenack 1968, A: 6, Eisenack 1963, A: 7, Umnova 1975, A, L: 8, Gorka 1969, A: 9, Konzalova- Mazankova 1969, (PR): 10, Vavrdova 1974, A: 11, Vavrdova 1982, H: 12, Martin 1969, (CA), Rh-T: 13, Martin 1974, (CA), Rh: 14, Elaouad-Debbaj 1981, (CA), A, p.d.: 15, Jardiné et al. 1974, (C ?A), (AeT), p.d. in part: 16, Deunff & Massa 1975, ?C, p.d., 7Rh: 17, Molyneux & Paris 1985, A, p.d.: 18, Hill et al. 1985, (RhAe), p.d.: 19, Bar & Riegel 1980, (AL), p.d.: 20, Brito 1967, L, p.d.: 21, Gray et al. 1985, L, p.d.: 22, Melendi & Volkheimer 1985, L: 23, Colbath in press, (CA), (PR), (RhAe): 24, Loeblich & Tappan 1978, (CA), (PR): 25, Loeblich & McAdam 1971, (CA), (PR): 26, Loeblich 1970, (PR): 27, Colbath 1979, (CA): 28, Johnson 1985, L: 29, Wright & Meyers 1981, (CA), p.d.: 30, Miller & Eames 1982, Rh: 31, Martin 1980, (PR): 32, Legault 1982, (CA), p.d.: 33, Staplin et al. 1965, (PR): 34, Cramer 1970, (AeT): 35, Duffield & Legault 1981, 1982, G, Rh-T: 36, Jacobson & Achab 1985, (PR): 37, Martin in press and personal observation, G (at Anticosti only), Rh-T. [Since submission of this paper, Whelan (1986) has commented briefly on the acritarchs from Dob’s Linn. ] 1981; Ross et al. 1982; Shaver 1985), palynological references for both late Edenian and Maysvillian strata in U.S.A. are included. In the Llandovery Series, acritarch data given for the Rhuddanian sometimes include those for the Aeronian and Telychian. Localities where the sections begin only with the Aeronian or Telychian are omitted here and may be found in Martin (in press). LATE ORDOVICIAN AND EARLY SILURIAN ACRITARCHS 301 Europe In Great Britain, no palynological work has been published on the Ashgill. The Ordovician— Silurian boundary stratotype strata at Dob’s Linn (Cocks 1985) are composed of condensed, deep-water, graptolitic shales, the base of the Llandovery being coincident with the base of the P. acuminatus Zone. The whole succession, from the Climacograptus peltifer Zone (early Caradoc) upwards, contains rare, blackish acritarchs, but these are too poorly preserved to provide useful information. The type Hirnantian (Hirnant Limestone) at Cwm Hirnant quarry, near Bala, North Wales, yielded rare acritarchs belonging to either poorly-defined or remnant Arenig—Llanvirn taxa (personal observation). The Caradoc Series (Costonian to Onnian stages) in the type area of Shropshire contains well preserved assemblages (Turner 1984) of Caradoc age, associated with others derived from Tremadoc and Arenig—Llanvirn deposits. Rhuddanian microfloras from near Llandovery are both poorly preserved and of low diversity but permit (Hill & Dorning in Cocks et al. 1984) the recognition of three biozones characterized, on the basis of published lists, by the successive appearance of taxa that, for the most part, are long-ranging in the Silurian or are left in open nomenclature. The top of the Rhuddanian there also contains reworked, pre-Caradoc Ordovician material (Martin in press). In the same region, and especially in the the Welsh Borderland (Hill 1974), partly published results for the Llandovery show, from the Aeronian onwards, a refined palynological zonation that may be compared with that outlined for Belgium (Martin 1969). Of particular significance are species of Domasia Downie, 1960 emend. Hill, 1974 and Dilatisphaera williereae (Martin) Lister, 1970. In the Massif of Brabant, Belgium (Martin 1974), moderately well preserved acritarchs, mostly long ranging and including some known from the Tremadoc to the Arenig—Llanvirn, are from boreholes. Parts of these rock successions are assigned a late Caradoc and/or Ashgill age on lithological and structural grounds in the absence of diagnostic macrofossils; those of the basal Rhuddanian are dated by graptolites and include strata of the P. acuminatus Zone. In the Baltic region (Gotland, Estonia, Latvia—Eisenack 1963, 1968; Umnova 1975), Poland (Gorka 1969) and Czechoslovakia (Konzalova-Mazankova 1969; Vavrdova 1974, 1982), as in Portugal (Elaouad-Debbaj 1981), data are relatively few for the Ashgill and absent for the Rhuddanian. The only Hirnantian acritarchs so far illustrated come from the Prague region (Vavrdova 1982). Africa and South America Microfloras from boreholes in north Africa are well preserved. At the Grand Erg Occidental in the Algerian Sahara (Jardiné et al. 1974) acritarch zone F corresponds to the Caradoc and perhaps Ashgill; it also contains taxa characteristic of the Arenig—Llanvirn and is present too in deposits of the Illizi Basin attributed doubtfully to the M. sedgwickii Zone of the Aeronian. In Libya (Deunff & Massa 1975; Molyneux & Paris 1985; Hill et al. 1985) acritarchs from the late Ordovician and early Silurian, cited and partially figured, are dated with particular reference to palynological data from western Europe and central U.S.A. In Deunff & Massa (1975) the list of taxa alleged to have been found in the early Rhuddanian C. vesiculosus Zone indicates a post-Llandovery age and is not considered further here. Acritarch data for the relevant interval in Ghana (Bar & Riegel 1980), Brazil (Brito 1967; Gray et al. 1985) and Argentina (Melendi & Volkheimer 1985) are dispersed and mainly without independent age control. The most noteworthy illustrated observation is that samples from Ghana said to occur at the Ordovician/Silurian boundary share only a single species, Dactylofusa marahensis Brito & Santos, 1965, with strata of the Maranhao Basin attributed to the Lower Silurian. In both cases the age is based on structural and palynological arguments. North America Publications referring to the eastern and central U.S.A. deal mainly with numerous new late Ordovician taxa from Oklahoma (Loeblich & McAdam 1971; Loeblich & Tappan 1978) and the Cincinnati area (Loeblich 1970; Loeblich & McAdam 1971; Loeblich & Tappan 1978; Colbath 1979); however, the acritarchs from the Richmondian Stage, which is correlated with part of the Ashgill Series, are from isolated samples. In the southern Appalachians (southwest 302 F. MARTIN Virginia, northwest Georgia and east Tennessee), a consistent acritarch correlation, based largely on new taxa, is documented (Colbath, in press) for the passage from Ordovician to Silurian; but the presence of the Gamachian and earliest Rhuddanian in the region is debat- able. An acritarch assemblage of undoubted Rhuddanian age in western New York State (Miller & Eames 1982) enables preliminary correlations to be made with assemblages in the southern Appalachians, Anglo-Welsh area and Belgium. A very few Llandovery, including perhaps Rhuddanian, acritarchs are known from central Pennsylvania (Johnson 1985). In eastern Canada, except for palynologically dated latest Caradoc or Ashgill strata in a borehole in the Labrador Sea (Legault 1982), data relate to the Province of Québec. Only reconnaissance studies are available for the pre-Hirnantian Ashgill of the Perce area (Martin 1980) in the Gaspé Peninsula. The White Head Formation at White Head (Lespérance 1985; Fig. 2 herein) has not yielded index acritarchs in the Hirnantian interval, and the basal Llando- very portion (base of Unit 6; personal observation) contains specimens deformed by crystal growth; some of the latter, for example Eupoikilofusa aff. E. ampulliformis, sensu Duffield & Legault 1981, are very characteristic of the Rhuddanian at Anticosti, from the base upwards of the Becscie Formation at Ellis Bay. At Anticosti an Ordovician/Silurian boundary stratotype was proposed (Barnes & McCracken 1981) in an allegedly continuous limestone-shale succession in the upper part of the Ellis Bay Formation (sensu Petryk 1981) at Ellis Bay. The base of the Silurian is marked by the appearance of the conodont Ozarkodina oldhamensis (Rexroad, 1967); Oulodus? nathani McCra- cken & Barnes, 1981 is an auxiliary indicator for the boundary. However, Lespérance (1985) places the boundary higher and in the Becscie Formation, on the assumption that the appear- ance of the trilobite Acernaspis coincides with the base of the P. acuminatus Zone. The shallow marine platform deposits there are very rich in microfloras and in micro- and macrofaunas, except graptolites (see Lespérance 1981 for numerous contributions and earlier references). On the whole, the Ashgill and Llandovery acritarchs of Anticosti are very well preserved and relatively abundant, but have been described only partially (Staplin et al. 1965; Cramer 1970; Duffield & Legault 1981, 1982), apart from strata dated as D. complanatus Zone, assigned to the early or middle Ashgill (Jacobson & Achab 1985). The Anticosti acritarchs The quality of the palynological material at Anticosti and its age control based on shelly macrofaunas and conodonts justify a preliminary synthesis. The ranges of some taxa there are compared (Fig. 2) with those from other regions. The compilation is based on the references given in the general distribution of data (Fig. 1) and for the post-Aeronian of the same regions, following those assembled by Martin (in press; explanation of Fig. 1). This restricted choice of taxa is conditioned by personal examination of twelve samples (see Appendix) from the upper part of Member 4 of the Ellis Bay Formation, of Gamachian age, to the upper part of the Jupiter Formation, correlated with the Telychian (C;) (Lespérance 1981). The choice could have been different, but in the present state of knowledge the comments would probably have been comparable with those below. The observations of Duffield & Legault (1981) are confirmed with regard to the change in composition of acritarch assemblages just above the base of the Silurian as defined on the basis of the appearance of diagnostic conodonts (Barnes & McCracken 1981) within Member 7 of the Ellis Bay Formation. If the correlation proposed by Lespérance (1985) is accepted, the major change in terms of appearance of new acritarch taxa occurs within the late Gamachian, rather than in the early Llandovery. At its type locality, on the west side of Ellis Bay, the entire member, | to 4m thick, is very poor in acritarchs. In particular, the locally developed bio- hermal bed, 1-5 to 2m thick, above the systemic boundary is sterile. Immediately above this bed, from the base of the Becscie Formation (sensu Petryk 1981; sample A2B7) onwards, the majority of taxa known from other regions and of Ordovician affinities are absent. Aremorica- nium squarrosum Loeblich & McAdam, 1971 (see synonymy in Jacobson & Achab, 1985: 171) is recognized in the early Richmondian, which is equated with latest Pusgillian to early Raw- 303 LATE ORDOVICIAN AND EARLY SILURIAN ACRITARCHS “SUOISAI JOYIO Ul SYOILILIO’ NSsOONUY poqoajas Jo sosuey 7 “sIYy “UO1}Ng!l4}3e a9Be snoiqnp :;joqwAs 91yde161}e1}SOU0IYDS € IIIOHSV ]@ aIDYMIaS|a a AYSAOQGNV11 jayye & ‘UOXe} JO p1od|as SNOIqnp :;oqwAs dIiydeib1}e43s uoxe} 4O 89U9a11ND9I0 OQUuOIYS J1IOjaq ~ ‘pajyelyUdsISjjipuNn sie sasayuyuoied ul syiun 1ysOol}uy UO = u a 2 = = - 4 m = > d1ydesBiyesySouosyo ‘Buiyep jeoibojoud\jed :*p'd ‘ueiuoaaq :q uoxe}, JO 89uUa1INIIO zi m S 2 z C o oO o > @) ‘MO|PN] :nqz ‘YOO]}UBM :M ‘UeIYSAJaYL :4 ‘UelUOIDY say puepueys “SONI E 2 z > z 2 ie = = > z = | < < Us > — ‘ 5 G By © ty : psepuejys ysiyig ~ z > > > > ueiueppnyy :yy ‘uel}UeUdIH :H ‘UeASY MEY :y ‘URIIIIBSNd id — = z = = = eys uediow 410 ‘“WiBusy :v ‘290pesed :D ‘90pe1ed-a9id ‘ueldIAOp1O :Od ea : EGRECIL U <== =a T 1-ev | nqe-1| | | | im AVAYFITIIM VYAVHdSILV IG pd(ntm) | M-2y | M-2V ML |M(Lewa(Lew) 1 ‘dS VISVWOG | | (Luw)ua]/SINHYOSITINdNY “a 4se VSNIOTINIOdNA uu(yo)) Mae Mev (444) "dS VIIWdOLOTAL | L ‘dS WNIGINAVHdSIDITdIL IN | | | | ‘dS VHYOHdOITVdOHY 32 AON dS ® NAD | | | “AON'dS WNIHALSO1TD03Hd | v-od AVNIdSILVOINd WNIGINAaVHdSILIVE =| af = a= oa T | | | | ae ‘pidy | | | wv | (ud) WNLdINDSNI WNIGINAVHdSOHLYO | (wo) | | v-od| v AYVINDNVLOAY WNIGINSaVHdSOHLYO | “p-dy | ) oe v(vo)| (ud) WNSOYYVNOS WNINVOINOWSYY m-od | | (144)| WASONIdSODITO WNIGIY4aVHdSOINOD uu M M-99| M | m(tev)/1(L4u)|,SISNSAGSIA H-VLVLIDIC VILNITNDOH, a-(12¥) a-9 |nq-yy| H-9 mM | M-4u 1-9 VLOW]Y SISVHdO11VxaIG (@) z m N wo | > 4 © Ee) 5 m| m alls Gel) JC o m | im z 2 423 D BS Cy Zils Oe alee lao al Sz > Seas > of |@s = 2 Se) ce) | eo a ae ¥ (VQVNV9 ‘D384ND) ILSODILNV 2 2 o bz |Sz2 ra) o >> m = | S 28 Zo = Cc Be c> cr > ’ = is a ra) r= ace || Sir eaten ae WOUS SHOYVLINOVY d319373S Sy |5 > Zz |5o D > = no = Zo r 2 z u Fas fe) uF ge > pars = > x = ‘vS'n vav (q4ed ul) > “NVO NVILVNNIONIO 304 F. MARTIN theyan by Barnes et al. (1981). The disappearance of Orthosphaeridium rectangulare (Eisenack) Eisenack, 1968 (Figs 4a, b; see synonymy in Elaouad-Debbaj 1981: 48) and of O. insculptum Loeblich, 1970 (Figs 3a, b) occurs within an unobserved interval in the Gamachian, between the upper parts of Member 5 (about 5 m below its top; sample A2B3) and Member 6 (0-3 m below its top; sample A2B4) of the Ellis Bay Formation. Baltisphaeridium plicatispinae Gorka, 1969 (Fig. 9) extends, according to Duffield & Legault (1981), into Member 7, below the biohermal bed. The appearance of taxa of Silurian affinities, which occurs mainly and progressively from the base of the Becscie Formation onwards, begins in the Gamachian, no later than the upper part of Member 5 (sample A2B3), source of the present example of Multiplicisphaeridium sp. 1, sensu Duffield & Legault 1981 (Fig. 16). The latter recalls the “M. forquiferum—M. forquillum’ group found by Cramer & Diez (1972) in the late Llandovery of Kentucky. Eupoikilofusa aff. E. ampulliformis (Figs 14a, b), which appears at the base of the Becscie Formation (sample A2B7), earliest Llandovery, is close to a Llandovery species known from the early Rhuddanian in Belgium (Martin 1974). The entry of Domasia Downie, 1960, emend. Hill 1974 (Fig. 6) and T ylotopalla Loeblich, 1970 (Fig. 10) on the one hand, and of Dilatisphaera williereae (Martin) Lister 1970 (Fig. 5) on the other, occurs in the Jupiter Formation at levels that are correlated (Barnes & McCracken 1981) respectively with the late Aeronian (C,—C,; sample A6A, about 3m above base of Member 3) and with the Telychian (C;; sample A7A1, 4 m below top of the Jupiter Formation). As yet no diacrodian has been identified from the upper part of the Gamachian, and no form suspected of being reworked from the Ordovician has been found in the Llandovery of Anticosti. The richness and variety of the microfloras in the Gamachian and Llandovery at Anticosti will lead inevitably to the introduction of new taxa, some of which will be index fossils. As an example, two forms from the Ellis Bay Formation (sample A2B3) are illustrated for the first time here and left in open nomenclature: Pheoclosterium sp. nov. (Figs 7a, b) and Gen. et sp. nov. cf. Rhopaliophora (Fig. 8). The only species formally assigned to the former genus, Pheo- closterium fuscinulaegerum Tappan & Loeblich, 1971, is characteristic of the late Ordovician. Its range (see Jacobson & Achab 1985 for all references) is from the Edenian of Indiana (Kope Formation; Tappan & Loeblich 1971; Colbath 1979) and from the Onnian, highest Caradoc, in Shropshire, England (upper part of Onny Shales; Turner 1984) to the Hirnantian in Czechoslo- vakia (Kosov Formation, Vavrdova 1982). The second acritarch, cf. Rhopaliophora, differs from that exclusively Ordovician genus in its opening and resembles ‘Hystrichosphaeridium’ wimani Figs 3-16 Acritarchs from Anticosti. All figured specimens are in the type fossil collection of the Geological Survey of Canada, Ottawa, and have numbers with the prefix GSC. Figs 3, 4, 7-9, 12, 15, 16: sample A2B3, Ellis Bay; Ellis Bay Formation, upper part of Member 5, Gamachian. Figs 11, 13, 14: sample A2B7, Ellis Bay; lowermost Becscie Formation, Llandovery, correlated with Rhuddanian, A, ,. Figs 5, 6, 10: sample A7A1, 4km southeast of Pointe Sud- Ouest; upper part of Jupiter Formation, Llandovery, correlated with Telychian, C;. Age assign- ments according to Lespérance (1981). Fig. 3 Orthosphaeridium insculptum Loeblich 1970. GSC 82877. Fig. 3a, x 400; Fig. 3b, enlargement, x 3000, of base of left process. Fig. 4 Orthosphaeridium rectangulare (Eisenack) Eisenack 1968. GSC 82878. Fig. 4a, enlargement, x 2000, of base of left lower process. Fig. 4b, x 200. Fig. 5 Dilatisphaera williereae (Martin) Lister 1970. GSC 82879, x 1000. Fig. 6 Domasia limaciformis (Stockmans & Williere) Cramer 1970. GSC 82880, x 500. Fig. 7 Pheoclosterium sp. nov. GSC 82881. Fig. 7a, enlargement, x 3000, of upper median processes. Fig. 7b, x 750. Fig. 8 Gen. et sp. nov. cf. Rhopaliophora sp. GSC 82882, x 300. Fig. 9 Baltisphaeridium plicatispinae Gorka 1969. GSC 82883, x 300. Fig. 10 Tylotopalla sp. GSC 82884, x 750. Figs 11, 12 ‘Hogklintia digitata—H. visbyensis.. Fig. 11, GSC 82885, x 250. Fig. 12, GSC 82886, x 100. Fig. 13 Goniosphaeridium oligospinosum (Eisenack) Eisenack 1969. GSC 82887, x 250. Fig. 14 Eupoikilofusa aff. E. ampulli- formis, sensu Duffield & Legault, 1981. GSC 82888. Fig. 14a, x 1000; Fig. 14b, enlargement, x 5000, of lower right part of vesicle. Fig. 15 Diexallophasis remota (Deunff) Playford 1977. GSC 82889, x 500. Fig. 16 Multiplicisphaeridium sp. I, sensu Duffield & Legault 1981. GSC 82890, x 500. LATE ORDOVICIAN AND EARLY SILURIAN ACRITARCHS 305 306 F. MARTIN Eisenack, 1968, determined by its author from the latest Ashgill of Gotland (Bornholmer Stufe F2 from an erratic boulder at Oil Myr). On Anticosti, in both the Ashgill and the Llandovery, there are geographically widespread forms with long stratigraphical ranges that are difficult to define because of their wide, contin- uous morphological variability within a single sample; examples are Diexallophasis remota (Deunff) Playford 1977 (Fig. 15) and the “Hogklintia digitata—H. visbyensis’ complex (Figs 11, 12). The recurrent abundance in certain Ordovician and Silurian strata, notably on Anticosti and in the Baltic region, of the latter complex and of, for instance, Goniosphaeridium oligospino- sum (Eisenack) Eisenack 1969 (Fig. 13) probably results from particular palaeoenvironmental conditions; the latter led Cramer & Diez (see 1974 for earlier references) to postulate a certain degree of provincialism linked to palaeolatitudes for Silurian acritarchs. Acritarch data for the latest Ordovician and earliest Silurian are as yet too disparate to permit reliable palaeogeographic reconstructions. Data from Anticosti indicate affinities and possibilities for correlation as follows. The Gamachian microfloras contain taxa known from the late Ordovician of central U.S.A. and/or the pre-Hirnantian Ashgill of Gaspé, and from the Ordovician of Europe (Baltic region and Portugal) and North Africa (Libya). In particular, the evolutionary scheme proposed by Loeblich & Tappan (1971) for the genus Orthosphaeridium Eisenack 1968, notably in part of the Cincinnatian of central U.S.A. and in the late Ashgill of the Baltic region, Gotland and Estonia, may be applied to the Gamachian of Anticosti and the late Ordovician of Portugal. The possibilities for correlation offered by the Llandovery acri- tarchs of Anticosti concern affinities with, principally and in decreasing order, the Gaspé area of Canada, England and Wales, Belgium and the U.S.A. In particular, the first occurrences of Domasia and of Dilatisphaera williereae, the levels of which are still inadequately known on Anticosti, should permit correlation with at least the Aeronian and the Telychian of the Anglo-Welsh area. Palynological data for the Rhuddanian of the latter area allow only a local zonation at present. Conclusions Owing to the dearth of published data, acritarchs have not been used directly as one of the criteria for the choice of an Ordovician-Silurian boundary during the activities of the I.U.GS. working group from 1974 to 1985. The Anticosti deposits are those likely to provide the most reliable palynological correlations, not only in the immediate vicinity of the systemic boundary but also at least from the early to middle Ashgill to the late Llandovery (Telychian, C;). This view is supported by the indication both of relatively continuous data and of direct correlations with the Gaspé area from the base of the Rhuddanian upwards, and the Anglo-Welsh area from the Aeronian upwards. Acknowledgements I am indebted to M. G. Bassett (National Museum of Wales, Cardiff), G. K. Colbath (Smithsonian Institution, Washington, D.C.) and J. A. Legault (University of Waterloo, Ontario, Canada) for critically reviewing the manuscript. Appendix Locality data for Anticosti Island, Province of Québec, Canada. All locality numbers in Lespeérance (1981: 1). Loc. A-2A: Pointe Laframboise area. Sample A-2A1: Ellis Bay Formation, Member 7, 0-40 m above oncolithic bed. Sample A-2A2: Becscie Formation, 0:60 m above base. Loc. A-2B: west side of Ellis Bay, section proposed as Ordovician—Silurian boundary stratotype by Barnes & McCracken (1981). Samples A-2B2 to A-2B6: Ellis Bay Formation; A-2B2: member 4, 3m below top of member; A-2B3: member 5, 5 m below top of member; A-2B4: member 6, 0:30 m below top of member; A-2B5 and A-2B6: member 7, respectively just above and 0:75 m above oncolithic bed. Samples A-2B7 to A-2B9: Becscie Formation. A-2B7: immediately above the biohermal level of LATE ORDOVICIAN AND EARLY SILURIAN ACRITARCHS 307 member 7 of the Ellis Bay Formation. A-2B8 and A-2B9: respectively 1-30 m and 25 m (approximately) above A-2B7. Loc. A-6A and sample A-6A: Cap Jupiter, north of mouth of Riviere Jupiter, Jupiter Formation, about 3m above base of member 3. Loc. A-7A: 4km southeast of Pointe du Sud-Ouest. Sample A-7A1: Jupiter Formation, 4m below its top. References Aldridge, R. J., Dorning, K. T., Hill, P. J., Richardson, J. B. & Siveter, D. J. 1979. Microfossil distribution in the Silurian of Britain and Ireland. Spec. Publs geol. Soc. Lond. 8: 433-438. Bar, P. & Riegel, W. 1980. Mickrofloren des hochsten Ordovizium bis tiefen Silurs aus der Unteren Sekondi-Serie von Ghana (Westafrika) und ihre Beziehung zu den Itaim-Schichten des Maranhao- Beckens in NE-Brasilien. N. Jb. Geol. Palaont. Abh., Stuttgart, 160: 42—60. Barnes, C. R. & McCracken, A. D. 1981. Early Silurian chronostratigraphy and a proposed Ordovician— Silurian boundary stratotype, Anticosti Island, Québec. In P. J. Lespérance (ed.), Field Meeting, Anticosti—Gaspe, Québec, 1981 2 (Stratigraphy and paleontology): 71-79. Montréal (I.U.G.S. Subcom- mission on Silurian Stratigraphy Ordovician—Silurian Boundary Working Group). , Norford, B. S. & Skevington, D. 1981. The Ordovician System in Canada, correlation chart and explanatory notes. Int. Un. geol. Sci., Ottawa, 8: 1—27. Bassett, M. G. 1985. Towards a ‘Common Language’ in Stratigraphy. Episodes, Ottawa, 8: 87-92. Brito, I. M. 1967. Silurian and Devonian Acritarchs from the Maranhao Basin, Brazil. Micropaleontology, New York, 13: 473-482. —— & Santos, A. S. 1965. Contribuigao ao conhecimento dos microfosseis Silurianos e Devonianos de Bacia do Maranhao, I. Notas prelim. Estud. Div. geol. min. Bras., Rio de Janeiro, 129: 1—29, 2 pls. Cocks, L. R. M. 1985. The Ordovician-Silurian Boundary. Episodes, Ottawa, 8: 98-100. —, Woodcock, N. H., Rickards, R. B., Temple, J. T. & Lane, P. D. 1984. The Llandovery Series of the type area. Bull. Br. Mus. nat. Hist., London, (Geol.) 38 (3): 131-182. Colbath, G. K. 1979. Organic-walled microphytoplankton from the Eden Shale (Upper Ordovician), Indiana, U.S.A. Palaeontographica, Stuttgart, (B) 171: 1-38. — 1986. Abrupt terminal Ordovician extinction in phytoplankton associations, southern Appalachians. Geology, Boulder, Colo., 14: 943-946. —— (in press). Systematics and biostratigraphy of organic-walled phytoplankton from the Upper Ordovician—Lower Silurian of the southern Appalachians, U.S.A. Contr. Am. Ass. Stratigr. Palynol., Dallas. Cramer, F. H. 1970. Distribution of selected Silurian acritarchs. Revta esp. Micropaleont., Madrid, num. extr. 203 pp., 23 pls. —— & Diez, M. de C. R. 1972. North American Silurian palynofacies and their spatial arrangement: acritarchs. Palaeontographica, Stuttgart, (B) 138: 107-180. 1974. Silurian acritarchs: distributions and trends. Rev. Palaeobot. Palynol., Amsterdam, 18: 137-154. Deunff, J. & Massa, D. 1975. Palynologie et stratigraphie du Cambro—Ordovicien (Libye nord- occidentale). C.r. hebd. Seanc. Acad. Sci., Paris, (D) 281: 21-24. Downie, C. 1984. Acritarchs in British stratigraphy. Spec. Rep. geol. Soc. Lond. 17. 26 pp. Duffield, S. L. & Legault, J. A. 1981. Acritarch biostratigraphy of Upper Ordovician—Lower Silurian rocks, Anticosti Island, Québec: Preliminary results. In P. J. Lespérance (ed.), Field Meeting, Anticosti— Gaspé, Québec, 1981 2 (Stratigraphy and paleontology): 91-99. Montréal (I.U.G.S. Subcommission on Silurian Stratigraphy Ordovician—Silurian Boundary Working Group). 1982. Gradational morphological series in early Silurian acritarchs from Anticosti Island, Québec. In B. Mamet & M. J. Copeland (comp. & ed.), Proc. 3rd N. Amer. Paleont. Conv., Montreal 1: 137-141, pl. 1. Eisenack, A. 1963. Mitteilungen zur Biologie der Hystrichospharen und iiber neue Arten. N. Jb. Geol. Palaont. Abh., Stuttgart, 118: 207-216. — 1968. Mikrofossilien eines Geschiebes der Borkholmer Stufe, baltisches Ordovizium, F2. Mitt. geol. StInst. Hamb. 37: 81-94. —— 1969. Zur Systematik einiger palaozoischer Hystrichospharen (Acritarcha) des baltischen Gebietes. N. Jb. Geol. Palaont. Abh., Stuttgart, 133: 245-266. Elaouad-Debbaj, Z. 1981. Acritarches de l’Ordovicien Supérieur du Synclinal de Bugaco (Portugal). Systématique—Biostratigraphie—-Intérét paléogéographique. Bull. Soc. géol. miner. Bretagne, Rennes, (C) 10 (2): 1-101. 308 F. MARTIN Gorka, H. 1969. Micro6rganismes de l’Ordovicien de Pologne. Palaeontol. pol., Warsaw, 22. 102 pp., 31 pls. Gray, J., Colbath, G. K., de Faria, A., Boucot, A. J. & Rohr, D. M. 1985. Silurian-age fossils from the Paleozoic Parana Basin, southern Brazil. Geology, Boulder, Colo., 13: 521—S2S. Hill, P. J. 1974. Stratigraphic palynology of acritarchs from the type area of the Llandovery and the Welsh Borderland. Rev. Palaeobot. Palynol., Amsterdam, 18: 11—28. ——, Paris, F. & Richardson, J. B. 1985. In B. G. Thusu & B. Owens (eds), Silurian Palynomorphs. Palynostratigraphy of North-East Libya. J. Micropalaeont., London, 4: 27-48. Holland, C. H. 1984. Steps to a standard Silurian. Proc. 27th Int. geol. Congr., Moscow, 1 (Stratigraphy): 127-156. Jacobson, S. R. & Achab, A. 1985. Acritarch biostratigraphy of the Dicellograptus complanatus graptolite zone from the Vauréal Formation (Ashgillian), Anticosti Island, Québec, Canada. Palynology, Dallas, 9: 165-198. Jardine, S., Combaz, A., Magloire, L., Peniguel, G. & Vachey, G. 1974. Distribution stratigraphique des Acritarches dans le Paleozoique du Sahara algérien. Rev. Palaeobot. Palynol., Amsterdam, 18: 99-130. Johnson, N. G. 1985. Early Silurian palynomorphs from the Tuscarora Formation in central Pennsylvania and their paleobotanical and geological significance. Rev. Palaeobot. Palynol., Amsterdam, 45: 307-360. Konzaloya-Manzankova, M. 1969. Acritarcha Evitt 1963 aus dem Under-Ashgill Bohmens. Palaeontogra- phica, Stuttgart, (B) 165: 81-92. Legault, J. A. 1982. First report of Ordovician (Caradoc—Ashgill) palynomorphs from Orphan Knoll, Labrador Sea. Can. J. Earth Sci., Ottawa, 19: 1851-1856. Lesperance, P. J. (ed.) 1981. Field Meeting, Anticosti—Gaspe, Quebec, 1981 1 (Guidebook). 56 pp. 2 (Stratigraphy and paleontology). 321 pp. Montréal (I.U.G.S. Subcommission on Silurian Stratigraphy Ordovician-Silurian Boundary Working Group). —— 1985. Faunal distributions across the Ordovician-Silurian boundary, Anticosti Island and Percé, Québec, Canada. Can. J. Earth Sci., Ottawa, 22: 838-849. Le Herisse, A. 1984. Upper Ordovician—lower Silurian organic walled microphytoplankton from the Nar Borehole, Gotland, Sweden. Jn J. Utting (ed.), Abst. 6th Intern. Palynol. Conf., Calgary 1984: 87. Lister, T. R. 1970. The acritarchs and the chitinozoa from the Wenlock and the Ludlow Series of the Ludlow and Millichope areas, Shropshire. Part 1. Palaeontogr. Soc. (Monogr.), London. 100 pp., 13 pls. Loeblich, A. R. 1970. Morphology, ultrastructure and distribution of Paleozoic Acritarchs. Proc. N. Amer. Paleont. Conv., Chicago 1969, (G): 705-788. & McAdam, R. B. 1971. North American species of the Ordovician Acritarch genus Aremoricanium. Palaeontographica, Stuttgart, (B) 135: 41—47. —— & Tappan, H. 1971. Two new Orthosphaeridium (Acritarcha) from the Middle and Upper Ordovician. Trans. Am. microsc. Soc., Lawrence, 90: 182-188. 1978. Some Middle and Late Ordovician Microphytoplankton from Central North America. J. Paleont., Tulsa, 52: 1233-1287. Martin, F. 1969. Les Acritarches de l’Ordovicien et du Silurien belges. Détermination et valeur strati- graphique. Mem. Inst. r. Sci. nat. Belg., Brussels, 160 (for 1968). 175 pp., 8 pls. — 1974. Ordovicien supérieur et Silurien inferieur 4 Deerlijk (Belgique). Palynofacies et microfacies. Mem. Inst. r. Sci. nat. Belg., Brussels, 174 (for 1973). 71 pp., 8 pls. —— 1980. Quelques Chitinozoaires et Acritarches ordoviciens supérieurs de la Formation de White Head en Gaspésie, Québec. Can. J. Earth. Sci., Ottawa, 17: 106-119. (in press). Silurian acritarchs. In M. G. Bassett & C. H. Holland (eds), A global Standard for the Silurian System. Cardiff. McCracken, A. D. & Barnes, C. R. 1981. Conodont biostratigraphy and paleoecology of the Ellis Bay Formation, Anticosti Island, Québec, with special reference to Late Ordovician—Early Silurian chrono- stratigraphy and the systemic boundary. Bull. geol. Surv. Can., Ottawa, 329 (2): 51-134, 7 pls. Melendi, D. L. & Volkheimer, W. 1985. Datos palinologicos del limite Ordovicico—Silurico de Talacasto. Rev. Tecn. Yacimientos Petrolif. Fisc. Bolivianos, Santa Cruz, 9 (for 1983): 157-163. Miller, M. A. & Eames, L. E. 1982. Palynomorphs from the Silurian Medina Group (Lower Llandovery) of the Niagara Gorge, Lewiston, New York, U.S.A. Palynology, Dallas, 6: 221-254. Molyneux, S. G. & Paris, F. 1985. Late Ordovician Palynomorphs. In B. G. Thusu & B. Owens (eds), Palynostratigraphy of North-East Libya. J. Micropalaeont., London, 4: 11-26. Petryk, A. A. 1981. Stratigraphy, Sedimentology, and Paleogeography of the Upper Ordovician—Lower Silurian of Anticosti Island, Québec. In P. J. Lespérance (ed.), Field Meeting, Anticosti—Gaspe, Quebec, 198] 2 (Stratigraphy and paleontology): 11-39. Montréal (I.U.G.S. Subcommission on Silurian Strati- graphy Ordovician-Silurian Boundary Working Group). LATE ORDOVICIAN AND EARLY SILURIAN ACRITARCHS 309 Playford, G. 1977. Lower to Middle Devonian acritarchs of the Moose River Basin, Ontario, Bull. geol. Surv. Can., Ottawa, 279. 87 pp., 20 pls. Rauscher, R. 1974. Recherches micropaléontologiques et stratigraphiques dans l’Ordovicien et le Silurien en France. Mem. Sci. géol., Strasbourg, 38 (for 1973). 124 pp., 12 pls. Rexroad, C. B. 1967. Stratigraphy and conodont paleontology of the Brassfield (Silurian) in the Cincinnati Arch area. Bull. Indiana geol. Surv., Bloomington, 36: 1—64, 4 pls. Ross, R. J. & 28 co-authors 1982. The Ordovician System in the United States. Correlation chart and explanatory notes. Int. Un. geol. Sci., 12. 73 pp. Shaver, R. H. (coord.) 1985. Midwestern Basin and Arches Region. Correlation of Stratigraphic Units of North America (COSUN A) Project. Tulsa, Am. Assoc. Petrol. Geol. (unpaginated). Staplin, F. L., Jansonius, J. & Pocock, S. A. J. 1965. Evaluation of some Acritarchous Hystrichosphere genera. N. Jb. Geol. Palaont. Abh., Stuttgart, 123: 167-201. Tappan, H. & Loeblich, A. R. 1971. Surface sculpture of the wall in Lower Paleozoic acritarchs. Micro- paleontology, New York, 17: 385-410. Turner, R. E. 1984. Acritarchs from the type area of the Ordovician Caradoc Series, Shropshire, England. Palaeontographica, Stuttgart, (B) 190: 87-157. Umnovya, H. N. 1975. The acritarchs of the Ordovician and Silurian from the Moscovian syneclise and from the Prebaltic. 166 pp., 20 pls. Moscow. [In Russian ]. Vayrdova, M. 1974. Geographical differentiation of Ordovician acritarch assemblages in Europe. Rev. Palaeobot. Palynol., Amsterdam, 18: 171-175. 1982. Recycled acritarchs in the uppermost Ordovician of Bohemia. Cas. Miner. Geol., Prague, 27: 337-345. Whelan, G. M. 1986. Acritarch and Chitinozoan distribution across the type Ordovician-Silurian bound- ary at Dobb’s Linn, Scotland. Abstr. Palaeont. Ass. a. Conf., Leicester, 1986: 23-24. Wright, R. P. & Meyers, W. C. 1981. Organic-walled Microplankton in the Subsurface Ordovician of Northeastern Kansas. Kansas geol. Surv. Subsurf. Geol., Tulsa, Ser. 4. 53 pp., 8 pls. %4 rs) Se Ae ee oe i Y= > ‘a hirepey ows se | - ot re ples eas m 02 eo. 717 we ri os ary =e. v ¥ a - . a bs aa = ky wet, oa ae wey. had Pes ) die = ~ le 544.5 ee oe aie: Sse ish atpaa’s eed aed oats i Ba at 4 a i) rica alias sehr 1< 't : a Hc my & agence he -% _ Su Te > bei Brachiopods across the Ordovician—Silurian boundary L. R. M. Cocks Department of Palaeontology, British Museum (Natural History), Cromwell Road, London SW7 5BD Synopsis Most of the late Ordovician brachiopod superfamilies also extend into the early Silurian, although the Gonambonitacea become extinct at or near the Ordovician—Silurian boundary and the earliest Cyrtiacea are found very close above it. Faunas close to the boundary are reviewed and listed, and the Hirnantian faunas of the latest Ordovician are found to be richer than the earliest Silurian Rhuddanian faunas in both abundance and diversity. Introduction At the time the Treatise on Invertebrate Paleontology brachiopod volume (Williams et al. 1965) was written, 44 brachiopod genera were recorded with ranges spanning the Ordovician— Silurian boundary, and in addition there were various families and subfamilies whose ranges spanned the boundary even if the recorded ranges of individual genera within them did not. The superfamilies involved are the Lingulacea, Trimerellacea, Discinacea, Craniacea, Orthacea, Enteletacea, Tripleciacea, Eichwaldiacea, Plectambonitacea, Strophomenacea, Davidsoniacea, Chonetacea, Porambonitacea, Pentameracea, Rhynchonellacea, Atrypacea and Athyridacea—a list which in itself demonstrates the morphological variability and diversity of the phylum in Ordovician—Silurian boundary times. However, rather than review each family, genus or species in turn here, it is more relevant to consider the brachiopod faunas actually recovered from strata near the boundary. In general the middle Ashgill was a period of great diversity among the brachiopods, but this diversity was reduced when the Rawtheyan endemic faunas, for example of North America (the late Richmondian) and Europe (e.g. the Boda Limestone of Sweden) gave way to the more cosmo- politan, and hence in total less diverse, faunas of Hirnantian times. Similarly, the profound effect of the Ordovician—Silurian boundary glacial episode made the subsequent recovery and build-up of the brachiopod faunas rather slow, and thus, even where the earliest Llandovery time is represented by rock (and not by the usual unconformity), the numbers and more particularly the diversity of the brachiopod faunas were rather poor. Latest Ordovician and earliest Silurian brachiopods In the following lists the records are reproduced of reliable determinations from relatively recent papers on brachiopods of Hirnantian and early Rhuddanian ages respectively. In most cases they are as the original authors determined them, but with ‘aff. or ‘cf’ omitted, and sometimes with genera or species updated by subsequent works. They are from the following authors and localities: A, uppermost Ellis Bay and lowermost Becscie Formations, Anticosti Island, Canada (Cocks & Copper 1981); B, Kosov Formation, Bohemia, Czechoslovakia (Marek & Havliéek 1967; Havliéek 1977); D, Durben Horizon, Kazakhstan, USSR (Nikitin et al. 1980); E, Lower Edgewood Group, Oklahoma, USA (Amsden 1974); G, High Mains Sand- stone and Lady Burn Formation, Girvan, Scotland (Cocks & Toghill 1973; Harper, this volume); H, St Martin’s Cemetery Horizon, Haverfordwest, Wales (Cocks & Price 1975); I, Hol Beck, England (Temple 1965); K, Kildare, Ireland (Wright 1968); L, Bronydd Formation, Llandovery, Wales (Cocks et al. 1984); M, persculptus and acuminatus Zones, Mirny Creek, north-east USSR (Koren et al. 1983); O, Langgyene and Langara Formations, Oslo—Asker district, Norway (Brenchley & Cocks 1982; Cocks 1982) and Myren Member (Baarli & Harper 1986); P, Unit 5, White Head Formation, Percé, Québec, Canada (Lespérance & Sheehan 1976, Bull. Br. Mus. nat. Hist. (Geol) 43: 311-315 Issued 28 April 1988 312 L. R. M. COCKS 1981); R, Varbola Formation, Estonia, USSR (Rubel 1970); S, Stawy, Poland (Temple 1965); V, Dalmanitina Beds, Vastergotland, Sweden (Bergstrom 1968); W, Hirnant Beds, Wales (Temple 1965); X, Hirnantian Beds, Keisley, England (Temple 1968); Y, Kuanyinchiao Beds, Yichang, China (Rong 1984a); Z, Artchalyk and Minkutchar Beds, Zeravshano-Gissar section, Altai Mountains, USSR (Nikiforova 1978). The latest Ordovician (Hirnantian) records from these localities are as follows: Lingulacea: Lingula sp. H, O; Lingulella sp. I, S; Palaeoglossa sp. V; Craniops/Paracraniops sp. H, O, V, X. Discinacea: Trematis norvegica Cocks O; Orbiculoidea concentrica (Wahlenberg) H, V, S; Orbiculoidea sp. O. | Craniacea: Acanthocrania sp. O, X; Philhedra grayii (Davidson) X; Philhedra sp. H, V; Philhedra? stawyensis Temple I, S; Philhedrella cribrum Temple X; Philhedrella sp. A, O. Orthacea: Comatopoma sororia Marek & Havli¢ek B; Comatopoma sp. O; Dolerorthis intermedius Nikiforova M; Dolerorthis praeclara Temple X; Dolerorthis savagei Amsden E; Dolerorthis sp. O; Geraldibella bella (Bergstrom) M, V; Geraldibella giraldi (Bancroft) H; Giraldibella subsilurica (Marek & Havli¢ek) B; Glyptorthis sp. G, O; Hesperorthis sp. M, O; Nicolella sp. O; Orthostrophella sp. E; Plaesiomys sp. G; Platystrophia sp. E, G, O; Skenidioides scoliodus Temple X; Skenidioides sp. H, O; Toxorthis mirabilis Rong Y; Toxorthis proteus Temple X. Enteletacea: Dalmanella biconvexa Williams H; Dalmanella cicatrica Nikitin D; Dalmanella edgewoodensis Savage E; Dalmanella pectinoides Bergstrom B, V; Dalmanella testudinaria (Dalman) A, B, H, I, K, M, O, P, S, V, W, Y; Dicoelosia sp. E, X; Diceromyonia? sera Amsden E; Draborthis caelebs Marek & Havli¢ek B, V, X, Y; Drabovia agnata Marek & Havli¢ek B; Drabovia westrogothica Bergstrom V; Drabovia sp. O, X; Dysprosorthis sinensis Rong Y; Epitomyonia sp. O; Hirnantia noixella Amsden E; Hirnantia sagittifera (M‘Coy) B, D, G, H, I, K, M, O, P, S, V, W, X, Y; Hirnantia sp. A; Horderleyella bouceki (Havliéek) S, W; Horderleyella fragilis Bergstrom V; Isorthis sp. M; Kinnella kielanae (Temple) B, P, S, V, W, X, Y; Leptoskelidion loci Cocks O; Leptoskelidion septulosum Amsden E; Mendacella? sp. E; Mirorthis mira Zeng Y; Onniella kalvoya Cocks O; Onniella? yichangensis Zeng Y ; Paucicrura sp. O; ‘Pionodema’ retusa Temple X; Ravozetina rava Marek & Havli¢ek B; Reuschella inexpectata Temple X; Trucizetina subrotundata Havli¢ek B; Trucizetina yichangensis Zeng Y; Visbyella? sp. [= Kayserella sp. nov. of Temple] X. Gonambonitacea: Kullervo? sp. O. Tripleciacea: Cliftonia psittacina (Dalman) B, H, K, O, V; Cliftonia obovata Chang Y; Cliftonia tubuli- striata (Savage) E; Cliftonia sp. D, M; Onychoplecia sp. X, Y; Oxoplecia sp. O; Triplesia protea Oradovskaya M;; Triplesia sanxiaensis Zeng Y; Triplesia sp. O. Plectambonitacea: Aegiromena convexa Chang Y; Aegiromena durbenensis Nikitin D; Aegiromena ultima Marek & Havli¢éek B, Y; Aegiromena sp. X; Anisopleurella novemcostata Nikitin D; Chonetoidea papil- losa (Reed) H; Eochonetes sp. G; Eoplectodonta nesnakomkaensis Oradovskaya M; Eoplectodonta rhom- bica (M‘Coy) O; Eoplectodonta oscitanda Cocks O; Eoplectodonta sp. D; Leangella cylindrica (Reed) O, V; Rugosowerbyella ambigua (Reed) D; Sampo sp. O; Sericoidea? O. Strophomenacea: Aphanomena parvicostellata Rong Y; Aphanomena schmalenseei Bergstrom V; Biparetis paucirugosus Amsden M; Eopholidostrophia sp. G; Eostropheodonta bublitschenki Nikitin D; Eostro- pheodonta hirnantensis (M‘Coy) including E. lucavica and E. siluriana A, B, G, I, K, M, O, P, S, V, W; Eostropheodonta whittingtoni Bancroft H; Katastrophomena sp. O; Kjaerina? sp. O; Kjerulfina? sp. V; Leptaena aequalis Amsden M; Leptaena martinensis Cocks H; Leptaena rugosa Dalman B, D, V; Leptaena sp. E, O; Leptaenopoma trifidum Marek & Havlitéek B, D, K, V, Y; Paromalomena polonica (Temple) B, D, I, S, X, Y; Rafinesquina? latisculptilis (Savage) E, M; Rafinesquina stropheodontoides (Savage) E; Rafinesquina ultrix Marek & Havliéek B, D; Rafinesquina urbicola Marek & Havliéek B, D; Titanomena grandis Bergstrom V. Davidsoniacea: Coolinia convexa (Savage) E; Coolinia dalmani Bergstrom A, O, V; Coolinia propinqua (Meek & Worthen) E; Coolinia sp. M, Y; Fardenia comes Marek & Havlitek B; Fardenia sp. G, X. Porambonitacea: Parastrophinella gracilis Oradovskaya M; Parastrophina sp. O. Pentameracea: Brevilamnulella kjerulfi (Kjaer) O; Brevilamnulella thebesensis (Savage) E, M; Brevilamnu- lella undatiformis Rozman M; Holorhynchus giganteus Kjaer O; Tcherskidium unicum (Nikolaev) M. Rhynchonellacea: Dorytreta sp. Y; Hypsiptycha sp. G; Rostricellula sp. G, O; Rhynchotrema? sp. M; Stegerhynchus concinna (Savage) E, M; Stegerhynchus? sp. E, O; Thebesia admiranda Oradovskaya M; Thebesia scopulosa Cocks O; Thebesia thebesensis (Foerste) E. Atrypacea: Eospirigerina gaspeensis (Cooper) M; Eospirigerina prisca Oradovskaya M; Eospirigerina putilla (Hall & Clarke) E; Eospirigerina sublevis Rozman M; Eospirigerina sp. G, O; ‘Homoeospira BRACHIOPODS ACROSS THE BOUNDARY 313 fiscellostriata Savage E; Plectatrypa sp. M; Protatrypa sp. X; Protozyga gastrodes Temple X; Zygospira fallax Marek & Havliéek B; Zygospira sp. O. Athyracea: Cryptothyrella crassa (J. de C. Sowerby) incipiens Williams G, H, K, Y; Cryptothyrella ovoides (Savage) E; Cryptothyrella terebratulina (Wahlenberg) M; Cryptothyrella sp. B, X; Hindella cassidea (Dalman) O, ?P, ?A; Hyattidina sp. M; Plectothyrella crassicostis (Dalman) [ex platystrophoides Temple] B, I, K, P, S, V, W, Y; Plectothyrella? mirnyensis Oradovskaya M. Eichwaldiacea: Dictyonella sp. E. The earliest Silurian (lower part of the Rhuddanian) records from these localities are: Lingulacea: Lingula sp. G. Discinacea: Orbiculoidea sp. H. Orthacea: Dolerothis plicata (J. de C. Sowerby) O; Dolerorthis sowerbyiana (Davidson) L; Dolerorthis sp. O, R; Giraldiella sp. L, Z; Hesperorthis imbecilla Rubel R; Platystrophia sp. R; Schizonema sp. L, O; Ptychopleurella sp. R; Skenidioides scoliodus Temple M; Skenidioides woodlandensis Reed O; Skeni- dioides sp. H, L, O. Enteletacea: Dalejina sp. R; Dicoelosia osloensis Wright O; Dicoelosia sp. L; Draborthis? sp. M; Epitomyonia sp. O; Fascifera sp. O; Howellites sp. O; Isorthis neocrassa Nikiforova Z; Isorthis prima Walmsley O; Isorthis sp. A; Kinnella sp. O; Onniella mediocra Rubel R; Ravozetina sp. L, O; Resserella sp. H, L; Reuschella sp. O; Visbyella sp. L. Tripleciacea: Triplesia sp. L, O. Plectambonitacea: Aegiria norvegica Opik O; Anisopleurella sp. L; Anisopleurella gracilis (Jones) H; Eoplectodonta duplicata (J. de C. Sowerby) L, O; Eoplectodonta sp. H; Leangella scissa (Davidson) L, O. Strophomenacea: Eopholidostrophia sp. A, L; Eostropheodonta sp. H; Furcitella sp. L; Katastrophomena sp. L; Leptaena aequalis Amsden M; Leptaena contermina Cocks A; Leptaena haverfordensis Bancroft O; Leptaena reedi Cocks L, O; Leptaena valentia Cocks L; Leptaena sp. H, O, R; Leptostrophia reedi (Bancroft) A; Leptostrophia sp. L. Davidsoniacea: Fardenia sp. G, L, R. Porambonitacea: Parastrophinella sp. Z. Pentameracea: Clorinda malmoyensis St Joseph Z; Clorinda undata (J. de C. Sowerby) H, L, O; Clorinda sp. R; Stricklandia lens (J. de C. Sowerby) H, L, O, R, Z; Virgiana sp. Z; Virgianella sogdianica Nikiforova & Sapelnikov Z. Rhynchonellacea: Rhynchotrema sp. L; Rhynchotreta? sp. G. Atrypacea: Alispira gracilis Nikiforova R; Clintonella aprinis (Verneuil) R; Clintonella sp. R; Eospirigerina porkuniana Rubel R; Idiospira sp. O; Eospirigerina sp. H, O, Z; Meifodia recta alia Nikiforova Z; Meifodia sp. L, O; Plectatrypa imbricata (J. de C. Sowerby) Z; Plectatrypa sp. L; Protatrypa malmoey- ensis Boucot, Johnson & Staton O, Z; Protatrypa sp. M; Protozyga sp. L; Zygospiraella sp. M, Z; Zygospiraella duboisi (Verneuil) R. Athyracea: ‘“Atrypina’ gamachiana Twenhofel A; Cryptothyrella angustifrons (Salter) L, G; Cryptothyrella crassa (J. de C. Sowerby) H, L; Cryptothyrella sp. A, R; ‘Hindella’ extenuata Rubel R; Hyattidina sp. M. From these lists it can be seen that the cited faunas carried 90 genera in the Hirnantian and only 54 in the early Rhuddanian, with 32 genera in common between the two lists. Part of this numerical discrepancy can be explained by the greater number of faunal lists available for beds of Hirnantian age (18), compared with only 8 for the early Rhuddanian; nevertheless that discrepancy can itself be explained by the fewer number of early Llandovery age faunas that can actually be found. Moreover, whereas the Hirnantian faunas can often be found in abun- dance (for example in China—Rong 1984a, b), the early Rhuddanian faunas are often very sparse both in numbers and diversity, and also in the actual size of the specimens, all of which explains why monographic treatment of them has been rather neglected, particularly by com- parison with the much richer and more diverse later Rhuddanian faunas, which are relatively well described (e.g. Temple 1970). In addition, presumably because of the glacially-caused eustatic lowering of sea level which peaked during the Hirnantian, there are many sections in which only the Hirnantian is represented by shelly deposits and with the beds above and below in which the only macrofossils are graptolites. Missing from both of the above lists are representatives of the Trimerellacea, Acrotretacea, Siphonotretacea and Chonetacea, all of which have reliable records from both late Ordovician and early Silurian rocks, but not from beds very close to the boundary; and from the early 314 L. R. M. COCKS Rhuddanian list the Craniacea and the Eichwaldiacea, which also yield representatives from later horizons in the Llandovery. The only brachiopod superfamily which appears to have become extinct at the end of the Hirnantian is the Gonambonitacea (although a few lower taxa such as the Trematidae also disappeared then); and the only new superfamily to appear anywhere near the base of the Silurian is the Cyrtiacea, whose earliest records, although not accurately dated in detail, come from beds in Tasmania extremely close to the boundary (Sheehan & Baillie 1981). In general, however, the degree of extinction across the boundary appears to have been far less than previously reported, largely because earlier studies have not concentrated on latest Ashgill and earliest Llandovery rocks. The extinctions at the end of the Hirnantian do not appear to have been greater than at the end Caradoc or end Rawtheyan. This is exemplified by a recent review of the atrypoids by Copper (1986), who states that only two genera, Idiospira and Cyclospira, may have become extinct near the boundary, and even these two have been reported (e.g. Baarli & Harper 1986) from early Silurian rocks. The strong ‘Silurian’ elements in the spire-bearer fauna, for example Hindella and Eospirigerina, actually appeared in late Rawtheyan times. Unfortunately no evolutionary gradation within a single genus has been adequately studied across the boundary, and thus no perfect recognition of the boundary by brachiopods is yet possible. The most striking changes in closely related groups are seen in the Pentameracea, which can be found in virtually rock-building abundance in some beds both above and below the boundary, although only rarely in the earliest Rhuddanian. In the Hirnantian, Holo- rhynchus, Brevilamnulella, and others dominate the fauna, whereas in the Rhuddanian their place is taken by Stricklandia, Clorinda, and a wide diversity of genera in the then tropical areas of the USSR (Nikolaev et al. 1977) and, rather later, Virgiana and Platymerella in the USA. In the east Baltic, Borealis is known from as low as the vesiculosus Zone (Boucot et al. 1969). The exact age, in terms of graptolite zones, of the various brachiopod faunas from near the systemic boundary, in particular the Hirnantia fauna, is also of great relevance in international correlation. In continuous sections, most Hirnantia faunas underlie beds bearing persculptus Zone graptolites, for example in the vast outcrop area in China, and in general the fauna is undoubtedly of extraordinarius Zone age or older; it spans four graptolite zones in China (Rong 1984b). However, in at least two places it occurs in beds with and above persculptus Zone graptolites. One is in Kazakhstan, USSR (Apollonov et al., this volume, p. 145), and the other is in the Lake District, England, where Locality 74/1 of Hutt (1974: 15) in Yewdale Beck, Cumbria (National Grid ref. SD 30739858) has yielded to J. E. Hutt (registered numbers BC 7217-7236), in order of abundance, Kinnella kielanae (Temple), Mirorthis mira Zeng, Plec- tothyrella crassicostis (Dalman), Cyclospira sp., Hirnantia sp. and other indeterminate orthids and dalmanellids, identified by the author and Rong Jia-yu. In addition the same bed has yielded many graptolites (J. E. Hutt, pers. comm. 1986), including Climacograptus medius Torn- quist, C. normalis Lapworth, C. miserabilis Elles & Wood, Glyptograptus persculptus (Salter), Diplograptus ex gr. modestus Lapworth and Monograptus ceryx Rickards & Hutt. These new records endorse the most preferable systemic boundary at the base of the acuminatus Zone. Acknowledgements I am most grateful to Rong Jia-yu and A. J. Boucot, who helpfully commented on the first draft of this paper. References Amsden, T. W. 1974. Late Ordovician and Early Silurian articulate brachiopods from Oklahoma, south- western Illinois and eastern Missouri. Bull. Okla geol. Surv., Norman, 119: 1-154, 28 pls. Baarli, B. G. & Harper, D. A. T. 1986. Relict Ordovician brachiopod faunas in the Lower Silurian of Asker, Oslo Region, Norway. Norsk geol. Tidsskr., Oslo, 66: 87-98. Bergstrom, J. 1968. Upper Ordovician Brachiopods from Vastergotland, Sweden. Geologica Palaeont., Marburg, 2: 1—35, 7 pls. BRACHIOPODS ACROSS THE BOUNDARY 315 Boucot, A. J., Kaljo, D. & Nestor, H. 1969. Stratigraphic range of the early Silurian Virgianiinae (Brachiopoda). Eesti NSV Tead. Akad. Toim., Tallinn, (Khim. Geol.) 18: 76-79. Brenchley, P. J. & Cocks, L. R. M. 1982. Ecological associations in a regressive sequence: the latest Ordovician of the Oslo—Asker District, Norway. Palaeontology, London, 25: 783-815, pls 85-86. Cocks, L. R. M. 1982. The commoner brachiopods of the latest Ordovician of the Oslo—Asker District, Norway. Palaeontology, London, 25: 755-781, pls 78-84. —— & Copper, P. 1981. The Ordovician-Silurian boundary at the eastern end of Anticosti Island. Can. J. Earth Sci., Ottawa, 18: 1029-1034. & Price, D. 1975. The biostratigraphy of the upper Ordovician and lower Silurian of south-west Dyfed, with comments on the Hirnantia fauna. Palaeontology, London, 18: 703-724, pls 81-84. —— & Toghill, P. 1973. The biostratigraphy of the Silurian rocks of the Girvan District, Scotland. Q. JI geol. Soc. Lond. 129: 209-243, pls 1-3. ——, Woodcock, N. H., Rickards, R. B., Temple, J. T. & Lane, P. D. 1984. The Llandovery Series of the type area. Bull. Br. Mus. nat. Hist., London, (Geol.) 38 (3): 131-182. Copper, P. 1986. Evolution of the earliest smooth spire-bearing atrypoids (Brachiopoda: Lissatrypidae, Ordovician-Silurian). Palaeontology, London, 29: 827-866, pls 73-75. Havliéek, V. 1977. Brachiopods of the order Orthida in Czechoslovakia. Rozpr. tustred. Ust. geol., Prague, 44; 1—327, pls 1-66. Hutt, J. E. 1974. The Llandovery graptolites of the English Lake District. Part 1. Palaeontogr. Soc. (Monogr.), London. 56 pp., 10 pls. Koren, T. N., Oradoyskaya, M. M., Pylma, L. J., Sobolevskaya, R. F. & Chugaeva, M. N. 1983. The Ordovician and Silurian boundary in the Northeast of the USSR. 208 pp., 48 pls. Leningrad, Nauka. [In Russian. | Lesperance, P. J. & Sheehan, P. M. 1976. Brachiopods from the Hirnantian stage (Ordovician-Silurian) at Perce, Québec. Palaeontology, London, 19: 719-731, pls 109-110. 1981. Hirnantian fauna in and around Perce, Québec. Jn P. J. Lespérance (ed.), Field Meeting, Anticosti—Gaspe, Quebec, 1981 2 (Stratigraphy and paleontology): 231-245. Montréal (I.U.G.S Sub- commission on Silurian Stratigraphy Ordovician—Silurian Boundary Working Group). Marek, L. & Havli¢ek, V. 1967. The articulate brachiopods of the Kosov Formation (Upper Ashgillian). Vést. UstFed. Ust. geol., Prague, 42 (4): 275-284, pls 1-4. Nikiforoya, O. I. 1978. [Brachiopods of the Tchasmankalon, Artchalyk and Minkutcher Beds.] In B. S. Sokolov & E. A. Yolkin (eds), Pogranichniye sloi ordovika i silura Altaye-Sayanskoy oblasti 1 Tyen- Shanya. Trudy Inst. Geol. Geofiz. Sib. Otdel., Moscow, 397: 102-125, pls 18-23. Nikitin, I. F., Popoy, L. E. & Rukavishnikova, T. B. 1980. Brachiopods. In M. K. Apollonov, S. M. Bandaletov & I. F. Nikitin (eds), The Ordovician—Silurian boundary in Kazakhstan. 300 pp., 56 pls. Alma-Ata. Nikolaev, A. A., Oradovskaya, M. M. & Sapelnikov, V. P. 1977. [Biostratigraphical review of the Ordovi- cian and Silurian pentamerids in the north-east USSR.] Trudy Inst. Geol. Geokhim. Akad. Nauk SSSR ural. nauch. Tsentr., Sverdlovsk, 126: 32-67, 11 pls. Rong Jia-yu 1984a. Brachiopods of latest Ordovician in the Yichang District, western Hubei, central China. In Nanjing Institute of Geology and Palaeontology, Academia Sinica, Stratigraphy and Palaeon- tology of Systemic boundaries in China. Ordovician-Silurian boundary 1: 111-190, pls 1-14. Anhui Sci. Tech. Publ. House. 1984b. Distribution of the Hirnantia fauna and its meaning. In D. L. Bruton (ed.), Aspects of the Ordovician System: 101-112. Universitetsforlaget, Oslo. Rubel, M. 1970. On the distribution of brachiopods in the lowermost Llandovery of Estonia. Eesti. NSV Tead. Akad. Toim., Tallinn, (Khim. Geol.) 19: 69-79. Sheehan, P. M. & Baillie, P. W. 1981. A new species of Eospirifer from Tasmania. J. Paleont., Tulsa, 55: 248-256, pl. 1. Temple, J. T. 1965. Upper Ordovician brachiopods from Poland and Britain. Acta palaeont. pol., Warsaw, 10: 379427, pls 1-21. —— 1968. The Lower Llandovery (Silurian) brachiopods from Keisley, Westmorland. Palaeontogr. Soc. (Monogr.), London. 58 pp., 10 pls. —1970. The Lower Llandovery brachiopods and trilobites from Ffridd Mathrafal, near Meifod, Mont- gomeryshire. Palaeontogr. Soc. (Monogr.), London. 76 pp., 19 pls. Williams, A. et al. 1965. In R. C. Moore (ed.), Treatise on Invertebrate Paleontology, H, Brachiopods. 927 pp. Lawrence, Kansas (Geol. Soc. America & Univ. Kansas Press). Wright, A. D. 1968. A westward extension of the upper Ashgillian Hirnantia fauna. Lethaia, Oslo, 4: 352-367. rs = rm - ae a. ay