INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly fi-om the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper aligmnent can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

UMI University Microfilms international A Bell & Howell Information C om pany 300 North Z eeb Road. Ann Arbor. Ml 48106-1346 USA 313.'761-4700 800.521-0600

Order Number 9228325

Studies in Proterozoic paleobiology from Spitsbergen and Artie C anada

Butterfield, Nicholas James, Ph.D.

Harvard University, 1992

Copyright ©1992 by Butterfield, Nicholas James. All rights reserved.

UMI 300 N. Zeeb Rd. Ann Arbor, MI 48106

HARVARD UNIVERSITY THE GRADUATE SCHOOL OK ARTS AND SCIENCES

THESIS ACCEPTANCE CERTIFICATE {To be placed in Original Copy)

The undersigned, nppoiiucd hy the

Division

Department of Organismic and Evolutionary Biology

Committee have examined a thesis entitled

"Studies in Proterozoic paleobiology from Spitsbergen and arctic Canada"

presented by Nicholas J. Butterfield candidate for the degree of Doctor of Philosophy and hereby certify that it is worthy_of acc^tanc

Sif;m uurc' Typed name Andrew H,,.. K,no.l.l ....

Signature / T / / / / Typed tiatne^. Ralph J

Signature T y^ed name Dpnald H P o s t e r /( Raymond Siever Date January 23, 1992

STUDIES IN PROTEROZOIC PALEOBIOLOGY FROM

SPITSBERGEN AND ARCTIC CANADA

A thesis presented

by

Nicholas James Butterfield

to

The Department of Organismic and Evolutionary Biology

in partial fulfillment of the requirements

for the degree of

Doctor of Philosophy

in the subject of Biology

Harvard University

Cambridge, Massachusetts

January 1992 11

© 1992 by Nicholas James Butterfield

All rights reserved Ill

ABSTRACT

Fossil assemblages from Spitsbergen, Baffin Island, and Somerset Island provide a much enhanced view of Proterozoic paleobiology. Shallow water shales, cherts, phosphorites, and carbonates in the 700-800 Ma Svanbergfjellet Formation, northeastern Spitsbergen, preserve 65 distinct fossil forms, including 4 species of multiceUular green algae (Siphonocladales and Chlorococcales), 3 multiceUular taxa of imknown taxonomic affiliation, 19 acritarch species, and numerous cyanobacteria and insertae sedis. An analysis of fossil taphonomy offers some useful taxonomic distinctions, but generally points to a pervasive 'over-splitting' of form taxa; several major taxonomic revisions are proposed. The SvanbergjQeUet fossils contribute significantly to the understanding of early multicellularity and to an increasingly detailed Neoproterozoic biostratigraphy. A shale sample from the Black Shale Member of the ca. 1250 Ma Agu Bay Formation (Fury and Hecla Group), northwestern Baffin Island, is dominated by small leiosphaerid acritarchs. Their very even distribution on bedding planes and a dearth of typical shallow-water forms (e.g., filaments, large and/or ornamented acritarchs) suggest an offshore deposition in water sufficiently deep to preclude a photosynthesizing benthos; a mid to outer shelf setting is supported by the local sedimentology. A review of Proterozoic fossil distribution permits the delineation of 5 depth-dependent paleoecological zones extending from restricted nearshore to basinal environments. Such recognition serves as a valuable measure of paleoenvironment and should substantiaUy refine any biostratigraphic signal. IV

Silidfied peri tidal carbonates from the 1270-725 Ma Hunting Formation on Somerset Island (arctic Canada) contain fossils of a well preserved bangiophyte red alga, Bangiomorpha antigua n. gen., n. sp. Evidence of transverse intercalary cell division in unlseriate filaments, and a subsequent longitudinal division to yield multiseriate filaments of radially-arranged wedge-shaped cells, shows it to be dosely allied to modem Bangia. Together with the Svanbergfjellet chlorophytes, the Hunting fossils push the record of taxonomically resolved metaphytes well back into the Proterozoic. ACKNOWLEDGEMENTS

I thank Andy Knoll for the inspiration and opportunity to pursue the study of early life, and Roger Buick, Stephen Grant and Julian Green for the enduring skepticism and collegiality necessary to carry it off. Keene Swett (University of Iowa) first introduced me to the business of systematic field work, and Jay Kaufman (then at the Biogeochemical Laboratories, Indiana University) that of stable isotope geochemistry. Fred Chandler (Geological Survey of Canada) provided sample materials of the Fury and Hecla Group and is largely responsible for its geological description (Chapter III); likewise, Andy Knoll, in the case of the Svanberg^ellet Formation (Chapter II). Rob Rainbird and Grant Young (University of Western Ontario) generously assisted with ongoing work on the Shaler Group, Victoria Island. I also thank Nikola Baumgarten, Steve Could, Don Pfister, Ralph Mitchell, Ray Siever, and sympathetic fellow graduate students for their encouragement, advice, and assistance.

Financially, I was supported from 1986 through 1990 by a Natural Sciences and Engineering Research Council of Canada (NSERC) post-graduate scholarship. Research funds were provided by the National Geographic, the Department of Organismic and Evolutionary Biology (Harvard University), the Geological Society of America, Monsanto Inc., and various NSF, NASA, and DOE grants to Andy Knoll. VI

TABLE OF CONTENTS

A bstract...... iü Acknowledgements ...... v Table of Contents ...... vi

I. INTRODUCTION ...... 1

II. PALEOBIOLOGY OF THE NEOPROTEROZOIC SVANBERGFJELLET FORMATION, SPITSBERGEN ...... 6 ABSTRACT ...... 6 INTRODUCTION ...... 8 MATERIALS AND METHODS ...... 9 GEOLOGICAL SETTING ...... 11 Paleoenvironments ...... 14 TAPHONOMY ...... 17 Recalcitrance ...... 19 Degradation ...... 21 Preservation ...... 22 Mineralization ...... 25 Fossil Analysis ...... 26 Thickness vs. opacity ...... 27 Compression ...... 28 TAXONOMY ...... 30 MULTICELLULARITY...... 32 Simple multicellularity ...... 33 Filamentous Chlorophyta ...... 34 vil

MULTICELLULARITY (cont.)

Complex multicellularity ...... 36 Other Proterozoic Occurrences ...... 38 Evolutionary Implications ...... 39 Metazoans(?) ...... 42 ACRITARCHS ...... 43 Biostratigraphy ...... 44 SYSTEMATIC PALEONTOLOGY ...... 46 nCURES ...... 143 REFERENCES ...... 199 APPENDIX A - List of Fossil Taxa ...... 227

III. PALEOENVIRONMENTAL DISTRIBUTION OF PROTEROZOIC MICROFOSSILS, WITH AN EXAMPLE FROM THE AGU BAY FORMATION, BAFFIN ISLAND ...... 230 ABSTRACT ...... 230 INTRODUCTION ...... 231 GEOLOGICAL SETTING ...... 232 Geochronology ...... 233 PALEONTOLOGY ...... 235 Paleobiology ...... 237 PALEOENVIRONMENTAL DISTRIBUTION ...... 239 CONCLUSION ...... 247 REFERENCES ...... 249 HGURES ...... 258 vin

IV. PALEOBIOLOGY OF THE PROTEROZOIC HUNTING FORMATION, SOMERSET ISLAND, ARCTIC CANADA ...... 270 ABSTRACT ...... 270 INTRODUCTION ...... 270 GEOLOGICAL SETTING ...... 271 PALEOBIOLOGY...... 274 Bangiomorpha antigua n. gen., n. sp ...... 275 SYSTEMATIC PALEONTOLOGY ...... 281 REFERENCES ...... 283 FIGURES ...... 290

V. APPENDIX - PUBLICATIONS

A. Published abstracts, etc. (1988-1992) 302

B. BUTTERFIELD, N. J., A. H. KNOLL, and K. SWETT. 1988. Exceptional preservation of fossils in an Upper Proterozoic shale. Nature 334:424-427. C BUTTERFIELD, N. J., A. H. KNOLL, and K. SWETT. 1990. A bangiophyte red alga from the Proterozoic of arctic Canada. Science 250:104-107. D. BUTTERFIELD, N. J. 1990. Organic preservation of non-mineralizing organisms and the taphonomy of the Burgess Shale. Paleobiology 16:272-286. E. BUTTERFIELD, N. J. 1990. A reassessment of the enigmatic Burgess Shale fossil Wiwaxia corrugata (Matthew) and its relationship to the polvchaete Canadia spinosa Walcott. Paleobiology 16:287-303. I. INTRODUCTION

Most of geological and bio-evolutionary time occurred prior to the Phanerozoic radiations of large organisms, yet the fossils used to resolve time and document evolutionary change belong overwhelmingly to this last one- eighth of Earth history. The probability, indeed necessity, of some significant, if cryptic interval of pre-Cambrian evolution has long been appreciated (Darwin, 1859; Walcott, 1910), but it is only in the last few decades that it has been pursued as a discipline in its own right. Following upon the early studies of the Gunflint Iron Formation (Barghoom and Tyler, 1965), Precambrian micropaleontology in North America came to be directed at the silidfied microbial mat assemblages of carbonate fades; the emerging pattern pointed to a very early establishment of modem prokaryote groups followed by a remarkable evolutionary (or at least morphological) stasis. In contrast, the impetus of early Soviet research wéis towards biostratigraphic correlation, hence its emphasis on the more age-delimited unicellular eukaryotes of siliddastic fades (Timofeev, 1966). More recent studies, including the present one, have begun to meld the materials and methods of these two 'schools', leading to a broader and much improved understanding of early life. The onset of the PhaneroZoic Eon is marked by the sudden appearance and radiation of biomineralized fossils; however, the full impact of the Precambrian/Cambrian transition is truly felt only in rare instances of exceptional preservation where non-mineralized organisms are preserved. These fossil Lagerstatten, such as the Middle Cambrian Burgess Shale, present a different, but assuredly a more accurate accounting of early Paleozoic life than that provided by the conventional fossil record (Appendix D, E). The same must certainly be true for the Proterozoic: the more representative fossil biota (with respect to evolutionary status) is not the 'conventional' record of microbial mats and unicellular eukaryotes, but that of exceptional Lagerstatte preservation. A few such assemblages are known from the pre-Ediacaran Proterozoic, most notably in shales of the Lakhandin and Miroedikha suites of northeastern Siberia (Timofeev et al., 1976). The 700-800 million-year-old (Ma) Svanberg^ellet Formation on northeastern Spitsbergen contains another (Appendix B); it is detailed in Chapter II. Even in cases of exceptional preservation a fossil biota cannot be read literally. There is a clear gradation from highly recalcitrant to effectively non- preservable biological structures, and it is the metabolically more active members of a community that tend towards the latter end of this scale; preservation potential becomes vanishingly small in the case of un walled protists or acuticular metazoans. In many, perhaps most instances, some amount of early degradation may be necessary to initiate fossilization, while differing circumstances may substantially alter the relative recalcitrance and/or appearance of various constituents. Taphonomy, the science of postmortem history and process, is clearly of central importance to paleobiological analysis (Chapter H; Appendix D). Because they cut through much of the taphonomic overprint obscuring primary structure, fossil LagerstStten should clearly stand as the principal reference points for taxonomic analysis. Large, well-preserved populations in the Svanbergfjellet assemblage are consequently applied to the revision and clarification of numerous taxa. As much as possible a natural classification is attempted. Simple structures are necessarily addressed by an artificial form taxonomy; nonetheless, an allowance for natural and taphonomic variation can yield useful taxonomic groups. Organisms and communities of organisms have always had discrete environmental ranges. Accurate reconstruction of the Proterozoic biosphere must therefore be prefaced by an analysis of paleoenvironmental distribution. Conversely, an understanding of fossil distribution can provide a high resolution measure of paleoenvironment. In Chapter III, a study of the paleontology and sedimentology of the ca. 1250 Ma Agu Bay Formation, northwestern Baffin Island is integrated with those of other fossiliferous Proterozoic units to derive a 5-part, basin-wide paleoenvironmental zonation. In the process of resolving depositional environment this scale also clarifies a number of seemingly anomalous fluctuations in the Proterozoic fossil record; most are a product of spatial (i.e., ecological) rather than temporal (i.e., evolutionary) change. Even so, there is excellent potential for at least a Neoproterozoic (1000-550 Ma) biostratigraphy when accompanied by suitable paleoenvironmental control (Chapters II, III). The fossil record provides much of the primary data on the timing and patterns of evolutionary history. Unfortunately, the taxonomy of Proterozoic fossils is often problematic and the first appearance of the major protistan lineages is certainly lost among the Acritarcha (organic-walled unicellular fossils of undeterminable taxonomic affinity). The best approximation will come from multiceUular fossils with their more complex and potentially diagnostic morphologies. MulticeUular eukaryotes are well known in latest Proterozoic (Ediacaran) rocks, but earlier examples are sparse and generally lack the distinctive characters or modern analogues necessary for phylogenetic analysis (Hofmann, 1985). In the absence of biochemical and molecular data, the classification of early multiceUular protists is largely dependent upon the identification of cellular patterns that reflect diagnostic cell division programs. On this basis, three multiceUular fossils in the Svanberg^ellet Formation are identified as siphonocladalean green algae (Chlorophyta) (Chapter II), while filaments in the 725-1250 Ma Hunting Formation, Somerset Island, can be ascribed unambiguously to the bangiacean red algae (Rhodophyta) (Chapter IV; Appendix C). Such taxonomic resolution is unprecedented for pre- Ediacaran fossils and these assignments provide significant new data on the timing and direction of early algal phylogeny. The Svanbergfjellet assemblage also includes a number of multiceUular fossils exhibiting substantial cellular (or possibly tissue) differentiation. Although of unknown taxonomic affiliation, these forms attest to a significantly pre-Ediacaran history of complex niulticellularity. Together with the algal record they invite speculation into the origin of multicellularity as a grade of organization (Chapter II). All three of the fossil assemblages addressed in this study come from limited samples and additional paleobiological, sedimentological, and geochemical investigation is clearly in order. Discovery of apatite nodules in the Svanbergfjellet Formation promises an absolute date for the sequence which, along with the detailed paleobiological and geochemical records of the Spitsbergen sequence as a whole (Knoll, 1991), should establish it as a key Neoproterozoic reference section. Moreover, biostratigraphic and chemostratigraphic correlations with other sequences, such as the minimally 723 ± 3 Ma Shaler Group, Victoria Island, arctic Canada (Butterfield and Rainbird, 1988; Heaman et al., 1990; Asmerom et al., 1991), promise an increasingly resolved, and global accounting of the late Proterozoic biosphere. REFERENCES

ASMEROM, Y., S. B. JACOBSEN, A. K. KNOLL, N. J. BUTTERFIELD, and K. SWETT. 1991. Strontium isotopic variations of Neoproterozoic seawater; Implications for crustal evolution. Geochirrüca et Cosmochimica Acta 55:2883-2894. BARGHOORN, E. S. and S. A. TYLER. 1965. Microorganisms from the Gunflint Chert. Science 147:563-577. BUTTERFIELD, N. J. and R. H. RAINBIRD. 1988. The paleobiology of two Proterozoic shales. Geological Society of America, Abstracts with Programs 20:A103. DARWIN, C. 1859. On the Origin of Species. J. Murray; London. HEAMAN, L. M., A. N. LeCHEMINANT, and R. H. RAINBIRD, R. H. 1990. A U-Pb baddeleyite study of Franklin igneous events, Canada. Geological Association of Canada, Program with Abstracts 15:A55. HOFMANN, H. J. 1985. Precambrian carbonaceous megafossils. Pp. 20-33. In D. F. Toomey and M. H. Nitecki (eds.), Paleoalgology: Contemporary Research and Applications. Springer-Verlag; Berlin. KNOLL, A. H. 1991. End of the Proterozoic Eon. Scientific American 265(4):64-73. TIMOFEEV, B. V. 1966. Mikropaleofitologicheskoe Issledovanie Drevnikh Svit. Nauka; Moscow. 147 pp. TIMOFEEV, B. V., T. N. HERMANN, and N. S. MIKHAILOVA. 1976. Mikrofitofossilii Dokembriia, Kembriia i Ordovika. Nauka; Leningrad. WALCOTT, C. D. 1910. Abrupt appearance of the Cambrian fauna on the North American continent. Cambrian Geology and Paleontology, II. Smithsonian Miscellaneous Collections 57:1-16. IL PALEOBIOLOGY OF THE NEOPROTEROZOIC SVANBERGFJELLET FORMATION, SPITSBERGEN

ABSTRACT

A fossil Lagerstàtte from the 700-750 Ma old Svanbergfjellet Formation of northeastern Spitsbergen offers a substantially enhanced view of late Proterozoic paleobiology. Fossils occur primarily as organic-walled compressions in shales and permineralizations in chert; secondary modes of preservation include bedding-plane imprints and mineral replacements in apatite and goethite(?). Taphonomic analysis reveals that preserved sheath and wall thicknesses vary with preservational mode; that the relative opacity of fossil walls commonly reflects original histological differences rather than wall thickness; that very robust-walled taxa such as Chuaria or Tawuia can be identified on the basis of their substantial sedimentary imprint; and that the flattening that attends compression fossils rarely distorts the two remaining dimensions. True multiceUular (including coenocytic) eukaryotes are a conspicuous component of the Svanbergfjellet assemblage: of 8 distinct taxa, one can be identified as a coenobial/colonial chlorococcalean and three as filamentous siphonocladaleans (Chlorophyta). Other forms are problematic, but several show significant cellular, or possibly tissue, differentiation. A review of Proterozoic multiceUular organisms indicates that a coenocytic grade of organization was common among early metaphytes and supports the view that cellularity is a derived condition in many "multiceUular" lineages. Nineteen acritarch taxa are preserved in the Svanbergfjellet sediments; ten of these show a readily identifiable ornamentation and contribute significantly to Neoproterozoic biostratigraphy; a world-wide and exclusively Late Riphean distribution of the large acanthomorphic acritarch Trachvhvstrichosphaera aimika identifies it as a particularly valuable index fossil. The Svanbergfjellet fossil assemblage preserves a total of 65 distinct forms of which 60 are treated taxonomically. As much as possible, principles of "natural" taxonomy are applied such that taphonomic and ontogenetic variants are declined separate taxonomic status. Major taxonomic revisions are offered for the acritarchs Trachvhvstrichosphaera and Chuaria. and for prokaryotic-grade filaments: Cephalonvx. Cvanonema, Oscillatoriopsis. Palaeolvngbya. Rugosoopsis. Siphonophvcus, Tortunema. and Veteronostocale. Newly erected taxa include eight new genera: Palaeastrum. Proterocladus, Segregocladus. Pseudotawuia, Valkvria. Cerebrosphaera, Osculosphaera. and Pseudodendron; fifteen new species in thirteen genera: Palaeastrum dvptocranum. Proterocladus major, P. minor. Segregocladus hermannae. Pseudotawuia birenifera. Valkvria borealis. Cerebrosphaera buickii. Osculosphaera hvalina. Pseudodendron anteridium. P. polvtaenium, Dictvotidium colandrum. Germinosphaera iankauskasii. Trachvhvstrichosphaera polaris. Siphonophvcus thulenemum and Digitus adumbratus; and seven new combinations: Leiosphaeridia wimanii. Eoentophysalis croxfordii, Cephalonvx geminatus. Oscillatoriopsis amadea. Siphonophvcus tvpicum. Siphonophvcus gracile, and Tortunema wemadskii. INTRODUCTION

Instances of exceptional fossil preservation are becoming increasingly appreciated as the best available measures of paleodiversity (Conway Morris, 1986). Accordingly, it is the rare fossil Lagerstatten such as the Burgess Shale or the Solenhofen Limestone that serve as the principal reference points in the reconstruction of evolutionary history. The late Proterozoic is among those intervals most calling for detailed Lagerstatte-type documentation as it holds the immediate forebears and, potentially, the explication of the Ediacaran/Cambrian radiation of large organisms (Knoll, 1991). Exceptional fossil preservation is known in rocks of this age but much of it is limited to prokaryotic microbial mat assemblages in relatively restricted carbonate facies (e.g., Schopf, 1968; Knoll, 1982); shale-hosted Lagerstatten are typically more diverse (e.g., Timofeev and Hermann, 1979; Hermann, 1981a; 1981b), but may also be taxonomically distinct. A reasonably representative accounting of Neoproterozoic paleobiology thus requires superior fossil representation in various taphonomic modes and from diverse paleoenvironments. Sediments of the 700-800 Ma old Akademikerbreen Group, northeastern Spitsbergen (Fig. 1), have proven a rich source of silidfied microfossils and related paleobiological data, particularly in the Draken Conglomerate (Knoll, 1982; Knoll et al., 1991) and overlying Blacklundtoppen formations (Knoll et al., 1989). Shale-hosted fossil assemblages are also reported from these units, but truly exceptional preservation in siliddastic fades appears to be limited to the immediately subjacent Svanbergfjellet Formation (Butterfield et al., 1988). As the Svanbergfjellet also preserves a diverse silidfied biota and various mineralized forms, it offers an unusually complete view of the late Proterozoic biosphere as it approaches the Ediacaran and Cambrian radiations. The Svanbergfjellet fossil Lagerstatte is here described with particular reference to paleoenvironmental distribution, taphonomic processes, multiceUular and unicellular eukaryotes, and various taxonomic reforms.

MATERIALS AND METHODS

Paleontological investigation of the Svanbergj^ellet Formation was directed primarily at organic-rich (i.e., dark) chert nodules in carbonate facies, and fine-grained siliddastic rocks. Cherts were examined in pétrographie thin sections cut perpendicular to bedding; of 39 samples, 22 proved to be fossiliferous, half of these exceptionaUy so (Fig. 2). Careful add (HF) maceration of fossiliferous chert samples yielded no intact fossil material. A standard palynological maceration of the SvanbergQellet shales produced a number of taxonomically depauperate fossil assemblages comprised largely of fragmentary material. Further investigation by way of bedding-parallel thin section revealed that most of this fragmentation was a product of the extraction procedure and that several shale horizons preserved abundant and diverse fossils. Indeed, the conventional palynological treatment of most Proterozoic shales may be responsible for a significant bias in the early paleontological record through its preferential recovery of simple and relatively robust fossils. Study in thin section also preserves biologically and paleoenvironmentaUy significant details of fossU associations, orientation, and bedding-plane distribution (e.g., Butterfield and Chandler, 1992; Chapter HI). Such being the case, all of the SvanbergfjeUet shales were initiaUy examined in pétrographie thin sections cut parallel to bedding. The principal advantage to maceration techniques lies in the ability to sample significant volumes of material. A simple, low manipulation 10 maceration procedure was therefore developed to enhance the recovery of rare fossil taxa, especially those too large or too delicate to be extracted by conventional maceration. The procedure involved submersing small (ca. 1 cm^) uncrushed subsamples of fossiliferous shale in concentrated hydrofluoric add (HF) where they were allowed to disaggregate with minimal agitation. Following two or three rinses in distilled water the preparations were examined for fossils with a binocular stereoscope using both transmitted and reflected light. Isolated fossils were collected by pipette, passed through two baths of distilled water, and transferred, always suspended in water, to microscope slide cover-slips. Removal of the water droplet results in a fossil adhering to the glass surface where it can then be prepared for light or scanning electron microscopy. The much improved photogenidty of these isolated fossils further encouraged the extraction of several spedmens initially identified in thin section (e.g.. Figs. 16.1,17.3,19.3, 25.5). With few exceptions (e.g., Segregocladus n. gen.. Digitus adumbratus n. sp.), all taxa recovered by add maceration were also recorded in thin section, thus confirming their syngenidty with surrounding sediments. The isolated fossil material is additionally distinguished from modem contaminants by its two-dimensionality and a distinct graphitic surface sheen derived from the low grade burial metamorphism; intact fossil material is more reflective than reconstituted kerogen (i.e, sapropel sheets). On average, the particulate organic material is light to medium brown in transmitted light. Eight of 17 sampled shale horizons contained fossils (Fig. 2). Two of these, P-2945 (Lower Dolomite Member) and 86-G-62 (Algal Dolomite Member), revealed exceptionally well preserved and diverse fossil assemblages, together representing a majority of Svanbergfjellet fossil diversity. These green to grey shales are fissile, finely laminated and extremely fine 11

grained, with most sediment grains measuring less than 1 pm in maximum dimension (Fig. 3.5). Compositionally they are almost entirely siliddastic (no carbonate) and contain less than 1.5% total organic carbon (TOC).

GEOLOGICAL SETTING

The Svanbergfjellet Formation is a distinctive succession of limestones, dolomites, and subordinate siliddastic fades within the Neoproterozoic Akademikerbreen Group, northeastern Spitsbergen (Fig. 1; Wilson, 1961; Knoll and Swett, 1990). Stratigraphically, it lies some 2000 meters below rocks of Early Cambrian age, and ca. 1000 meters below a late Proterozoic (Varanger) tillite. Radiometric dates place few constraints on the age of the Svanbergfjellet Formation, but microfossils in the formation (and in the overlying Draken Conglomerate; Knoll et al., 1991) indicate a younger Late Riphean age, ca. 700-800 Ma. Biostratigraphic and chemostratigraphic (6"C and ®^Sr/“Sr; Knoll et al., 1986; Derry et al., 1989; Asmerom et al., 1991) correlation with the better dated Shaler Group of northwestern Canada (minimally 723 ±3 Ma old; Heaman et al., 1990) corroborate and potentially constrain this estimate. The Svanbergfjellet Formation is nearly 600 m thick in its southernmost exposures at Svanbergfjellet nurtatak, but thins to an estimated 100 m at Kluftdalen, ca. 150 km to the north (Fig. 1; Wilson, 1961). Despite the apparent thickness variation, the same four members can be distinguished throughout the outcrop area (Fig. 2). The basal Lower Dolomite Member lies conformably above the mixed limestones and dolomites of the upper Grusdievbreen Formation. This member, ca. 150 m thick throughout its area of exposure, save for Kluftdalen, 12 consists predominantly of stratiform microbialites and associated intraclastic grainstones, dolomicrites, and scattered low domal or digitate stromatolitic bioherms (Figs. 3.3, 3.4, 3.6). Most carbonates are dolomite, although subordinate thicknesses of limestones occur; a 4-8 meter sequence of quartz arenite and shale near the top of the member constitutes the only significant incursion of siliciclastic rocks. At Svanbergfjellet this latter unit contains centimeter- to dedmeter-sized apatite nodules that could potentially yield a precise radiometric (U-Pb) age for the sequence (Oosthuyzen and Burger, 1973; Slansky, 1986); correlative shales at Polarisbreen contain one of the two prindpal shale-hosted fossil assemblages in the formation (P-2945). Intraclastic carbonates of the Lower Dolomite Member preserve abundant silicified microfossils. Fenestrae and erosional surfaces up to several cm deep document a frequent subaerial exposure of Lower Dolomite carbonates (Knoll and Swett, 1990) and the succession is interpreted as a tidal flat/lagoonal complex, not dissimilar from that described for the overlying Draken Conglomerate (Knoll et al., 1991). The member is capped by a 5-8 m thick Miniaria biostrome (Fig. 3.2) that can be traced from Svanbergfjellet north to at least Geerabukta. The succeeding Lower Limestone Member is a ca. 150 m thick transgressive sequence. It consists predominantly of dark grey to black, laminated to dedmeter-scale beds of caldlutite that commonly contain microspar-filled syneresis cracks. Microbially laminated carbonates are rare within this member, as is chert; however, at least two thin horizons of silicified stratiform stromatolites contain microfossils. Most Lower Limestone deposition took place below fair weather wave base, following the broad trend for deeper water carbonates in the late Proterozoic to be limestone rather than dolomite (Knoll and Swett, 1990). 13

The most distinctive unit of the Svanbergfjellet Formation is the Algal Dolomite Member, which consists of orange-weathering stromatolitic dolomites interbedded with green to black (rarely red) siliciclastic mudstones (Fig. 3.1). Much of the thickness variation recorded for the formation as a whole is accounted for by this member, which varies from 250 m in the south to just 50 m at Kluftdalen; at Geerabukta, where many of the fossils reported here were collected, the member is 100 m thick. Algal Dolomite Member stromatolites consist of inzeriform and tungussiform columns that form domes or extensive biostromes with billowy surfaces; as the member's name implies, all carbonates are dolomitic. Despite the conspicuous nature of the stromatolites in outcrop, mudstones (and rare quartzose sandstones) constitute about half of the member's thickness. The mudstones drape stromatolites, forming knife-sharp lithological contacts (Fig. 3.1), and constitute the principal locus of microfossil preservation in the formation (86-G-62). Algal Dolomite Member sedimentation appears to have taken place largely in quiet subtidal environments, although chaotic blocks within several biostrome horizons indicate that strong storms affected the bottom.

The Upper Limestone Member comprises the uppermost unit of the formation. Like the underlying Algal Dolomite Member, this succession thins markedly from more than 100 m in the south to a few meters at Kluftdalen. Dark grey to black limestones, much like those of the Lower Limestone Member, predominate. This upper unit is distinguished by its greater proportion of dolomitic grainstones and rudites and carbonaceous shales, which increase in abundance toward the top of the member. A regionally extensive bench of quartz arenite marks the top of the formation (although not at Geerabukta); Svanbergfjellet rocks are overlain abruptly by the intraclastic dolomitic grainstones of the Draken Conglomerate. 14

Pàleoenvironments

Overall, the Svanberg^ellet Formation records a variety of coastal marine depositional environments ranging from supratidal to subtidal below wave base. This spectrum of depositional environments is matched by fossil assemblages which document aspects of biological activity over much of this habitat range. Indeed, paleontological data may often provide as much or more resolution as sedimentary fabrics in determining pàleoenvironments (e.g.. Knoll et al., 1991). This is especially true of shale facies which commonly lack diagnostic sedimentary structures (Butterfield and Chandler, 1992; Chapter III). In carbonate facies most Svanbergfjellet microfossils are found in silicified flake conglomerates in which the clasts are often the ripped up fragments of microbial mat. This microbialite grainstone was deposited in lower intertidal to shallow subtidal environments as indicated by the minimal clast transport (clasts are typically poorly sorted, angular and of a high aspect ratio; Figs. 3.3, 3.4) and the co-occurrence of in situ columnar stromatolites with several centimeters of synoptic relief (Fig. 3.6). Paleontologically, the setting is represented by the large ornamented acritarchs Cymatiosphaeroides and Trachvhvstrichosphaera which occupy tlie interstitial spaces of grainstones and often appear to have acted as individual clastic particles. Some interstices further show evidence of microbial recolonization, thus recording some degree of post-depositional biological activity. Geopetally oriented populations of Sphaerophvcus in the matrix of sample 86-SP-8 (Fig. 20.20), and interstitial filaments in sample 86-G-15 (Figs. 23.2-23.4) were almost assuredly photosynthetic and point to relatively extended periods of quiescence well within the photic zone. In both instances this secondary growth would have contributed to the final stabilization of the sediments. 15

The sedimentary and paleontological fabrics within the constituent clasts of the microbialite grainstones further reveal their diverse, if broadly peri tidal, provenance. Apart from non-fossiliferous clasts of (or after) carbonate microspar, two clast types dominate most of the grainstones. The most common is constructed of densely interwoven mats of Siphonophvcus typicum n. comb, with interspersed populations of Mvxococcoides (Fig. 3.3). These clasts are entirely free of mineral grains, show no evidence of sedimentary compaction, and only vague lamination; both the sedimentological and paleoecological evidence point to a high intertidal origin (cf.. Knoll et al., 1991). The other principal clast type is characterized by a distinctive fabric of crinkly sapropelic laminae that incorporate substantial amounts of clastic material, including flakes of the dense, non-laminated microbial mat type (Fig. 3.4). Filamentous and other microbial mat fossils are uncommon as primary constituents, the most conspicuous fossils being the large ornamented acritarchs Trachvhvstrichosphaera and Cvmatiosphaeroides. Given the common occurrence of these same acritarchs in the grainstone matrix, the paleontological signature clearly supports the textural evidence for a shallow to intermediate subtidal derivation of these laminated clasts. Other paleontologically distinct, but less common microbialite intraclasts are suggestive of various micro-environments within the intertidal zone. Some are constructed entirely of dense colonial populations of spheroidal microfossils such as Eoentophvsalis croxfordii n. comb. (Fig. 20.1) or Eoentophvsalis belcherensis (Fig. 20.4) and compare closely with modem intertidal/supratidal entophysalid mats (cf., Golubic and Hofmann, 1976). A somewhat different setting is suggested in sample 86-SP-8 where most clasts are dominated by Sphaerophvcus parvum. Stalk-forming Polvbessurus bipartitus was also an intertidal inhabitant (Knoll et al., 1991), but it occurs as 16 in situ crusts in the Lower Limestone Member (86-P-89, P-3400) rather than in redeposited clasts. The localized shallowing of this otherwise deep-water unit (as inferred from the presence of Polvbessurus and associated microbial mat constituents) is supported by the co-occurrence of fenestrae infilled with botryoidal, concentrically layered cements. Other fossiliferous cherts in the Lower Limestone Member likewise represent in situ sedimentation; however, their even lamination (P-3075) and/or exclusively planktic fossil assemblages (P-3085) clearly identify them as subtidal deposits. All of the fossiliferous shales in the Svanbergfjellet Formation were deposited in relatively shallow water, as evinced by their close proximity to, and intercalation with stromatolitic carbonates in uniformly shallow water sequences. Moreover, a non-random bedding plane distribution of microfossils reveals a significant representation of benthic organisms (including laterally extensive microbial mats), typical of low energy, photic-zone environments (Butterfield and Chandler, 1992; Chapter HI). Even so, the distinct distributions of fossils in the two principal fossil-bearing Svanberg^ellet shales (P-2945 and 86-G-62) suggest important differences in paleoenvironment that are not reflected sedimentologically. For example, where sample P-2945 is characterized by an abundant, but relatively low diversity assemblage of spheroidal and ornamented acritarchs (including Trachvhvstrichosphaera). 86- G-62 acritarchs exhibit low dominance and extremely high diversity. Significantly, the size frequency distributions of planktic leiosphaerids in the two units are markedly dissimilar (Fig. 4). The limited development of filamentous microbial mats or other benthos in P-2945 suggests that these distributional differences derive from its more distal paleoenvironment relative to that of 86-G-62 (zone 3 vs. zone 2 in the scheme of Butterfield and Chandler, 1992 [Chapter IE]). 17

In Proterozoic successions, silicified carbonate and shale facies typically represent distinct pàleoenvironments; however, they broadly intersect in shallow subtidal settings (Knoll, 1984; Butterfield and Chandler, 1992; Chapter in). Fossil taxa common to both subtidal cherts and shales in the Svanbergfjellet Formation include Palaeolvngbva, Rugosoopsis, several species of Siphonophvcus, the large acanthomorphic acritarch Trachvhvstrichosphaera, and likely some leiosphaerid acritarchs. In addition, species of Obruchevella, Brachvpleganon and Veteronostocale that were originally described from silicified carbonates are here preserved in shale. Taphonomic selectivity of the two systems will account for some of the remaining differences; however, much of it surely derives from distinct micro-environments within the shallow subtidal zone. Note, for example, that many of the shared taxa were prokaryotic and likely to have had relatively wide ecological tolerances and distributions.

TAPHONOMY

Accurate reconstruction of a fossil as an individual and ecologically interactive organism requires a thorough understanding of its preservational history. Paleobiological investigation thus merges with sedimentology, low temperature geochemistry and, ultimately, the manner in which fossils are retrieved and studied. In the absence of significant biomineralization, Proterozoic fossil taphonomy approximates that of non-mineralizing organisms of Phanerozoic age (Butterfield, 1990), but with some important exceptions. Prior to the Ediacaran-Cambrian radiation of bioturbating metazoans there would have been a markedly enhanced potential for "stagnation "-type fossil preservation (Seilacher et al., 1985), a feature reflected in the Svanbergfjellet 18

Formation by the preservation of finely laminated sedimentary fabrics (Fig. 3.5). Moreover, a predominantly abiological Proterozoic silica cycle in which silica was precipitated as a semi-evaporitic phase in peri tidal environments (Maliva et ah, 1990) provided a unique taphonomic window onto near-shore microbial mats and associated biotas; this view was effectively eliminated by the Cambrian radiation of mineralizing sponges and radiolarians, and the transfer of silica deposition to offshore environments. On the other hand, a relatively late evolution of recalcitrant, aromatically-based biopolymers (e.g., sclerotin, lignin) would skew the record in favor of Phanerozoic organic-walled fossils. The preservation of pre-Ediacaran fossils followed a number of taphonomic pathways, including compressions and impressions (molds) in shale, early diagenetic permineralization, and mineral replacement. The Svanbergfjellet Formation is remarkable in that it includes examples of all these taphonomic modes, thus offering an unusually diverse view of late Proterozoic life. A majority of the Svanbergfjellet fossils are composed of original (although now kerogenized) organic carbon and owe their preservation to a coincidence of relative recalcitrance of the original structures and more or less anti-biotic depositional circumstances. These have previously been considered with respect to the organic preservation of non-mineralizing animals (Butterfield, 1990) and permineralized land plants and peats (Knoll, 1985). I continue the discussion here as they concern taphonomic processes in the Proterozoic. 19

Recalcitrance

The preservable constituents of non-mineralizing organisms are clearly in the extracellular biopolymers making up cell walls, sheaths, and cuticles (Hanic and Craigie, 1969; Gunnison and Alexander, 1975; Butterfield, 1990); the negligible preservation potential of cytoplasmic constituents is reflected in their typical absence or severely atrophied appearance in organically preserved fossils (Figs. 5.7, 7.1, 7.5). During the Proterozoic the most recalcitrant biopolymers were probably strongly aliphatic, oxygen-poor compounds such as sporopollenin. The widespread occurrence of sporopollenin in modern dinoflagellate cysts and prasinophyte phycoma lends support to the interpretation of most Proterozoic acritarchs as their functional, and possibly taxonomic analogues (Tappan, 1980). However, sporopollenin is also an important constituent of the vegetative cell walls of several chlorophyte algae (e.g., Atkinson et al., 1972; Marchant, 1977) and points to a more heterogenous pool of potential sources for organic-walled microfossils (Butterfield and Chandler, 1992; Chapter HI). The marked similarity between Svanbergfjellet Palaeastrum n. gen. and extant, sporopollenin-containing Coelastrum and Pediastrum (Chlorophyta) suggests that this fossil owes its preservation to such "vegetative" sporopollenin. Sporopollenin is not the Only recalcitrant compound produced by protistan-grade organisms. Biopolymers that satisfy the degradational tests for sporopollenin (resistance to acetolysis and saponification) but with distinct chemical structures are important cell wall constituents of various chlorophyte algae, including the hydrocarbon-rich Botrvococcus braunii (Berkaloff et al., 1983; Puel et al., 1987). These and other "alghumins" (= "algaenans") often appear to be more recalcitrant than even sporopollenin (Tegelaar et al., 1989) 20

and may contribute significantly to sedimentary kerogen (Philp and Calvin, 1976; Hatcher et al., 1983; Zelibor et al., 1987). While themselves comprising only a minor proportion of the cell wall, such compounds may serve to shield more labile cell wall constituents (e.g., cellulose) by altering or blocking sites of enzyme activity (Alexander, 1973; Zelibor et al., 1987). Comparably recalcitrant, but chemically distinct residues are also reported from the envelopes (cell walls and sheaths) of some cyanobacteria (e.g., Chalansonnet et al., 1987); unlike many of their eukaryotic counterparts, these biopolymers exhibit little structural integrity and can not independently preserve morphology. Envelope glycolipids, peptidoglycans, and lipopolysaccharides exhibit moderate chemical recalcitrance (Tegelaar et al., 1989) and undoubtedly contribute to the organic preservation of Proterozoic cyanobacteria. It is indeed relative rather than extreme recalcitrance that should guide taphonomic analysis, particularly in cases of exceptional preservation where fossil representation is skewed away from a simple view of resistant cysts and spores. Relative recalcitrance is of course dependent upon numerous interacting factors, but will clearly include circumstances of deposition, taxonomy (although in no immediately obvious pattern; Hanic and Craigie, 1969; Foree and McCarty, 1970; Gunnison and Alexander, 1975; Chalansonnet et al., 1987; Zelibor et al., 1988), and the physiological state of a prospective fossil. For example, very young and very old algal cultures may be particularly resistant to decay (Jewell and McCarty, 1971), while some cyanobacteria and algae can remain viable for extended periods if buried alive (Gunnison and Alexander, 1975; Birch et al., 1983); this delay in degradation might allow a prospective fossil the grace period necessary for its full isolation in the sediment. 21

Degradation

All naturally occurring organic compounds are naturally biodegradable (Alexander, 1973); preservation of organic-walled fossils thus requires that the normal cycle of decomposition be prematurely and wholly terminated. Prior to the introduction of macrophagous animals such degradation would have been dominated by mechanical fragmentation, photolysis (Kiebar et al., 1990), and heterotrophic and autolytic biochemistry. Geochemical signatures indicate that the biogeochemical cycling of carbon has been active since the early Archean, yet direct evidence of early heterotrophic microbes is rare (Lanier, 1988). Svanbergfjellet Brachvpleganon (Figs. 22.10, 22.11) were previously proposed as fossil heterotrophs (Butterfield et al., 1988) but a possible autotrophic metabolism cannot be ruled out. The same is true for the Siphonophvcus-tvpe filaments that commonly line the body cavity of Valkyria n. gen. (Fig. 10.8). Unambiguous evidence of heterotrophy in the Svanbergfjellet biota occurs as micro-trace fossils in the walls of various acritarchs (Figs. 12.8,13.2,19.5). These distinctive circular perforations are clearly not primary, nor could they have been formed through sedimentary compaction since they typically pass through only one wall of the now double-walled compressions. The exact nature of the responsible heterotroph(s) is not known, but the larger perforations (up to 50 pm diameter in some leiosphaerids) are conceivably the product of colonial, degradative bacteria; the 2-3 pm holes in Cerebrosphaera n. gen. (Fig. 12.8) appear to have had a distinct origin. While heterotrophic bacteria are central to an operative carbon cycle, a number of studies suggest that particulate organic carbon (POC), the 'stuff of organic-walled fossils, serves as a particularly poor microbial substrate (Burns, 22

1979; Fallon and Brock, 1979; Karl et al., 1988). Microbial biodégradation appears instead to be centered on the very finest POC and/or dissolved organic carbon (Fallon and Brock, 1979). Indeed, axenic (bacteria-free) cultures of degrading algae show much the same degree of particle reduction as those seeded with heterotrophic bacteria (Foree and McCarty, 1970). The clear implication is that organisms are inherently self-destructing, presumably as a result of autolytic enzymes. Autolytic degradation results from the activation of enzymes involved in the remodelling or removal of cell walls as they accommodate growth, division, and the release of gametes and zygotes (Ferris et al., 1988; Matsuda, 1988; Butterfield, 1990). The specific capacity of autolysins to degrade otherwise recalcitrant cell walls, their immediate post­ mortem availability, and their potential accumulation in sediments (Bums, 1979; Matsuda, 1988) suggests they represent a fundamental barrier to organic- walled fossil preservation. The role of autolytic enzymes in cyanobacterial taphonomy may be reflected in the preferential preservation of sheaths over the seemingly more recalcitrant cell walls (Golubic and Barghoorn, 1977; Bauld, 1981): where the latter are subject to autolytic restructuring, mucilaginous sheaths are simply vacated or plastically re-formed and thus are not accompanied by specific autolysins.

Preservation

Early isolation by burial is necessary, though not in itself sufficient, for the preservation of organic-walled fossils. The exceptional preservation often encountered in silicified 'flake conglomerates' (e.g.. Knoll, 1982; Mendelson and Schopf, 1982; present study) suggests a bias in favor of rapid sedimentation, particularly when removed from sites of concentrated biological activity. By 23

the same token, planktic microorganisms might be systematically under­ represented in the fossil record due to negligible sinking rates. Accelerated deposition of such forms is nevertheless effected through attachment to larger sinking particles (Lochte and Turley, 1988) including, in siliciclastic environments, the mutual flocculation of plankton and clays (Avnimelech et al., 1982). In this latter case, particle formation is dependent upon available cations (divalent cations more readily induce flocculation than monovalent cations), total organic carbon (high TOC inhibits flocculation; Theng, 1979), and the physiological state of interacting organisms (e.g., Dawson et al., 1981; Fattom and Shilo, 1984). Once deposited, sediment interment fundamentally alters the degradational context of a prospective fossil. Fine grained (or very early mineralized) sediments can, for example, prevent the access of heterotrophic microbes (Fontes et al., 1991) and/or their respective electron acceptors (e.g., oxygen; Revsbech et al., 1980), while the sealed-in metabolites of early biodégradation may inhibit further heterotrophic activity (Butterfield, 1990). However, as discussed above, exclusion of microbial activity is not synonymous with the cessation of biodégradation; nor is degradation significantly impeded by the imposition of anaerobic conditions (Foree and McCarty, 1970; Fallon and Brock, 1979; Allison, 1988; Butterfield, 1990) or moderate reductions in temperature (Foree and McCarty, 1970; Kidwell and Baumiller, 1990; Wiebe et al., 1992). Preservation of organic-walled fossils requires a rapid and profound interference with all enzyme-mediated biochemistry. Organic-walled fossils in the Svanbergfjellet Formation are occasionally preserved in dolomite microspar (Fig. 23.11) or permineralized in large euhedral dolomite crystals (Fig. 3.7); most, however, occur either as 24 compressions in shale or as permineralizations in early diagenetic silica. These two principal taphonomic modes, however, differ fundamentally in both process and overall fossil constitution. Silicified fossils, for example, seem not to be extractable by acid dissolution of the chert matrix and owe their integrity more to the marked stability of chert than to the actual preservation of organic material. In contrast, the generally coherent and add-extractable fossils in shale derive from the exceptional preservation of organic structure -per se. The marked anti-enzymatic activity of particular clay-organic systems has long been recognized by soil scientists (Bums, 1979; Theng, 1979) and appears to be an important mechanism in the organic preservation of shale- hosted fossils (Butterfield, 1990). Degradative enzymes are potentially inhibited by adsorption onto and within clay minerals, with the overall efficacy dependent upon clay type, available exchange cations, pH, and total organic carbon (TOC); maximum preservation potential is expected to occur with expanding, montmorillonite-type clays, monovalent exchange cations, acid pH, and relatively low TOC (Butterfield, 1990). Other combinations will also undoubtedly induce fossilization; however, both the fossil record and theoretical considerations point to the central importance of low organic carbon to sediment ratios (low TOC). In the Svanbergfjellet shales, loss of fossil definition and increased inter-fossil fusion ("natural vulcanization") is related directly to increasing TOC, such that organic rich horizons (> 1.5% TOC) preserve only extremely robust taxa such as Chuaria and Tawuia. It is with respect to TOC that fossil preservation in early diagenetic silica departs most notably from that in shales. Such permineralization appears to result from a particular affinity of dissolved monosilicic adds with the exposed functional groups of degrading organic compounds, hence its particular assodation with high TOC (Leo and Barghoorn, 1976; Knoll, 1985); 25

indeed, it is the enhanced chemical reactivity of the system, rather than its inhibition (as with shales), that leads to increased preservation potential in chert. Together with a predominantly peritidal locus of early diagenetic silica during the Proterozoic (Maliva et al., 1990), this basic disparity of process strongly influenced the environments and ecosystems sampled by the two taphonomic modes. While silica preferentially preserved the organic-rich microbial mat assemblages of peritidal environments, shale facies would have represented less restricted and largely subtidal ecosystems. The two are not entirely exclusive and they broadly intersect in shallow subtidal environments (p. 17).

Mineralization.—Fossil preservation through early diagenetic mineral replacement is an important taphonomic mode for Phanerozoic fossil Lagerstatten (Allison, 1988), yet it represents only a minor component of the Proterozoic record (e.g. Wang et al., 1983; Knoll et al., 1991). In the Svanbergfjellet Formation it accounts for two filamentous but otherwise problematic forms. In one instance, bedding-parallel thin sections of an Algal Dolomite Member shale (86-G-30) reveal abundant, often thickly entangled cylindrical structures that pétrographie and ED AX analysis show to be composed of a yellow, anisotropic, iron-rich mineral, possibly g^aethite (Figs. 26.3-26.6). The occasional inclusion of intact organic-walled filaments attests to the biogenicity of these fossils; however, the mineral overgrowths have significantly distorted the original morphology (see systematic section). The other occurrence is of vertically-onented and branched(?) filamentous structures in shale-hosted, early diagenetic apatite nodules of the Lower Dolomite Member (SV-3; Figs. 23.6, 23.7). The preserved lamination in the nodules appears to be that of the uncompressed shale and thus provides a unique view 26

of the third (i.e., vertical) dimension in fine-grained siliciclastic facies. While the taxonomic affiliation of these poorly preserved filaments is unclear, their habit is clearly biological. Evidence of primary biomineralization among the Svanbergfjellet fossils is equivocal. Siphonophvcus-like filaments in the interstitial spaces of a silicified microbialite grainstone (86-G-15) are marked by intervals of densely packed, sub-micron-sized particles that create a distinctive banded pattern (Figs. 23.2-23.4) broadly reminiscent of the biologically-induced biomineralization of some extant filamentous microbes (e.g., Scvtonema iulianum). However, these particles do not now show any optical activity beyond that of the replacive chert and they conceivably represent a simple organic encrustation.

Fossil Analysis

Taphonomy can be addressed deductively as question of materials and process, such as emphasized in the above discussion; alternatively, the analysis can begin with a fossil and work back to reconstruct the living organism. Under this latter approach, fossil interpretation will derive largely from taphonomic studies on modem analogues (e.g., actualistic experimentation) or comparison between different preservational modes (e.g., contemporaneous shale- and chert-hosted fossil assemblages). Moreover, secondary taphonomic features may themselves provide significant information about primary structure, such as its relative plasticity or general constitution (Butterfield, 1990). The two approaches are of course not mutually exclusive, and the mental exercise of laying on a taphonomic overprint will surely intersect with attempts to peel it back from fossil material. 27

Thickness vs. opacity—V^aW thickness is commonly referred to in taxonomic discussions of organic-walled microfossils, although it is sure to be under some degree of taphonomic control. Thus, the thick extracellular sheaths that typify many permineralized fossils are invariably collapsed as compressions, and, in some cases, appear to be reduced to a simple shagreenate surface texture (e.g., Germinosphaera iankauskasii n. sp., Trachvhvstrichosphaera aimika. Digitus adumbratus n. sp.). The thickness of true cell wall appears to be less affected by taphonomic mode (e.g., the thin innermost wall of permineralized T. aimika is much the same as that preserved in shale; Fig. 18.11); however, it will certainly vary with the overall quality of preservation. Despite this taphonomic factor, the original nature of fossil cell walls can sometimes be inferred independently through an examination of the enclosing sediments. Thus, the thick and resilient walls of Chuaria. Tawuia and Cerebrosphaera n. gen. may be recognized by their prominent (and potentially diagnostic) impressions onto otherwise flat bedding planes, even in the absence of original organic material (e.g.. Figs. 8.5, 8.7, 8.8; Gussow, 1975). In contrast, originally thin-walled or otherwise insubstantial structures leave little or no sedimentary imprint (Figs. 8.1, 8.6). Wall thickness can of course be measured directly under SEM (Figs. 13.8, 23.8). Thickness might also be inferred from the relative opacity of organic structures; however, a systematic carbonization of organic-walled fossils with increasing metamorphic grade clearly rules out the universal application of such a measure. There are also clear taxonomic and histological differences in fossil opacity (Butterfield, 1990). In the Svanbergfjellet, for example, Chuaria, Tawuia. and Cerebrosphaera n. gen. are the only fossils regularly opaque to the light source of a standard laboratory microscope. Granted, these taxa are all relatively thick-walled, but close examination reveals that this is not the 28 direct cause of their optical density. A mounted ca. 3 pm thick fragment of Ç circularis wall was ground down incrementally with 600 grit polishing compound, yet it remained opaque until all but removed; the material making up the wall is inherently dark. Similarly, the outer(?), ca. 1 pm thick layer of Tawuia is opaque, yet a lightly bonded 3.5 pm thick inner(?) layer remains remarkably translucent (Figs 8.4, 23.8). Pronounced histological differences in opacity are also noted in the various structures of Valkvria n. gen., Proterocladus n. gen., Palaeastrum n. gen., and Trachvhvstrichosphaera polaris n. sp. Thus, while relative opacity does not necessarily translate to relative thickness, it may be of some histological and, thereby, taxonomic significance.

Compression.—The projection of three dimensional structure onto the two dimensions of a bedding plane distinguishes shale-hosted carbonaceous fossils from most other preservational modes. Under compression, organisms with an originally circular cross-section might be expected to increase their overall dimensions by a factor of tc/2 (= 57%) (Hofmann and Aitken, 1979; Zhang Z., 1982; Schopf, 1992); such, however, seems rarely to be the case. Comparison of two- and three-dimensionaliy preserved graptolites (Briggs and Williams, 1982) and actualistic compression studies on various three-dimensional "fossils" (Harris, 1974; Rex and Chaloner, 1983; Rex, 1983) reveals that most flattened fossils can be viewed as a relatively undistorted two-dimensional image of the original structure, the equivalent of a photograph. Permineralized and compression fossils will thus be broadly equivalent, at least with respect to absolute dimensions.

The most obvious way to accommodate the surplus wall of the third dimension during flatting is through folding (Harris, 1974: type 3 failure), as is 29 apparent in many (most?) acritarchs. Alternatively, the excess may be taken up through a plastic thickening of the walls (type 4 failure), a mode suggested here for the very likely quite plastic sheaths of cyanobacteria. Unfolded compression fossils might also deform plastically through lateral expansion (type 5 failure); however, the difficulty in achieving this behavior experimentally (Harris, 1974) and the closely comparable diameters of Siphonophvcus robustum and S. typicum n. comb, in Svanbergfjellet cherts and shales militate against such an interpretation. In any event, collapse due to decay (which does not involve lateral expansion) will significantly precede sedimentary compaction in most sedimentary regimes (Conway Morris, 1979; Briggs and Williams, 1981). Some Svanbergfjellet fossils do show evidence of compression-induced distortion. The robust, non-folding walls of Cerebrosphaera n. gen. typically accommodated flattening through the formation of radial fractures (Figs. 12.3- 12.6), a habit betraying the primary rigidity of its walls (see also, Dictvotidium colandrum n. sp.. Figs. 14.3,14.4) and accompanied by substantial lateral expansion (Harris, 1974; type 2 failure). In a comparable manner, the single radial split that characterizes many Leiosphaeridia wimanii n. comb, acritarchs results in an artificially elongated (ellipsoidal) compression fossil (Figs. 13.5, 13.6). Shape distortion may also occur in unsplit folded compressions, but the mean diameter is expected to approximate that of the original structure.

The points raised here are of course a minority of those that figure into the occurrence and final appearance of a fossil. Further study, especially of fossil Lagerstatten and their particular depositional, biogeochemical, and temporal contexts, should significantly enhance paleobiological resolution. In summary, a detailed understanding of taphonomic processes is necessary to 30

appreciate the form and degree of post-mortem information loss, to accurately assess that which remains, and to serve as a guide in the search for further fossil occurrences.

TAXONOMY

Systematic taxonomy is the framework supporting most biological and paleobiological investigation. Among living organisms it is based largely on degree of phenotypic similarity with a "type" (and its range of intra-specific variation), despite the widespread acceptance of the biological species concept. The same is true (necessarily) for fossil species, but here a "natural" classification may become obfuscated by taphonomic alteration and a lack of close modern analogues, especially in the case of Proterozoic forms. Insofar as we are interested in the fossil record as a guide to evolutionary history and paleobiology in general, every attempt should be made to look upon fossils as real organisms preserved in various stages of ontogeny and postmortem decay. Significant details of Proterozoic fossil biology can often be determined through a consideration of functional morphology and preserved 'behavior' (e.g., the coloniality of Brachvpleganon or the phototropism of Polvbessurus and Siphonophvcus), as well as the intra-spedfic variation revealed by large populations. Thus, on the basis of their unique processes and a clear morphological continuum, eight previously named 'spedes' of the acanthomorphic acritarch Trachvhvstrichosphaera are now recognized as developmental variants of a single biological entity, even though the higher- order taxonomy remains unclear. Conversely, a judidous recourse to type materials may help to resolve relatively meaningless 'wastebasket' taxa into well-defined groups (e.g., Chuaria/ Leiosphaeridia/ Cerebrosphaera n. gen.). In 31

a few instances, fossils share a sufficient complement of characters with living analogues that they can be treated under a fully natural and modem taxonomy, as is the case for the cyanobacteria Polvbessurus, Eoentophvsalis and Obruchevella, and the Svanbergfjellet siphonocladaleans Segregocladus n. gen. and Proterocladus n. gen. An artificial "form taxonomy" may be inavoidable for very simple Proterozoic fossils such as spheroids and filaments. Such treatment will clearly conflate disparate natural taxa (compare, for example, modem Oscillatoria with Beggiatoa. or Phormidium with Chloroflexus): however, broadly meaningful groups can often be defined (e.g., regularly recurring size-frequency distributions among Leiosphaeridia or Siphonophvcus - see systematic section). On the other hand, a taxonomic recognition of taphonomic variants needlessly over-estimates paleodiversity (Knoll and Golubic, 1979); the practice of treating silicified and shale-hosted fossils under separate taxonomies, for example, is now recognized as largely misleading (Pjatiletov, 1988; Jankauskas, 1989; Knoll et al., 1991). To be useful, fossil taxa must accommodate a degree of natural variation (e.g., Shukovsky and Halfen, 1976; Haxo et al., 1987) and taphonomy (e.g., Golubic and Barghoorn, 1977; Horodyski et al., 1977) comparable to that observed in their modem analogues. A relatively conservative taxonomic policy has been taken in the present study under the assumption that it will at least provide a reliable estimate of minimum diversity. Filamentous microfossils of prokaryotic aspect comprise one of a number of taxonomic quandaries in Proterozoic taxonomy; several hundred 'species' have been erected to describe a relatively limited range of morphology. Their abundance and diverse preservation in the Svanbergfjellet Formation offer an excellent opportunity to review their overall taxonomy. As strict form taxa, eight basic forms (genera) can be recognized: 32

1) Osdllatoriopsis - unsheathed cellular trichômes with cells equal to or wider than long; 2) Cvanonema - unsheathed cellular trichomes with cells longer than wide; 3) Veteronostocale - unsheathed cellular trichomes severely constricted at the intercellular septa; 4) Palaeolvngbva - cellular trichomes within a single outer sheath; 5) Tortunema - pseudo-septate filaments; 6) Cephalonvx - 'pseudo-cellular' filaments; 7) Rugosoopsis - cellular or acellular filaments with a double outer sheath, the outer one having a transverse fabric; and 8) Siphonophvcus - smooth-walled filamentous sheaths. Major synonymies and systematic revisions of these taxa are presented in the Systematic Paleontology section.

MULTICELLULARITY

The advent of eukaryotic multicellularity represents one of the fundamental advances in the history of life on Earth. How and why some organisms came to be constructed of numerous integrated units is an ongoing and fascinating subject of investigation (Sharp, 1934; Cavalier-Smith, 1978; Buss, 1987; Kauffman, 1987; Bonner, 1988; Kaplan and Hagemann, 1991). A significantly pre-Cambrian record of multicellular life is suggested by both molecular phylogenetic studies (Runnegar, 1982; Sogin, 1989) and basic evolutionary theory; however, the detailed paleontological evidence necessary to confirm and constrain this conjecture has remained largely elusive. The Svanbergfjellet shales (especially those of the Algal Dolomite Member) offer a new and substantially enhanced view of multicellular life 100-150 Ma prior to the Ediacaran-Cambrian radiations of large animals. In addition to fossils, a search for pre-Ediacaran multicellularity requires a strict definition of the term, and an appreciation of how such a habit might 33 be recognized based solely on fossil morphology. In the first instance, 'multicellular', as here applied, refers only to eukaryotic organisms; prokaryotes can indeed be constructed of multiple cells, but such interaction has never led to significant morphological differentiation or evolutionary innovation (Awramik and Valentine, 1985). Fundamentally unicellular organisms (prokaryotic or eukaryotic) that have simply agglomerated, or divided but failed to separate are likewise excluded from the discussion. Conversely, demonstrably coenocytic (i.e., multi-nucleate but non-septate) forms are included under the assumption that they represent either a primitive or derived state of multicellularity. Relatively large size may serve as an accessory criterion for multicellularity but is neither necessary or sufficient for identifying the condition. True multicellular fossils are expected to exhibit signs of cellular differentiation, intercellular communication, higher-order (i.e., emergent) structure, and/or morphological comparison with extant organisms that are accepted as being multicellular.

Simple Multicellularity.—Of the seven taxa of unambiguously multicellular organisms preserved in the Svanbergfjellet Formation, the simplest is Palaeastrum n. gen. (Figs. 5.1-5.3). At first glance simply an agglomeration of spheroidal uni cells, its distinct and regular differentiation of inter-cellular attachment structures (plaques) shows Palaeastrum to have had a colonial/coenobial grade of multicellular organization closely comparable to the coenobia of some extant chlorococcalean green algae (e.g., Pediastrum or Coelastrum - Fig. 5.4). In contrast, species of Germinosphaera are not multiple-celled and, strictly speaking, are classified among the Acritarcha. The random arrangement of their one to several open-ended and occasionally branched 34 tubular processes nevertheless points to an active growth habit, similar to that of germinating zoospores in some modern filamentous protists. The xanthophyte alga Vaucheria. for example, reproduces asexually by means of large (ca. 100 pm) spheroidal zoospores (actually multi-nucleate coenocytes) that germinate one or more filamentous, sometimes branched primordia (Fig. 16.7) in a fashion indistinguishable from that seen in G. fibrilla n. comb. (Fig. 17). The known (cf., Oomycota) and probable convergence upon this relatively simple habit precludes a positive assignment of germinosphaerids to any particular protistan lineage; their principal significance lies in documenting a fully coenocytic grade of multicellularity in the Svanbergfjellet assemblage.

Filamentous Chlorophyta (Green Algae).—Some Svanbergfjellet fossils stand as true multicellular protists without accompanying argument. The uniseriate filaments of both Proterocladus n. gen. and Segregocladus n. gen. are constructed of multiple cells separated by differentiated inter-cellular septa, express a higher order morphological complexity through lateral branching, and can be quite large (a specimen of P. major n. sp. extends for ca. 1 cm in bedding-parallel thin-section). Uniseriate cellular filaments are of course distributed widely among protistan-grade organisms; however, both of these fossils are distinguished by unusually long and markedly irregular cell lengths. In combination with various details of branching, septum formation, and apparent reproductive structures, such a pattern is particularly characteristic, indeed diagnostic of the modem siphonocladalean chlorophyte Cladophoropsis. Both Proterocladus and Segregocladus are consequently assigned to the Siphonocladales (cellular, multi-nucleate Ulvophyceae). Cladophoropsis owes its distinctive cellular habit to "segregative cell division" wherein the cytoplasm of an essentially coenocytic (i.e., multi- 35 nucleate) filament cleaves at various intervals, lays down an intervening septum, and initiates a single lateral branch at the uppermost end of the newly defined cell (Borgesen, 1913). Septum formation may occur without branching, and branches may themselves undergo further segregative cell division; however, a branch typically remains in full cytoplasmic communication with its parent cell, i.e., a septum is not formed below the branching point in the parental axis. This is very much the pattern observed in Segregocladus n. gen. In contrast, both species of Proterocladus n. gen. occasionally exhibit sub­ branch septum formation (Figs. 6.6, 7.1), and P. major n. sp. includes rare specimens with multiple lateral branches on a single cell (Figs. 6.2, 6.10) or with branches not immediately associated with a septum. In these respects, Proterocladus would appear to be more closely affiliated with the diverse branching habits of Cladophora (cf., van den Hoek, 1984). A distinction has been made between those chlorophytes that undergo segregative cell division (Siphonocladales) and the more regularly dividing (and divided) forms, the Cladophorales. Ultrastructural and molecular analyses now show these features to be non-diagnostic at the ordinal level (CXKelly and Floyd, 1984; van den Hoek, 1984; Zechman et al., 1990) and the Cladophorales is thus subsumed into the Siphonocladales (O'Kelly and Floyd, 1984: p. 135). As a consequence, the mosaic of Cladophoropsis and Cladophora characters exhibited by Proterocladus n. gen. become fully accommodated by the order. Indeed, van den Hoek (1984) argues, on morphological grounds, that Cladophoropsis and Cladophora are sufficiently alike to be merged as a single genus. In any event, the large cell volumes and clear siphonocladalean affiliations of both Segregocladus n. gen. and Proterocladus can be used to infer the multi-nucleate, semi-coenocytic nature of their respective cells. 36

Also deriving from the molecular data is evidence that the Class Ulvophyceae Mattox and Stewart, 1984 is polyphyletic (Zechman et ah, 1990). Interestingly, the monophyletic subset of the Ulvophyceae that includes the Siphonocladales is very much that of the Class "Bryopsidophyceae", which had been delineated largely on morphological criteria (Round, 1963). The possibility that non-cytoplasmic constituents can accurately reflect higher order taxonomy is especially reassuring to paleontologists.

Complex multicellularity.—The multicellular fossils discussed thus far are notable for their lack of significant cellular differentiation, surely the necessary prerequisite to the evolution of tissue-grade plants, animals, and, in some sense, fungi. Complex multicellular fossils are a conspicuous, if taxonomically problematic component of the Svanbergfjellet assemblage and provide a unique view onto the status of early intra-organismal specialization. The most common (and most complex) such form is Valkvria n. gen., a large (up to 1 mm long) sausage-shaped fossil with peculiarly lobate lateral 'axes', a medial stripe, and a variety of associated vesicular structures (Figs. 9-11); at least six discrete "cell types" are recognizably preserved (see systematic section). Using Bonner's (1988: p. 128) "number of cell types" as a measure of relative complexity or "grade of organization", Valkvria rates as at least as complex as the most differentiated modem algae or fungi. Moreover, it is not immediately clear whether some or all of these structures are differentiated single cells, or the product of a number of specialized cells. In the latter case they would then be of a tissue grade of organization and Valkvria would represent a level of complexity otherwise not recognized until the appearance of Ediacaran faunas, some 100 Ma into the future. 37

The problem of cell versus tissue recognition is also encountered in the problematic macrofossil Pseudotawuia n. gen. with its terminal pair of large reniform structures (Fig. 8.1). It is possible that the various components of Pseudotawuia represent single, albeit highly differentiated cells; however, the combination of both large size and apparent bilateral symmetry point intriguingly (although not conclusively) in the direction of tissue-grade complexity and metazoan affiliation. In this regard it is important to recall that most animal cells and tissues are not preservable as organic-walled fossils and that preservation will be limited largely to cuticle and other secreted, extracellular 'tissues' (Butterfield, 1990). The macroscopic 'problematicum' Tawuia occurs world-wide in late Proterozoic sediments (Hofmann, 1985a; 1985b), including those of the Svanbergfjellet Formation. It does not exhibit the localized differentiation of Pseudotawuia n. gen. (the vague terminal structures present in some Tawuia are mechanically induced; Hofmann, 1985a). However, close examination of well preserved Svanbergfjellet material shows it to have a distinctive bi-layered wall construction; a ca. 1 pm thick, opaque, brittle layer is lightly bonded to a ca. 3.5 pm thick, translucent and flexible inner(?) stratum (Figs. 8.4, 23.8). Along with its considerable size, such marked histological differentiation establishes Tawuia as a complex, and probably multicellular, organism. Absence of any preserved cellularity suggests that Tawuia cells were unwalled, or that the organism was fundamentally coenocytic; the very robust, and apparently continuous wall argues against a metazoan affiliation. 38

Other Proterozoic Occurrences

Multicellular fossils of varying degrees of taxonomic and temporal resolution are preserved elsewhere in the Proterozoic. Uniseriate and multiseriate filaments from the ca. 1250-725 Ma Hunting Formation in arctic Canada are identified unambiguously as bangiophyte red algae (Butterfield et al., 1990; Chapter IV); like the Svanbergfjellet siphonocladaleans, they are identified on the basis of diagnostic cell division patterns. The branched, largely aseptate filaments of Palaeovaucheria clavata Hermann. 1981(a) from the Late Riphean Lakhandin suite of Siberia are clearly multicellular (multi- nucleate) and are tentatively accepted as a Vaucheria-like alga (Chromophyta), despite the indistinguishable morphology of some Oomycota. Unbranched and apparently coenocytic filaments are known from ca. 1800 Ma deposits in China (Hofmann and Chen, 1981) and ca. 1400 Ma deposits in both China and Montana (Walter et al., 1976; Du et al, 1986; Walter et al., 1990), but their taxonomie affiliations remain speculative. The same is true for the widely distributed tawuids and longfengshanids of the pre-Vendian Neoproterozoic (Hofmann, 1985a; 1985b), and a variety of putative algal (Timofeev et al., 1976; Timofeev and Hermann, 1979; Jankauskas, 1989) and fungal (Hermann, 1979) remains in the Lakhandin sediments. Problematic bedding plane markings in middle Proterozoic sandstones of Western Australia (Grey and Williams, 1989) and Montana (Horodyski, 1982) are conceivably the imprints of a relatively large seaweed, and a possible mineralized metaphyte is reported from late Proterozoic carbonates in southeastern California (Horodyski and Mankiewicz, 1990). Fossils and ichnofossils of multicellular organisms of course abound in latest Proterozoic (Ediacaran) rocks, although probable algal forms are 39 relatively few and morphologically simple (e.g., Gnilovskaya, 1988; Zhang Y., 1989; Grant et al., 1991; Chen and Xiao, 1991).

Evolutionary Implications

While admittedly patchy, there are some interesting aspects to the pre- Ediacaran record of multicellular fossils. For example, of those forms that can be reasonably classified into extant taxonomic groups, all are algal. And, with the addition of the Svanbergi^ellet assemblage, there is now strong paleontological evidence for the presence of all three principal algal clades, the Rhodophyta (Butterfield et al., 1990; Chapter IV), Chromophyta (Hermann, 1981), and Chlorophyta (present study) by at least 750 Ma ago. Moreover, each of these three lineages had independently evolved a true multicellular organization by that time, although of a rather simple grade. Conspicuously absent is evidence of parenchymatous or pseudoparenchymatous construction and its inherent capacity for producing large and elaborate algal structures, such as appear in the early Paleozoic (e.g., Walcott, 1919). Indeed, even the simple, regularly septate filaments that dominate (at least numerically) most modem algal communities are rare in the Proterozoic, and entirely absent in the Svanbergfjellet assemblage. Although data are limited, the most common 'grade' of pre-Ediacaran multicellularity appears to have been either coenocytic (e.g., Palaeovaucheria. Germinosphaera. Tawuia, Longfengshania Du, 1982, Majaphvton Timofeev and Hermann, 1976, Archaeoclada Hermann, 1989, Variaclada Hermann, 1989, Caudosphaera Hermann and Timofeev, 1989) or semi-coenocytic (i.e., very large, multi-nucleate cells such as occur in Segregocladus n. gen. and Proterocladus n. gen.). The exceptions to this generalization include the agglomerative multicellularity of Palaeastrum n. sp. 40 and Eosaccharomvces Hermann, 1979, the intercalary cell division of the Canadian bangiophyte, (Butterfield et al., 1990; Chapter IV), and the possibly prokaryotic "Gunflintia" barghoomii, Mai thy, 1975, Trachythrichoides Hermann, 1976, and Lomentunella Hermann, 1981. Notably, all probable metaphytes of middle Proterozoic age appear to have been coenocytic and show no evidence of regular septation (e.g., Walter et al., 1976; Hofmann and Chen, 1981; Du et al., 1986; Walter et al., 1990). The possibility of an evolutionary sequence leading from coenocytic to multiple-celled multicellular organisms is particularly intriguing in light of recent (Kaplan and Hagemann, 1991) and not so recent (Sharp, 1934) inquiries into the applicability of conventional "cell theory" to the understanding of multicellularity. In essence, cell theory holds that a multicellular body is (both developmentally and evolutionarily) an aggregate of unicells, each of the same rank and the equivalent of a unicellular organism. The alternative "organismal theory" would argue that it is the whole 'multicellular' body that equates with a unicellular organism, and that large size and morphological differentiation are independent of, and are likely to have preceded cellularity. Such independence is supported by experimental studies wherein growth and development may proceed even when cell division has been suppressed (Sharp, 1934; Kaplan and Hagemann, 1991), and by the natural occurrence of diverse coenocytic organisms that converge upon complex, cellular-grade and tissue-grade forms (e.g., the coenocytic chlorophytes Brvopsis and Caulerpa). Early coenocytes would have generally come under various selective pressures for internal partitioning (e.g., mechanical support; Kaplan and Hagemann, 1991), hence the typical situation of large organisms being constructed of multiple, semi-independent units, the cells. Because a primitive cell division program is unlikely to have closely synchronized cytokinesis (septation) and 41 karyogenesis (mitosis), the cells of early multiple-celled multicellular organisms are expected, under this evolutionary scenario, to have been multi-nucleate and of irregular size (such as found in the Siphonocladales). Regular cellularity would then arise as developmental programs became more refined. As discussed above, the limited fossil record suggests that such refinements appeared relatively late in the Proterozoic. A coenocytic to cellular evolutionary sequence is supported by molecular phylogenetic analysis (rRNA) within at least one major branch of the green algae. The consensus phylogenetic tree of Zechman et al. (1990: fig. 3) shows that regularly septate Cladophora and Chaetomorpha form a sister group to irregularly septate Cladophoropsis, and that the Siphonocladales as a whole is sister group to the coenocytic Caulerpales and Dasycladales. Whether this pattern holds true for other multicellular lineages awaits further analysis in groups that retain both coenocytic and cellular forms. Among the fungi, for example, it is the regularly septate and strictly dikaryotic Basidiomycota that are thought to be the most highly derived, followed by the often irregularly multinucleate Ascomycota; the coenocytic Zygomycota and Chytridiomycota appear to be most primitive (Tehler, 1988; Bruns, 1991). Initial molecular analyses broadly support this evolutionary sequence (Bowman et al., 1992). There are, of course, other means of becoming multicellular. Inherently unicellular organisms can agglomerate and fuse to form a higher order structure (e.g., the cellular slime molds), or fail to separate following cell division to yield structured multicellular colonies (e.g. the coenobial green algae, including Volvox and the Svanbergfjellet fossil Palaeastrum n. gen.; Figs. 5.1-5.3). The question is whether such interactions ever gave rise to subsequent evolutionary innovations, as cell theory would hold, or whether they represent an independent and essentially dead-end strategy for increasing 42

size without a concomitant increase in complexity (Bonner, 1988). Particularly in the light of accruing fossil evidence, "it is not at all clear that these afford the key to the evolution of [multicellular] organisms in general" (Sharp, 1934: p. 22). Agglomeration/non-separation also appears to be the principal mode of "multi-cellularity" among prokaryotes (e.g., myxobacteria, cyanobacteria, actinomyctes) and may account for their relatively limited morphological evolution (cf., Awramik and Valentine, 1985). hietazoansi?).—The early and broadly contemporaneous fossil occurrence of rhodophyte, chromophyte, and chlorophyte algae is in keeping with various molecular phylogenetic analyses suggesting that these three clades diverged from a common ancestor at or around the same time (Perasso et al., 1989; Bhattacharya et al., 1990). Interestingly, the molecular data at this level of resolution do not readily distinguish the divergence of the metazoan clade(s) (Sogin, 1989; Hendriks et al., 1991); animals too are likely to have had a significant pre-Ediacaran history. With respect to the fossil record, putative pre-Ediacaran metazoans have been limited largely to various vermiform compressions in the late Proterozoic Liulaobei and Jiuliqiao Formations of China (Zheng, 1980; Wang, 1982; Sun et al., 1986), but absence of any diagnostic metazoan features precludes a final assessment of these fossils. The same is true for the Svanbergfjellet fossils Valkvria n. gen. and Pseudotawuia n. gen., although they do offer the additional feature of clearly differentiated cells or tissues (Figs. 8.1, 9-11). Because non-mineralized metazoan body fossils are rarely preserved and are subject to ambiguous interpretation, the search for early animal life has come to be directed at the potential trace fossil record, a program flawed on two counts. Firstly, the formation of sediment-displadng trace fossils would 43

imply the presence of a relatively sophisticated locomotory apparatus such as a hydraulic skeleton, yet the earliest animals were almost certain to have been acoelomate and relatively lethargic. Secondly, early energetic animals are likely to have been small meiofauna which have been shown to leave little or no perceptible sedimentary trace, at least in microbial mats (Farmer, 1989). The undisrupted lamination of the Svanbergfjellet shales (Fig. 3.5) thus does not necessarily imply an absence of metazoans and the more promising approach for documenting early metazoan history probably lies in the discovery and careful interpretation of body fossils.

ACRITARCHS

A significant proportion of the Svanbergfjellet fossils are solitary organic-walled vesicles of indeterminate taxonomic affinity—acritarchs. Most are likely to be the cysts, spores or vegetative unicells of various eukaryotic algae, although prokaryotes, protozoans, or even metazoans (e.g., Kuc, 1972) might also be represented. The most common acritarchs in the Svanbergfjellet Formation (as for the Proterozoic in general) are simple, relatively thin-walled spheroids (leiosphaerids) ranging in size from a few microns to over a millimeter in diameter; their classification is based on broad modalities in size frequency distribution (Fig. 4; Jankauskas, 1989). While of some practical value, such taxonomy is inherently artificial as indistinguishable vesicles would have derived such diverse forms as Germinosphaera. Osculosphaera n. gen., Pterospermopsimorpha. and Trachvhvstrichosphaera. When examined in situ (in bedding parallel thin section) leiosphaerids also exhibit diverse 'behaviors': where some populations are randomly distributed on bedding planes, implying a planktic origin, others occur as discrete, localized 44 populations, suggestive of benthic growth (Butterfield and Chandler, 1992; Chapter HI). A conspicuous fraction of Svanbergfjellet acritarchs feature processes or other distinctive surface ornamentation, and are therefore amenable to a considerably more precise, albeit 'form' taxonomy. Ten such morphologically complex taxa have been recorded in shales and cherts of the Lower Dolomite, Lower Limestone, and Algal Dolomite members. As in other Neoproterozoic successions, they typically occur in shallow, open-water environments (Butterfield and Chandler, 1992; Chapter III) with occasional transport into more restricted (e.g.. Fig. 15.3) or deeper-water facies.

B iostratigraphy

The fossil record serves its greatest practical role by providing information on the relative and, by extension, absolute ages of sedimentary sequences. Biostratigraphic zonation of the Proterozoic, however, has been frustrated by the extreme morphological conservatism of both prokaryotes and leiosphaerid acritarchs, the signature fossils of the Precambrian record. Distinctive, morphologically complex acritarchs are of relatively recent discovery, but promise a much enhanced resolution of at least Neoproterozoic time (Butterfield and Knoll, 1989; Knoll and Butterfield, 1989). The rich assortment of ornamented acritarchs in the Svanbergfjellet Formation contributes significantly to this growing record and corroborates and extends various stratigraphie trends seen elsewhere. One of the most distinctive and widely distributed of these late Proterozoic acritarchs is Trachvhvstrichosphaera aimika. now known from at least 14 localities worldwide, including the Svanbergfjellet Formation (see 45 systematic section). Without exception, it occurs in rocks of Late Riphean age and appears to be an excellent index fossil for the pre-Vendian Neoproterozoic. In silicified carbonate facies, T. aimika (= T. vidalii) regularly co-occurs with another distinctive acritarch, Cvmatiosphaeroides kullingii (Knoll and Calder, 1983; Knoll, 1984, Knoll et al., 1991; present study); their similar co-occurrence in the Tindir Group of the western Yukon (Allison and Awramik, 1990) argues compellingly against the purported early Cambrian or Vendian age of those sediments, in agreement with independent chemostratigraphic data (Kaufman et al., 1992). Trachvhystrichosphaera aimika also occurs in cherts and shales of the minimally 723 ± 3 Ma old Wynniatt Formation of arctic Canada (Butterfield and Rainbird, 1988; Heaman et al., 1990), thus providing an absolute age tiepoint for the taxon. As the Wynniatt and Svanbergfjellet also share similar species of Comasphaeridium and Germinosphaera, as well as distinctive S'^C and ®^Sr/®*Sr geochemical signatures (Asmerom et al. 1991), the two sequences can be considered closely correlative; the Svanbergfjellet Formation is likely to be ca. 700-750 Ma old. Ornamented acritarchs of pre-Ediacaran age are often very much larger than Paleozoic forms, typically measuring several hundreds of microns (vs. tens of microns) in diameter (Zang and Walter, 1989; Butterfield and Knoll, 1989). Thus, even in the absence of detailed taxonomy, this 'grade of organization' may serve as a coârse biostratigraphic marker (e.g.. Knoll and Ohta, 1988). It is not the case, however, that the smaller, 'Paleozoic-aspecf acritarchs are exclusively Paleozoic. The Svanbergfjellet assemblage includes species of Comasphaeridium. Dictvotidium. and Goniosphaeridium that would not appear out of place in Cambrian rocks. Their 'unusual' occurrence here is partly explained by the combination of exceptional preservation and the non- disruptive search procedure, i.e, in bedding-parallel thin section; even so, they 46

stand in marked contrast to their robust and widely recoverable Paleozoic counterparts. The full biostratigraphic potential of these and other distinctive Proterozoic acritarchs awaits comparable studies elsewhere (e.g., Butterfield and Rainbird, 1988), particularly as they may be corroborated by various chronometric (Heaman et al., 1990) and chemostratigraphic (Asmerom et al., 1991) techniques.

SYSTEMATIC PALEONTOLOGY

The Svanbergfjellet fossils are classifiable at various taxonomic levels and with varying degrees of confidence. Six broad categories are used in the following systematic treatment: 1) multicellular eukaryotes classified under m odem protistan taxonomy (p. 47); 2) multicellular eukaryotes - incertae sedis (p. 54); 3) unicellular eukaryotes - incertae sedis (= Acritarcha; p. 62); 4) unambiguous cyanobacteria (p. 93); 5) probable cyanobacteria (p. 98); and 6) overall incertae sedis, in which all higher-order taxonomy is uncertain (p. 130). An index of all fossil taxa is given in Appendix A.

All type and illustrated specimens are housed in the Paleobotanical Collections of the Harvard University Herbaria and assigned Harvard University Paleobotanical Collection (HUPC) numbers. 47

Domain EUCARYA Woese, Kandler, and Wheelis, 1990 Division CHLOROPHYTA Pascher, 1914 Class CHLOROPHYCEAE Kützing, 1845 Order CHLOROCOCCALES Fritsch, 1935 Genus PALAEASTRUM n. gen. Type species.—Palaeastrum dvptocranum n. sp. Diflgwosfs.—Colonial, spheroidal to ellipsoidal cells with prominent intercellular attachment discs; discs circular with a reinforced rim. Colonies monostromatic. Discussion.—Palaeastrum n. gen. differs fundamentally from simple pluricellular aggregates such as Ostiana (Figs. 5.6-5.9). The intercellular attachment discs are not simply the product of cell-cell contact but fully differentiated structures involved in the maintenance of colony structure. In many instances the fossil is represented almost solely by these distinctively arrayed discs (Fig. 5.2), the undifferentiated cell walls having considerably less preservation potential. The attachment discs and mode of colony formation in Palaeastrum n. gen. have close morphological analogues among extant chlorococcalean green algae. Pediastrum and Coelastrum. for example, form multicellular coenobia in which the cells are attached to one another by oriented and differentiated attachment 'plaques'. In Coelastrum, the plaques are thickened overall but particularly so at their perimeters (Fig. 5.4; Marchant, 1977), much as in the fossil. Interestingly, both Pediastrum and Coelastrum are characterized by walls containing sporopollenin (Atkinson et al., 1972; Marchant, 1977), a feature that has considerably enhanced their likelihood of preservation (Gray, 1960; Evitt, 1963a). 48

Palaeastrum n. gen. colonies are monostromatic and vary from ca. 15 cells (Fig. 5.1) to several lOO's of cells, traceable in bedding-parallel thin section for over 1 mm. Unless its colonies were very large and most of the specimens represent transported fragments, it is unlikely that Palaeastrum had a determined size (i.e., cell number). In contrast, the true coenobia of Pediastrum (monostromatic) and Coelastrum (spheroidal; Fig. 5.4) have a set number of cells. Palaeastrum is nevertheless sufficiently comparable to these modern forms to establish its similar grade of multicellular organization, and to infer its probable affiliation to the chlorococcalean chlorophytes. The integrated colonies of Eosaccharomvces from the Neoproterozoic of Siberia (Hermann, 1979) are broadly comparable to those of Palaeastrum. They differ in their lack of differentiated intercellular attachment structures, their distinctive 'streaming' habit, and larger average cell size. A single Svanbergfjellet 'colony' containing several intimately interconnected cells up to 70 pm across might be assignable to Eosaccharomvces (Fig. 5.5). Efymo/ogy.—From the Greek palaios - ancient and astron - star, with reference to its antiquity and the morphological comparison of the fossil with extant Coelastrum and Pediastrum.

PALAEASTRUM DYPTOCRANUM n. sp. Figures 5.1-5.3 Diagnosis.—A species of Palaeastrum with cells 15-25 pm diameter. Colonies monostromatic and of various sizes. Description.—Spheroidal to ellipsoidal cells 12-20 pm diameter (% = 16.7 pm; s.d. = 2.0 pm; n = 20) connected to one another by 3-6 (typically 4) attachment discs. Resultant colonies comprising lO's to lOO's of cells arranged in a single layer. Attachment discs robust, with a reinforced rim; 4-11 pm diameter (% = 49

6.7 |im; s.d. = 1.9 jam; n = 32). Extracellular sheath(s) absent. Etymology.—IrTom the Greek dypto - dive and kranos - helmet, with reference to the diving helmet appearance of individual cells. Material.—7 colonies from 3 shale horizons: 86-G-62, 86-G-61, B-2-2 Holotype.—HUPC 62708; Figure 5.1; Slide 86-G-62-46; England-Finder coordinates: M-48-1.

Class ULVOPHYCEAE Mattox and Stewart, 1984 Order SIPHONOCLADALES Oltmanns, 1904 Genus PROTEROCLADUS n. gen. Type species.—Pro terocl a dus major n. sp. Diflgwosis.—Multicellular, uniseriate and occasionally branched filaments with pronounced intercellular septa. Cells thin-walled, psilate and cylindrical; length highly variable, but typically much longer than wide. Branches usually subjacent to a septum in the primary axis and themselves often septate; sometimes isolated by a second, subjacent septum in the main axis. Apical terminations simply rounded or clavate. Discussion .--Proterocladus n. gen. has close morphological analogues among the extant Chlorophyta, in particular with the branching septate filaments of Cladophoropsis and Cladophora (Fig. 7.4). Some simple Rhodophyta have a broadly comparable morphology (e.g., Rhodochorton); however, the marked variance in Proterocladus cell length is especially reminiscent of that observed in Cladophoropsis, while its branching patterns and septal structure are comparable to those of Cladophora (these two extant chlorophytes are closely related members of the Siphonocladales; p. 35). The robust, two-dimensional septa of Proterocladus n. gen. show less variation in diameter than the thin-walled, often expanded cells and are 50 therefore used as the principal measure of filament width. A discrete size break at ca. 10 pm and subsidiary qualitative differences warrant the recognition of two species, P. major and P. minor. Proterocladus is distinguished from Segregocladus n. gen. by its substantially more robust and more frequent septa. Etymology.-From the Greek proteros - earlier and klados - branch, with reference to its branching habit and similarity to modem Cladophora.

PROTEROCLADUS MAJOR n. sp. Figures 6.1-6.10, 7.3 Diagnosis.—A species of Proterocladus with septa 10-30 pm diameter. Filaments typically constricted at septa. Description.—Multicellular, uniseriate, occasionally branched filaments. Cells cylindrical, 13-52 pm wide (% = 32 pm; s.d. = 8 pm; n = 162) and 52-731 pm long (x = 322 pm; s.d. = 143 pm; n = 77) (inclusion of specimens with only one septum expands the range in cell length to 52-990+ pm [x = 320+ pm; s.d. = 158+ pm; n = 153]). Septa circular and robust, 10-32 pm wide (% = 20; s.d. = 4; n = 158). Branches usually subjacent to a septum in the primary axis and sometimes fully isolated by a second septum; one to several branches per cell. Branch diameter 50-150% that of the 'primary^ axis (x = 80%). Apical terminations occasionally clavate. Discussion.—Proterocladus major n. sp. typically occurs as transported fragments of a few to ca. 20 cells but whole thalli were sure to have formed conspicuous tufts or algal carpets; one specimen (Fig. 6.7) extends for nearly 1 cm in bedding-parallel thin section. Another shows what appear to be six separate filaments radiating from the apex of a single cell (Fig. 6.10), reflecting its substantial capacity for dense colonial growth. 51

The septa of P. major n. sp. were occasionally preserved in parallel-to- bedding orientation where they can be confirmed as being circular (Figs. 6.1, 6.9). In one instance a septum appears to be perforated by a central pore (Fig. 6.1) inviting comparison with the primary pit connections of red algae, or with the enigmatic chlorophyte Smithsoniella Sears and Brawley, 1982. Another specimen, however, shows no such structure (Fig. 6.9), and the pore may be simply a reflection of a centripetal septum formation such as occurs in Cladophora, or a taphonomic feature. Etymology.—With reference to its large size. Material—162 cells measured from ca. 50 thallus fragments in shale horizon 86-G-62. Holotype.—HLSFC 62709; Figure 6.3; Slide 86-G-62-53; England-Finder coordinates; L-11-2.

PROTEROCLADUS MINOR n. sp. Figures 7.1-7.2 Diagnosis.—A species of Proterocladus with septa 3-7 pm diameter. Filaments rarely constricted at septa. Descripfzon.—Multicellular, uniseriate, occasionally branched filaments. Cells cylindrical, 5-12 pm wide (x = 6.8 pm; s.d. = 1.5 pm; n = 20) and 30-160 pm long (x = 91 pm; s.d. = 40 pm; n = 12) (inclusion of specimens with only one septum expands the range in cell length to 30-175+ pm [% = 94+ pm; s.d. =41+ pm; n = 20]). Septa robust, 3-7 pm wide (x = 5.4 pm; s.d. = 1.0 pm; n = 19). Branches subjacent to a septum in the primary axis and usually isolated by a second, subjacent septum. Branches approximately the same diameter as the primary axis. Cell contents sometimes preserved. Apical terminations rounded. 52

Discussion.—In addition to its smaller dimensions, Proterocladus minor n. sp. is distinguished from P. major n. sp. by an absence of significant constrictions at septa and the approximately equal diameters of branches and primary axes. Like Segregocladus hermannae n. sp., P. minor occasionally preserves cell contents as dark, elongate inclusions; it is distinguished from S. hermannae by its prominent and relatively numerous septa. Efymo/ogy.—With reference to its small size relative to the type species. Material—20 cells measured from 5 thalli in shale horizon 86-G-62. Holotype.—HUFC 62710; Figure 7.1; Slide 86-G-62-15M; England-Finder coordinates: N-24-0.

Genus SEGREGOCLADUS n. gen. Type specfgs.—Segregocladus hermannae n. sp. Diagnosis.—Uniseriate, occasionally branched filaments with infrequent and insubstantial septa. Cells much longer than wide. Branches subjacent to a single septum in the parental axis and freely communicating with the parental cell; maximally, one branch per cell. Some septa not associated with branches. Rare conical protrusions in the lateral wall. Discwssiow.—The branches of Segregocladus n. gen. always occur immediately subjacent to a septum in the primary axis (Fig. 7.7), but without a second septum that would isolate it from the parental cell. In combination with its infrequent and irregularly spaced septa such a pattern is indistinguishable from that arising from the "segregative cell division" of the modem siphonocladalean green alga, Cladophoropsis (p. 34). The only lateral outgrowth in Segregocladus not obviously associated with a septum is a single conical protrusion (Fig. 7.6); it is indistinguishable from the zoospore release structures that occur on Cladophoropsis (Borgesen, 1913: p. 48, fig. 33; O'Kelly 53 and Floyd, 1984). Together, these various cellular features permit Segregocladus to be assigned with some confidence to the Siphonocladales. Segregocladus n. gen. differs from Proterocladus n. gen. by its less frequent and considerably less robust septa. The branches of Palaeovaucheria Hermann (1981a) are not accompanied by septa. Etymology—From, the Latin, segrego - set apart and Greek, klados - branch, with respect to the inferred segregative cell division of the branched fossil.

SEGREGOCLADUS HERMANNAE n. sp. Figures 73-1.7 Diagnosis.—A species of Segregocladus with filaments 7-14 pm diameter. Description.—Dnisenate, occasionally branched, thin-walled filaments, 7-14 pm diameter. Septa uncommon and insubstantial; usually associated with branches. Maximally, one branch per cell which diverges at right angles from the 'primary' axis; branches with a slight basal constriction but otherwise of approximately the same diameter as the parent axis. Rare, conical protrusions in the lateral wall. Cell contents sometimes preserved as dark rod-shaped inclusions. Apical terminations simply rounded. Etymology.—In honor of Dr. Tamara Hermann for her pioneering work in Proterozoic paleontology. Material—A single thallus from shale horizon 86-G-62. Holotype.—HUPC 62711; Figure 6.5; Slide 86-G-62-93M; England- Finder coordinates: P-44-0. 54

INCERTAE SEDIS PSEUDOTAWUIA n. gen. Type spgcies.—Pseudotawuia birenifera n. sp. Dfflgwos/s.—Thin-walled tomaculate macrofossil with a terminal pair of dark reniform structures. Etymology.—With respect to its superficial resemblance to T

PSEUDOTAWUIA BIRENIFERA n. sp. Figure 8.1 Diagnosis.—A species of Pseudotawuia ca. 2 mm wide and ca. 1 cm long. DescrfpfiOM.--Thin-walled organic film, 1.84 mm wide and 9.45+ mm long, bearing a bilaterally(?) symmetrical pair of dark, ca. 0.75 mm long reniform structures at one end. DfscMssfow.—Pseudotawuia birenifera n. sp. differs conspicuously from Tawuia dalensis Hofmann, 1979. It is defined by a thin organic film that has left no sedimentary imprint (co-occurring Cerebrosphaera n. gen. leave prominent imprints) and, at one end, bears a symmetrical pair of dark reniform structures. These are unrelated to the mechanically induced 'disc-like terminal structures' on some of the Little Dal Tawuia (cf., Hofmann, 1985a), and, if they are indeed bilaterally symmetrical, represent a grade of organization generally thought to be limited to Ediacaran and younger organisms. The combination of thin-walled cuticle, elongate shape, and paired terminal structures in P. birenifera invites comparison with simple vermiform metazoa; however, the absence of any unambiguous metazoan features requires that it be classified as incertae sedis with both its physiology and higher order taxonomy undetermined. 55

Etymology.—VroTn the Latin bi - two, ren - kidney and fero - bear, with reference to the paired reniform structures. Material—A single specimen from shale horizon 86-G-30 Holotype.—HUFC 62741; Shale specimen 86-G-30-3BP.

Genus TAWUIA Hofmann, 1979

Pumilibaxa ZHENG, 1980, p. 61. Arciforma Y AN, 1984 (in Wang, 1984), p. 79 (nom. nud.). Baculiformis YAN, 1984 (in Wang, 1984), p. 79 (nom. nud.). Cylindraceusa YAN, 1984 (in Wang, 1984), p. 79 (nom. nud.). Ephippiodeusa YAN, 1984 (in Wang, 1984), p. 79 (nom. nud.). Fusiphvsa YAN, 1984 (in Wang, 1984), p. 79 (nom. nud.). Valvaphvsa YAN, 1984 (in Wang, 1984), p. 79 (nom. nud.). Bipatinella ZHENG, 1984 (in Wang, 1984), 79 (nom. nud.). Claviforma ZHENG, 1984 (in Wang, 1984), p. 79 (nom. nud.). Conicina ZHENG, 1984 (in Wang, 1984), p. 79 (nom. nud.). Fengyangella ZHENG, 1984 (in Wang, 1984), p. 79 (nom. nud.). Linguiformis ZHENG, 1984 (in Wang, 1984), p. 78 (nom. nud.). Liulaobeia ZHOU, 1984 (in Wang, 1984), p. 79 (nom. nud.). Glossophyton DUAN AND DU, 1985, (in Xing et al., 1985) p. 72. Mesonactus FU, 1989, pp. 74, 77.

Tachymacrus FU, 1989, pp. 74, 77. 5fa

Type species.—Tawuia dalensis Hofmann, 1979, p. 158-160. Discussion.—Most thick-walled, more or less sausage-shaped (tomaculate) Proterozoic macrofossils can be reliably assigned to the form genus Tawuia. The principal difficulty in its identification lies in an apparent morphological gradation into ellipsoidal and circular forms. Thus, while Hofmann (1985a; 1985b) considers genera such as Shouhsienia and Ellipsophvsa to be short T. dalensis, Zhang R. et al. (1991) prefer to retain these ellipsoidal fossils as a distinct form. It is worth noting in this regard that the distortion of a spheroid under compression will often yield an ellipsoid with a lengthrwidth ratio of ca. 1.5 (Harris, 1974: text-figs. 5.C-5.H, 6); or considerably more if the vesicle is split (Harris, 1974: p. 139). Unlike Tawuia, Shouhsienia tends to have a single terminal split, suggesting that much of its aspect ratio is taphonomically induced and that its taxonomic affiliation is with the spheroidal acritarchs, e.g., Leiosphaeridia wimanii n. comb. (Figs. 13.5-13.6). On the other hand, unsplit ellipsoids with a length:width ratio as small as 2.5 (Fig. 8.4-8.5) can, on independent structural grounds, be confidently ascribed to Tawuia (see below). A single species of Tawuia is recognized by Hofmann (1985a); others may exist but they await rigorous documentation. Like T. siniensis Duan, 1982, Svanbergfjellet Tawuia are markedly smaller and less curved than the mean/mode of the Little Dal type material; however, the size range and habit of both these populations falls entirely within that of T. dalensis (Hofmann, 1985a), thus leaving the putative second species without diagnostic (i.e., unique) characters. In view of the relatively small numbers of Svanbergfjellet Tawuia and the difficulty in assessing discrete species from the literature, only its generic-level synonymy is considered here. 57

TAWUIA DALENSIS Hofmann, 1979 Figures 8.2-S.5, 23.8 Description.—Large tomaculate fossils with more or less parallel sides and rounded ends. Complete length 3.07-8.57 mm (x = 4.96 mm; n = 3); width 0.94-1.90 mm (% = 1.39 mm; s.d. = 0.30 mm; w = 11); length:width ratio 2.5-6.0 (% = 3.7; n = 3); occasionally somewhat narrower medially, and/or towards one end. Wall bi-layered: 'primary' outer(?) layer opaque, brittle, and minimally 1 pm thick; 'secondary' inner(?) layer translucent, flexible, and ca. 3.5 pm thick (possibly 3.5/2 = 1.75 pm thick). Walls typically with transverse wrinkles. Primary splits absent.

D jscmss/o « .“ The carbonaceous walls of Svanbergfjellet T. dalensis are very well preserved and can be freed from their shale matrix by careful acid maceration. SEM (Fig. 23.8) and transmitted light microscopy (Fig. 8.4) of a number of these isolated specimens reveals a compound wall structure: a 1 pm thick, brittle and opaque wall is lightly bonded to a 3.5 pm thick, but remarkably flexible translucent layer. This translucent layer appears to have been internal to the dark one (it was not found on the back side of the specimen in Fig. 8.3 after add dissolution of the shale matrix) and may thus represent the two fused sides of a collapsed inner cylinder; it nevertheless comprises the bulk of a Tawuia fossil and is capable of maintaining its form, even in the absence of the more conspicuous opaque wall (Fig. 8.4). The substantial overall thickness of T. dalensis (minimally 5.5 pm) is comparable to that of Chuaria drcularis and undoubtedly contributes to its similar occurrence as imprints and compressions in diverse sediments. The observation of two discrete wall layers contributes significant new detail in delineating T. dalensis morphology, but little towards a resolution of its overall biology. If, however, the outer wall can be shown to be 58 fundamentally opaque (p. 27), then it would be unlikely that T. dalensis carried on a photosynthetic metabolism, at least in this particular habit. Its large size and complex histology suggests that it was a true multicellular, possibly coenocytic, organism; otherwise, its taxonomic affiliations remain obscure. Svanbergfjellet T. dalensis are commonly associated with large discoidal (spheroidal) fossils. Some of these are possibly Chuaria (although not Ç drcularis), but many are clearly Cerebrosphaera n. gen.; for example, the ca. 420 pm diameter fossil just off the end of the T. dalensis specimen in Figure 8.2. Whether these co-occurrences represent a meaningful association (i.e. a 'Chuaria-Tawuia assemblage') is uncertain due to small sample size; Cerebrosphaera is certainly far more widespread than Tawuia in the Svanbergfjellet shales. Material-16 specimens (including 5 counterparts) from shale horizons 86-G- 62 and 86-G-61. 4 specimens isolated by add maceration: 2 mounted on glass slides, 2 for SEM.

VALKYRIA n. gen. Type species.~Valkvria borealis n. sp. Diflgnosfs.—Thin-walled tubular bodies with rounded ends that bear lateral extensions. Lateral axes unirambus or branched; attached only to the central portion of the main body. Circum-terminal dark/opaque discs and a vaguely defined medial stripe are common; a large, centrally-positioned sac-like structure and a sub-terminal partition in the main body occasionally present. Multiple spedmens may diverge radially from a dark central body. Dfscussiow.—That Valkvria n. gen. was a complex, multicellular organism is uncontroversial; however, a survey of possible modem analogues yields no 59 bauplan-levei comparisons. The least sensational, though not obviously correct interpretation would ally it to some of the complex algae that develop both axial filaments of "unlimited" growth and lateral filaments of limited or determinate growth: e.g., Batrachospermum (Rhodophyta), Draparnaldia, Drapamaldiopsis (Chlorophyta, Chaetophorales), Dasvcladus, Batophora (Chlorophyta, Dasycladales). These modem forms are nevertheless distinct in that they are permanently attached to a substrate, have their lateral axes regularly arranged in whorls on the main axis, and bear their reproductive structures in or on the lateral axes. They furthermore lack structures that would preserve as a medial stripe or the various vesicular constituents of Valkvria, and they are all fundamentally larger than the fossil. Species of Valonia (Siphonocladales) satisfy some of these concerns (e.g., smaller size, irregular lateral axes), but otherwise offer no more tenable a comparison. Etymology.—After the Valkyries of Norse mythology; the armed warrior- maidens of Odin, choosers of the slain.

VALKYRIA BOREALIS n. sp. Figures 9.1-9.5, 10.1-10.8,11 Diagnosis.—A species of Valkvria with the central body 100-1000 pm long and 20-200 pm wide.

Description .-Thin-walled tubular bodies with rounded ends 164-930 pm long (% = 465 pm; s.d. = 200 pm; n = 25) and 25-182 pm wide (x = 95 pm; s.d. = 29 pm; n = 79), that bear 1-14 lateral extensions; entire specimens with a mean aspect ratio of 5.5:1 {s.d. = 1.2:1; n = 25). Lateral axes occur only on the central portion of the main body, never the ends, with the attachment points marked by prominent circular scars, 5-34 pm diameter (% = 17 pm; s.d. = 5 pm; n = 78). Lateral axes typically of a lobate construction; hollow and apparently non- 60 septate; uniramous or branched. The main body commonly bears one to several ca. 30 pm wide, circum-terminal dark/opaque circular bodies (present in 41 of 85 specimens) and a more or less vaguely defined longitudinal medial stripe (present in 24 specimens); a number of specimens have a terminal portion of their main body separated by a prominent partition. Three specimens bear a large, centrally-positioned thick-walled vesicle and associated structures. In one instance, three 'whole' specimens of V. borealis n. sp. diverge radially from a large (ca. 450 pm diameter), ill-defined opaque body. A number of specimens preserve entangled Siphonophvcus septatum within the main body. P/scMssfow.—Valkvria borealis n. sp. preserves at least six distinct cell (or tissue) types; 1) the thin-walled, often lightly shagreenate main body; 2) hollow, uniramous or branched (Figs. 10.1,10.4, 10.6,10.7) lateral axes, typically of a lobate construction (Figs 9.3-9.5) but occasionally more straight- walled (Figs. 10.4,10.7); 3) dark to opaque circular bodies (spheroids?) that occur near one or both ends of the main body (Figs. 9.1, 9.3-9.5, 10.2, 10.3); 4) a longitudinal stripe within the main body which sometimes appears to terminate at one of the dark, circum-terminal discs (Figs. 9.1-9.2, 10.3); 5) a large, centrally-positioned, thick-walled vesicle that may occupy a substantial proportion of the main body; one end this vesicle appears to differentiate a narrow finger-like projection (Figs. 10.1,10.5) and the other a curved discoidaK?) structure (Fig. 10.5); and 6) a large, poorly defined opaque ovoid with a radial texture to which several V. borealis 'individuals' may be attached at one end (Fig. 11). These six of course represent only those 'cell types' that were both preservable and physically contiguous with recognizable Valkvria fossils; the true number must have been higher (e.g., gametes, alternate generations, etc.). 61

Most V. borealis n. sp. are 'entire' and lack obvious attachment structures, thus suggesting they led a free-floating, planktic existence. On the other hand, an occurrence of ca. 12 specimens localized in ca. 25 mm^ of a single bedding-parallel thin section points to their at least occasionally gregarious habit. Three of these specimens are attached at one end to a large, poorly defined opaque ovoid (Fig. 11) which conceivably represents a germinating or developing reproductive body. This gregarious population is further characterized by a prevalence of individuals in which the distal (in the case of the attached specimens) ca. Vs of the main body is separated by a prominent crosswall and contains elevated amounts of condensed organic matter (Figs. 10.2,11). Fifteen of the 85 measured specimens of V. borealis n. sp. contain substantial accumulations of Siphonophvcus septatum filaments within the main body and, in one instance, appear to extend into a lateral axis (Fig. 10.8). Their occurrence here is clearly not just happenstance and they possibly represent degrading heterotrophic microbes. Alternatively, the filaments may have been symbiotic within (or parasitic upon) the larger organism. Etymology.—From the Latin borealis - northern, with reference to the high latitude of the fossil occurrence. Material.—80 specimens isolated by add maceration and mounted on glass slides; 5 individual spedmens and a 'colony' consisting of ca. 12 individuals identified in bedding-parallel thin section. All from shale horizon 86-G-62. Holotype.-HUPC 62712; Figure 9.1; Slide 86-G-62-5M; England-Finder coordinates: L-29-1. 62

Group ACRITARCHA Evitt, 1963 Genus CEREBROSPHAERA n. gen. Type species.—Cerebrosphaera buickii n. sp. Diflgwosfs.—Spheroidal vesicles with regularly and prominently wrinkled walls. Wrinkles sinuous: anastomosing, interfingering, or rarely sub-parallel, but never intersecting. Vesicle walls ca. 1.5 pm thick, inelastic, and often opaque. Enveloping thin-walled sheath sometimes present.

D/ scmssio «.—The diagnostic wrinkling of Cerebrosphaera n. gen. walls is possibly taphonomic, but if so, is derived from a unique underlying wall construction. Its widespread occurrence in Svanbergfjellet shales and common co-occurrence with typical (i.e., unwrinkled) leiosphaerids rules out localized taphonomy as an explanation for its texture. Cerebrosphaera is further distinguished from leiosphaerids by an enveloping sheath, and its thick, inelastic walls. This latter quality is expressed in its lack of secondary folds and a proclivity for radial fracture under sedimentary compaction (Figs. 12.3, 12.4-12.6; cf., Harris, 1974: type 2 failure). The cerebroid wrinkling of Cerebrosphaera walls clearly preceded their fracture, suggesting that the wrinkles may in fact have been primary. Small (15-150 pm diameter), radially split, and usually opaque spheroidal microfossils with a rough surface have previously been classified as Turuchanica Rudavskaja. The recent subsumption of Turuchanica into Leiosphaeridia by Jankauskas (1989), and, more importantly, the absence of regular wrinkling in Turuchanica. militates against its comparison with Cerebrosphaera n. gen. More reminiscent is a population of thick-walled, 80- 500 pm diameter, and occasionally ensheathed spheroids described as Chuaria globosa Ogurtsova and Sergeev, 1989, from the Late Riphean of southern Kazakhstan. These silicified fossils are difficult to compare with the 63

Svanbergfjellet compressions, but they appear to have similarly convoluted walls; their three-dimensional preservation further suggests that the wrinkles were primary or a very early diagenetic feature. Etymology .—’From the Latin cerebrum - brain and sphaera - ball, with reference to the cerebroid convolutions of its walls.

CEREBROSPHAERA BUICKII n. sp. Figures 12.1-12.8

Leiosphaeridia sp. cf. L. atava. KNOLL, SWETT AND MARK, 1991, p. 558, fig. 21.2-21.3.

Diagnosis.—A species of Cerebrosphaera between 100 and 1000 pm diameter. Descr/pf/ow.—Spheroidal microfossils, 100-960 pm diameter = 365 pm, s.d. = 151 pm; n = 115), with the primary wall prominently and regularly convoluted. Wrinkles sinuous; anastomosing, interfingering, or rarely sub-parallel, but never intersecting. Psilate walls ca. 1.5 pm thick, often opaque, and relatively inelastic; radial fractures common; secondary folds absent. Thin-walled, leiosphaerid-like enveloping sheath occasionally present. Discussion .—Cerebrosphaera buickii n. sp. is conspicuous in most of the fossiliferous Svanbergfjellet shales, as well as in the immediately overlying Draken Formation (Knoll et al., 1991). Its distinctive wrinkling pattern allows the positive identification of even isolated fragments (Fig. 12.2). Reflected light (Figs. 12.1,12.5) or SEM (Figs. 12.3,12.7, 12.8) may be necessary to recognize opaque material, while transmitted light reveals the nature of the thin structureless envelope (Fig 12.4). 64

The primary wall of Ç. buickii n. sp. may be translucent when isolated in single layers (Fig. 12.2), but whole specimens are typically opaque as the two sides of the spheroid double the overall thickness (Figs. 12.4, 12.6). Under SEM the substantial thickness of Ç. buickii wall can be measured directly (Fig. 12.8), or by halving the width of the tightest wrinkles (Fig. 12.7); both methods indicate a single wall thickness of ca. 1.5 pm. In contrast, the walls of Chuaria circularis are 2-3 pm thick yet were originally much more pliant (i.e., non­ fracturing). The SEM of the fractured Ç. buickii specimen in Fig. 12.8 further reveals numerous 2-3 pm diameter, rounded perforations, possibly the result of heterotrophic activity. Etymology.—In recognition of the unrelenting advice and skepticism of Dr. Roger Buick, who first identified these fossils as pyrite framboids. He subsequently suggested the generic name. Material.~120 specimens from 5 shale horizons: P-2945; 86-G-33; 86-G-62, 86- G-61, 86-G-28, 86-G-30; 12 in bedding-parallel thin section, 6 as exposed bedding-plane compressions. Of 102 isolated specimens, 18 are mounted for SEM, 84 on glass slides for light microscopy. Holotype.—HUPC 62713; Figure 12.4-5; Slide P-2945-47M; England-Finder coordinates: S-33-4.

Genus CHUARIA Walcott, 1899

Fermoria CFIAPMAN, 1935, p. 114-115. Protobollela CHAPMAN, 1935, p. 117. Non Chuaria wimani BROTZEN, 1941, p. 258-259. (= Leiosphaeridia). Nec Chuaria globosa OGURTSOVA AND SERGEEV, 1989, p. 121. (= Cerebrosphaera?). 65

Type species.—Chuaria circularis, Walcott, 1899, p. 234-235. Diagwosis.—Thick-walled (single wall > 2 pm thick) spheroidal vesicles up to 5000 pm diameter. Walls psilate, opaque. Radial splits absent. Envelope absent. P/scMssfo».—Chuaria has been widely discussed and variously interpreted, major reviews being offered by Ford and Breed (1973) and Hofmann (1985a; 1985b). Duan (1982) and Sun (1987a) notwithstanding, the consensus view of Chuaria is of a hollow, organic-walled spheroid of unknown but probably algal affinities; i.e. an acritarch. This interpretation is supported by the Svanbergfjellet material. An unambiguous Ç circularis specimen (Fig. 13.7 - SEM, Fig. 13.8 - reflected light) reveals in cross-section (Fig. 13.9) the two well- defined sides of a compressed vesicle separated by an internal space. In its current usage Chuaria has come to serve simply as a (form) taxonomic pigeonhole for all relatively large, spheroidal acritarchs (Jankauskas, 1989). While such application may be useful for a broad characterization of Proterozoic fossil assemblages, it is clear that a disparate array of natural taxa are unnecessarily subsumed into the genus. Large (ca. 1 mm) spheroidal acritarchs in the Svanbergfjellet Formation can be classified into at least three readily distinguishable genera: Leiosphaeridia. Cerebrosphaera and Chuaria. Only the latter two have walls substantial enough to leave a significant sedimentary imprint. Chuaria is further distinguished from Cerebrosphaera by its thicker, entirely opaque, and more pliant walls, its simple folds, and its lack of an enveloping sheath. Perhaps the most reliable taxonomic feature of Chuaria is its substantial wall thickness. Unfortunately, this character is usually given in relative terms or, if quantified, the method of measurement left unstated. SEM of well preserved, extractable material from the upper Svanbergfjellet shales permits 66 its direct and unambiguous measurement. For example, the 770 pm C. circularis mentioned above (Fig. 13.7-13.8) has a single wall thickness of 2.1 pm (Fig. 13.9), while fragments of a larger, 3.5 mm diameter specimen have single walls ca. 3.0 pm thick. Both the above measurements were taken from very well preserved specimens with uneroded, psilate surfaces. Most fossils, however, will have been subject to considerably greater degradation such that preserved wall thickness may be substantially reduced, or lost altogether (e.g., Gussow, 1973). In these cases vesicle wall thickness and folding pattern can often be more reliably inferred from an examination of the enclosing sediments. During burial, the 4-6 pm thick Chuaria (up to at least 12 pm thick after folding; Harris, 1974: text-fig. 5) interfered profoundly with the otherwise smooth bedding planes. The resulting imprints bear clear witness to the robust constitution of the interred vesicle, even in the absence of the original wall material (Fig. 8.7-S.8; Walcott, 1899; Ford and Breed, 1973). Conversely, the absence of significant sedimentary molding is reasonable grounds for questioning an assignment to Chuaria. Thinner-walled Leiosphaeridia wimanii n. comb., for example, leave a much less substantial imprint on enclosing sediments (Fig. 13.6; see below), while the large (4.05 mm) spheroidal fossil in Figure 8.6 is preserved solely as a thin organic film. The walls of all unambiguous Ç circularis in the Svanbergfjellet Formation are opaque to the transmitted light of a standard laboratory microscope. As discussed in the text (p. 28), this is not a direct function of wall thickness, nor can it be ascribed to differential taphonomy; rather, it appears to be an inherent property of Chuaria wall structure. Opacity is also characteristic of the Chuaria type material (Walcott, 1899: p. 234) and conspedfic Fermoria (Chapman, 1935: p. Ill), hence their early comparison 67 with various phosphatic fossils (Ford and Breed, 1973). No such confusion attended L. wimanii n. comb. (= Ç wimani) which is fundamentally translucent. In the absence of any convincing data to the contrary, Chuaria is currently represented by a single species, Ç circularis, although future analysis may yield additional well-defined groups. As with any taxonomic assignment, identification of a Chuaria species requires the positive documentation of diagnostic characters present in type materials; specimens lacking such features are best recorded as Chuaria sp. Thus, it is not always clear that fossils reported as Chuaria circularis are in fact resolvable to the species level (e.g., material reported by Hofmann, 1977), and it is indeed possible that the Chuaria of the oft-cited Chuaria-Tawuia assemblage (e.g., Duan, 1982; Hofmann, 1985a; 1985b) may not in fact be Chuaria circularis. Chuaria wimani Brotzen, 1941 is clearly misplaced in the genus (see Leiosphaeridia wimanii n. comb.), and Ç. globosa Ogurtsova and Sergeev, 1989 is described as having a sheath, unlike the type species.

CHUARIA CIRCULARIS Walcott, 1899 Figures 8.7, 8.8,13.7-13.9

Chuaria circularis. FORD AND BREED, 1973, p. 539-542 (partim). Chuaria annularis ZHENG, 1980, p. 59-60, PI. 1, fig. 3-4. Chuaria minima MATTHY AND SHUKLA, 1984, p. 146-148 (partim). Chuaria multirugosa DU. 1985, p. 70, PI. 16, fig. 3 (in Xing et al., 1985). Shouhsienia multirugosa. DU AND TIAN, 1985, PI. 1, fig. 16-17. 68

Diagnosis.—A species of Chuaria with well defined concentric wrinkles or folds around its periphery; 400-5000 pm diameter. Description.—Flattened spheroidal fossils preserved as very thick-walled (2-3 pm single-wall thickness), opaque, kerogenous vesicles, or as imprints/molds on shale bedding planes, 0.43-3.50 mm diameter {x = 1.76 mm; s.d. = 0.93 mm; n = 20). Concentric wrinkles/folds around the periphery with sufficient relief to leave a substantial imprint on bedding planes. Surfaces psilate. Fractures, where present, tangentially rather than radially oriented. Envelope absent. Discussion.—The circumscription of Ç circularis has been variously discussed in terms of its overall diameter, wrinkling patterns, and wall thickness. Vidal and Ford (1985) further suggested that preservational mode, in this case as add resistant kerogen, be applied as a diagnostic character. While certainly useful in addressing the question of biogenicity, this latter condition cannot be applied to the diagnosis of biologically meaningful fossil spedes (Sun, 1987b: p. 351), and, in any event, the original carbonaceous constitution of Ç drcularis has already been diagnosed at the 'Group' (Acritarcha) level. In the Svanbergfjellet shales Ç drcularis occurs most commonly as bedding plane imprints/ compressions with the retention of various proportions of the carbonaceous wall (Fig. 8.7-8.8), sometimes induding an 'organic stain' (Fig. 8.8). These structures are unquestionably biogenic and reliably assigned to Ç drcularis, but they are not add resistant. Concentric peripheral folding is certainly not unique to Ç. drcularis but it is a prominent feature of the type material and offers an unambiguous means for subdividing the genus. Spedmens lacking such folds may constitute a separate natural spedes (although they are sure to be absent in three- dimensionally preserved Ç drcularis. if such exists). For compression fossils, this characteristic folding may reveal something of the original nature of the 69

walls, with a complete absence of folds suggesting a considerably more plastic wall than in Ç circularis (c.f., Harris, 1974: type 4 or 5 failure). In any event, Ç circularis differs fundamentally from both Cerebrosphaera buickii, in which the thinner walls tend to fracture rather than fold, and Leiosphaeridia wimanii n. comb., where the thinner and translucent walls exhibit a pattern of folding no more regularly concentric than that found in smaller leiosphaerids. Chuaria circularis is generally accepted as being less than 5 mm in diameter, however the lower size limit is not all clear. Ford and Breed (1973) have suggested an arbitrary lower size limit of 0.5 mm, and Jankauskas et al. (1987) one of 300 pm (subsequently altered to 1000 pm; Jankauskas, 1989), while Vidal and Ford (1985) have included specimens as small as 70 pm in the species. The Svanberg^ellet population of unambiguous Ç. circularis comprises 20 specimens ranging from 0.43 mm to 3.50 mm in diameter, with a mean of 1.81 mm. This compares well with the type material and I suggest a lower size limit of 400 pm for the species. Thick-walled (or at least opaque) spheroids as small as 190 pm across also occur in the Svanbergfjellet shales but are distinguished from Ç circularis by their lack of concentric wrinkling. If these forms can be shown to be Chuaria, they await rigorous definition as a species other than circularis. Material—20 specimens (19 on bedding planes, 1 isolated and mounted for SEM) from 4 shale horizons: 86-G-28, 86-G-33, 86-G-61, 86-G-62. An additional 75 specimens lack various features diagnostic of Ç circularis and are here considered Chuaria spp. (32 on bedding planes; 32 isolated and mounted on glass slides) from 5 shale horizons: 86-G-28, 86-G-33, 86-G-61, 86-G-62, P-2945. 70

Genus COMASPHAERIDIUM Staplin, Jansonius, and Pocock, 1965 Type spec/es.--Comasphaeridium cometes (Valensi, 1949), p. 18.

COMASPHAERIDIUM sp. Figure 14.5 Descnpffon.--Thin-walled vesicles, ca. 50 pm diameter, that bear abundant, straight to slightly curved, solid processes, ca. 0.2 wide and up to 11 pm long. Discussion.-Comasphaendium is rare in the Svanbergfjellet assemblage, represented by just two unambiguous specimens. They are broadly comparable to middle Silurian Ç sequestratus Loeblich; however, final assignment awaits more and better preserved material. Svanbergfjellet Comasphaeridium are decidedly unlike the much larger (100-300 pm diameter) forms described from the Vendian of China (Zhang Z., 1984; Yin, 1985a), but closely resemble specimens from the Late Riphean Wynniatt Formation, Victoria Island, arctic Canada (Butterfield and Rainbird, 1988). Material.—!, possibly 3 specimens from shale horizon 86-G-62.

Genus CYMATIOSPHAEROIDES Knoll, 1984, emend. Knoll et al., 1991 Type spgç/es.—Cvmatiosphaeroides kullingii Knoll. 1984, p. 153.

CYMATIOSPHAEROIDES KULLINGn Knoll, 1984, emend. Knoll et al.,1991 Figures 15.1-15.5

Trachvsphaeridium laufeldi. KNOLL AND CALDER, 1983, PI. 60, fig. 1-2. Cvmatiosphaeroides kullingii KNOLL, 1984, p. 153, fig. 9A-C; ALLISON AND AWRAMIK, p. 287, fig. 11.1-11.3; KNOLL, SWETT, AND MARK, 1991, p. 557, figs. 4.4, 4.6. 71

Unnamed form D KNOLL, 1984, p. 151, fig. 9H. Pterospermopsimorphid type A KNOLL, 1984, fig. TP.

Description.—Spheroidal vesicles, 32-345 pm diameter (% = 129 pm; s.d. = 86.4 pm; n = 51), with abundant, narrow (ca. 1 pm) solid processes that support a single or multilaminated envelope. Maximum overall dimension 49-390 pm (x = 154 pm; s.d. = 99 pm; n = 53). Up to 12 envelopes surround a vesicle (x = 3), sometimes producing a massive, space-filling mucilage. Outer envelopes typically compressed and poorly defined; occasionally separated from one another by process-supported intervals similar to that between the vesicle and inner envelope. Two vesicles rarely within a common envelope. Discussion.— Cvmatiosphaeroides and Ç kullingii were recently rediagnosed by Knoll et al. (1991) who recognized the multiple layering of the envelopes. The Svanbergfjellet population generally supports their emendations but reveals that both the number of enveloping layers (up to at least 12; Fig. 15.1- 15.2) and maximum dimensions (up to 400 pm) must be increased to encompass the considerable variability of the species. Other than a slight mode at ca. 100 pm (maximum O.D), the relatively large population of Svanbergfjellet Ç kullingii shows no obvious size-class separation. In combination with its often prolific envelope/ mucilage production and wide range in superficial morphology, this continuous size distribution suggests that Ç kullingii was an actively growing vegetative structure rather than a dormant cyst or spore. Such an interpretation is supported by evidence that Ç kullingii actively attached itself to substrates (Fig. 15.2; Knoll et al., 1991: fig. 4.4) and the rare occurrence of two vesicles within a common outer envelope. Cvmatiosphaeroides kullingii occurs primarily in shallow subtidal environments, often acting as a clast within 72 micTobialite grainstones; a single specimen embedded in dense filamentous mat (Fig. 15.3) indicates its occasional transport to intertidal/supratidal settings (cf.. Knoll et al., 1991). In addition to the Svanbergfjellet occurrences, Ç. kullingii is found in the immediately overlying Draken Conglomerate Formation, the Hunnberg and Rysso Formations on adjacent Nordaustlandet, and the upper Tindir Group of northwestern Canada. In silicified carbonate environments it always co-occurs with the large process-bearing acritarch Trachvhvstrichosphaera. A variety of chemostratigraphic and biostratigraphic data point that this Cvmatiosphaeroides-Trachvhvstrichosphaera assemblage as diagnostic for the latter part of the Late Riphean (cf., Kaufman et al., 1992). Material—56 specimens recorded from 6 silicified carbonate samples: 86-G-8, 86-G-9, 86-G-14, 86-G-15, P-2664, P-2628. An additional specimen is preserved in carbonate in sample 86-G-15.

Genus DICTYOTIDIUM, Eisenack, 1955 Type spgc/gs.—Dictvotidium dictvotum (Eisenack, 1938), p. 27-28.

DICTYOTIDIUM COLANDRUM n. sp. Figures 14.1-14.4, Diagnosis.—A species of Dictvotidium with a robust reticulate framework defining 1.5 pm - 4 pm wide polygonal fields. Short solid processes project from the intersecting points of the polygons and support thin membranes above the primary reticulum. Overall diameter, 30-60 pm. Dgscr/pfiow.—Spheroidal microfossils, 30-60 pm diameter (% = 33 pm; s.d. = 8 pm; n = 8), defined by a robust reticulum of 1.5-4 pm wide polygonal fields. Short (ca. 1 pm) solid spines project from the intersecting points of the 73

polygons and support thin membranes above the primary reticulum.

D iscmss/qw .—Dictvotidium colandrum n. sp. is distinguished from other species in the genus by its robust reticulum and apparent absence of vesicle wall within the polygonal fields. In combination, these two features probably contributed to the unusual pre-burial rigidity of D. colandrum, as reflected in the occurrence of isolated fragments in thin section (Figs. 14.3-14.4) The short protruding spines of D. colandrum n. sp. are fully accommodated by Dictvotidium; however, the faint intervening membranes (Fig. 14.1) are not stated as a feature of any previously named species. The Cambrian genus Acrum has broadly reminiscent membranes (Downie, 1982: fig. 3) but lacks spines. An unnamed problematic microfossil from the Neoproterozoic Visingso Beds of Sweden (Vidal, 1976: fig. 23A-23D) may prove to be a specimen of D. colandrum. Efymo/ogy.—With reference to its pronounced colander-like structure. Material—9 specimens from shale horizon 86-G-62 Holotype.-HUPC 62714; Figure 14.1; Slide 86-G-62-45; England-Finder coordinates: N-31-4.

Genus GERMINOSPHAERA Mikhailova, 1986, emend. Type species. Germinosphaera bispinosa Mikhailova. 1986, p. 33. Emended Diagnosis.—Spheroided' vesicles bearing one to six hollow, freely communicating, and typically open-ended filamentous processes with rare branches. Multiple processes positioned equatorially on the vesicle or not, but otherwise randomly distributed. Discussion .—Germinosphaera was established to accommodate distinctive, process-bearing spheroids from the Upper Riphean Dashkin Suite of Siberia; depending upon the number of processes, the four described specimens were 74 placed in one of two spedes, unispinosa or bispinosa. The considerably larger Svanbergfjellet population (41 specimens) reveals that process number and length are highly variable in Germinosphaera and that this variation is almost certain to be intraspecific. Thus, the two very similar sized original species are subsumed into one, G. bispinosa, while a Svanbergfjellet population with substantially larger vesicles represents a legitimate second species, G. fibrilla n. comb. A new form with a distinct wall texture and process habit may eventually warrant separate generic status, but is here classified as G. iankauskasii n. sp. The more or less random distribution of hollow, open-ended processes characteristic of Germinosphaera suggests it was an actively growing vegetative structure, rather than a dormant cyst or spore. It has a close morphological (but not necessarily taxonomic) analogue in the germinating zoospores of the modem xanthophyte alga, Vaucheria. The large (ca. 100 pm diameter) asexual spores of Vaucheria establish new filamentous thalli by germinating one to several equatorial, sometimes branching primordia (Fig. 16.7). Under this 'germinating zoospore' interpretation Germinosphaera would clearly derive from simple Leiosphaeridia-type acritarchs.

GERMINOSPHAERA BISPINOSA Mikhailova, 1986, emend. Figures 16.4-16.5

Germinosphaera unispinosa MIKHAILOVA, 1986, p. 33, fig. 5.

Emended Diagnosis.—A spedes of Germinosphaera with psilate vesicles between 13 and 35 pm diameter. Processes 2.5-3.5 pm wide and, when multiple, arranged equatorially on a veside. 75

Description.--Thm-'waUeà psilate vesicles, 18-35 pm diameter (% = 25 pm; s.d. = 5 pm; n = 14), bearing 1 to 4 filamentous processes, 2.5-S.5 pm wide and up to 185 pm long. Processes hollow, communicating freely with vesicle, and typically open ended. Multiple processes arranged equatorially on a vesicle. Discussion.--A majority of Svanbergfjellet G. bispinosa bear a single filamentous process; however, one otherwise indistinguishable specimen has four such extensions arranged at ca. 90° intervals around the vesicle. Multiple- processed G. bispinosa also occur in the broadly correlative Wynniatt Formation, arctic Canada (unpublished data). Material.~14 specimens from shale horizon 86-G-62.

GERMINOSPHAERA FIBRILLA (Guyang, Yin, and Li, 1974) n. comb. Figures 17.1-17.8

Ooidium fibrillum GUYANG, YIN, AND LI, 1974, p. 120, PI. 47, fig. 4. Archaeohvstrichosphaeridium truncatum GUYANG, YIN AND LI, 1974, p. 77, PI. 27, fig. 19.

Phvcomvcetes sp., YIN L., 1985, PI. 4, fig. 19. Germinosphaera tadasii WEISS, 1989 (partim), p. 143, PI. 47, figs. 3-4 (in Jankauskas, 1989).

Diagnosis.—A spedes of Germinosphaera with psilate vesicles between 45 and 125 pm diameter. Processes 3-5 pm wide and, when multiple, arranged equatorially on the veside. Description.—Thin-y/a\ied psilate vesides, 49-120 pm diameter (x = 87 pm; s.d. = 21 pm; n = 18), bearing 1-4 filamentous processes, 3-5 pm wide and up to 90 pm long. Processes hollow, communicating freely with the vesicle, and 76

typically open ended; occasionally branched. Multiple processes arranged equatorially on a vesicle. Discussion.—Unlike the type species, G.. fibrilla n. comb, commonly bear two, three, or four processes. In specimens with more than two processes, their diagnostic equatorial positioning also becomes clear, with the third (and fourth, if present) process never originating from within the perimeter of the flattened vesicle. Such an orientation suggests that G. fibrilla was a benthic organism in which the primary growth plane was horizontal; a similar arrangement is observed in the germinating zoospores of surface-growing Vaucheria. Branching processes in G. fibrilla (Fig. 17.6) lend further support to the 'germinating zoospore' interpretation. The type specimen of G. fibrilla n. comb., from the Sinian of southwestern China, was clearly misplaced as a species of Ooidium. With its ca. 85 pm diameter vesicle and 3, ca. 4 pm wide, equatorially arranged processes it is all but indistinguishable from the Svanbergfjellet material. Material.~18 specimens from shale horizon 86-G-62.

GERMINOSPHAERA JANKAUSKASH n. sp. Figures 16.1-16.3

Diagnosis.—A species of Germinosphaera with shagreenate walls and vesicles between 45 and 90 pm diameter. Processes 5-10 pm wide and randomly distributed.

Description.—Shagreenate-walled vesicles, 46-86 pm diameter (x = 71 pm; s.d. = 14 pm; n = 6), that bear 2-7 randomly distributed processes, 5-10 pm wide and up to 32 pm long. Processes hollow, communicating freely with the vesicle, and typically open ended; sometimes with a slight terminal expansion. 77

Discussion.—G. iankauskasii n. sp. is distinguished from other species of Germinosphaera by its shagreenate wall texture (possibly the remnants of an outer mucilaginous layer), significantly broader processes, and a random positioning of processes on the vesicle (vs. constrained to one plane). This latter condition is best observed in the type specimen (Fig. 16.1) where, in addition to the four 'equatorial' processes, there are three that clearly originate from within the perimeter of the flattened vesicle. The indeterminate number and positioning of processes in G. iankauskasii n. sp. suggest a functional interpretation similar to that of other Germinosphaera species, i.e, a germinating propagule, broadly comparable to the zoospores of Vaucheria (Fig. 16.7). Unlike these other taxa, however, G. iankauskasii primordia appear to have had no preferred orientation, and growth evidently proceeded in three dimensions. Etymology.—In recognition of Lithuanian paleontologist. Dr. Tadas Jankauskas. Material.-6 specimens from shale horizon 86-G-62. Holotype.—HUPC 62715; Figure 16.1; Slide 86-G-62-12M; England-Finder coordinates: L-31-3.

Genus GONIOSPHAERIDIUM Eisenack, 1969 Type species.—Goniosphaeridium polvgonale (Eisenack, 1931), p. 113.

GONIOSPHAERIDIUM sp. Figures 14.6-14.7 Description .—Moderately dark-walled vesicles, ca. 24 pm diameter, bearing numerous blunt-tipped processes. Processes ca. 1.5 pm broad by 3 pm long, apparently hollow, and communicating freely with the vesicle; rarely branched(?). 78

Discussion.—These small, 'Paleozoic aspect' acanthomorphic acritarchs (p. 45) do not readily recall any named species of Goniosphaeridium; they do, however, conform to its generic diagnosis of psilate or shagreenate vesicles with hollow, unconstricted, and simple processes. Closer overall comparisons can be found among species of Baltisphaeridium (e.g., B. brevispinosum (Eisenack)) or Goreonisphaeridium (e.g., G. echinodermum (Stockmans and Willière)), but the processes of these forms are, respectively, isolated from the central vesicle, and solid. A single, apparently branching process in one Svanbergfjellet specimen (Fig. 14.7) and a pylome-like structure in the other (Fig. 14.6) further complicate the taxonomy of these fossils. Material—2 specimens from shale horizon 86-G-62.

Genus GORGONISPHAERIDIUM Staplin, Jansonius, and Pocock, 1965 Type specfes.—Goreonisphaeridium winslowii Staplin et al., 1965, p. 193.

GORGONISPHAERIDIUM sp. Figures 14.9-14.10 Dgsm'ptfoM.—Approximately 1 quadrant of a thick-walled spheroidal vesicle bearing abundant, closely packed processes. Processes solid, tapering from 1.5 to 1.0 pm and ca. 2 pm long. Full vesicle diameter calcuated to be ca. 250 pm. Discussion.—This isolated fragment compares closely with the single 'G. maximum' described from the overlying Draken Conglomerate (Knoll et al., 1991). The principal difference lies in the shorter processes of the Svanbergfjellet specimen, possibly a consequence of transport erosion. Material—A single specimen in shale horizon 86-G-63. 79

Genus LEIOSPHAERIDIA Eisenack, 1958 Type specfgs.—Leiosphaeridia baltica Eisenack, 1958, p. 2-3. Discussion.-Jankauskas et al. (1987; 1989) offered a comprehensive revision of Proterozoic leiosphaerid acritarchs, dividing the smooth-walled forms into four basic species based on wall thickness and size-class: L. minutissima - thin- walled, < 70 pm; L. tenuissima - thin-walled, 70-200 pm; L. crassa - thicker- walled, < 70 pm; L. jacutica - thicker-walled, 70-800 pm. There are some difficulties with this taxonomy, particularly in dealing with larger forms, and in assessing wall thicknesses; overall, however, it is proving to be both a useful and not altogether artificial classification. All four basic species are found in the Svanbergfjellet shales and there does appear to be a break in size- frequency distributions on or around 70 pm (Fig. 4).

LEIOSPHAERIDIA CRASSA (Naumova, 1949), emend. Jankauskas, 1989 Figures 16.6, 23.11 Discussion .—Moderately thick-walled leiosphaerid acritarchs less than 70 pm wide are abundant in Svanbergfjellet shales. In sample 86-G-62 (Algal Dolomite Member) they show a clearly defined, right-skewed size-frequency distribution with a single mode at ca. 17 pm (x = 19.1 pm; s.d. = 10.7; n = 1065; Fig. 4). Leiosphaerid size-frequency in the Lower Dolomite Member shale (P- 2945) is distinct, but likewise suggests an important break at ca. 70 pm (Fig. 4). L. crassa occurs variously as isolated individuals, large populations covering entire bedding planes, or very localized associations; in this latter conformation it dearly grades into sheet-like Ostiana. Some spedmens show unambiguous release structures (Fig. 16.6) suggesting, but not necessarily proving (Butterfield and Chandler, 1992; Chapter III) their role as cysts or 80 spores; others were presumably the prerequisite stage to smaller germinosphaerids (Figs. 16.4-16.5,17). Mflferffl/.—Abundant in most fossilferous shale horizons. 1220 measured specimens.

LEIOSPHAERIDIA JACUTICA (Timofeev, 1966), emend. Mikhailova and Jankauskas, 1989 Figure 16.8 Discwssiow.—Svanbergfjellet L. jacutica range from 71 to 796 pm diameter and a population isolated by acid maceration shows a broad, right-skewed distribution centered on ca. 150 pm (x = 232; s.d. = 116; « = 150). This broad modality suggests the prevalence of a single natural taxon; however, L. jacutica is likely to encompass a disparate variety of fossil organisms. In the Svanbergfjellet assemblage such a form would clearly have given rise to Germinosphaera fibrilla n. comb.. Osculosphaera hvalina n. sp. (Fig. 15.11), and Trachvhvstrichosphaera aimika (Fig. 18.1); T. aimika may also 'return' to a leiosphaerid habit through erosion of its processes (e.g.. Fig. 13.1). Material—252 measured specimens from 5 shale horizons: 86-G-62, 86-G-61, 86-G-33; 86-G-28; P-2945. 150 specimens isolated by acid maceration.

LEIOSPHAERIDIA TENUISSIMA Eisenack, 1958 Figure 16.9 Discwssiow.—These thin-walled leiosphaerids are distinguished from L. jacutica by the relative transparency and (original) flexibility of their walls. Svanbergfjellet L. tenuissima range from 92 pm to 624 diameter (considerably larger than the 200 pm upper limit suggested in Jankauskas, 1989) and show a right-skewed size-frequency distribution centered broadly on 150 pm {x = 237 81

|im; s.d. = 129; n = 21). The marked similarity between the distributions of L. tenuissima and L. jacutica casts some doubt on the purported distinction between the two species. Material.—21 specimens recorded from three shale horizons: 86-G-62, 86-G- 63, P-2945; all isolated by acid maceration.

LEIOSPHAERIDIA WIMANII (Brotzen, 1941) n. comb. Figures 13.4-13.6

Das Fossil aus der Visingsogruppe, WIMAN, 1894, PI. 5, figs. 1-5. Chuaria wimani BROTZEN, 1941, p. 258-259. Kildinella maena TIMOFEEV, 1969, p. 14, PI. 6, figs. 4-5. Shouhsienia shouhsienensis. ZHANG R., FENG, MA, XU, AND Y AN, 1991, p. 120, PI. 1, figs. 16-26 (see for extended synonymy).

Diagnosis.—A species of Leiosphaeridia 800-2500 pm diameter. Walls translucent, psilate to finely textured, commonly with medial split release structures. Folds simple. Description.—Psilate or finely textured vesicles, 0.76-2.48 mm diameter (x = 1.27 mm; s.d. = 0.44 mm; n = 14), commonly with a single medial split. Walls thin, leaving little or no imprint on enclosing sediments, and typically translucent in transmitted light. Folds simple. Discussion.—While the stated size range for L. wimanii n. comb, is admittedly arbitrary, these large, relatively thin-walled spheroids are clearly something other than the tail end of the L. jacutica distribution; the vesicles, for example, commonly show biologically mediated splits (Wiman, 1894; Timofeev, 1969; Zhang R. et al., 1991). In the Svanbergfjellet material this feature is indicated 82 by the smooth, rounded edges of the splits (Fig. 13.4-13.6) and evidence that compressional folding took place after the wall had ruptured. The type material of L. wimanii n. comb., from the Late Proterozoic Visingso Beds (Wiman, 1894), was initially described as a species of Chuaria (Brotzen, 1941). Eisenack (1951) subsequently referred it to Leiosphaera (= Leiosphaeridia), although he later reversed this decision, considering it to be a chitdnous foram (Eisenack, 1966). More recently. Ford and Breed (1973), Vidal and Ford (1985), and Jankauskas (1989) subsumed C. wimani (including Kildinella maena Timofeev, 1969) into Ç circularis. These authors notwithstanding, a close analysis of the essential features of Chuaria circularis (see above) and those of the Visingso fossils originally described as Ç wimani show the two forms to have little in common other than their millimetric dimensions. Chuaria is always opaque, and is sufficiently thick-walled to leave a substantial imprint on bedding planes; excystment/release structures do not occur in the type material (Ford and Breed, 1973: p. 546) or in the Svanbergfjellet populations. In contrast, the Visingso fossils commonly exhibit medial splits, are translucent, even through three superimposed layers (Wiman, 1894: pi. 5, figs. 1-3; Eisenack, 1966: fig.2; Timofeev, 1969; 1970: pi. 1, figs. A, B), and leave only the slightest, if any, perceptible imprint on bedding plane surfaces; SEM of a split 1.25 mm wide Visingso 'Ç wimani' (middle Visingso Beds sample courtesy of G. Vidal) shows its wall to be ca. 1 pm thick, half that of a considerably smaller Svanbergfjellet Ç circularis (Figs. 13.7-13.9). Apart from their greater overall diameter these fossils are indistinguishable from other leiosphaerids and are therefore revised as L. wimanii n. comb. Svanbergfjellet L. wimanii n. comb, are likewise characterized by biologically mediated medial splits and relatively thin walls that leave only slight impressions on bedding surfaces (Fig. 13.6). There is some amount of 83 variation in their folding patterns and relative opacity that may warrant further taxonomic distinction; however, these differences are here considered to be accommodated by L. wimanii. The orthography of the specific epithet, wimani, is corrected in accordance with ICBN Article 32.5. Material—14 specimens from 4 shale horizons: 86-G-62, 86-G-61, 86-G-33, P-2945.

LEIOSPHAERIDIA spp. Figures 13.2-13.3 Discussion—The Svanbergfjellet assemblage unquestionably includes many more leiosphaerid species than can be determined from size and wall opacity/thickness alone. The difficulty lies in unambiguously diagnosing their essential features. Two leiosphaerid forms are readily distinguished in the Algal Dolomite Member shales, but only on the basis of what appear to be taphonomically induced surface textures. In the one case, dark-walled vesicles, 181-488 pm diameter (% = 319 pm; s.d. = 116 pm, n = 6), are perforated by numerous holes with a distinctive bridged habit (Fig. 13.2); in the other, thin- walled vesicles, 344-1325 pm diameter (x = 798 pm; s.d. = 333 pm, n = 7), have a shagreenate to velutinous wall structure and commonly include a central, longitudinal crease or inclusion (Fig. 13.3). These taphonomicC?) fabrics may well represent underlying taxonomic differences but, in the absence of supporting primary features, cannot be legitimately applied to taxonomic diagnoses. 84

Genus OSCULOSPHAERA n. gen. Type species.--Osculosphaera kulgunica (Jankauskas, 1980) n. comb., p. 192. Diagnosis.—Psilate spheroidal vesicles with a single, well-defined circular opening. Discussion.—While Osculosphaera n. gen. clearly derives from a simple leiosphaerid precursor, its unique and readily identifiable release structure represents a substantial qualitative, and therefore generic-level distinction from Leiosphaeridia. The separation is also dictated by practical concerns as there are now at least two discrete size classes of the form: the type species, from Late Riphean shales of the Southern Urals, is typically 15-35 pm diameter (Jankauskas, 1980a) while the Svanbergfjellet population ranges from 35 to 131 pm. Both species appear to have somewhat rigid walls, expressed in O. kulgunica n. comb, compressions by their characteristic radial fractures, and in three dimensionally preserved O. hvalina n. sp. by its remarkably consistent sphericity and rare angular indentations. Svanbergfjellet Osculosphaera n. gen. are more or less flask-shaped in lateral aspect (Figs. 15.6-15.8,15.9) inviting comparison with vase-shaped microfossils such as Melanocvrillium Bloeser, 1985. This latter form, however, is characterized by very thick walls and various drcumoral ornamentation. Hvalocvrillium Allison, 1989 more closely approximates the wall structure of Osculosphaera. but its long cylindrical habit is entirely distinct from spheroidal Osculosphaera. Etymology.-Trom the Latin osculum - mouth and sphaera - ball. 85

OSCULOSPHAERA HYALINA n. sp. Figures 15.6-15.10 Diagnosis.—A spedes of Osculosphaera with hyaline walls; vesicles 35-150 pm diameter. Description.—Thin-walled psilate spheroidal vesicles, 35-131 pm diameter (% = 78 pm; s.d. = 29 pm; « = 11), with a prominent drcular opening 30-45% the diameter of the vesicle. Wall around the opening typically projected outward forming a short 'oral collar'; rarely curled inward. P/scMSsfow ."Osculosphaera hvalina n. sp. co-occurs a with a population of undifferentiated spheroids (Fig. 15.11) closely comparable in wall structure and size-frequency distribution (22-163 pm diameter; x = 79 pm; s.d. = 35 pm; n = 70). These are clearly the antecedent form of O. hvalina, although in a strict form taxonomy they would presumably be assigned to spedes of Leiosphaeridia. In one instance an O. hvalina vesicle contains multiple, ca. 13 pm diameter spheroids, pointing to its role as a reproductive structure. The 'oral collar' O. hvalina n. sp. is insuffident to recover the entire spheroid palinspastically, indicating that some of the wall has been subject to enzymatic removal. Actual restructuring of the wall, however, appears to have been minimal, with the orifice rim typically showing no secondary reinforcement; the prindpal alteration may have been an induced flexibility in the circum-oral region (Fig. 15.9). Etymology.—From the Greekhyalinos - of glass, in reference to the translucent, somewhat rigid walls. Material.—11 specimens in two chert samples (from a single horizon): P-3085, 86-P-82. 71 assodated entire spheroids. Holotype.—HUPC 62716; Figure 15.6; Slide P-3085-1A; England-Finder coordinates: P-43-2. 86

Genus PTEROSPERMOPSIMORPHA Timofeev, 1966 Type speci'gs.—Pterospermopsimorpha pileiformis Timofeev, 1966, p. 34. D/scwssion .--Probable eukaryotic spheroids with envelopes are relatively uncommon in Svanbergfjellet biota, but given their wide range in size (17-213 pm) and wall structure are sure to represent a number of Pterospermopsimorpha species. Conversely, the marked size discrepancy between the inner and outer vesicles illustrates the potential for an artificial inflation of leiosphaerid species diversity. Some silicified pterospermopsimorphids are also likely to be degradational (or developmental) variants of Cvmatiosphaeroides (see above).

PTEROSPERMOPSIMORPHA PILEIFORMIS Timofeev, 1966 Figure 14.8 Descriphbn.“ Smooth-walled spheroid preserved within a larger smooth- walled spheroidal envelope. Inner vesicle, 25 pm diameter; outer vesicle, 43 pm diameter; both with simple folds. Material—A single unambiguous specimen in shale horizon 86-C-62. Five additional specimens assigned to Pterospermopsimorpha spp.

Genus TRACHYHYSTRICHOSPHAERA Timofeev and Hermann, 1976, emend.

Non Trachvhvstrichosphaera bothnica TYNNI AND DONNER, 1980, p. 11, PI. 2, figs. 15-17.

Nec Trachvhvstrichosphaera parva MIKHAILOVA. 1989, p. 47, PI. 2, figs. 2-3 {in Jankauskas, 1989).

Nec Trachvhvstrichosphaera truncata HERMANN AND JANKAUSKAS, 1989, p. 48, PI. 2, figs. 5-6 {in Jankauskas, 1989). 87

Type spec/gs.—Trachyhvstrichosphaera aimika Hermann, 1976, p. 48 (in

Timofeev et al., 1976). D/flgwos/s.—Relatively large spheroidal vesicles bearing one to many irregularly distributed hollow processes. Vesicle wall psilate to (degradationally) shagreenate, or covered with closely-packed, short solid spines. Processes communicating freely with the vesicle but otherwise highly variable; occurring as small cones, relatively long tubes, and variously expanded, constricted and/or bifurcated forms. Outer sheath/matrix variable: absent to diaphanous (shagreenate) to robust; bi-layered in silicified specimens. Discussion.—As much as anything, it is the marked variability of Trachyhystrichosphaera (within a readily identifiable habit) that distinguishes it from other acanthomorphic acritarchs. Recognition of systematic trends within a population (including a size range of well over an order of magnitude - Knoll et al., 1991), and evidence of 'germinating' processes (Fig. 18.11) argues strongly for considering its variation as an ontogenetic series. Thus, in their early stages with only a few very small processes present, vesicles may be all but indistinguishable from unomamented spheroids (Fig. 18.1). Additional processes were apparently initiated throughout a vesicle's development such that any one fossil will usually preserve processes in various stages of 'maturity' (Figs. 19.3,18.5). After the appearance of at least some processes, extracellular envelope material was secreted by the central vesicle (compare Figs. 19.3 and 19.5; or Figs. 18.3 and 18.4; note also the specimens in which the sheath is not significantly separated from the main vesicle and does not encompass the processes - Figs. 18.4,18.9-18.10). A fully enclosing envelope is found only in mature(?) Trachvhvstrichosphaera. which are also characterized by considerably more uniform and evenly distributed processes (Fig. 19.5; Jankauskas, 1989: pi. 2. fig. 1) than is otherwise expressed. 88

Differential taphonomy adds a further level of 'variation' to Trachyhystrichosphaera. Thus, many of the supposed distinctions in wall texture can be attributed to the variable preservation of an extracellular sheath (Pjatiletov, 1988; Jankauskas, 1989; pi. 1, figs. 7, 8). Moreover, the unusual terminal expansion of some processes appears to coincide with their penetration through and beyond a confining envelope (Jankauskas, 1989: pi. 1, fig. 7), suggesting a secondary derivation; the processes of some Svanbergfjellet specimens are markedly constricted where they penetrate the outer membrane (Fig. 18.9). In classifying T rach vhvstrichosphaera the basic principles of form taxonomy must be adhered to; however, the recognition of its broad developmental and taphonomic variation allows some degree of natural taxonomy to be applied. Thus, undetermined or indeterminate features such as process number and morphology (within limits) are not considered useful taxonomic characters. Similarly, most previously named Trachvhystrichosphaera species and specimens preserve some evidence of an extracellular sheath (Pjatiletov, 1988), while those few that don't are otherwise indistinguishable from their ensheathed counterparts in the same assemblages. Thus, sheath occurrence in Trachvhystrichosphaera appears not to be taxonomically significant. Of the 11 published species of Trachvhvstrichosphaera, 3 appear to be misplaced in the genus: T. bothnica. T. parva (both are very small vesicles with ambiguously preserved processes), and T. truncata (solid, evenly distributed processes). Another 6 appear to be reconcilable to the type species, T. aimika, while a second legitimate species, T. polaris n. sp., is described below. 89

TRACHYHYSTRICHOSPHAERA AIMIKA Hermann, 1976, emend. Figures 18.1-18.11,19.7

Trachvhystrichosphaera aimika HERMANN, 1976, p. 48, Pl. 19, figs. 6, 8, Pl. 20, figs. 1-3 {in Timofeev et al., 1976); JANKAUSKAS, 1978, p. 915, figs. 1.1-1.5; YIN C., 1985, Pl. 4, fig. 3. Nucellohvstrichosphaera megalea TIMOFEEV AND HERMANN, 1976, p. 47- 48, Pl. 19, fig. 1, Pl. 20, figs. 4-5 (in Timofeev et al. 1976). ?Trachvhvstrichosphaera sp., TYNNI AND DONNER, 1980, p. 11, Pl. 2, fig. 20. Trachvhystrichosphaera vidalii KNOLL, 1984, p. 154-156, figs. 8A-K; KNOLL AND CALDER, 1983, p. 493, Pl. 58, figs. 9-10; ALLISON AND AWRAMIK, 1989, p. 287-288, figs. 9-10; JANKAUSKAS, 1989, p. 48, PI. 2, fig. 1; KNOLL, SWETT AND MARK, 1991, figs. 4.8, 6.4, 7.4-7.8. Trachvhystrichosphaera. BUTTERFIELD AND RAINBIRD, 1988, p. A103. Trachvhystrichosphaera megalia PJATILETOV, 1988, p. 82, PI. 1, fig. 5, PI. 2, fig. 5 Trachvhystrichosphaera membranacea PJATILETOV, 1988, p. 81-82, PI. 1, figs. 1-4, PI. 2, fig. 4. Trachvhystrichosphaera sp., ZANG, 1988, PI. 42, figs. A-D. Trachvhystrichosphaera magna ALLISON, 1989, p. 288, figs. 11.5-11.8 (in Allison and Awramik, 1989). Trachvhystrichosphaera cvathophora HERMANN, 1989, p. 47, PI. 1, fig. 7 (in Jankauskas, 1989). Trachvhystrichosphaera stricta HERMANN, 1989, p. 47, PI. 2, figs. 4, 7-8 (in Jankauskas, 1989). Trachyhystrichosphaera cf. magna, SERGEEV, 1991, p. 92, fig. 2. 90

Emended Diagnosis.-A species of Trachvhystrichosphaera with the primary vesicle wall (originally) psilate. Processes 3-8 pm diameter. Description.—Smooth walled vesicles 113-702 pm diameter (x = 264 pm; s.d. = 127 pm; n = 123), bearing one to many hollow, unevenly distributed processes. Extracellular envelope usually present but of variable character; silicified specimens preserve two distinct enveloping layers: a thick, relatively dense inner layer and a thin, diffluent, and apparently tacky outer layer. Processes 3- 8 pm in maximum diameter and of varying morphology: small simple cones, longer tubes, and variously expanded, constricted and/or basally bifurcated forms; up to 45 pm long. Pisci/ssioM.—Trachvhystrichosphaera aimika is readily identified by its variable process morphology (though always between 3 and 8 pm diameter) and thin psilate walls. A relatively thick extracellular sheath usually surrounds the vesicle and close examination of silicified specimens show it to be bi-layered; the tacky outer layer apparently served to attach the vesicle to solid substrates (Fig. 18.11). Like Cvmatiosphaeroides kullingii, T. aimika thus appears to have had at least an intermittently benthic habit. Silicified T. aimika differs only superficially from the more common shale-hosted material and shows a closely comparable size-frequency distribution {x = 265 pm; s.d. = 137 pm; n = 10). More difficult to identify are forms with very short or eroded processes (Figs. 18.1,18.7); in the latter case, specimens may only be identifiable by the distinctive 'arrangement' of perforations left in a vesicle wall (Fig. 13.1). The morphological range of T. aimika is further expanded by vesicles with marked, more or less medial constrictions (Fig. 18.7), and a single specimen showing evidence of binary fission; two T. aimika within a common envelope have been recorded in the overlying Draken Conglomerate Formation (A. H. Knoll, pers. comm., 1991). 91

All the above synonymized species of Trachvhystrichosphaera can be recognized as developmental and/or taphonomic variants of a single biological entity, T. aimika. However, examination of photographs of the type specimen of Nucellosphaeridium bellum (Timofeev, 1969: pi. 6, fig, 6; 1970, pi. 1, fig. C) strongly suggests that it too is conspedfic (cf., Jankauskas 1989: pi. 2, fig. 8). If confirmed, T. aimika would be a junior synonym of 'T. bella'. Trachvhystrichosphaera aimika has been reported from a number of Siberian sequences, Kazakhstan, the Southern Urals, Finland, China, Australia, various formations on Svalbard, and from the Yukon Territory and Northwest Territories of Canada. In all cases it can be reliably constrained to the Late Riphean, making it a valuable index fossil for this time period. Material—125 specimens; 10 from silidfied carbonate samples 86-G-8 and 86- G -15,115 from shale horizons 86-G-62 and P-2945.

TRACHYHYSTRICHOSPHAERA POLARIS n. sp. Figures 19.1-19.6,19.8 Diagnosis.—A spedes of Trachvhystrichosphaera with the primary vesicle covered by dosely packed, short solid spines. Prindpal processes hollow, 3-5 pm diameter. Description.—Vesicles, 95-235 pm diameter (% = 160 pm; s.d. = 33 pm; n = 20), covered with closely packed, short, solid spines and bearing several randomly distributed larger processes. Processes hollow, 3-5 pm diameter, and of variable morphology: tubular, ampulliform, or gradually expanding; in mature(?) spedmens, up to 55 pm long, terminally flared, and confluent with an outer envelope. Shorter, solid spines are thin and hair-like on small sp>ecimens to relatively thick (ca. 1 pm) and straight on larger forms. Extracellular envelope present or absent. 92

Discussion.—The principal (hollow) processes of T. polaris n. sp. show very much the same range of morphology and distribution as those of T. aimika. Similarly, an external envelope is usually, but not necessarily (Figs. 19.1-19.3, 19.6) present. Thus, the two taxa differ primarily in the distinctive 'bristled' surface of T. polaris. Subsidiary differences include a more restricted size range for T. polaris, and its somewhat darker (thicker?) walls. Its outer envelope appears to be more clearly delineated than in most T. aimika (Figs. 19.5,19.8), but comparable envelopes do occur in the type species (Fig. 19.7; Jankauskas, 1989: pi. 2, fig. 1). Fully enveloped specimens are characterized by more uniform and regularly distributed principal processes than unsheathed forms, suggesting they represent the mature and/or fully encysted stage of the life cycle. This 'mature' envelope is then subject to both biologically mediated breakdown (Fig. 19.5; note the rounded perforations that were formed prior to compaction) and mechanical erosion (Fig. 19.4). Etymology.—With reference to its northerly occurrence and the star-like appearance of the type specimen. Material—21 specimens from shale horizon 86-G-62. Holotype.—HUFC 62717; Figure 19.5; Slide 86-G-62-1M; England-Finder coordinates: U-30-3. 93

Domain BACTERIA Woese, Kandler, and Wheelis, 1990 Kingdom EUBACTERIA Woese and Fox, 1977 Phylum CYANOBACTERIA Stanier et al., 1978 Class COCCOGONEAE Thuret, 1875 Order CHROOCOCCALES Wettstein, 1924 Family CHROOCOCCACEAE Nageli, 1849

Genus GLOEODINIOPSIS Schopf, 1968, emend. Knoll and Golubic, 1979 Type species.--Gloeodiniopsis lamellosa Schopf, 1968, p. 684.

GLOEODINIOPSIS LAMELLOSA Schopf, 1968, emend. Knoll and Golubic, 1979 Figure 20.8 Discussion .—Knoll et al. (1991) considered as Gloeodiniopsis only those populations of spheroidal microfossils that exhibit both multiple envelopes and two to eight daughter cells within a common envelope. Spheroids with more than one envelope in the SvanbergQellet assemblage occur intermittently in low diversity microbial mat intraclasts; however, multiple cells with multiple envelopes are rare. Of the latter, one specimen is associated with a large population of envelope-free or single-enveloped Mvxococcoides, suggesting some genetic or ontogenetic relationship between the two. Material.—Two specimens from two chert horizons: 86-G-8, 86-G-9.

Family ENTOPHYSALIDACEAE Geitler, 1925 Genus EOENTOPHYSALIS Hofmann, 1976, emend. Mendelson and Schopf, 1982 Type species.—Eoentophvsalis belcherensis Hofmann. 1976, p. 1070-1072. 94

EOENTOPHYSALIS BELCHERENSIS Hofmann, 1976 Figures 20.4, 20.5

Eoentophvsalis cumulus KNOLL AND GOLUBIC, 1979, p. 148-149, figs. 2E, 3.

Description.—Spheroidal vesicles, 3-5 pm diameter, arranged in small pustulose clusters or extensive colonies of several hundred vesicles. Diads and tetrads commonly isolated by an enveloping sheath. Discussion.—Except for the luxuriant mat-forming habit, Svanbergfjellet E. belcherensis exhibit most of the variation found in the type populations. Furthermore, they show no clear morphological break between small isolated colonies (Fig. 20.5) and incipient mats (Fig. 20.4). Golovenoc and Belova (1984) have suggested that the more isolated colonial forms be assigned to a separate genus, Eogloeocapsa: however, the type specimen of Eoentophvsalis is already of this particular habit (Hofmann, 1976: pi. 6, fig. 13), making such a designation untenable. Eoentophvsalis cumulus differs only in age from E. belcherensis (Sergeev and Krylov, 1986) and is thus its junior synonym. Material—Scattered colonies in chert samples 86-G-4 and P-2664.

EOENTOPHYSALIS CROXFORDH (Muir, 1976) n. comb. Figure 20.1

Ameliaphvcus croxfordii MUIR, 1976, p. 155, figs. 6L-M, 7A-B.

Description.—Massive pluricellular aggregations of close-packed, often compressionally distorted vesicles; usually separated from one another by a 95 substantial (ca. 2 pm thick) extracellular layer. Vesicles equidimensional to slightly elongate, 8-16 pm diameter (x = 12 pm). Discussion.—A number of intraclasts in the Lower Dolomite microbialite grainstones are constructed exclusively of E. croxfordii and, with dimensions of up to 1.5 mm, can comprise several thousand individual vesicles. Mutual compression has resulted in a typically polygonal cellular outline, although the vesicles themselves are usually separated by a relatively thick mucilaginousC?) envelope. This intervening layer is sometimes absent between paired smaller cells suggesting that growth occurred by means of binary fission. The fragmentary nature of the present material precludes a positive determination of overall colony morphology; however, the occasional suggestion of a radial fabric (Fig. 20.1) may be reflecting an involvement in mammillate microbial mat formation, such as that reported for modern Entophvsalis (e.g., Golubic and Hofmann, 1976). Alternatively, they may derive from more complex pluri- or multi-cellular structures such as Thallophvca ramosa Zhang Y., 1989; some higher order organization is suggested in several of the Svanbergfjellet specimens where cells surround and define what appear to nave been internal spaces. Eoentophvsalis croxfordii n. comb, differs from the mat-forming variant of E. belcherensis primarily in the considerably larger dimensions of its cells, those of the type species having a mean diameter of just 3.9 pm. Eoentophvsalis vudomatica Lo, 1980 and E. areata Mendelson and Schopf, 1982 (Fig. 20.2) have more comparable cell dimensions (ca. 12 pm) but fail to express a dense colonial/mat-forming habit. Eoentophvsalis dismallakesensis Horodyski and Donaldson, 1980 and E. magna McMenamin et al., 1983 lack both the size and the habit of E. croxfordii. Material—6 occurrences in 3 chert samples: 86-G-9, 86-G-14, 86-G-15. 96

Order PLEUROCAPSALES Geitler, 1925 Family DERMOCARPACEAE Geitler, 1925 Genus POLYBESSURUS Fairchild, 1975, ex Green et al., 1987 Type specfes.--Polvbessurus bipartitus Fairchild, 1975, ex Green et al., 1987, p. 938.

POLYBESSURUS BIPARTTTUS Fairchild, 1975, ex Green et al., 1987 Figures 21.3, 21.6-21.7 Discussion.—This distinctive stalk forming cyanobacterium is found in the Lower Limestone Member where it occurs both in small isolated colonies associated with abundant Siphonophvcus typicum n. comb., and as massive crust-forming accumulations. In the latter case it occupies, almost to the exclusion of other taxa, a ca. 2 cm high by 5 cm wide mound; the remaining diversity in this crust is limited to localized populations of S. tvpicum and various spheroids. The paleobiology of P. biparti tus was discussed in detail by Green et al. (1987). Its pattern of cell division (by baeocyte production), and thereby its ordinal-level classification, was inferred from the geometrical arrangement of close packed stalks. The Svanbergfjellet material supports these previous interpretations and possibly offers direct confirmation of the reproductive mode. Midway up and within the Polvbessurus stalk shown in Figure 21.7 is a cluster of ca. 12 pm spheroids much smaller than the cell that would have formed the stalk. These cells likely represent the baeocytes (cell fission without intervening growth) inferred by Green et al. (1987: p. 935). Branching of Polvbessurus stalks was also considered possible but not observed in the East Greenland fossils (Green et al., 1987); it is rarely encountered in the Svanbergfjellet populations. 97

Polvbessurus is proving to have had both wide geographical and temporal distribution in the Proterozoic. It has now been reported from two ca. 725-1250 Ma sequences from the Canadian Arctic (Butterfield et al., 1990; Hofmann and Jackson, 1991), the Middle Riphean of the Southern Urals (Sergeev, 1991), 750-850 Ma sections in South Australia (Fairchild, 1975), and in 700-800 Ma old rocks from East Greenland and Spitsbergen (Green et al., 1987; Knoll et al., 1991; present study). Material.—Massive population in chert sample 86-P-89; also in P-3400.

Class HORMOGONEAE Thuret, 1875 Order OSCILLATORIALES Copeland, 1936 Family OSCILLATORIACEAE Kirchner, 1898 Genus OBRUCHEVELLA Reitlinger, 1948, emend. Knoll, 1992 Type spgcfes.-Obruchevella delicata Reitlinger, 1948, p. 78. Discussion.—The record of helically coiled microfossils was recently reviewed by Knoll (1992), with a recommendation that all such forms deriving from intact extracellular sheaths (as opposed to split sheaths or cellular trichomes) be placed in the form genus Obruchevella. The artificial distinction of three- dimensional (= Obruchevella) vs. two-dimensional preservation (= Volyniella) was earlier recognized by Jankauskas (1989).

OBRUCHEVELLA BLANDITA Schenfil, 1980 Figures 22.1-22.6

Obrachevella condensata LIU , 1984, p. 177, PI. 1, fig. 10 (in Liu et al., 1984). Volvniella glomerata lANKAUSKAS. 1980b, p. 112, PI. 12, fig. 18-19. Glomovertella glomerata TANKAUSKAS. 1989, p. 108-109, PI. 31, fig. 8-10. 98

Description.—Nonseptate tube wound into a regular helix with adjacent coils in close contact. Filament diameter constant at 1.5 |im; outside diameter (O.D.) of helices 6-27 |im (% = 15.9 }im; s.d. = 3.9 pm; n = 150). Helix O.D. usually uniform but can contract by half over a translational distance of 10 pm. PfscMssfon.—Obruchevella blandita was originally described from silicified, three-dimensionally preserved material from the Upper Riphean of the Yenisey Ridge region, Siberia. Its two-dimensional counterpart is common in Svanbergfjellet shales where it occurs as intact units up to 100 pm long (Fig. 22.1); as disaggregating (Fig 22.2) and partially disaggregated fragments (Fig. 22.3-22.6); and as isolated ring structures (Fig. 22.5). The Svanbergfjellet population differs from the type material in its slightly narrower filament diameter (1.5 pm vs. 2.1-2.2 pm), greater range of helix O.D. (6-27 pm vs. 18-20 pm), and by the occasional occurrence of thick external annulations (Figs. 22.3, 22.6); these features are here considered reasonably accommodated by O. blandita. Disaggregating specimens furthermore appear indistinguishable from Glomovertella glomerata (cf., Jankauskas, 1980b; 1989). Our understanding of the overall habit of O. blandita corresponds closely with Tynni and Donner's (1980: p. 13-14) reconstruction of the larger 'Volvniella' cvlindrica. Material.—several hundred specimens from four shale horizons: 86-G-28, 86-G-30, 86-G-61, 86-G-62.

Phylum CYANOBACTERIA(?) Stanier et al., 1978 Class COCCOGONEAE(?) Thuret, 1875 Order CHROOCOCCALES(?) Wettstein, 1924 Family CHROOCOCCACEAE(?) Nageli, 1849 99

Genus EOSYNECHCXIOCCUS Hofmann, 1976, emend. Golovenoc and Belova, 1984

Non Eosvnechococcus eloneatus GOLOVENOC AND BELOVA, 1984, p. 29, PI.

2, fig. 5.

Type species.--Eosvnechococcns moorei. Hofmann, 1976, p. 1057-1058. Discussion.—Eosynechococais was originally erected for ellipsoidal microfossils with a length to width ratio of ca. 2:1 (Hofmann, 1976). Subsequently named species (excluding E. eloneatus) increased this ratio to 3.6:1 (Golovenoc and Belova, 1984: Table 2), which is here taken as the upper limit for the genus. The asymmetrical cell division(?) and markedly longer aspect ratio of E. elongatus (6:1) place it well outside reasonable limits for the genus.

EOSYNEGHOCOCCUS MOOREI Hofmann, 1976 Figure 23.10 Description.—Ellipsoidal vesicles approximately twice as long as wide (ca. 6 x 3 pm) and typically linked end to end; outer envelope absent. Material.—Several small populations in chert sample P-2664.

Genus SPHAEROPHYCUS Schopf, 1968 Type species.-Sphaerophvcus parvum Schopf, 1968, p. 672. 100

SPHAEROPHYCUS PARVUM Schopf, 1968 Figures 20.12-20.20 Descripfion.—Spheroidal vesicles, 2-3.5 pm diameter, usually without enveloping sheaths. Solitary or grouped into small regular colonies of 2-16 cells.

D jscmsszow .—Svanbergfjellet S. parvum closely approximate the size and structure of the type Bitter Springs material, the only significant difference being their tighter, more regular packing of pluricellular colonies. In one sample, 86-SP-S, they also exhibit an interesting behavioral aspect, occurring on and along the exposed surfaces of intraclastic grains (Fig. 20.20). Such distribution indicates that they actively colonized the recently exposed surfaces, while a geopetal preference for one surface (presumably the upper) supports their interpretation as photosynthesizers. These epilithic populations are almost certainly related to the less conspicuous S. parvum that occur within intraclasts of the same sample. Material.—Sewered hundred specimens in chert sample 86-SP-8; others in P-2664 and 86-G-15.

SPHAEROPHYCUS spp. Figure 20.11

D iscwsszom .—Colonial fossils comparable to named species of Sphaerophvcus but of considerably larger dimensions are occasionally encountered in Svanbergfjellet cherts. A distinctive population of ca. 16 pm diameter unicells, diads and tetrads occurs in close association with crust-forming Polvbessurus biparti tus in sample 86-P-89. 101

Class HORMOCONEAE(?) Thuret, 1875 Order OSCILLATORIALES(?) Copeland, 1936 Family OSCILLATORIACEAE(?) Kirchner, 1898 Genus CEPHALONYX Weiss, 1984

Calvptothrix. JANKAUSKAS, 1980, p. 107. Striatella ASSEJEVA, 1983, p. 6 {in Assejeva and Velikanov, 1983). Non Striatella MÀDLER, 1964, p. 189. Arthrosiphon WEISS, 1984, p. 106. Contextuopsis HERMANN, 1985, p. 150 {in Sokolov and Iwanowski, 1985). Rectia TANKAUSKAS. 1989, p. 121.

Type spgcies.—Cephalonvx coriaceus (Assejeva, 1983) n. comb., p. 6 {in Assejeva and Velikanov, 1983). Dia^osis.—Unbranched filamentous sheaths with annular thickenings that reflect the original positioning of cells (pseudo-cellular). Annulations typically of greater diameter than the intervening regions of the sheath. Discussion .-Pseudo-cellular fossil filaments were first described by Assejeva (1983) and given the binomial Striatella coriacea: however, a previous usage of Striatella by Madler (1964) has taxonomic priority (ICBN Article 64.1). The earliest legitimate generic name (ICBN Article 57.1) that can be applied to these fossils is Cephalonvx Weiss, 1984, a pseudo-cellular filament from the Late Riphean Miroedikha Suite, Siberia. Consequently, the type species is reconstituted as Cephalonvx coriaceus (Assejeva. 1983). Pseudo-cellular Cephalonvx is potentially confused with Palaeolvnebva; however, the regular annulations of Cephalonvx are simply externaK?) thickenings in a filamentous sheath, not true cellular remains (although they 102

do appear to reflect the original positioning of cells). In contrast, the annulations in pseudo-septate Tortunema represent the original positioning of intercellular septa, while the transverse fabric of the outer sheath of Rugosoopsis is entirely independent of the trichome.

CEPHALONYX GEMINATUS (Jankauskas, 1980) n. comb. Figure 23.1

Calvptothrix geminata TANKAUSKAS, 1980, p. 107-108, PI. 12, fig. 20.

DescrzpfioM.—Filamentous sheath, 17 pm wide, with prominent, 4-5 pm long annulations separated by 1-2 pm long thin-walled intervals; pseudo-cellular annulations marginally wider than intervening areas. Intact terminus simply rounded. Discussion.—Reexamination of the type material of Calvptothrix Schopf, 1968 shows it to be a true cellular trichome that, at least in places, retains a relatively thick mucilaginous sheath. It is clearly unrelated to pseudo-cellular Cephalonvx (including 'Calvptothrix' geminata). The single specimen of Svanbergfjellet Ç. geminatus n. comb, is marginally wider than the size range suggested by the original description (17 pm vs. 13-15 pm), but it clearly shows the same arrangement of pseudo-cellular thickenings and the thin inter-annular regions of the holotype. Material—A single specimen from shale horizon 86-G-62. 103

Genus CYANONEMA Schopf, 1968, emend.

Non Cvanonema disjoncta OGURTSOVA AND SERGEEV, 1987, p. 111-112, PI. 9, fig. 3-4. (= Oscillatoriopsis).

Type species.--Cvanonema attenuata Schopf, 1968, p. 670. Emended DwgTiosis.—Unbranched uniseriate cellular trichomes in which cell length is greater than cell diameter (length:width > 1). Enveloping sheath absent. Not at all to moderately constricted at septa. Discwssion.—The distinction between 'cell length < diameter' (Oscillatoriopsis) vs. 'cell length > diameter' (Cvanonema) is certain to be artificial to some extent; it nevertheless serves as an unambiguous character in form taxonomy. Cell length to width ratios are similarly applied in the botanical classification of m odem cyanobacteria (e.g., Geitler, 1925). In addition to the type, four species of Cvanonema have appeared in the literature: Ç disiuncta has cells shorter than wide and therefore belongs in Oscillatoriopsis. while the others, Ç inflatum Oehler, 1977, Ç. minor Oehler, 1977, and Ç ligamen Zhang Y., 1981 would appear to be accommodated by Ç attenuata.

CYANONEMA sp. Figure 24.10 Discussion.—A single trichome in the upper Svanbergfjellet shales has cells ca. 7 pm wide by 12 pm long and is therefore classified as Cvanonema. Its considerably larger dimensions exclude it from Ç attenuata. Material.-One specimen from shale horizon 86-G-62. 104

Genus OSCILLATORIOPSIS Schopf, 1968, emend.

Halvthrix SCHOPF, 1968, p. 678, PI. 77, fig. 7. Non Oscillatoriopsis? hubeiensis YIN AND LI, 1978, p. 88-89, PI. 7, fig. 9. Nec Oscillatoriopsis robusta HORODYSKI AND DONALDSON, 1980, p. 149-152, fig. 13H. (= Palaeolvnebva). Nec Oscillatoriopsis bothnica TYNNI AND DONNER, 1980, p. 15, PI. 7, fig. 83. (= Tortunema). Nec Oscillatoriopsis constricta TYNNI AND DONNER, 1980, p. 15, PI. 7, figs. 82, 85-86. (= Tortunema). Nec Oscillatoriopsis maena TYNNI AND DONNER, 1980, p. 14, PI. 6, figs. 64- 66, 68-70. (= Tortunema). Nec Oscillatoriopsis bacillaris HERMANN. 1981b, p. 121, PI. 12, fig. 13. (= Tortunema). Nec Oscillatoriopsis nochtuica YAKSCHIN. 1981, p. 31, PI. 11, fig. 3 (in Yaks chin and Luchinina, 1981). Nec Oscillatoriopsis tomica YAKSCHIN. 1981, p.32, PI. 12, figs. 1, 2, 5 (in Yaks chin and Luchinina, 1981). Nec Oscillatoriopsis acuminata XU. 1984, pp. 218, 312, PI. 1, figs. 3-4, 6. (= Siphonophvcus). Nec Oscillatoriopsis disciformis XU, 1984, pp. 218-219, 313, PI. 3, fig. 7. (= Siphonophvcus). Nec Oscillatoriopsis elabra XU 1984, pp. 219, 313, PI. 2, figs. 6, 8A. (= Siphonophvcus). Nec Oscillatoriopsis hemisphaerica XU. 1984, pp. 218, 312, PI. 1, figs. 7-8, PI. 2, fig. 12. (= Siphonophvcus). 105

Nec Oscillatoriopsis tuberoilata XU, 1984, pp. 219, 313, PI. 1, figs. 1, 2. (= Siphonophvcus). Nec Oscillatoriopsis funiformis RAGOZINA, 1985, p. 142-143, PI. 59, figs. 2-3 (in Sokolov and Iwanowski, 1985). Nec Oscillatoriopsis rhomboidalis SIVERZEVA, 1985, p. 143, PI. 60, figs. 4, 6-7 (in Sokolov and Iwanowski, 1985). (= Pomoria). Nec Oscillatoriopsis hechiensis LIU AND LI, 1986, p. 266, PI. 1, fig. 2. Nec Oscillatoriopsis subtilis ZHANG P., ZHU, AND SONG, 1989, p. 323-324, PI. 1, figs. 3, 4. (= Cvanonema). Nec Oscillatoriopsis aneusta (Kolosov, 1984) JANKAUSKAS, 1989, p. 116. (= Tortunema).

Type spgc/es.--Oscillatoriopsis obtusa Schopf, 1968, p. 667. Emended Dwgnosfs.—Unbranched, uniseriate cellular trichomes in which cell length is less than or equal to cell diameter (length:width < 1). Enveloping sheath absent. Not at all to moderately constricted at septa. PzscMsszoM.—Oscillatoriopsis is here emended to include only those filamentous microfossils that can be confidently identified as cellular trichomes (vs. pseudo-septate sheaths - see Tortunema). lack an enveloping sheath (us. sheathed - see Cephalonvx. Palaeolvnebva. Rugosoopsis), have a cell lengthzwidth ratio less than or equal to 1 (vs. > 1 - see Cvanonema). and are not severely constricted at intercellular septa (vs. severely constricted - see Veteronostocale). all these being characteristics of the type species. Through failing to satisfy one or more of these criteria, all the above listed species are hereby removed from Oscillatoriopsis. Given the marked variation recorded in terminal cell morphology, trichome length, and septal constriction of both living and moribund 106 oscillatorian cyanobacteria (Hofmann, 1976; Shukovsky and Halfen, 1976; Horodyski et al., 1977; Golubic and Barghoorn, 1977; Haxo et al., 1987), such features are not considered significant in considering the generic-level taxonomy of fossil forms; in a natural taxonomy they may even find limited application to species differentiation. Thus, a number of previously named genera fall clearly within the morphological limits of Oscillatoriopsis. although, on the basis of unique subsidiary features, some of these may stand as distinct species within the genus (e.g., a new combination, O. nodosa, could accommodate the oscillatoriopsan Halvthrix nodosa Schopf. 1968). The isodiametric cell dimensions of O awramikii Wang et al., 1983 and the larger O. cuboides Knoll et al., 1988 are also sufficiently distinctive to warrant their retention as separate species. Most other 'species' of Oscillatoriopsis have cells considerably wider than long. Over 75 species of Oscillatoriopsis or Oscillatoriopsis as herein emended appear in the literature; 56 are accepted as belonging to the genus. Much of this purported diversity derives from relatively few and often very localized assemblages (Schopf, 1968; Schopf and Blade, 1971; Knoll, 1981; Zhang P., 1981; 1982), with almost all 'spedes' known from unique or very few specimens. The recognition of substantial intra-spedfic and taphonomic variation now dictates a more conservative taxonomy. The delineation of Osdllatoriopsis spedes should be based primarily on the identification of discrete size dasses as determined from large populations (with the widest point of a trichome taken as the closest approximation of its original width). Most of the 112 recorded Svanbergfjellet specimens of Osdllatoriopsis are drcumscribed by a unimodal size-frequency distribution between 3 and 8 pm diameter (= O. obtusa). The outliers from this distribution differ also in subsidiary qualitative characters, thus supporting their assignment to two 107 other size classes (form species) that appear in the literature, O. amadea n. comb, and O. longa. In summary, I recognize four basic species of Oscillatoriopsis with diameters less than 25 pm: O. minuta n. comb., 1-3 pm wide; O. obtusa, 3-8 pm wide; O. amadea n. comb., 8-14 pm wide; and O. longa, 14-25 pm wide.

OSCILLATORIOPSIS MINUTA (Schopf, 1968) n. comb.

Cephalophvtarion minutum SCHOPF, 1968, p. 669-670, PI. 78, fig. 9-12. Contortothrix vermiformis SCHOPF. 1968, p. 671, PI. 79, fig. 7-8. Anabaenidium Tohnsonii SCHOPF, 1968, p. 680-681, PI. 81, fig. 4. Archaeonema loneicellularis SCHOPF. 1968, p. 678-679, PI. 80, fig. 11.

Diagnosis.~A species of Oscillatoriopsis with trichomes 1-3 pm diameter. Discussion.—Some Svanbergfjellet trichomes measure as little as 3 pm but they clearly represent the lower limit of the O. obtusa distribution. In contrast, the small oscillatoriopsans that occur in the Ross River locality of the Bitter Springs Formation (Schopf, 1968) appear to represent a distinct size class and warrant separate species status within Oscillatoriopsis (both Contortothrix and Anabaenidium are structurally indistinguishable from Oscillatoriopsis. while their coiled habit must be considered taxonomically unreliable in light of their rare occurrence; reexamination of the type specimen of Archaeonema shows it to be a poorly preserved O. minuta n. comb.). Oscillatoriopsis minuta has not been observed in the Svanbergfjellet assemblage. 108

OSCILLATORIOPSIS OBTUSA Schopf, 1968, emend. Figures 24.1-24.5, 24.11

Caudiculophvcus rivularioides SCHOPF, 1968, p. 679-680, PI. 79, figs. 3-6. Cephalophvtarion grande SCHOPF, 1968, p. 669, Pl. 78, figs. 1-4. Caudiculophvcus acuminatus SCHOPF AND BLACIC, 1971, p. 951, Pl. 105, fig. 7. Cephalophvtarion constrictum SCHOPF AND BLACIC, 1971, p. 943-944, Pl. 105, figs. 1, 9. Cephalophvtarion delicatulum SCHOPF AND BLACIC, 1971, p. 946, Pl. 108, fig. 7. Cephalophvtarion grande (Schopf, 1968) SCHOPF AND BLACIC, 1971, p. 956, Pl. 106, figs. 1-2, 4, 9-10. Cephalophvtarion laticellulosum SCHOPF AND BLACIC, 1971, p. 944, Pl. 105, figs. 2, 6, Pl. 106, fig. 3. Cephalophvtarion variabile SCHOPF AND BLACIC, 1971, p. 944-946, Pl. 107, figs. 2, 3, 5, 8, Pl. 108, fig. 3. Filiconstrictosus diminutus SCHOPF AND BLACIC, 1971, p. 948-950 (partim), Pl. 106, figs. 7, 8. Filiconstrictosus maiusculus SCHOPF AND BLACIC, 1971, p. 947-948, Pl. 105, fig. 8. Oscillatoriopsis breviconvexa SCHOPF AND BLACIC, 1971, p. 943, Pl. 105, fig. 5. Palaeolvnebva minor SCHOPF AND BLACIC, 1971, p. 942-943, Pl. 105, fig. 4. Partitiofilum eonevloides SCHOPF AND BLACIC, 1971, p. 947, Pl. 105, fig. 3, Pl. 106, fig. 6. Oscillatoriopsis schopfii OEHLER. 1977, p. 344, fig. 13A. 109

Oscillatoriopsis psilata MAITHY AND SHUKLA, 1977, p. 179, Pl. 2, fig. 12. Oscillatoriopsis anshanensis YIN, 1979, p. 47, Pl. 2, figs. 10, 12. Oscillatoriopsis curta HORODYSKI AND DONALDSON, 1980, p. 149, figs. 13A-G. Cephalophvtarion taenia ZHANG Y., 1981, p. 493, Pl. 1, figs. 8-10. Oscillatoriopsis iixianensis ZHANG P., 1981, p. 254, Pl. 1, figs. 1, 2, 4. Oscillatoriopsis luozhuangensis ZHANG P., 1981, p. 255. Pl. 1, figs. 5, 8. Oscillatoriopsis qingshanensis ZHANG P., 1981, p. 256, Pl. 1, figs. 3, 6. Cephalophvtarion turukhanicum WEISS, 1984, p. 104-105, Pl. 9, fig. 4. Filiconstrictosus eniseicutn WEISS. 1984, p. 105, Pl. 9, fig. 5. Cephalophvtarion piliformis MIKHAILOVA, 1986, p. 35, figs. 8-9. Oscillatoriopsis parvula LIU AND LI, 1986, pp. 265, 266, 270, Pl. 1, fig. 4. Primorivularia dissimilara HERMANN, 1986, p. 39, figs. 13-14. Primorivularia absoluta HERMANN. 1986, p. 39-40, fig. 12. Cvanonema disiuncta OGURTSOVA AND SERGEEV, 1987, 111-112, Pl. 9, figs. 3-4. Obconicophvcus minor YIN, 1987, p. 479, Pl. 28, figs. 5, 8.

Emended Diagnosis.~A spedes of Osdllatoriopsis with trichomes 3-8 pm diameter; cells wider than long. Descrzpfiow.—Cellular trichomes, 3-8 pm wide (% = 5.8 pm; s.d. = 1.2 pm; n = 109), with no extracellular envelope. Complete trichomes from 50 to 200 pm long; usually occurring as isolated individuals, but in one instance as a loose assodation of ca. 50 trichomes extending over ca. 10 mm^ of a shale bedding plane. Trichome termini blunt, rounded, or tapered; typically one end rounded and the other gradually tapering. Cells 1.5 - 3.5 pm long, rarely constricted at septa. 110

D/scwssio«.—Svanbergfjellet O. obtusa exhibit considerable variation in terminal cell morphology and overall length, but their otherwise uniform size (width) and habit argue convincingly against these features being applied to species-level taxonomy. Likewise, cell length seems to be somewhat variable, differing by up to 1 pm within a trichome and up to 2 pm between trichomes. All of the above synonymized species fall within the size limits defined by the large, unimodal population of Svanbergfjellet O. obtusa. Given the disparate taphonomic and geologic histories that these various 'species' represent, their morphological variation does not exceed what might be reasonably expected from a single Oscillatoria-like precursor. On the other hand, the fundamental simplicity of such fossil trichomes precludes a natural taxonomy, and O. obtusa may well represent a diverse, though now undifferentiable assortment of taxa and physiologies (p. 31). Material.~l09 trichomes from three shale horizons; 86-G-62, 86-G-28, P-2945.

OSCILLATORIOPSIS AMADEA (Schopf and Blacic, 1971) n. comb. Figure 23.5

Obconicophvcus amadeus SCHOPF AND BLACIC, 1971, p. 950, PI. 107, fig. 1. Oscillatoriopsis media MENDELSON AND SCHOPF, 1982, p. 64-65, PI. 4, figs. 3, 5-6. Oscillatoriopsis acuta ZHANG P., 1982, pp. 36, 40, PI. 1, fig. 4. Oscillatoriopsis doliocellularis ZHANG P., 1982, pp. 37, 40, PI. 1, fig. 6. Oscillatoriopsis formosa ZHANG P., 1982, pp. 37, 40, PI. 1, fig. 9. Oscillatoriopsis taimirica SCHENFIL, 1983, p. 473, PI. 1, fig. 2. Oscillatoriopsis connectens ZHANG P., 1987. p. 268. PI. 1, figs. 5-6. Cephalophvtarion maiesticum ALLISON, 1989, p. 273, figs. 8.7-8.S. I ll

Diagnosis.—A spedes of Osdllatoriopsis with trichomes 8-14 pm diameter; cells wider than long. DescrzpfioM .—Sheathless, non-tapered cellular trichome, 10 pm wide; cells ca. 4 pm long and narrowly separated from adjacent cells. Discussion.—A single specimen of Oscillatoriopsis in the Algal Dolomite Member falls between the size distributions of O. obtusa and the larger O. longa (see below). Given its additional qualitative differences (all cells separated and lack of terminal taper) it is reasonably assigned to a separate taxon, O. amadea n. comb, (alternatively, this spedmen may have derived from a pseudo-cellular sheath such as Cephalonvx (cf.. Fig. 23.1). The two original Bitter Springs spedmens of Osdllatoriopsis amadea (Obconicophvcus amadeus) do not provide a useful size range for the species. On purely artifidal grounds the size range falling between that of O. obtusa and O. longa might be suggested as it offers approximately the same degree of variation in trichome width as that documented for O. obtusa. That the 8-14 pm size range drcumscribes a natural taxonomic grouping nevertheless awaits confirmation from a single large population.

Material—A single spedmen in shale horizon 86-G-62.

OSCILLATORIOPSIS LONGA Timofeev and Hermann, 1979, emend. Figures 24.6-24.7

Osdllatoriopsis longus TIMOFEEV AND HERMANN, 1979, p. 139, PI. 29, figs. 3,4. Oscillatoriopsis major LIU. 1982, p. 148, PI. 11, fig. 1 Hyalothecopsis nanshanensis ZHANG P., 1982, pp. 35-36, 39, PI. 1, fig. 2. Hvalothecopsis sinica ZHANG P., 1982, pp. 35, 39, PI. 1, fig. 1. 112

Halvthrix leningradica SCHENFIL, 1983, p. 473, PL 1, fig. 1. Osdllatoriopsis variabilis STROTHER. KNOLL, AND BARGHOORN, 1983, p. 26, PI. 3, figs. 3-6,11. Osdllatoriopsis princeps ZHANG P. AND Y AN, 1984, pp. 198, 203, PI. 1, fig. 6. Osdllatoriopsis aculeata ZHANG P. AND Y AN, 1984, pp. 199, 203, PI. 1, fig. 7. Osdllatoriopsis connectens ZHANG P. AND GU, 1986, p. 15-16, PI. 1, fig. 10. Osdllatoriopsis strictura ZHANG P. AND GU, 1986, p.l6, PI. 1, fig. 8. Osdllatoriopsis valida ZHANG P. AND GU, 1986, p. 16, PI. 2, figs. 1, 4. Osdllatoriopsis planaria ZHANG P. AND GU, 1986, p. 16, PI. 1, fig. 12. Partitiofilum cungusum MIKHAILOVA, 1989, p. 118, PI. 27, fig. 4 (in Jankauskas, 1989).

Emended Diagnosis.~A spedes of Osdllatoriopsis with trichomes 14-25 pm diameter; cells wider than long. Discussion.—As with O. amadea. a natural accounting of O. longa suffers from inadequate population sizes and/or detailed documentation. The upper size limit of 25 pm is taken from that originally suggested for O. longa and Parti tiofilum tungusum; the lower is suggested by the dearth of reported trichomes 13-14 pm wide, and the absence of trichomes >14 pm wide in most assemblages containing smaller osdllatoriopsan^. That the type specimen of O. longa is from shale rather than three dimensionally preserved in chert is not considered significant (p. 28). Instances of cellular disaggregation show the Svanbergfjellet specimens to be true trichomes rather than pseudoseptate sheaths. Niaterial.—S trichome fragments, from shale horizons 86-G-61 and 86-G-28. 113

Genus PALAEOLYNGBYA Schopf, 1968, emend.

Paleolvngbva HERMANN, 1974, p. 8-9. Rhicnonema HOFMANN, 1976, p. 1053. Non Palaeolvngbva minor SCHOPF AND BLACIC, 1971, p. 942-943. PI. 105, fig. 4. (= Osdllatoriopsis). Nec Palaeolvngbva sinica YIN AND LI, 1978, p. 89, PI. 7, fig. 10. Nec Palaeolvngbva spiralis WANG, ZHANG, AND GUO, 1983, p. 162-164, figs. 22.1-22.3, 22.7, 22.9. (= Obruchevella).

Type spgdes.—Palaeolvngbva barghoomiana. Schopf 1968, p. 665-666. Emended Dw^osfs.—Unbranched, uniseriate, multicellular trichomes with a single extracellular envelope. Discussion.—As a form taxon, Palaeolvngbva will certainly intergrade with other filamentous microfossils (cf., Hoftnann, 1976); I suggest that the name be reserved for smooth-walled filamentous sheaths that contain a regular array of prominently preserved cells. Thus, the occasional preservation of cells in Siphonophvcus (Fig. 27.4), Tortunema (Fig. 27.1), or Rugosoopsis (Fig. 25.1- 25.2) do not warrant their indusion in Palaeolvngbva. Similarly, the regular helices of P. spiralis Wang et al., 1983, along with its ambiguously preserved cells, argue for its transfer to Obruchevella. Cells within filamentous sheaths often show signs of severe shrinkage and their dimensions cannot be considered as useful characters, even within a form taxonomy. Uncollapsed sheath diameter is therefore the prindpal criterion for determining spedes of Palaeolvngbva. 114

PALAEOLYNGBYA CATENATA Hermann, 1974 Figures 25.6-25.7

Osdllatoriopsis robusta HORODYSKI AND DONALDSON, 1980, p. 149-152, fig. 13H. Palaeolvngbva maxima ZHANG Y., 1981, p. 495, Pl. 2, figs. 4, 6-7. Doushantuonema peatii ZHANG Z., 1981, p. 204, Pl. 1, figs. a-c.

Diagnosis.—A spedes of Palaeolvngbva with sheaths 10-25 pm diameter. DascripfioM.—Filamentous sheaths, 8-16 pm wide {x = 12.1 pm; s.d. = 2.5 pm; n = 8), with a uniseriate array of partially degraded cells. Discussion.—Originally described from relatively few specimens, the size range of P. catenate may be subject to some future modification; the holotype measures ca. 15 pm across. Svanbergfjellet P. catenate occur in both shales (Fig. 25.7) and subtidal silidfied carbonates (Fig. 25.6). Material.—3 specimens from shale horizon 86-G-62; 5 in chert sample P-2628.

PALAEOLYNGBYA HEBEIENSIS Zhang and Yen, 1984, emend. Figures 25.5, 25.8

Palaeolvngbva conicus LIU AND LI, 1986, pp. 266, 270, PI. 1, fig. 5.

Palaeolvngbva sphaerocephala HERMANN AND PYLINA, 1986, p. 38, figs. 2-9 (in Hermann, 1986).

Emended Diagnosis.—A spedes of Palaeolvngbva with sheaths 34-46 pm diameter. 115

Description.—Filamenious sheaths, 36-46 wide (x = 42.3 pm; s.d. = 3.8 pm; n = 4), with a uniseriate array of shrunken or partially degraded cells. PfscwsszoM.—Palaeolvngbva hebeiensis was originally described from a single indifferently-preserved fragment from the Mesoproterozoic Gaoyuzhuang Formation, China. The diameter of its "hyaline" sheath was not stated, but appears to be little more than that of the cells (ca. 35 pm). Palaeolvngbva sphaerocephala is of this same size class (34-42.5 pm) and is therefore a junior synonym of P. hebeiensis. Palaeolvngbva hebeiensis can clearly grade into the form genera Rugosoopsis (Fig. 25.5: note the single transverse wrap outside the sheath), Tortunema (e.g., Hermann, 1986: fig. 7-8), and, of course, Siphonophvcus. The specimen in Figure 25.8 is unique in that the trichome has remained fully intact rather than separating into the more typical series of isolated cells. Material.—4 specimens from shale horizon 86-G-62.

PALAEOLYNGBYA spp. Figure 25.9 Discussion.—Narrow, transversely banded filaments are conspicuous in the Svanbergfjellet shales, generally co-occurring with a variety of sheathed filamentous taxa; in some instances they can be observed to be physically contiguous. Close examination shows them to be uniseriate trichomes of shrunken cells about which an enveloping sheath has collapsed, much as would be expected if cells had been preserved within the collapsed portion of the sheath shown in Figure 24.8. As the dimension of shrunken cells cannot be taken as a reliable taxonomic character, these Palaeolvngbva fossils are not classifiable to the species level. Material.—4 measured specimens from shale horizon 86-G-62. 116

Genus RUGOSOOPSIS Timofeev and Hermann, 1979, emend.

Plicatidium JANKAUSKAS, 1980, p. 109, PI. 12, fig. 15.

Type specfes.--Rugosoopsis tenuis Timofeev and Hermann, 1979, p. 139. Emended Diagnosis.—Bi-lnyered filamentous sheaths; inner sheath smooth or pseudoseptate; outer sheath with a prominent transverse fabric. Descn'ptfo«.--Siphonophvcus-like or Tortunema-like filamentous sheaths, wholly or partially enclosed in a second layer having a pronounced transverse fabric. Atrophied cells occasionally preserved within the inner sheath. Dzscwsszo».—Differential preservation of the outer sheath results in a wide, but superficial variation in Rugosoopsis form. When largely retained it imparts a finely plicated or cellular texture to filaments (Figs. 25.1-25.2). More commonly, and to varying degrees, this outer layer 'unravels', exposing portions of the underlying pseudo-septate or smooth-walled sheaths (Figs. 25.3-25.5); the unraveled sheath often remains associated as an entangled 'filamentous' halo (Figs. 25.9, 27.1, 27.2). A clear gradation of Rugosoopsis into Tortunema (Fig. 27.2), Siphonophvcus (Fig. 25.1-25.4), and Palaeolvngbva (Fig. 25.5: note the single outer 'wrap' at the arrow) suggests that these various forms may all belong to the same natural taxon. Multiple extracellular sheaths are characteristic of a number of modern filamentous cyanobacteria, most notably species of Scvtonema and Lvngbva: as with fossil Rugosoopsis. these multiple sheaths may not all be of a similar construction. For example, in their examination of degrading L. aestuarii at Laguna Mormona, Horodyski et al. (1977) distinguished an inner pliant-walled sheath from a clearly more rigid outer layer (Horodyski et al., 1977: fig. 6M; compare with Fig. 25.3). Interestingly, the degrading outer layer of L. aestuarii 117 appears to develop a substantial Rueosoopsis-like transverse fabric (Horodyski et al., 1977: fig. 6Q). Most Rugosoopsis are thus reasonably interpreted as the extracellular sheaths of oscillatoriacean cyanobacteria. Pjatiletov (1988) emended Rugosoopsis to include Plicatidium Jankauskas, 1980, and, on the basis of large populations, found a distinct break in sheath size-frequency distribution at ca. 50 pm diameter: those less than 50 pm were considered R. tenuis, those larger, R. latus (Jankauskas, 1980). He named an additional large (60-150 pm) form, R. rugososiusculus, based on purported differences in the transverse banding.

RUGOSOOPSIS TENUIS Timofeev and Hermann, 1979, emend. Figures 25.1-25.4, 27.2

Tubulosa corrugata ASSEIEVA, 1982, p. 13, PI. 2, figs. 10,11. Karamia costata KOLOSOV, 1984, p. 41-42, PI. 6, fig. 1. Karamia iazmirii. KOLOSOV, 1984, p. 40-41 (partim). PI. 5, figs. 1, 2.

Karamia segmentata KOLOSOV, 1984, p. 40, PI. 3, fig. 1, PI. 4, fig. 1. Siphonophvcus costatus. YIN, 1987, p. 480, PI. 11, figs. 1-4.

Emended Diagnosis.—A spedes of Rugosoopsis with sheaths less than 60 pm diameter. DiscMssion.—Svanberg^ellet Rugosoopsis range from 7 pm to 57 pm in width with a mode at ca. 30 pm {x = 29.1 pm; s.d. = 13.4 pm; n = 56), but little indication that more than a single population is represented. This substantially increases the 30-37.5 pm range proposed in its original description (Timofeev and Hermann, 1979), and marginally exceeds that suggested by Pjatiletov (1988). Most R. tenuis have been reported from shale fades; however, two 118 small (ca. 10 pm diameter) specimens were recorded from silicified carbonates in the Lower Dolomite and Algal Dolomite members; the silicified Siphonophvcus specimen in Figure 21.5 has associated material that may represent a disaggregated outer sheath. The above synonymized taxa share with R. tenuis a transverse banding outside an extracellular filamentous sheath that is unrelated to the cellular trichome (cf., Cephalonvx). Material.—54 specimens from shale horizons 86-G-62 and 86-G-61; 2 from chert samples P-2664-4A and 86-G-14-1A.

Genus SIPHONOPHYCUS Schopf, 1968, emend. Knoll et al. 1991

Non Siphonophvcus attenuatum WEISS. 1989, p. 121-122, PI. 25, fig. 6-7 (in Jankauskas, 1989). (= Cephalonvx).

Type gpeczes.—Siphonophvcus kestron Schopf. 1968, p. 671.

D zscmssiom .—Following a number of taxonomic revisions (e.g., Zhang Z., 1982, Pjatiletov, 1988; Knoll et al., 1991) all unbranched, smooth-walled, (originally) tubular sheaths have come to be classified under the form genus Siphonophvcus. Species are, in turn, defined on the basis of discrete size (width) classes as determined from reasonably large populations (Knoll et al., 1991). Although correct in principle, these above revisions were typically carried out at the genus level thereby failing to recognize the taxonomic priority of some of the synonymized species (e.g., S. tvpicum n. comb, and S. gracile n. comb.). The SvanbergQellet fossil assemblage includes at least six species of Siphonophvcus. all but the smallest of which are interpreted as cyanobacterial. 119

SIPHONOPHYCUS THULENEMUM n. sp. Figure 22.9 Diagnosis.—A spedes of Siphonophvcus ca. 0.5 pm diameter.

Description.—UnhTanched, smooth-walled filamentous microfossils 0.5 pm diameter and up to several hundred microns long. Typically gregarious, forming both small sub-parallel assodations and relatively extensive mats. Discussion.—Archaeoirichion contortum has traditionally been the taxon to which sub-micron diameter fossil filaments were assigned; however, reexamination of the type shows it to be a spedmen of Siphonophvcus septatum (see below). Siphonophvcus thulenemum n. sp. represents a smaller size dass quite distinct from S. septatum or any other previously described taxa, and its repeated and well-defined occurrences in bedding-parallel thin section show it not to be a degradational variant of some larger form; its width does not vary appredably from 0.5 pm. As with other small spedes of Siphonophvcus (Knoll et al. 1991), physiologic and higher taxonomic determination of S. thulenemum must remain speculative. Etymology.—VTom. the Greek thoule - northernmost, and nema - thread. Material.—Three populations (several hundred filaments) from shale horizon 86-C-62. Holotype.—HUPC 62718; Figure 22.9; Slide 86-G-62-46; England-Finder coordinates: R-28-3.

SIPHONOPHYCUS SEPTATUM (Schopf, 1968) Knoll et al., 1991 Figures 22.7-22.8, 10.8

Archaeotrichion contortum SCHOPF. 1968, p. 686, PI. 86, fig. 1-2. Eomvcetopsis? campvlomitus LO, 1980, p. 143-144, PI. 1, fig. 9-11. 120

Allachiunica tenuiuscula KOLOSOV, 1982, p. 80, PI. 14, fig. 1. Tudomophvton microscopicum KOLOSOV. 1982, p. 75, PI. 11, fig. 1. Siphonophvcus chuii LIU, 1982, p. 148, PI. 11, figs. 5, 8, 11. Eophormidium capitatum XU, 1984, pp. 219, 314, PI. 1, fig. 5. Archaeotrichion lacerum HERMANN, 1989, p. 88, PI. 39, fig. 4 (in Jankauskas, 1989).

Description .--Unbranched, aseptate, smooth-walled filamentous microfossils 1- 2 pm diameter. Discussion.—Siphonophvcus septatum in the Svanbergfjellet Formation is found only in shale facies where it occurs as isolated individuals (Fig. 22.8); in loose, sometimes regularly criss-crossing associations; or, in one instance, as a tightly intertwined triple helix (Fig. 22.7; cf., Flagellis Assejeva, 1982). Siphonophvcus septatum filaments are also repeatedly found lining a central space in multicellular Valkvria n. gen., suggesting their possible role as biodegraders. Alternatively, they might be interpreted as symbiotic with (or parasitic upon) the larger organism. In any event, this variety of habits clearly illustrates the incorporation of diverse organisms within the form taxon S. septatum. Reexamination of the type specimen of Archaeotrichion contortum shows it, where not collapsed, to be a full micron in diameter and therefore a junior synonym of S. septatum. The type material of A. lacerum likewise has an original diameter of at least 1 pm, while its diagnosed intertwining habit does not appear to warrant separate species designation. Material.—! relatively extensive 'mat' populations and several isolated specimens from shale horizon 86-G-62; 15 occurrences within Valkvria n. gen. 121

SIPHONOPHYCUS ROBUSTUM (Schopf, 1968) Knoll et al., 1991 Figures 26.1, 26.7

Eomvcetopsis filiformis SCHOPF, 1968, p. 685-686, Pl. 82, figs. 1, 4, Pl. 83, figs. 5-8. Archaeonema longicellularis, SCHOPF AND BARGHOORN, 1969, p. 117, Pl. 21, figs. 1-4, Pl. 22, figs. 2-4. Eomvcetopsis psilata MAITHY AND SHUKLA, 1977, p. 180, Pl. 2, fig. 15. Beckspringia communis LICARI, 1978. p. 779-780. Pl. 1, figs. 3-7. Acranella granulata KOLOSOV, 1982, p. 81, Pl. 13, fig. 1. Allachiunica daedalea KOLOSOV. 1982, p. 79-80, Pl. 12, fig. 2. Tudomophvton minisculum KOLOSOV, 1982, p. 76, Pl. 11, fig. 2. Eomvcetopsis polesicus ASSETEVA. 1983, p. 170, Pl. 7, fig. 12. Eophormidium liangii XU, 1984, pp. 219, 314, Pl. 3, figs. 3-4. Eophormidium semicirculare XU. 1984, pp. 219, 314-315, Pl. 2, figs. 1-2. Osdllatoriopsis acuminata XU, 1984, pp. 218, 312, Pl. 1, figs. 3-4, 6. Osdllatoriopsis disdformis XU, 1984, pp. 218-219, 313, Pl. 3, fig. 7. Osdllatoriopsis elabra XU 1984, pp. 219, 313, Pl. 2, figs. 6, 8A. Osdllatoriopsis hemisphaerica XU, 1984, pp. 218, 312, Pl. 1, figs. 7-8, Pl. 2, fig- 12. Osdllatoriopsis tuberculata XU, 1984, pp. 219, 313, Pl. 1, figs. 1-2. Schizothropsis caudata XU. 1984, pp. 219, 315, Pl. 3, figs. 1-2, 10, Pl. 2, figs. 3-4.

Description.—Unbranched, aseptate, smooth-walled filamentous microfossils 2- 3 pm diameter.

Discussion .—Siphonophvcus robustum commonly serves as an accessory builder in silicified FYoterozoic microbial mat communities (e.g., Schopf, 1968; 122

Green et al., 1989; Knoll et al., 1991). It seems not to occur in Svanbergfjellet cherts, however it is the principal constituent of shale facies microbial mats, the entangled filaments commonly extending for nearly a centimeter in both bedding-parallel thin sections (Fig. 26.1) and macerated material. Siphonophvcus robustum also occurs as the sole constituent,of circular (originally spheroidal?) structures up to several hundred microns in diameter (Fig. 26.7) which may be broadly comparable to modem Nostoc balls. The ca. 4 mm spheroidal compression in Figure 8.6 appears to have a similar filamentous construction, illustrating the potential confusion of such compound structures with large acritarchs (cf.. Sun, 1987a). Material—27 mat-forming populations extending laterally for more than 1 mm (22 in thin section, 5 isolated by maceration), from three shale horizons: 86-G- 62, 86-G-61, 86-G-30. Three spheroidal 'colonies' from 86-G-62.

SIPHONOPHYCUS TYPICUM (Hermann, 1974) n. comb. Figures 26.2, 26.8-26.9, 23.2-23.4

Leiothrichoides tipicus HERMANN, 1974, p. 7, PI. 6, figs. 1-2. Eomvcetopsis cvlindrlca MAITHY, 1975, p. 140, PI. 4, figs. 27-28. Eomvcetopsis rugosa MAITHY, 1975, p. 140, PI. 4, figs. 25-26. Eomvcetopsis pflueii MAITHY AND SHUKLA, 1977, p. 180, PI. 2, fig. 16. Siphonophvcus crassiusculum HORODYSKI. 1980, p. 656, PI. 1, figs. 6-7. Eomvcetopsis rimata TANKAUSKAS. 1980b, p. 111-112, PI. 12, fig. 11. Eomvcetopsis? siberiensis LO, 1980, p. 139-143, PI. 1, figs. 1-8. Siphonophvcus inomatum ZHANG Y., 1981, p. 491-493, PI. 1, figs. 1, 3-5. Tudomophvton vulgatum KOLOSOV. 1982, p. 77, PI. 11, fig. 4, PI. 12, fig. 1. Tudomophvton multum KOLOSOV, 1982, p. 76-77, PI. 11, fig. 3. 123

Sacharia crassa KOLOSOV, 1982, p. 79. PI. 12, fig. 3. Uraphvton rectum KOLOSOV, 1982, p. 82, PI. 13, fig. 2. Siphonophvcus hughesii NAUTIYAL, 1982, p. 175, figs. lA-G. Eomvcetopsis pachvsiphonia ZHU, 1982, p. 6, PI. 1, figs. 1-4. Eomvcetopsis crassiusculum (Horodyski, 1980). ZHANG Z., 1982, p. 455-456, PI. 47, figs. 3-6, 9-13. Eomvcetopsis crassus YIN, 1985b, p. 180, PI. 1, figs. 1-2, 5-6, PI. 2, fig. 9.

Descrzph'on.“ Unbranched, aseptate, smooth-walled filamentous microfossils 4- 8 pm diameter. Some grainstone interstitial specimens with discrete thickened or darkened intervals. Discussion .—Siphonophvcus tvpicum is the dominant mat builder in stromatolitic facies of the Svanbergfjellet Formation. It occurs most commonly as extremely dense, unoriented and entangled filamentous sheaths (Fig. 26.8), sometimes forming over 1 cm thick accumulations of sediment-free, probably supratidal, microbial mat. With the input of clastic sediments a more oriented fabric of alternating vertical and horizontal filaments is developed (Fig. 26.9); the higher taxonomic diversity of these sediment-bearing mats suggests deposition in lower tidal flat environments (Knoll et al., 1991). Both of these microbial mat fabrics are recorded primarily from the intraclasts of silicified microbialite grainstones, thus their paleoenvironmental interpretation is inferential (p. 15). The interstitial spaces of these redeposited clasts represent a further microenvironment exploited by S. tvpicum. Here the filaments tend to be solitary, very straight, and commonly exhibit 10-100 pm long intervals of 'pointilistically^ darkened (biomineralized?) intervals (Figs. 23.2-23.4; p. 26). Siphonophvcus tvpicum is less regularly encountered in Svanbergfjellet shales but does occasionally occur as entangled mats (Fig. 26.2). It is readily 124

distinguished from co-occurring S. robustum by its thicker, seemingly more rigid walls, as well as its larger size. The above synonymy reflects the taxonomic adjustments necessitated by recent generic revisions of Siphonophvcus (e.g., Pjatiletov, 1988; Knoll et al., 1991). As Leiothrichoides is now a junior synonym of Siphonophvcus. S. tvpicum n. comb, is the first named species defining a size range between that of S. robustum and S. kestron (S. tvpicum holotype ca. 6.5 pm wide). The paratype offered by Timofeev and Hermann (1979) somewhat exceeds the upper size limit characteristic of S. tvpicum (= S. inornatum; Knoll, 1982: text, fig. 3) and is more appropriately assigned to S. kestron. Material—Abundant populations from 7 chert horizons: 86-G-8; 86-G-9; 86-G- 14; 86-G-15; 86-P-89; P-2664; P-3400. Eight populations from shale horizons 86- G-62 and 86-G-30.

SIPHONOPHYCUS KESTRON Schopf, 1968 Figure 21.4

Leiothrichoides tvpicus. TIMOFEEV AND HERMANN, 1979, PI. 29, fig. 1. Siphonophvcus beltensis HORODYSKI. 1980, p. 654-656, PI. 1, fig. 4. Eurvaulidion cvlindratum LO, 1980, p. 144-146, PI. 2, figs. 1-3. Siphonophvcus indicus NAUTIYAL. 1980, p. 3, fig. lA. Tudomophvton unifarium KOLOSOV, 1982, p. 78, PI. 11, fig. 5, PI. 12, fig. 4. Uraphvton distinctum KOLOSOV, 1982, p. 81-82, PI. 14, fig. 2. Uraphvton evolutum KOLOSOV. 1982, p. 82-83, PI. 14, fig. 3, PI. 15, fig. 1. Siphonophvcus laishuiensis ZHANG P. AND YAN, 1984, pp. 198, 203, PI. 1, fig. 3. 125

Eomvcetopsis contorta ZHU, 1984, pp. 173-174, 183, Pl. 3. figs. 1-3, 6 (in Zhu et al., 1984). Eomvcetopsis lata GOLOVENOC AND BELOVA, 1985, p. 99, Pl. 7, fig. 4. Siphonophvcus sinensis ZHANG Z., 1986, pp. 32, 36, Pl. 1, figs. 1, 3, Pl. 2, fig. 4.

Descripfiow.—Unbranched, aseptate, smooth-walled filamentous microfossils, 8- 16 pm diameter. PzscwsszoM.—Siphonophvcus filaments with diameters between 8 and 16 pm are relatively rare in the Svanbergfjellet assemblage. Moreover, those that are recorded, both from chert and shale, are associated with filamentous material that may represent a disaggregated outer sheath comparable to that of Rugosoopsis. Material.—Rare specimens in chert sample P-2628 and shale horizon 86-G-62.

SIPHONOPHYCUS GRACILE (Pjatiletov, 1980) n. comb. Figures 27.4, 25.8-25.9

Leiothrichoides gracilis PJATILETOV. 1980, p. 16-17, PI. 4, figs. 4-5. Uraphvton crassitunicatum KOLOSOV. 1982, p. 83, PI. 15, fig. 3. Uraphvton lenaicum KOLOSOV. 1982, p. 83, PI. 15, fig. 2. Siphonophvcus capitaneum NYBERG AND SCHOPF, 1984, p. 753, fig. HE. Eomvcetopsis grandis PJATILETOV, 1988, p. 68-69, PI. 7, fig. 2.

Diagnosis.—A species of Siphonophvcus 16-32 pm diameter. Description.—Unbranched, aseptate, smooth-walled filamentous microfossils, 16-32 pm diameter (x = 22.9 pm; s.d. = 5.7 pm; n = 21). 126

Discussion.—A number of larger diameter Siphonophvcus species have appeared in the literature but it is often not clear what, if any, biologically meaningful size distributions are being delineated. Leiothrichoides gracilis is described as less than 26 pm diameter (holotype ca. 17 pm), S. capitaneum as 24-33 pm, and Eomvcetopsis grandis as greater than 16 pm wide. In light of the size distribution of the Svanbergfjellet population, the size range 16-32 pm may circumscribe a possible natural grouping; it has the added, if entirely artificial advantage of maintaining the delineation of Siphonophvcus species by the geometric increase of their size (diameter) ranges: S. septatum - 1-2 pm, S. robustum - 2-4 pm, S. tvpicum n. comb. - 4-8 pm, S. kestron 8-16 pm, and S. gracile n. comb. - 16-32 pm. Three additional taxa S. punctatus Maithy, 1975, Taenitrichoides iarvschevicus Assejeva, 1983 and S. latus Yin, 1987 presently identify larger smooth-walled filaments, 40-50 pm, 35-100 pm, and 30-50 pm wide, respectively. Material—21 specimens from shale horizon 86-G-62.

Genus TORTUNEMA Hermann, 1976, emend.

Non Tortunema eniseica HERMANN. 1976, p. 40, PI. 12, fig. 4 (in Timofeev et al., 1976). (= Siphonophvcus).

Type spgcfes.—Tortunema wemadskii (Schepeleva, 1960) n. comb., p. 170. Emended Diagnosis.—VnhTnnched filamentous sheaths with transverse lines or thickenings that reflect the original positioning of intercellular septa (pseudo­ septate). Dzscwss/on.—Pseudo-septate sheaths have commonly been confused with cellular trichomes and in many instances the distinction may not be at all clear. 127

The former are often distinguishable by the presence of cellular residues that have moved freely within the sheath (Fig. 27.1; Hermann, 1986: fig. 7-8); by patterns that grade longitudinally (and even laterally) into unambiguous Siphonophvcus-tvpe sheaths (Fig. 27.2); by a tendency for differential 'septal' prominence or 'missing septa' (Fig. 24.8); and/or by degradational collapse of the sheath rather than a disaggregation into separate cells (Figs. 27.3, 24.8). These criteria, however, may not decide all cases of apparently septate filaments: for example, had the Tortunema specimen in Figure 27.1 not retained its few 'transported' cells, it clearly would have been classified as Osdllatoriopsis. In light of the well-documented preservational bias of cyanobacterial sheaths over actual cell constituents (Golubic and Barghoorn, 1977; Horodyski et al., 1977; Bauld, 1981), pseudoseptate Tortunema is the preferred interpretation for otherwise ambiguous spedmens; Osdllatoriopsis/Cvanonema/Veteronostocale are most usefully reserved for those filaments with positive evidence of true cellular preservation. The earliest named genus to which pseudo-septate fossils might be ascribed is Osdllatoriites Zalessky, 1926; however, the large size (200 pm wide) and ambiguous illustration of the type spedes, O. bertrandi (Jurassic, Volga Basin), militate against its being applied to the much smaller Proterozoic occurrences. Thus, Kolosov (1984) reformulated Osdllatorites wemadskii Schepeleva, 1960 as the type spedes of Botuobia Pjatiletov, 1979, with both he (Kolosov) and Jankauskas (1989) distinguishing Botuobia from Oscillatoriopsis primarily on the basis of size. The two genera are indeed distinct, but not because of generically insignificant differences in dimension: Botuobia wemadskii. indeed all the named spedes of Botuobia. are clearly extracellular sheaths bearing only the imprints of the cell septa they once contained. As Tortunema sibirica Hermann, 1976 is an earlier named taxon of pseudo-septate 128 filaments (similarly identified by its often 'missing septa'), Botuobia becomes its junior synonym and the type species reverts to T. wemadskii n. comb. Under this revision at least four species of Tortunema are recognized: T. angusta (Kolosov, 1984) n. comb (sheaths less than 10 pm; T. wemadskii (Schepeleva, 1960) n. comb, (sheaths 10-25 pm); T. patomica (Kolosov, 1982) n. comb, (sheaths 25-60); T. magna (Tynni and Donner, 1980) n. comb, (sheaths 60-100pm).

TORTUNEMA WERNADSBOI (Schepeleva, 1960) n. comb. Figures 27.1-27.3, 24.8

Osdllatorites wemadskii SCHEPELEVA. 1960, p. 170 Tortunema sibirica HERMANN, 1976, p. 40, PI. 12, figs. 2-3 {in Timofeev et al., 1976). Tortunema sibirica (Hermann. 1976) HERMANN, p. 123 {in Jankauskas, 1989), PI. 29, figs. 2, 4, 6, 10. Botuobia vermiculata PTATILETOV. 1979, p. 13, PI. 4, figs. 6-9; KOLOSOV, 1984, p. 43-44, PI. 7, fig. 2. Osdllatoriopsis bothnica TYNNI AND DONNER, 1980, p. 15, PI. 7, fig. 83. Osdllatoriopsis constrida TYNNI AND DONNER, 1980, p. 15, PI. 7, figs. 82, 85-86. Botuobia wemadskii (Schepeleva, 1960) KOLOSOV, 1984, p. 44-46, PI. 7, fig. 3. Botuobia immutata KOLOSOV. 1984, p. 46, PI. 8, fig. 1. Tortunema cellulaefera PJATILETOV. 1988, p. 79-80, PI. 7. fig. 3-4.

Diagnosis.—A spedes of Tortunema with pseudo-septate sheaths 10-25 pm diameter. 129

Description.—Pseudo-septate filamentous sheaths, 16-27 pm wide (x = 20.1 pm; s.d. = 4.3 pm; n = 8), occasionally enclosing disarrayed condensed cell remnants. PfscMss/ow.—Tortunema was originally erected to describe septate (pseudo­ septate), S-curved filaments that taper towards both ends. As with cellular trichomes, the limited terminal narrowing of these filaments is likely to be of intra-specific and/or taphonomic origin, while the curved habit is not at all constant and thus not taxonomically useful. The distinguishing features of T. wemadskii n. comb, are therefore its pseudo-septate nature and size (width). Interestingly, the 10-25 pm size range offered for T. sibirica Hermann, 1976 is identical to that of Botuobia wemadskii (cf., Jankauskas, 1989: pp. 101). 'Cell' length is not considered a meaningful taxonomic character for pseudoseptate filaments at either the genus or species level, it being recognized as highly irregular within such forms (Pjatiletov, 1979; Kolosov, 1984). M a teria ls specimens from shale horizon 86-G-62.

Family NOSTOCACEAE(?) Kützing, 1843

Genus VETERONOSTOCALE Schopf and Blade, 1971, emend. Type spedes.—Veteronostocale amoenum Schopf and Blade, 1971, p. 950-951. Emended Diagnosis.—Unbranched, uniseriate, cellular trichomes constructed of spheroidal to subspheroidal cells and having no extracellular sheath. Discussion.—Given its Anabaena-like 'string of beads' appearance, Veteronostocale is perhaps the most likely candidate for nostocacean cyanobacteria in the Proterozoic fossil record, although comparable forms are also found in the Osdllatoriaceae (e.g., Pseudanabaena). Veteronostocale is represented by at least two spedes, V. amoenum and a somewhat larger form, V. copiosus Ogurtsova and Sergeev, 1987. 130

VETERONOSTOCALE AMOENUM Schopf and Blacic, 1971 Figure 24.9

Filiconstrictosus diminutus SCHOPF AND BLACIC, 1971 (partim), p. 948-950, Pl. 106, fig. 5.

Descripf ion .--Spheroidal to ellipsoidal cells up to 4 pm wide and 6 pm long, linked into a sinuous trichome ca. 140 pm long. Material—A single specimen from shale horizon 86-C-62.

INCERTAE SEDIS Genus BRACHYPLECANON Lo, 1980 Type spgc/es.—Brachypleeanon khandanum Lo, 1980, p. 156. Discussion.—Small rod-shaped Proterozoic microfossils have been ascribed to Eosvnechococcus Hofmann, 1976, Archaeoellipsoides Horodyski and Donaldson, 1980, Brachypleganon Lo, 1980, and Bactrophycus Zhang Y., 1985. On the basis of an apparently continuous morphological gradient Golovenoc and Belova (1984) subsumed Archaeoellipsoides into Eosvnechococcus and erected a number of new species such that the taxon now includes forms with an aspect ratio as high as 3.6:1 (see Eosvnechococcus. above). In contrast, Brachvpleganon has a length to width ratio of ca. 7:1 (the 4-15:1 aspect of Bactrophycus suggests that it may be a junior synonym of Brachvpleganon). Brachvpleganon thus stands as a legitimate form genus. More interestingly, an appreciation of its preserved behavior supports its distinction as a natural taxon. Unlike the end-to-end linked strings or closely packed aggregations typical of Eosvnechococcus (Hofmann, 1976), Brachvpleganon usually occurs as loosely aggregated and oriented, unattached colonies (Lo, 131

1980); within the form, this habit is at least as telling as measured differences in aspect ratio. Svanbergfjellet Brachvpleganon are preserved in shales rather than the silicified carbonate of the type material, however both the individual rod morphology and the colonial organization are closely comparable.

BRACHVPLEGANON KHANDANUM Lo, 1980 Figures 22.10-22.11 Descrfpffow.—Rod-shaped microfossils 1.5 pm wide by 6-16 pm long (% = 10.1 pm; s.d. = 2.2 pm; n = 150); ca. 7 times longer than wide. Ends rounded but not significantly tapered. Envelope absent. Usually occurring in loosely oriented populations (colonies) of up to ca. 100 individuals. Discussion.—The combination of rod shape and oriented colonies was earlier taken as implying a heterotrophic physiology for these fossils (Butterfield et al., 1988). These are now considered as insufficient to rule out a possible photosynthetic metabolism and B. khandanum is classified as incertae sedis. Material.—18 populations from shale horizon 86-G-62.

Genus CHOLOROGLOEAOPSIS Maithy, 1975

Polvsphaeroides contextus HERMANN, 1976, p. 42-43, PI. 14, figs. 3-4 {in Timofeev et al., 1976).

Type species.—Cholorogloeaopsis zairensis Maithy, 1975, p. 139. Discussion.—h\ addition to Cholorogloeaopsis. spheroidal microfossils arrayed into filament-like colonies are found in Polvsphaeroides Hermann, 1976, Cvanothrixoides Golovenoc and Belova, 1985, and Zinkovioides Hermann, 1985. Cholorogloeaopsis differs from the type species of Polvsphaeroides (P. 132

filliformis) in having no enveloping sheath; P. contextus. however, has no sheath and is therefore subsumed into Cholorogloeaopsis.

CHOLOROGLOEAOPSIS ZAIRENSIS Maithy, 1975 Figure 20.9

Polvsphaeroides biseritus LIU, 1985, p. 65, PI. 7, figs. 14-15 (in Xing et al., 1985).

Description.—Spheroidal to moderately ellipsoidal cells forming an elongate colony ca. 15 pm (2 cells) wide and 365 pm long. Maximum cell dimensions 7- 13 pm (% = 9.6 pm; s.d. = 1.6 pm, n = 30). Cells typically with dark inclusion. Terminal cells unordered. Enveloping sheath absent. Discussion.—Svanbergfjellet Ç zairensis correspond closely to the type material from the Upper Group of the Bushimay System, Zaire, as well as to synonymous P. biseritus. Material.~A single colony from shale horizon 86-G-62.

Genus DIGITUS Pjatiletov, 1980 Type species.—Digitus fulvus Pjatiletov, 1980, p. 68 (in Pjatiletov and Karlova, 1980). Discussion.—Both Digitus and Brevitrichoides Jankauskas, 1980(b) are form taxa describing smooth-walled filamentous microfossils preserving both ends (i.e., entire filaments). Both taxa might be described as 'short filaments' or 'long rods' and both their absolute and relative (aspect ratio) dimensions overlap. If a generic-level distinction were to be made it would be in the overall narrower aspect and the slightly more tapered ends of Digitus. 133

DIGITUS ADUMBRATUS n. sp. Figures 7.S-7.9 Diagnosis.—A species of Digitus ca. 35 pm wide and 250-500 pm long. Surface shagreenate. Descr/ptiow.—Rod-shaped entire filaments terminally tapered for ca. 10% of their length. Surface shagreenate but otherwise unadorned. Width, 34-40 pm (x = 36 pm, S.D = 2 pm, n = 5); length, 293-468 pm {x = 373 pm, s.d. = 30 pm, n = 5). Discussion.—Di^tus adumbratus n. sp. falls entirely outside the size range of previously named species of Digitus (maximum width and length of D. fulvus are, respectively, 25 pm and 125 pm) and therefore warrants separate species status. The shagreenate surface texture of D. adumbratus is also distinctive and likely represents the remnants of a mucilaginous envelope. Etymology.—From the Latin adumbratus - vaguely outlined, overshadowed, with reference to the shagreenate surface texture. Material.—5 specimens from shale horizon 86-G-62. Ho/ofype.-HUPC 62719; Figure 7.9; Slide 86-G-62-133M; England-Finder coordinates: N-29-2.

Genus MYXOCOCCOIDES Schopf, 1968 Type species.—Mvxococcoides minor Schopf, 1968, p. 676.

MYXOCOCCOIDES MINOR Schopf, 1968 Figure 20.3 Descripfiow.—Spheroidal microfossils, 8-18 pm diameter, commonly aggregated into loose colonies of several to ca. 20 cells. Colonies typically embedded in a thin organic matrix. 134

Discussion .--These populations compare closely with the type Mvxococcoides populations except that their size range (within even a single colony) spans that of all three of the originally named Bitter Springs species. In lieu of a major revision of small spheroidal microfossils (which would include at least 28 described species of Mvxococcoides) this population is assigned to M. minor, with the suggestion that the size range of the taxon be altered to include cells 8-18 pm in diameter. Material.—5 colonies and numerous isolated individuals recorded in chert sample P-2664.

MYXOCOCCOIDES CANTABRIGIENSIS Knoll, 1982 Figures 20.6, 20.10 Description.—Relatively thick-walled spheroidal microfossils, 11-20 pm diameter, typically clustered into loose colonies of several to several tens of cells. Envelope absent. Material.—Abundant colonies in chert sample 86-G-14; other scattered occurrences.

MYXOCOCCOIDES spp. Figure 20.7 Discussion.—A variety of smooth-walled spheroidal microfossils occur in silicified carbonates of the Svanbergfjellet Formation. Most are thin-walled, colonial (but without an enclosing matrix), and occur within dense, dominantly filamentous microbial mat. They range in size from 8 to 50 pm diameter (typically 10-30 pm) and are here ascribed to Mvxococcoides spp. Also included are spheroids with a single envelope (Fig. 20.7), but otherwise lacking features diagnostic of Gloeodiniopsis (see above). 135

Genus OSTIANA Hermann, 1976 Type specfes."Ostiana microcystis Hermann, 1976, p. 43 {in Timofeev et al., 1976).

OSTIANA MICROCYSTIS Hermann, 1976 Figures 5.6-S.9 Description.—Closely packed sheets to loose associations of cells, one or (rarely) two layers thick; up to 2100 pm broad. Cells commonly deformed due to mutual compression and often with dark inclusions. Cell diameter 11-27 pm with a prominent mode at 16 pm. Extracellular matrix absent. DzscwssfoM.—Svanbergfjellet O. microcystis isolated by acid maceration occur in three distinct habits that may at some time warrant separate taxonomic designations: 1) tightly packed, polyhedral, and inclusion-bearing cells that compare well with the type material (Figs. 5.7, 5.8); 2) closely packed but undistorted thin-walled cells that construct bi-Iayered sheets (Fig. 5.9); and 3) more loosely packed sheets of undistorted thicker-walled cells (Fig. 5.6). The first form is by far the most common and is also regularly encountered in bedding-parallel thin sections of shale. More loosely bonded sheets of these same cells are also observed in thin section, and a continuum from highly integrated sheets through to localized associations of isolated cells is readily apparent. Thus, while O. microcystis appears superficially to be multicellular— the sheets do act as a unit, wrinkling and folding upon themselves (Figs. 5.7, 5.8) and, in several instances, include seemingly structured apertures (Fig. 5.7)- -it clearly represents a simple pluricellular isensu Awramik and Valentine, 1985) construct of aggregated or unseparated unicells. Monostromatic cellular colonies are characteristic of a number of extant cyanobacterial taxa, including Microcrocis. Holopedia and Merismopedium 136

(Frank and Landman, 1988). As with O. microcystis, the cells of these taxa may be close-packed and polygonally deformed and/or less closely associated such that peripheral cells are often separated from the main colony. They differ markedly from Svanbergfjellet Ostiana in their much smaller cell size (2.5-6.8 pm) and by the presence of a thin mucous layer around the colony (such a layer does occasionally occur in the type material of Ostiana). Ostiana microcystis was almost certainly photosynthetic, but with its relatively large cells it is not immediately obvious whether it was prokaryotic or eukaryotic. Material—27 specimens isolated by acid maceration and 42 in bedding- parallel thin section; an additional 63 loose colonies or associations (of up to several thousand cells) of similar cells are also recorded. From shale horizons 86-G-62 and 86-G-61.

Genus PSEUDODENDRON n. gen. Type species.—Pseudodendron anteridium n. sp. Diagnosis.—Uniseriate or multiseriate, false-branched, sometimes anastomosed filamentous microfossils enclosed in an extracellular sheath. Filaments often tapered, with a sub-micron longitudinal striation or fabric. No preserved cellularity. Discussion.—Branching in filamentous microfossils, particularly if it is both 'false' (i.e., does not involve the trichome) and anastomosing, conceivably derives from a diagenetic fusing of simple filaments. That this is not the case in Pseudodendron n. gen. is shown by the clearly differentiated (biologically mediated) gussets present at most branching points, and the regularly recurring habit of its higher order structure. The combination of a fine longitudinal fabric and a very small lower size range suggests that the larger filaments of Pseudodendron n. gen. may 137 have been multiseriate. Compound filaments have been described from several Proterozoic fossil assemblages (e.g., Hermann, 1974; Horodyski and Donaldson, 1980; Kolosov, 1984), but none shows the true 'false-branching' of the Svanbergfjellet populations. Modern Schizothrix-type cyanobacteria are both multiseriate and false-branching and provide a possible modern analogue for Pseudodendron. Pseudodendron n. gen. is broadly comparable to the diagnosis of the putative Proterozoic metaphyte Ulophvton Timofeev and Hermann, 1979. The type material of the latter, however, is clearly unrelated, and is possibly not even of primary origin. A second species, U. longiscapus Hermann, 1989 {in Jankauskas, 1989) is unquestionably biogenic and may be a species of Pseudodendron. Talakania Kolosov, 1983 is also multiseriate and (rarely) branched; it differs from the present material in its lack of an enveloping sheath and its clearly cellular constitution (Kolosov, 1984: pis. 20-22). The Svanbergfjellet assemblage includes two species of Pseudodendron n. gen., P. anteridium n. sp. and P. polvtaenium n. sp., based primarily on differences in the outer envelope. A number of branched filaments that exhibit an overall similar habit, but with no evidence of a differentiated envelope or longitudinal striation are here classified as Pseudodendron sp. (e.g.. Fig. 23.9). Etymology—From the Greekpseudes - false, and dendron - tree, with reference to the false-branching habit.

PSEUDODENDRON ANTERIDIUM n. sp. Figures 28.1-28.7, 28.10

Branched filamentous structure NYBERG AND SCHOPF, 1984, p. 770, fig. IIF. 138

Diagnosis.—A species of Pseudodendron with discrete, clearly defined axes and false branches, the latter typically reinforced on the inside angle of junctions by a prominent concave gusset. Description.—Fûamenis 3 to 60 pm wide and up to 2.7 mm long, often tapering, with a single or several smaller filament(s) contained within an outer sheath; a longitudinal ca. 1 pm striation sometimes present. False branches well defined, commonly diverging at angles of less than 45° from the main axis and with the inside angle supported by a prominent concave gusset contiguous with the primary sheath. At least two orders of branching present. Separate axes sometimes grown back together. Discussion .--Pseudodendron anteridium n. sp. is a conspicuous component of the Lower Dolomite Member shale assemblage (P-2945), with specimens recovered as single, relatively nondescript filaments; sparsely to regularly branched forms; and filaments terminating (or originating?) in a bulbous expansion (Fig. 28.3). In the Upper Algal Dolomite shale it is less common but in one instance clearly exhibits its capacity to rejoin separated axes (Fig. 28.4). That the branching of P. anteridium n. sp. is 'false' is indicated by the confluence of lateral axes into a primary axis without their actual incorporation into its central 'trichome'; rather, they maintain their individuality and continue on in parallel formation within the outer sheath of the main filament (Fig. 28.7). Like the outer envelope, the filament cores appear to be composed of extracellular sheath material as indicated in their occasional confluence with the primary envelope (Fig. 28.5); actual trichomes seem not to be preserved. Svanbergfjellet P. anteridium n. sp. compare closely with silicified fossils described from the 680-790 Ma old Min'yar Formation of the southern Urals (Nyberg and Schopf, 1984). These similarly multiseriate, ensheathed, and false- branched filaments contain up to 36 narrow (2.0-5.0 pm) tubular sheaths. 139

Etymology.—From the Greek, anteridion - support, buttress, with reference to the conspicuous branch reinforcements. Material.—S>7 examples, 9 in bedding-parallel thin section, 78 isolated by acid maceration. From two shale horizons: P-2945, 86-G-62. Holotype.-HUFC 62720; Figure 28.10; Slide P-2945-7M; England-Finder coordinates: L-41-3.

PSEUDODENDRON POLYTAENIUM n. sp. Figures 21.1, 28.8-28.9 Diagnosis.—A species of Pseudodendron with anastomosed filaments and a limited external envelope. Confluent filaments result from both true 'false- branching' and apparent longitudinal fragmentation. Descrzpfiow.—False-branched and anastomosed filaments, 1-40 pm wide (typically ca. 20 pm), often occurring in matted associations up to 2.2 mm in maximum dimension. Filaments with a distinct longitudinal fabric. Enveloping extracellular sheath present but often inconspicuous.

D/ scmssiow .—The outer envelope of P. polvtaenium n. sp. is considerably reduced from that of the type species and lacks its prominently reinforced junctions (Fig. 28.8). Where the envelope is absent the separation (and rejoining) of filaments appears to be random (Fig. 21.1) and may represent the simple disaggregation of a wider 'thallus' along its longitudinal fabric. Thus much, but certainly not all, of the reticulation observed in P. polvtaenium may be an artifact of taphonomy. Etymology.—Erom the Greek, poly - many, and tainia - ribbon. Material.—20 specimens (some as extensive mats) from shale horizon 86-G-62. Holotype.—HUPC 62721; Figure 28.8; Slide 86-G-62-202M; England-Finder coordinates: 0-42-3. 140

Filament-Bearing Body Figure 21.2 Description.—Dark central body, 105 x 35 pm, bearing several radially oriented, fibrous filaments; filaments up to 400 pm long and distally expanded (from 13 to 30 pm wide). Dfscwssiow.—This unique specimen appears to differentiate two distinct cell or tissue types, thereby suggesting a relatively complex grade of organization. Conversely, it might be interpreted as two quite unrelated structures that were fused through sedimentary compaction. Both its regular radiating pattern and the uniqueness of the two components argue against this latter alternative. Material—A single specimen from shale horizon 86-G-62.

Sub-Vertical Branched Tubes Figures 21.5, 21.8-21.10 Description.—Gregarious, more or less vertically oriented, and commonly branched tubes, 14-40 pm diameter (% = 22 pm; s.d. = 6 pm; n = 25); moderately to highly sinuous, often with localized constrictions and/or expansions. Walls dark and particulate, 1-2 pm thick. Discussion.—The gregarious behavior of these oriented tubes imposes a conspicuous sub-vertical fabric along a single ca. 2 mm laminae of a silicified, flat-laminated carbonate (P-3075). Immediately associated spheroidal microfossils are typical of most silica permineralized fossils; thus, the dark particulate tube walls are not simply a product of differential taphonomy— indeed, it is not entirely clear whether they are carbonaceous. In the absence of preserved cellular structure the derivation of these branched tubes remains speculative. The default, but not necessarily correct interpretation is that they represent false branching prokaryotes, comparable to 141 modem scytonematacean cyanobacteria. Equally plausible alternatives include mineral overgrowths of simple filamentous precursors (see below), branching eukaryotic algae, or even metazoan traces. Vertically oriented, but less sinuous and branched filaments are preserved as 13-60 pm wide casts in the immediately overlying Draken Formation (Knoll et al., 1991). Material—Abundant in silicified carbonate sample P-3075.

Sub-Vertical Branched(?) Filaments in Apatite Figures 23.6, 23.7 Description.—Gregarious, often branched(?) sub-vertical filaments preserved in apatite nodules; 8-35 pm diameter. D/scwssion.—Abundant, poorly preserved filamentous structures are concentrated along discrete laminae, or may be more generally dispersed, in shale-hosted apatite nodules of the Lower Dolomite Member. They are clearly biogenic, but the apparent branching may well be a product of diagenesis. Material—Common in nodular apatite samples SV-2 and SV-3.

Mineralized Filaments - Goethite(?) Figures 26.3-26.6 Description.-Solitary to thickly amassed sinuous 'filaments' preserved three- dimensionally as mineral overgrowths (goethite?) on Siphonophvcus-1 ike sheaths, or entirely mineralic. Diagenetic fusion of intersecting filaments common. Diameter variable but less than 25 pm. Dfscwssion .—Mineralized filaments are found on exposed bedding planes (Fig. 26.3) and in bedding-parallel thin sections (Figs. 26.4-26.6) of a single Algal Dolomite Member shale (86-G-30). That they originated as biological structures is evident from both their pronounced matting habit (Fig. 26.6) and 142 the occasional preservation of internal organic-walled sheaths (Fig. 26.4). It is also clear, however, that much of the morphology of these fossils is secondary; it is the mineral overgrowth that accounts for most of their girth, and that has fused separate filaments to give them the appearance of branching (Fig. 26.5- 26.6). The yellow mineral defining these fossils is translucent, anisotropic, and iron-rich (EDAX analysis), possibly goethite. In addition to its precipitation on and within filamentous organisms, it occurs as small veins and localized crystals throughout the sample, a habit identifying the filament rinds as a product of early diagenesis rather than primary biomineralization. Possible eukaryotic algal filaments reported from the Late Proterozoic Chuar Group by Horodyski and Bloeser (1983) exhibit a number of features reminiscent of the Svanbergfjellet fossils. Although somewhat larger (40-130 pm wide), these filaments likewise occur three-dimensionally on shale bedding planes, exhibit a sinuous and pseudo-branching habit, and sometimes preserve an internal cylindrical structure (original sheath?). It is thus possible that they too represent mineralic overgrowths of considerably smaller Siphonophvcus- like filaments and as such are less convincingly interpreted as eukaryotic. Maferifl/.—Abundant examples in shale horizon 86-G-30. 143

Figure 1. Map of Spitsbergen and neighboring islands, showing the principal outcrop localities of Svanbergfjellet Formation: G = Geerabukta; P =

Polarisbreen; SV = Svanbergfjellet; K = Kluftdalen. Densely stippled areas in the enlargements indicate Akademikerbreen Group outcrop. The stratigraphie column represents the late Proterozoic (Lomfjorden Supergroup) and early

Paleozoic sedimentary successions in northeastern Spitsbergen; the

Svanbergfjellet Formation is marked with a star. c c 8000 o S 7000

6000

5000 # o SV

4000

3000

2000

100 km 1000

2 km

OGCHAUUKTA

FAKSEFJELLET

OnACOtSEM

FAKSEVAGEN 145

Figure 2. Stratigraphie column of the Svanbergfjellet Formation as measured at Geerabukta, showing the positions of significantly fossiliferous horizons.

The level of samples from other localities (e.g., "P", "SP", "SV", "B") are inferred.

Sample numbers in parentheses represent fossiliferous units not figured in the present work. C T3*TJ SVANBERGFJELLET FORMATION S

3 O co o o g o i (/) o o o z _L_ m LOWER DOLOMITE MEMBER LOWER LIMESTONE MEMBER ALGAL DOLOMITE Mbr s> S) N

“TT" T tT" M "W 00 09 77 O0T3O0 00 00 00 0000 0 0 9 roio 7coco 7 O) o> o 0) 0) § 0) 0) p o ? Q roo) 9 S 9 I o o o m 9 0 0 00 a s CO CO IS s I o CO S 00O) 9 42k.CO ro o ' o ro cn CO o 147

Figure 3. Scale bar in 2 equals 6.2 cm for 2; 15 mm for 6; 2.5 mm for 3-4; 120 pm for 5, 7.

1. Outcrop of the Algal Dolomite Member. The massive units are stromatolitic dolomite separated by, and intercalated with fine grained (and fossiliferous) siliciclastic rocks.

2. Miniaria stromatolites from the top of the Lower Dolomite Member.

3. 4. Silicified intraclastic grainstone from the Lower Dolomite Member; 86- G-8-2a. The large clast in Fig. 3.3 and a majority of the smaller clasts in both Figs. 3.3 and 3.4 are of massive, uncompressed, and sediment-free microbial mat. The large clast in Fig. 3.4 is of the clastic-rich laminated mat type; note the incorporation of various intraclasts, including shards of the sediment-free mat.

5. Perpendicular to bedding thin section of fossiliferous shale sample 86-G- 62 (Algal Dolomite Member); note the very fine grain size and undisrupted lamination.

6. Silicified columnar stromatolite and intraclastic grainstone, from the Lower Dolomite Member; 86-G-15-2A.

7. Polvbessurus bipartitus preserved in large diagenetic dolomite crystals with pronounced cleavage; Lower Limestone Member (86-P-89; England Finder co-ordinates: S-64-0). %

a m

IÂ

<««• 149

Figure 4. Size frequency distribution of leiosphaerid acritarchs less than 100 pm diameter, from the two principal fossiliferous shales in the Svanbergfjellet

Formation: 86-G-62 (Algal Dolomite Member) and P-2945 (Lower Dolomite

Member). Counts were taken from bedding parallel thin sections (9 of 86-G-62 and 3 of P-2945) and included all apparently planktic (i.e., non-clustered) specimens. Note the differing vertical scales of the two histograms. 300-

2 0 0 - 86-G-62

n =1084

10 0 -

30-, P-2945 n= 217

2 0 -

10-

0 5 10 19 20 20 30 38 40 40 80 85 OO 00 70 70 00 00 OO OO 100 /iiti 151

Figure 5. Scale bar in 1 equals 20 |im for 1-4; 50 pm for 5-6, 9; 125 pm for 7-8.

Figure captions include HUPC number, slide identification, and "England

Finder" location coordinates (in parentheses). All fossil specimens from shale.

Palaeastrum dvptocranum n. gen., n. sp. 1. HUPC 62708; 86-C-62-46 (M-48-1); Holotype. 2. HUPC 62722; 86-G-62-10 (L-38-3); differentiated attachment discs (plaques); the thinner cell walls have been largely degraded. 3. HUPC 62723; 86-G-61-1 (R-53-2); portion of a ca. 1 mm wide colony.

Coelastrum proboscideum (Chlorophyta) 4. SEM (courtesy of H. J. Marchant).

Eosaccharomyces(?) sp. 5. HUPC 62724; 86-G-62-14 (Q-28-3); note the pronounced, but undifferentiated inter-cellular attachments.

Ostiana microcystis 6. HUPC 62725; 86-G-62-190M (0-14-1); thick-walled (dark), loosely aggregated morph. 7. HUPC 62726; 86-G-62-193M (0-28-0); monostromate morph; note wrinkles and apparently structured aperture. 8. HUPC 62727; 86-G-62-190M (Q-42-0); monostromate morph; folded over on itself. 9. HUPC 62728; 86-G-62-40M 0-36-4); bi-stromate morph.

153

Figure 6. Scale bar in 5 equals 45 for 1-3; 60 for 9; 100 for 4-8,10. Figure captions include HUPC number, slide identification, and "England Finder" locaHon coordinates (in parentheses). All specimens from shale.

Proterocladus major n. gen., n. sp. 1. HUPC 62732; 86-G-62-22M (Q-33-1); note central 'pore' in the septum adjacent to the branch. 2. HUPC 62733; 86-G-62-18M (G-37-0); cell with two branches, one of which is septate and entire. 3. HUPC 62709; 86-G-62-53 (L-11-2); Holotype; in bedding-parallel thin section. 4. HUPC 62734; 86-G-62-17M (5-36-0); one entire cell. 5. HUPC 62735; 86-G-62-90M (G-31-0); three entire cells. 6. HUPC 62736; 86-G-62-76M (0-25-0); three entire cells - two with branches. 7. HUPC 62737; 86-G-62-10 (P-25-0); one entire cell - in bedding-parallel thin section. 8. HUPC 62738; 86-G-62-19M (U-36-3); terminal cell with rounded end. 9. HUPC 62739; 86-G-62-87M (L-27-0); collapsed filament with septum preserved parallel to bedding. 10. HUPC 62740; 86-G-62-79M (0-37-1); single cell with plexus of 5 branches. ■ •*

r SLJ" Vr: 8 r t ■> 4 .

VC 155

Figure 7. Scale bar in 5 equals 50 pm for 1-3, 5; 185 pm for 4; 20 pm for 6-7;

75 pm for 8-9. Figure captions include HUPC number, slide identification, and

"England Finder" location coordinates (in parentheses). All fossil specimens from shale.

Proterocladus minor n. gen., n. sp. 1. HUPC 62710; 86-G-62-15M (N-24-0); Holotype. Note the isolation of branches by immediately suprajacent and subjacent septa in the main axis; also, the cytoplasmic residues and rounded terminal cell of the lower branch. 2. HUPC 62729; 86-G-62-14M (L-40-2); the lateral branch in this specimen lacks a subjacent septum in the in the main axis.

Proterocladus major n. gen., n. sp. 3. HUPC 62730; 86-G-62-16M (L-41-4); with lateral branch.

Cladophora sp. 4. With branches and robust septa.

Seereeocladus hermannae n. gen., n. sp. 5-7. HUPC 62711; 86-G-62-93M (P-44-0); Holotype. 5, note the rod-shaped cytoplasmic residues, and rare septa not associated with branches. 6, detail of a conical structure similar to zoospore release structures in modem Cladophoropsis. 7, detail of a branch showing the single (and relatively insubstantial) adjacent septum (compare with that of Proterocladus - Fig. 7.3).

Digitus adumbratus n. sp. 8. HUPC 62731; 86-G-62-119M (G-36-1). 9. HUPC 62719; 86-G-62-133M (N-29-2); Holotype. ■Sr

\

. ,. x-^>k 8

K.

' ■ " 4iR' 157

Figure 8. Scale bar in 1 equals 1.5 mm. Figure captions include HUPC number and sample identification. All specimens from shales.

Fseudotawuia birenifera n. gen., n. sp. 1. HUPC 62741; 86-G-30-3BP; Holotype; note the insignificant sedimentary imprint of this specimen relative to that of Tawuia: also, the large size and apparent bilateral symmetry of the terminal structures.

Tawuia dalensis 2. HUPC 62742; 86-G-62-1BP; with associated Chuaria sp. and Cerebrosphaera buickii n. sp. 3. HUPC 62743; 86-G-62-3S; now isolated and mounted for SEM. 4. HUPC 62744; 86-G-62-108M (K-32-0); macerated specimen mounted for light microscopy. Note the translucent layer remaining where the opaque wall has broken away. 5. HUPC 62745; 86-G-61-1BP; 'Pumilabaxa' morph.

Large spheroidal compression 6. HUPC 62746; 86-G-61-3BP; note the three radially oriented compression(?) splits. A faint filamentous texture suggests a possible comparison to modem Nostoc balls.

Chuaria circularis 7. HUPC 62747; 86-G-61-2BP; note substantial sedimentary imprint, but the absence of most of the original wall. 8. HUPC 62748; 86-G-28-2BP; with associated 'organic stain'. W -"".".'-\.V'^ 159

Figure 9. Scale bar in 2 equals 150 pm. Figure captions include HUPC number, slide identification, and "England Finder" location coordinates (in parentheses). All specimens from shale.

Valkvria borealis n. gen., n. sp. 1. HUPC 62712; 86-G-62-5M (L-29-1); Holotype; with medial stripe and dark terminal structures. 2. HUPC 62749; 86-G-62-7M (M-29-4); with vague medial stripe. 3. HUPC 62750; 86-G-62-6M (R-29-4); with terminal structures and well preserved, lobate lateral axis. 4. HUPC 62751; 86-G-62-2M (P-36-0); with terminal structures. 5. HUPC 62752; 86-G-62-58M (L-32-0); with terminal structures. f t i 161

Figure 10. Scale bar in 7 equals 50 for 1, 4, 6; 80 pm for 2, 5, 7-8; 125 pm for 3. Figure captions include HUPC number, slide identification, and

"England Finder" location coordinates (in parentheses). All specimens from shale.

Valkvria borealis n. gen., n. sp. 1. HUPC 62753; 86-C-62-236M (M-41-0); note the large central vesicle with pointed terminal structure and a possible orifice in the adjacent outer wall; lateral axes branched. 2. HUPC 62754; 86-G-62-73M (N-21-3); 'partitioned' terminal segment with inclusions; compare with Fig. 11. 3. HUPC 62755; 86-G-62-64M (P-41-0); with medial stripe and terminal discoidal structure. 4. HUPC 62756; 86-G-62-41M (Q-19-1); branched, lateral axes; note the prominent (and diagnostic) attachment scars. 5. HUPC 62757; 86-G-62-10M (N-29-3); fragment with large central vesicle; one end pointed, the other with a curled structure. 6. HUPC 62758; 86-G-62-4M (N-31-4); branched lateral axis, possibly with second order branching; apparently septate. 7. HUPC 62759; 86-G-62-58M (K-40-4); branched and 'entire' lateral axis; non-septate. 8. HUPC 62760; 86-G-62-73M (S-29-2); specimen in which the lumen of the main body has been occupied by filamentous micro-organisms; they also appear to extend up one of the lateral axes (upper right). /

m )

> 1

' / & # j y 163

Figure 11. ColoniaK?) Valkyria borealis n.gen, n. sp. (HUPC 62907); drawn from shale thin-section 86-G-62-35 (England Finder coordinates: H-52-0). Note the partitioned distal segments of the two principal specimens are the same as the isolated structure in Fig. 10.2. I 1 lut / 0 0 1

UJUJ s o 165

Figure 12. Scale bar in 3 equals 200 pm for 1, 4-6; 80 pm for 2-3; 15 pm for 7;

7.5 pm for 8. Figure captions include HUPC number, sample identification, and "England Finder" location coordinates (in parentheses). All specimens from shales.

Cerebrosphaera buickii n. gen., n. sp. 1. HUPC 62761; 86-G-28-1BP; bedding plane specimen with unusual sub­ parallel / sub-radial wrinkling pattern. 2. HUPC 62762; P-2945-61M (N-30-2); transmitted light micrograph of isolated wall fragment showing typical wrinkling pattern. 3. HUPC 62763; 86-G-33-2S; SEM photograph of whole specimen; note the compression induced radial splits and the absence of secondary folds. 4. 5. HUPC 62713; P-2945-47M (S-33-4); Holotype; with outer sheath and profound radial splitting. 4, (transmitted light), emphasizing the thin- walled envelope and nearly opaque main vesicle. 5, (reflected light), showing the characteristic surface convolutions. 6. HUPC 62764; 86-G-62-232M (R-15-4); note the opacity of the fossil where the two sides of the spheroid are superimposed. 7. HUPC 62765; 86-G-33-1S; SEM detail of the interfingering and anastomosing surface convolutions. 8. HUPC 62766; 86-G-33-1S; SEM detail of single-wall cross-section; note the non-psilate inner surface. The rounded perforations are not characteristic of the taxon and appear to be the result of heterotrophic activity, possibly predatory. n 167

Figure 13. Scale bar in 5 equals 200 pm for 1-3; 300 pm for 4-8; 3 pm for 9.

Figure captions include HUPC number, slide identification, and "England

Finder" location coordinates (in parentheses). All specimens from shales.

Trachvhvstrichosphaera aimika 1. HUPC 62767; 86-G-62-230M (0-24-2); although no unambiguous processes are preserved on this specimen, the size and 'unusual' arrangement of perforations are particularly characteristic of T. aimika. The processes were apparently removed by erosion.

Leiosphaeridia spp. 2. HUPC 62768; 86-G-62-172M (K-35-3); a large dark-walled spheroidal acritarch with distinctively perforated walls. 3. HUPC 62769; 86-G-62-71M (M-24-4); a large shagreenate spheroidal acritarch with a dark elongate inclusion(?).

Leiosphaeridia wimanii n. comb. 4. HUPC 62770; 86-G-62-238M (K-20-2); note the smooth-edged split that appears to have been in place before flattening. 5. FIUPC 62771; 86-G-62-182M (S-27-0); semi-translucent specimen with biologically mediated medial split. 6. HUPC 62772; 86-G-61-3BP; bedding-plane specimen with biologically mediated medial split.

Chuaria circularis 7-9. HUPC 62773; 86-G-62-1S; 7, SEM photograph; 8, reflected light photograph of same specimen; 9, SEM of cross sectional fracture surface.

169

Figure 14. Scale bar in 9 equals 20 pm for 1-9; 50 pm for 10. Figure captions include HUPC number, slide identification, and "England Finder" location coordinates (in parentheses). All specimens in shale.

Dictvotidium colandrum n. sp. 1. HUPC 62714; 86-G-62-45 (N-31-4); Holotype; note the faint membranes between the short, protruding spines. 2. HUPC 62774; 86-G-62-51 (K-47-3). 3. HUPC 62775; 86-G-62-12 (T-27-3); isolated fragment. 4. HUPC 62776; 86-G-62-12 (N-32-1); broken specimen illustrating the characteristic, pre-burial brittleness.

Comasphaeridium sp. 5. HUPC 62777; 86-G-62-34 (Q-51-2).

Goniosphaeridium sp. 6. HUPC 62778; 86-G-62-16 (S-16-4); with a possible pylome. 7. HUPC 62779; 86-G-62-40 (R-51-4); note the apparent branching of one process.

Pterospermopsimorpha pileiformis 8. HUPC 62780; 86-G-62-11 (V-36-1).

Goreonisphaeridium sp. 9,10. HUPC 62781; 86-G-63-3 (H-37-2); fragment.

171

Figure 15. Scale bar in 2 equals 25 for 1, 3; 75 |im for 2; 60 yim for 4-5; 50 pm for 6-11. Figure captions include HUPC number, slide identification, and

"England Finder" location coordinates (in parentheses). All specimens in chert.

Cymatiosphaeroides kullingii 1. HUPC 62782; 86-G-14-1A (J-43-1); with ca. 12 extracellular layers forming a massive, space-filling mucilage; in the matrix of an intraclastic grainstone. 2. HUPC 62783; 86-G-14-1A (F-72-4); with ca. 12 relatively compressed extracellular layers. Note the attachment to a grainstone intraclast. 3. HUPC 62784; 86-G-15-1 (L-64-3); within a dense microbialite intraclast. 4. HUPC 62785; 86-G-8-2B (U-64-0); in a laminated intraclast. 5. HUPC 62786; 86-G-8-2B (U-61-4); in a laminated intraclast.

Osculosphaera hvalina n. gen., n. sp. 6. HUPC 62716; P-3085-1A (P-43-2); Holotype. 7. HUPC 62787; 86-P-82-1B (X-47-2). 8. HUPC 62788; P-3085-1E (U-48-4). 9. HUPC 62789; P-3085-1C (G-63-3); with the 'oral' collar curled inward. 10. HUPC 62790; P-3085-1D (P-65-4). 11. HUPC 62791; P-3085-1E (E-62-2); co-occurring spheroid of the same size and structure as O. hvalina.

173

Figure 16. Scale bar in 2 equals 30 pm for 1-3; 20 pm for 4-6; 140 pm for 7; 40

pm for 8-9. Figure captions include HUPC number, slide identification, and

"England Finder" location coordinates (in parentheses). Figures 1-7, 8-9 from

shale.

Germinosphaera iankauskasii n. sp. 1. HUPC 62715; 86-G-62-12M (L-31-3); Holotype; note the several processes that originate from within the perimeter of the vesicle. 2. HUPC 62792; 86-G-62-30M (J-33-1). 3. HUPC 62793; 86-G-62-155M (N-35-1).

Germinosphaera bispinosa 4. HUPC 62794; 86-G-62-28 (0-19-2). 5. HUPC 62795; 86-G-62-14 (J-41-2).

Leiosphaeridia crassa 6. HUPC 62796; 86-G-62-6 (E-25-4); with medial splits.

Vaucheria sessilis 7. Germinating zoospore of the modem chromophyte alga showing multiple and occasionally branched filament primordia.

Leiosphaeridia jacutica 8. HUPC 62797; 86-G-62-10 (S-33-3).

Leiosphaeridia tenuissima 9. HUPC 62798; 86-G-62-71M (M-16-3). Û'* 7.'

5 I ■ ^ •& I

■ 0 f ' • A » . : 175

Figure 17. Scale bar in 5 equals 50 |rm. Figure captions include HUPC number, slide identification, and "England Finder" location coordinates (in parentheses). All specimens from shale.

Germinosphaera fibrilla n. comb. 1. HUPC 62799; 86-G-62-97M (L-38-3). 2. HUPC 62800; 86-G-62-24M (H-31-2). 3. HUPC 62801; 86-G-62-27M (0-38-4). 4. HUPC 62802; 86-G-62-23M Q-30-3). 5. HUPC 62803; 86-G-62-25M (N-28-3). 6. HUPC 62804; 86-G-62-100M (0-29-3); with branching process. 7. HUPC 62805; 86-G-62-28M (0-30-0). 8. HUPC 62806; 86-G-62-33M (K-36-3); Holotype. A. 177

Figure 18. Scale bar in 11 equals 75 pm for 1-6, 8-10; 150 pm for 7; 25 pm for

11. Figure captions include HUPC number, slide identification, and "England

Finder" location coordinates (in parentheses). Figures 1-7 from shales; 8-11 in chert.

Trachvhvstrichosphaera aimika 1. HUPC 62807; 86-G-62-163M (B-31-2); note the small conical process at the arrow. 2. HUPC 62808; 86-G-62-159M (0-15-2). 3. HUPC 62809; 86-G-62-32M (F-45-4). 4. HUPC 62810; P-2945-1M (K-30-3); with partial envelope. 5. HUPC 62811; 86-G-62-31M (G-33-0); note the various process morphs. 6. HUPC 62812; P-2945-2M (L-33-1); with remnants of outer sheath material. 7. HUPC 62813; 86-G-62-158M (L-26-3); large specimen with medial constriction; processes inconspicuous. 8. HUPC 62814; 86-G-15-2A (S-35-1); with bifurcated processes and partial sheath; in microbialite grainstone. 9. HUPC 62815; 86-G-15-2A (0-38-3); specimen attached to a tabular intraclast in microbialite grainstone. 10,11. HUPC 62816; 86-G-15-2A (Q-35-2); note the processes extending through the outer sheath. Detail in Fig. 18.11 shows an incipient process and three discrete wall layers: a thin, inner wall contiguous with the processes; a thick and intermediate layer (= sheath?); and an outer diffluent layer that appears to have served in attaching the vesicle to its substrate (far right). g p B bw 179

Figure 19. Scale bar in 5 equals 50 for 1, 2, 4; 80 |im for 3, 5-8. Figure captions include HUPC number, slide identification, and "England Finder" location coordinates (in parentheses). All specimens from shale.

Trachyhvstrichosphaera polaris n. sp. 1. HUPC 627817; 86-G-62-11M (Q-28-0); with hair-like surface ornamentation. 2. FIUPC 62818; 86-G-62-168M (0-32-0). 3. HUPC 62819; 86-G-62-205M (N-33-2); note the various process morphs. 4. HUPC 62820; 86-G-62-40 (0-24-3); in bedding-parallel thin section; note the remnants of envelope at process termini. 5. HUPC 62717; 86-G-62-1M (U-30-3); Holotype. 6. HUPC 62821; 86-G-62-206M (P-34-2). 8. HUPC 62823; 86-G-62-101M (H-37-0); with envelope.

Trachvhvstrichosphaera aimika 7. HUPC 62822; 86-G-62-102M (K-33-0); note the absence of short, solid spines on the inner vesicle.

181

Figure 20. Scale bar in 20 equals 50 |iM for 1, 4, 6, 20; 20 pm for 2, 3, 5, 7-19.

Figure captions include HUPC number, slide identification, and "England

Finder" location coordinates (in parentheses). Figure 9 in shale, others in chert.

Eoentophysalis croxfordii n. comb. 1. HUPC 62824; 86-G-15-1 (E-59-2).

Eoentophysalis sp. 2. misplaced.

Myxococcoides minor 3. HUPC 62825; P-2664-4A (J-55-4).

Eoentophysalis belcherensis 4. HUPC 62826; 86-G-8-2B (E-61-3). 5. HUPC 62864; 86-G-4-1A (A-54-3).

Myxococcoides cantabrigiensis 6. HUPC 62827; 86-G-14-1A (P-49-4). 10. HUPC 62831; P-2664-4A (D-58-3).

Myxococcoides sp. 7. HUPC 62828; 86-G-14-1A (T-72-1).

Gloeodiniopsis lamellosa 8. HUPC 62829; 86-G-9-2C (C-56-1).

Cholorogloeaopsis zairensis 9. HUPC 62830; 86-G-62-44 (N-50-3).

Sphaerophycus sp. 11. HUPC 62832; P-3085-1B (L-52-0).

Sphaerophycus parvum 12-20. In chert thin-section 86-SP-8-1. 12, HUPC 62833 (R-60-4); 23, HUPC 62834 (M-60-2); 14, HUPC 62835 (G-43-2); 15, HUPC 62836 (K-55-4); 16, HUPC 62837 (P-62-4); 17, HUPC 62838 (K-55-4); 18, HUPC 62839 (K-54-4); 19, HUPC 62840 (H-54-0) - note sheath; 20, Population of S. paryum colonizing the surface of a microbialite intraclast (P-62-4). 8 #

10 • I. • » #

12 13 14 15

% ê I

16 17 ' 18 î 19 11

2 0 183

Figure 21. Scale bar in 2 equals 75 for 1, 7, 9-10; 120 pun for 2-3; 50 pim for

4-6, 8. Figure captions include HUPC number, slide identification, and

"England Finder" location coordinates (in parentheses). Figures 1 and 2 from shale; 3-10 in chert.

Pseudodendron polvtaenium n. gen., n. sp. 1. HUPC 62721; 86-G-62-202M (0-42-3); Holotype.

Filament-bearing body 2. HUPC 62841; 86-G-62-70M (L-27-0).

Polvbessurus bipartitus 3. HUPC 62842; 86-P-89-1 (H-43-4). 6. HUPC 62845; 86-P-89-1C (U-37-4). 7. HUPC 62846; 86-P-89-1C (T-68-3); note the cluster of probable baeocytes within the stalk on the right.

Siphonophvcus kestron 4. HUPC 62843; P-2628-1A (V-48-1); with associated filamentous material similar to disaggregating Rugosoopsis sheath.

Sub-vertical branched tubes 5. HUPC 62844; P-3075-1F (S-47-3). 8. HUPC 62847; P-3075-1A (Z-55-3). 9. HUPC 62848; P-3075-1C (N-60-4); showing characteristic sinuosity. 10. HUPC 62849; P-3075-1A (Z-55-4). I

n 185

Figure 22. Scale bar in 1 equals 20 pm for 1-10, 50 pm for 11. Figure captions include HUPC number, slide identification, and "England Finder" location coordinates (in parentheses). All specimens in shale.

Obruchevella blandita 1. HUPC 62850; 86-G-62-1 (P-48-2); all coils in close contact; note the tubular cross-section on the exposed terminal coil. 2. FIUPC 62851; 86-G-30-4 (N-27-4); disaggregating specimen. 3. HUPC 62852; 86-G-28-2 (N-27-3); oblique view of fragment. 4. HUPC 62853; 86-G-62-16 (L-18-1); cross-sectional view. 5. HUPC 62854; 86-G-62-11 (N-22-4); single isolated coil. 6. HUPC 62855; 86-G-61-4 (S-44-1); with annular thickening.

Siphonophvcus septa turn 7. HUPC 62856; 86-G-62-24 (K-37-3); filaments entwined as a triple helix. 8. HUPC 62857; 86-G-62-24 (P-47-1).

Siphonophvcus thulenemum n. sp. 9. HUPC 62718; 86-G-62-46 (R-28-3); Holotype.

Brachvpleganon khandanum 10. HUPC 62858; 86-G-62-47 (N-50-4). 11. HUPC 62859; 86-G-62-10 (R-15-2); aggregated and loosely oriented colony. o

m. 187

Figure 23. Scale bar in 6 equals 20 pm for 1-5,10; 375 pm for 6; 185 pm for 7;

75 pm for 9; 50 pm for 11; 3.5 pm for 8. Figure captions include HUPC number, sample identification, and "England Finder" location coordinates (in parentheses). Figures 1, 5, 8-9 from shale; 2-4,10 in chert; 6-7 in apatite; 11 in dolomitic microspar.

Cephalonyx geminatus n. comb. 1. HUPC 62860; 86-G-62-17 (M-45-4).

Siphonophvcus-like filaments with localized encrustations 2, 3. HUPC 62861; 86-G-15-2A (Z-30i). 4. HUPC 62862; 86-G-15-3 (0-49-1).

Oscülatoriopsis amadea n. comb. 5. HUPC 62863; 86-G-62-65 (N-30-3).

Sub-vertical branched(?) filaments in shale-hosted apatite nodules. 6,7. HUPC 61374; 86-SV-3-1A. 6, (0-41-2); 7, (R-42-1).

Tawuia dalensis 8. HUPC 62865; 86-G-62-2S; SEM of a broken edge of Tawuia (believed to be the counterpart of the specimen in Fig. 8.3), showing the two discrete wall components. Where fiie thin layer is opaque and brittle, the imderlying(?) thicker stratum is remarkably translucent and flexible (determined prior to mounting for SEM).

Pseudodendron(?) sp. 9. HUPC 62866; 86-G-62-94M (M-34-0).

Eosvnechococcus moorei 10. HUPC 62867; P-2664-2A (P-67-0).

Leiosphaeridia crassa 11. HUPC 62868; 86-G-9-3 (S-56-2); in a dolomitized intraclast. FI

li

S

^ 10 * 189

Figure 24. Scale bar in 11 equals 25 pm. Figure captions include HUPC

number, slide identification, and "England Finder" location coordinates (in

parentheses). All specimens from shales.

Oscillatoriopsis obtusa 1. HUPC 62869; 86-G-62-11 (N-27-2); note that filament separation occurred between, rather than across cells. 2. 86-G-62 (destroyed); both ends rounded. 3. HUPC 62870; 86-G-62-1 (R-38-2); one end rounded, the other narrowed to a point. 4. HUPC 62871; 86-G-62-40 (L-46-4); both ends rounded. 5. HUPC 62872; 86-G-62-65 (R-30-2); one end markedly narrowed. 11. HUPC 62878; 86-G-62-43 (P-12-2); entire specimen with one end rounded, the other pointed. Note the cytoplasmic residues preserved within individual cells.

Oscillatoriopsis longa 6. HUPC 62873; 86-G-61-7 ((3-45-0). 7. HUPC 62874; 86-G-61-7 ((>45-0).

Tortunema wemadskii n. comb. 8. HUPC 62875; 86-G-62-41M (M-35-1); detail of Fig. 27.3; note that the pseudoseptate sheath collapsed as a unit rather than as individual cells.

Veteronostocale amoenum 9. HUPC 62876; 86-G-62-6 (G-39-2).

Cvanonema sp. 10. HUPC 62877; 86-G-62-10 (P-22-1). 9 . 1

.4 ,

V # X \

f 191

Figure 25. Scale bar in 5 equals 50 |im for 1, 2, 4-7, 9; 80 for 3, 8. Figure captions include HUPC number, slide identification, and "England Finder" location coordinates (in parentheses). Figures 1-5, 7-9 from shale; 6 in chert.

Rugosoopsis tenuis 1. HUPC 62879; 86-C-62-126M (J-36-1); with several unoriented cell remnants within the sheath. 2. HUPC 62880; 86-G-62-126M (K-40-3); with internal cells; note the apparent cellularity resulting from the transverse fabric of the outer sheath. 3. HUPC 62881; 86-G-62-148M (P-43-3); outer sheath partially removed. 4. HUPC 62882; 86-G-62-120M (J-28-4); outer sheath largely removed.

Palaeolvngbva hebeiensis 5. HUPC 62883; 86-G-62-132M (0-32-2); note the single transverse 'w rap' at the arrow, indicating the prior presence of a Rugosoopsis-like outer sheath.

Palaeolvngbva catenata 6. HUPC 62884; P-2628-1A (D-58-1); in chert. 7. HUPC 62885; 86-G-62-67 (P-11-4); in shale.

Palaeolvngbva/Rugosoopsis/Siphonophvcus 8. HUPC 62886; 86-G-62-93M (P-46-0); various filamentous form taxa that may represent a single biological entity. Note the narrow cellular tridiome within the otherwise Siphonophvcus gracile n. comb, filament. 9. HUPC 62887; 86-G-62-115M (M-29-4); series of atrophied cells within a collapsed filamentous sheath. The associated diaphanous material appears to be disaggregated Rugosoopsis-type sheath. :/î« * 5 r ï' MIÆ 193

Figure 26. Scale bar in 8 equals 80 pm for 1-2, 5, 7-9; 200 pm for 3; 50 pm for

4, 6. Figure captions include HUPC number, slide identification, and "England

Finder" location coordinates (in parentheses). Figures 1-7 from shales; 8, 9 in chert.

Siphonophvcus robustum 1. HUPC 62888; 86-G-62-60 (N-29-3); in situ mat in bedding-parallel thin section of shale. 7. HUPC 62893; 86-G-62-173M (K-34-2); small Nostoc-ball-like aggregation of filaments.

Siphonophvcus tvpicum n. comb. 2. HUPC 62889; 86-G-30-6 (N-16-2); in situ mat in bedding-parallel thin section of shale. 8. HUPC 62894; 86-G-14-1A (R-65-4); dense, largely unoriented filamentous mat with no input of sedimentary particles; intraclast in silicified intraclastic grainstone. 9. HUPC 62895; 86-G-15-1 (T-46-3); alternating vertically and horizontally oriented filamentous mat with significant input of allochthonous sediments; intraclast in grainstone.

Filaments mineralized by goethite(?) 3. HUPC 62890; 86-G-30-4BP; bedding plane specimen. 4. HUPC 62891; 86-G-30-3 (0-37-1); bedding-parallel thin section showing simple Siphonophvcus filaments with thick mineral overgrowths. 5. FIUPC 62892; 86-G-30-5 (G-38-3); bedding-parallel specimen exhibiting diagenetically induced 'branching'. 6. 86-G-30 (destroyed); entangled, mat-like habit in bedding-parallel thin section.

195

Figure 27. Scale bar in 2 equals 100 |im. Figure captions include HUPC number, slide identification, and "England Finder" location coordinates (in parentheses). All specimens from shale.

T ortunema/Rugosoopsis/Siphonophvcus 1. HUPC 62896; 86-G-62-147M (R-21-2); pseudoseptate filament (Tortunema) containing freely transported cell residues and a halo of Rugosoopsis-like outer sheath material. 2. HUPC 62897; 86-G-62-233M (N-37-0); pseudoseptate filament (Tortunema) with localized sections that could be classified as Siphonophvcus or Rugosoopsis: note the halo of disaggregating outer sheath material. 3. HUPC 62875; 86-G-62-41M (M-35-1); pseudoseptate filament (Tortunema) with intact outer sheath. Unlike the transverse fabric typical of Rugosoopsis. this outer layer appears to be oriented longitudinally. 4. HUPC 62898; 86-G-62-113M (L-30-0); Siphonophvcus gracile n. comb., with dispersed cells. i 197

Figure 28. Scale bar in 4 equals 220 pm for 1-3; 50 pm for 4, 8-10; 80 pm for 5-

7. Figure captions include HUPC number, slide identification, and "England

Finder" location coordinates (in parentheses). All specimens from shales.

Pseudodendron anteridium n. gen., n. sp. 1. HUPC 62899; P-2945-9M (N-29-0); tapering filaments of various diameters, with a common base. 2. HUPC 62900; P-2945-4M (N-34-3); with reinforced branch junctions. 3. HUPC 62901; P-2945-13M (0-30-0); with terminal swelling. 4. HUPC 62902; 86-G-62-33 (R-49-4); illustrating the capacity to fuse separated axes. 5. HUPC 62903; P-2945-5M (K-29-2); with four subordinate branches; note the confluence of the branches with the outer sheath of the main axis. 6. HUPC 62904; P-2945-74M (L-18-4). 7. P-2945 (destroyed); note that the 'cores' of the secondary axes do not fuse with one another, nor with that of the main axis. 10. HUPC 62720; P-2945-7M (L-41-3); Holotype.

Pseudodendron polvtaenium n. gen., n. sp. 8. HUPC 62905; 86-G-62-96M (L-39-2); note the slight envelope, just discernable at the branch junction. 9. HUPC 62906; 86-G-62-237M (K-17-0); detail of a 1.25 mm long, otherwise unbranched filament; note the thin enveloping sheath and three branches. f 199

REFERENCES

ALEXANDER, M. 1973. Nonbiodegradable and other recalcitrant molecules. Biotechnology and Bioengineering 15:611-647. ALLISON, C. W. and S. M. AWRAMIK. 1989. Organic-walled microfossils from earliest Cambrian or latest Proterozoic Tindir Group rocks, northwest Canada. Precambrian Research 43:253-294. ALLISON, P. A. 1988. Konservat-Lagerstatten: cause and classification. Paleobiology 14:331-344. ASMEROM, Y., S. B. JACOBSEN, A. H. KNOLL, N. J. BUTTERFIELD, and K. SWETT. 1991. Strontium isotopic variations of Neoproterozoic seawater; Implications for crustal evolution. Geochimica et Cosmochimica Acta 55:2883-2894. ASSEJEVA, E. A. 1982. Novye planktonnykh vodoroslei venda Volyno-Podolii. Pp. 5-16. In Yu. V. Teslenko (ed.), Sistematika i Evoliutsiia Drevnikh Rastenii Ukrainy. Naukova Dumka; Kiev. ASSEJEVA, E. A. 1983. Stratigraficheskoe znachenie pozdnedokembriiskikh mikrofossilii iugo-zapada Vostochno-Evropeiskoi platformy. Pp. 148-180. In V. A. Riabenko (ed.), Stratigrafiia i Formatsii Dokembriia Ukrainy. Naukova Dumka; Kiev. ASSEJEVA, E. A. and V. A. VELIKANOV. 1983. Novaia nakhodka iskopaemykh fitoostatkov v liadovskikh sloiakh venda Podolii, verkhnii Dokembrii. Pp. 3-8. In O. S. Vialov (ed.), Iskopaemaia Fauna i Flora Ukrainy. Naukova Dumka; Kiev. ATKINSON, A. W., Jr., B. E. S. GUNNING, and P. C. L. JOHN. 1972. Sporopollenin in the cell wall of Chlorella and other algae: 200

ultrastructure, chemistry, and incorporation of ^^C-acetate, studied in synchronous cultures. Planta (Berlin) 107:1-32. AVNIMELECH, Y., B. W. TROEGER, and L. W. REED. 1982. Mutual flocculation of algae and clay: evidence and implications. Science 216:63-65. AWRAMIK, S. M. and J. W. VALENTINE. 1985. Adaptive aspects of the origin of autotrophic eukaryotes. Pp. 11-21. In B. H. Tiffney (ed.). Geological Factors and the Evolution of Plants. Yale University Press; New Haven. BAULD, J. 1981. Geobiological role of cyanobacterial mats in sedimentary environments: production and preservation of organic matter. BMR Journal of Australian Geology and Geophysics 6:307-317. BERKALOFF, C., E. CASADEVALL, C. LARGEAU, P. METZGER, S. PERACCA, and J. VERLET. 1983. The resistant polymer of the walls of the hydrocarbon-rich alga Botrvococcus braunii. Phytochemistry 22:389-397. BHATTACHARYA, D., H. J. ELWOOD, L. J. GOFF, and M. L. SOGDM. 1990. Phylogeny of Gracilaria lemaneiformis (Rhodophyta) based on sequence analysis of its small subunit ribosomal RNA coding region. Journal of Phycology 26:181-186. BIRCH, P. B., J. O. GABRIELSON, and K. S. HAMEL. 1983. Decomposition of Cladophora: I. Field studies in the Peel-Harvey estuarine system. Western Australia. Botanica Marina 26:165-171. BLOESER, B. 1985. Melanocvrillium, a new genus of structurally complex late Proterozoic microfossils from the Kwagunt Formation (Chuar Group), Grand Canyon, Arizona. Journal of Paleontology 59:741-765. BONNER, J. T. 1988. The Evolution of Complexity by Means of Natural Selection. Princeton University Press; Princeton. 260 pp. 201

B0RGESEN, F. 1913. The marine algae of the Danish West Indies. Part I, Chlorophyceae. Dansk Botanisk Arkiv 1(4). 158 pp. BOWMAN, B. H., J. W. TAYLOR, A. G. BROWNLEE, J. LEE, S. LU, and T. J. WHITE. 1992. Molecular evolution of the Fungi: relationship of the basidiomycetes, ascomycetes, and chytridiomycetes. Molecular Biology and Evolution 9:285-296. BRIGGS, D. E. G. and S. H. WILLIAMS. 1981. The restoration of flattened fossils. Lethaia 14:157-164. BROTZEN, F. 1941. Nâgra bidrag till visingsoformationens stratigrafi och tektonik. Geologiska Foreningens Forhandlinger 63:245-261. BRUNS, T. D., T. J. WHITE, and J. W. TAYLOR. 1991. Fungal molecular system.atics. Annual Review of Ecology and Sysiematics 22:525-564. BURNS, R. G. 1979. Interaction of microorganisms, their substrates and their products with soil surfaces. Pp. 109-138. In D. C. Ellwood, J. Melling, P. Rutter (eds.). Adhesion of Microorganisms to Surfaces. Academic Press; London. BUSS, L. W. 1987. The Evolution of Individuality. Princeton University Press; Princeton. 201 pp. BUTTERFIELD, N. J. 1990. Organic preservation of non-mineralizing organisms and the taphonomy of the Burgess Shale. Paleobiology 16:272-286. BUTTERFIELD, N. J. and F. W. CHANDLER. 1992. Paleoenvironmental distribution of Proterozoic microfossils, with an example from the Agu Bay Formation, Baffin Island. Palaeontology (in press). BUTTERFIELD, N. J., A. H. KNOLL, and K. SWETT. 1988. Exceptional preservation of fossils in an Upper Proterozoic shale. Nature 334:424-427. 202

BUTTERFIELD, N. J. and A. H. KNOLL. 1989. Morphologically complex fossils through the Late Proterozoic: their application to biostratigraphy and paleobiology. Geological Association of Canada, Program with Abstracts 14:A100. BUTTERFIELD, N. J., A. H. KNOLL, and K. SWETT. 1990. A bangiophyte red alga from the Proterozoic of arctic Canada. Science 250:104-107. BUTTERFIELD, N. J. and R. H. RAINBIRD. 1988. The paleobiology of two Proterozoic shales. Geological Society of America, Abstracts with Programs 20:A103. CAVALIER-SMITH, T. 1978. Nuclear volume control by nucleoskeletal DNA, selection for cell volume and cell growth rate, and the solution of the DNA C-value paradox. Journal of Cell Science 34:247-278. CHALANSONNET, S., C. LARGEAU, E. CASADEVALL, C. BERKALOFF, G. PENIGUEL, and R. COUDERC. 1987. Cyanobacterial resistant biopolymers. Geochemical implications of the properties of Schizothrix sp. resistant material. Organic Geochemistry 13:1003-1010. CHAPMAN, F. 1935. Primitive fossils, possibly atrematous and neotrematous Brachiopoda, from the Vindhyans of India. Geological Survey of India, Records 69:109-120. CHEN M. and XIAO Z. 1991. Discovery of the macrofossils in the upper Sinian Doushantuo Formation at Maiohe, eastern Yangtze Gorges. Scientia Geologica Sinica 10:317-324. CONWAY MORRIS, S. 1979. Middle Cambrian polychaetes from the Burgess Shale of British Columbia. Philosophical Transactions of the Royal Society of London B 285:227-274. CONWAY MORRIS, S. 1986. The community structure of the Middle Cambrian Phyllopod Bed (Burgess Shale). Palaeontology 29:423-467. 203

COPELAND, J. J. 1936. Yellowstone thermal Myxophyceae. Annals of the New York Academy of Science 36:1-232. DAWSON, M. P., B. A. HUMPHREY, and K. C. MARSHALL. 1981. Adhesion: a tactic in the survival strategy of a marine vibrio during starvation. Current Microbiology 6:195-199. DERRY, L. A., L. S. KETO, S. B. JACOBSEN, A. H. KNOLL, and K. SWETT. 1989. Sr isotopic variations in Upper Proterozoic carbonates from Svalbard and East Greenland. Geochimica et Cosmochimica Acta 53:2331-3239. DOWNIE, C. 1982. Lower Cambrian acritarchs from Scotland, Norway, Greenland and Canada. Transactions of the Royal Society of Edinburgh: Earth Sciences 72:257-285. DU R. 1982. The discovery of the fossils such as Chuaria in the Qingbaikou system in Northwestern Hebei and their significance. Geology Review (Beijing) 28:1-7. DU R. and HAN L. 1985. Algal macrofossils from the Qingbaikou System in the Yanshan Range of north China. Precambrian Research 29:5-14. DU R., HAN L., and LI H. 1986. Discovery of megafossils in the Gaoyuzhuang Formation of the Changchenian System, Jixian. Acta Geologica Sinica 1986(2):115-120. DUAN C. 1982. Late Precambrian algal megafossils Chuaria and Tawuia in some areas of eastern China. Alcheringa 6:57-68. EISENACK, A. 1951. Über Hystrichosphaerideen und andere Kleinformen aus Baltischem Silur und Kambrian. Senckenbergiana Lethaea 32:187-204. EISENACK, A. 1955. Chitinozoen, Hystrichospharen und andere Mikrofossilien aus dem Beyrichia-Kalk. Senckenbergiana Lethaea 36:157-188. 204

EISENACK, A. 1958. Tasmanites Newton 1875 und Leiosphaeridia n. g. aïs Gattungen der Hystrichosphaeridea. Palaeontographica A 110:1-19. EISENACK, A. 1966. Über Chuaria wimani Brotzen. Neues Jahrbuch für Geologie und Palâontlogie, Monatshefte 1966(l):53-56. EISENACK, A. 1969. Zur Systematik einiger palâozoischer Hystrichospharen (Acritarcha) des baltischen Gebietes. Neues Jahrbuch für Geologie und Palâontologie, Abhandlungen 133:245-266. EVm, W. R. 1963a. Occurrence of freshwater alga Pediastrum in Cretaceous marine sediments. American Journal of Science 261:890-893. EVITT, W. R. 1963b. A discussion and proposals concerning fossil dinoflagellates, hystrichospheres, and acritarchs, II. Proceedings, National Academy of Science, U.S.A. 49:298-302. FAIRCHILD, T. 1975. The geological setting and paleobiology of a Late Precambrian stromatolitic microflora from South Australia. Unpublished Ph.D. dissertation. University of California, Los Angeles. 272 pp. FALLON, R. D. and T. D. BROCK. 1979. Decomposition of blue-green algal (cyanobacterial) blooms in Lake Mendota, Wisconsin. Applied and Environmental Microbiology 37:820-830. FARMER, J. D. 1989. Observations on the distribution and abundance of some modem laminated microbial mats and the late Proterozoic decline in stromatolite diversity and abundance. Geological Society of America, Abstracts with Programs 21:A146-A147. FATTOM, A. and M. SHILO. 1984. Hydrophobicity as an adhesion mechanism of benthic cyanobacteria. Applied and Environmental Microbiology 47:135-143. 205

FERRIS, F. G., W. S. FYFE, and T. J. BEVERIDGE. 1988. Metallic ion binding by Bacillus subtilis: implications for the fossilization of microorganisms. Geology 16:149-152. FONTES, D. E., A. L. MILLS, G. M. HORNBERGER, and J. S. HERMAN. 1991. Physical and chemical factors influencing transport of microorganisms through porous media. Applied and Environmental Microbiology 57:2473-2481. FORD, T. D. and W. J. BREED. 1973. The problematical Precambrian fossil Chuaria. Palaeontology 16:535-550. FOREE, E. G. and P. L. McCARTY. 1970. Anaerobic decomposition of algae. Environmental Science and Technology 4:842-849. FRANK, T. H. and A. G. LANDMAN. 1988. Morphology and taxonomy of Microcrocis sp. (Chroococcales, Cyanophyta) from Lake Vechten, The Netherlands, with a note on its observed movements. Archiv für Hydrobiologie 78(Supplement):397-410. FRITSCH, F. E. 1935. The Structure and Reproduction of the Algae; Volume 1. Macmillan; New York, and The University Press; Cambridge. FU J. 1989. New materials of late Precambrian Huainan biota fossil in Shouxian, Anhui. Acta Palaeontologica Sinica 28:72-78. GEITLER, L. 1925. Cyanophyceae. Pp. 1-450. In A. Pascher (ed.). Die Süsswasserflora Deutschlands, Osterreichs, und der Schweiz 12. GNILOVSKAYA, M. B. (ed ). 1988. Vendotaenids of the East European Platform. Nauka; Leningrad. 144 pp. GOLOVENOC, V. K. and M. Yu. BELOVA. 1984. Rifeiskie mikrobioty v kremniakh iz Billiakhskoi serii Anabarskogo podniatiia. Pédeontologicheskii Zhumal 1984(4):23-32. 206

GOLOVENOC, V. K. and M. Yu. BELOVA. 1985. Rifeiskie mikrobioty v kremniakh Yeniseiskogo kriazha. Paleontologicheskii Zhurnal

1985(2):94-103. COLUBIC, S. and E. S. BARCHOORN. 1977. Interpretation of microbial fossils with special reference to the Precambrian. Pp. 1-14. In E. Fliigel (ed.), Fossil Algae. Springer-Verlag; Berlin. GOLUBIC, S. and H. J. HOFMANN. 1976. Comparison of Holocene and mid- Precambrian Entophysalidaceae (Cyanophyta) in stromatolitic algal mats: cell division and degradation. Journal of Paleontology 50:1074-1082. GRANT, S. W. F., A. H. KNOLL, and G. J. B. GERMS. 1991. Probable calcified metaphytes in the latest Proterozoic Nama Group, Namibia: origin, diagenesis, and implications. Journal of Paleontology 65:1-18. GRAY, J. 1960. Fossil chlorophycean algae from the Miocene of Oregon. Journal of Paleontology 34:453-463. GREEN, J. W., A. H. KNOLL, S. GOLUBIC, and K. SWETT. 1987. Paleobiology of distinctive benthic microfossils from the upper Proterozoic Limestone-Dolomite "Series", central East Greenland. American Journal of Botany 74:928-940. GREEN, J. W., A. H. KNOLL, and K. SWETT. 1988. Microfossils from oolites and pisolites of the Upper Proterozoic Eleonore Bay Group, central East Greenland. Journal of Paleontology 62:835-852. GREEN, J. W., A. H. KNOLL, and K. SWETT. 1989. Microfossils from silicified stromatolitic carbonates of the Upper Proterozoic Limestone-Dolomite 'Series', central East Greenland. Geological Magazine 126:567-585. 207

GREY, K. and I. R. WILLIAMS. 1990. Problematic bedding-plane markings from the Middle Proterozoic Manganese Subgroup, Bangemall Basin, Western Australia. Precambrian Research 46:307-328. GUNNISON, D. and M. ALEXANDER. 1975. Resistance and susceptibility of algae to decomposition by natural microbial communities. Limnology and Oceanography 20:64-70. GUSSOW, W. C. 1973. Chuaria sp. cf. Ç circularis Walcott from the Precambrian Hector Formation, Banff National Park, Alberta, Canada. Journal of Paleontology 47:1108-1112. HANIC, L. A. and J. S. CRAIGIE. 1969. Studies on the algal cuticle. Journal of Phycology 5:89-102. HARRIS, T. M. 1974. Williamsoniella lienieri: its pollen and the compression of spherical pollen grains. Palaeontology 17:125-148. HATCHER, P. G., E. C. SPIKER, N. M. SZEVERENYI, and G. E. MACIEL. 1983. Selective preservation and origin of petroleum-forming aquatic kerogen. Nature 305:498-501. HAXO, F. T., R. A. LEWIN, K. W. LEE, and MING-REN LI. 1987. Fine structure and pigments of Oscillatoria (Trichodesmium) aff. thiebautii (Cyanophyta) in culture. Phycologia 26:443-456. HEAMAN, L. M., A. N. LeCHEMINANT, and R. H. RAINBIRD, R. H. 1990. A U-Pb baddeleyite study of Franklin igneous events, Canada. Geological Association of Canada, Program with Abstracts 15:A55. HENDRIKS, L., R. De BAERE, Y. VAN de PEER, J. NEEFS, A. GORIS, and R. De WACHTER. 1991. The evolutionary position of the rhodophyte Porphvra umbilicalis and the basidiomycete Leucosporidium scottii among other eukaryotes as deduced from complete sequences of small ribosomal subunit RNA. Journal of Molecular Evolution 32:167-177. 208

HERMANN, T. N. 1974. Nakhodki massovykh skoplenii trikhomov v rifee. Pp. 6-10. In B. V. Timofeev (éd.), Mikrofitofossilii Proterozoia i Rannego Paleozoia SSSR. Nauka; Leningrad. HERMANN, T. N. 1979. Nakhodki gribov v rifee. Pp. 129-136. In B. S. Sokolov (éd.), Paleontologiia Dokembriia i Rannego Kembriia. Nauka; Leningrad. HERMANN, T. N. 1981a. Nitchatye mikroorganizmy Lakhandinskoi svity reki Mai. Paleontologicheskii Zhumal 1981(2):126-131. HERMANN, T. N. 1981b. Nakhodki nitchatykh vodoroslei v Miroedikhinskoi svite verkhnego dokembriia. Paleontologicheskii Zhurnal 1981(4):118-121. HERMANN, T. N. 1986. O nakhodkakh sinezelenykh nitchatykh vodoroslei v verkhnem dokembrii (Miroedikhinskoi svite). Pp. 37-40. In B. S. Sokolov (éd.), Aktualnye Voprosy Sovremennoi Paleoalgologii. Naukova Dumka; Kiev. HOFMANN, H. J. 1976. Precambrian microflora, Belcher Islands, Canada: significance and systematics. Journal of Paleontology 50:1040-1073. HOFMANN, H. J. 1977. The problematic fossil Chuaria from the late Precambrian Uinta Mountain Group, Utah. Precambrian Research 4:1-11. HOFMANN, H. J. 1985a. The mid-Proterozoic Little Dal macrobiota, Mackenzie Mountains, north-west Canada. Palaeontology 28:331-354. HOFMANN, H. J. 1985b. Precambrian carbonaceous megafossils. Pp. 20-33. In D. F. Toomey and M. H. Nitecki (eds.), Paleoalgology: Contemporary Research and Applications. Springer Verlag; Berlin. HOFMANN, H. J. and J. D. ATTKEN. 1979. Precambrian biota from the Little Dal Group, Mackenzie Mountains, northwestern Canada. Canadian Journal of Earth Sciences 16:150-166. 209

HOFMANN, H. J. and CHEN J. 1981. Carbonaceous megafossils from the Precambrian (1800 Ma) near Jixian, northern China. Canadian Journal of Earth Sciences 18:443-447. HOFMANN, H. J. and G. D. JACKSON. 1991. Shelf-facies microfossils from the Uluksan Group (Proterozoic Bylot Supergroup), Baffin Island, Canada. Journal of Paleontology 65:361-382. HORODYSKI, R. J. 1980. Middle Proterozoic shale-facies microbiota from the lower Belt Supergroup, Little Belt Mountains, Montana. Journal of Paleontology 54:649-663. HORODYSKI, R. J. 1982. Problematic bedding-plane markings from the Middle Proterozoic Appekunny Argillite, Belt Supergroup, northwestern Montana. Journal of Paleontology 56:882-889. HORODYSKI, R. J., B. BLOESER, and S. VONDER HAAR. 1977. Laminated algal mats from a coastal lagoon. Laguna Mormona, Baja California, Mexico. Journal of Sedimentary Petrology 47:680-696. HORODYSKI, R. J. and B. BLOESER. 1983. Possible eukaryotic algal filaments from the late Proterozoic Chuar Group, Grand Canyon, Arizona. Journal of Paleontology 57:321-326. HORODYSKI, R. J. and J. A. DONALDSON. 1980. Microfossils from the Middle Proterozoic Dismal Lakes Group, arctic Canada. Precambrian Research 11:125-159. HORODYSKI, R. J. and C. MANKIEWICZ. 1990. Possible Late Proterozoic skeletal algae from the Pahrump Group, Kingston Range, southeastern California. American Journal of Science 290-A:149-169. JANKAUSKAS, T. V. 1978. Rastitelnye mikrofossilii iz rifeiskikh otlozhenii luzhnogo Urala. Doklady Akademii Nauk SSSR 242:913-915. 210

JANKAUSKAS, T. V. 1980a. Shisheniakskaia mikrobiota verkhnego rifeia luzhnogo Urala. Doklady Akademii Nauk SSSR 251:190-192. JANKAUSKAS, T. V. 1980b. Novye vodorocli iz verkhnego rifeia luzhnogo Urala i Priuralia. Paleontologicheskii Zhurnal 1980(4):107-113. JANKAUSKAS, T. V. (éd.). 1989. Mikrofossilii Dokembriia SSSR. Nauka; Leningrad. 188 pp. JANKAUSKAS, T. V., N. S. MIKHAILOVA, and T. N. HERMANN. 1987. V vsesoiuznyi kollokvium po mikrofossiliiam dokembriia SSSR. Izvestiia Akademii Nauk SSSR, Seriia Geologicheskaia 1987(9):137-139. JEWELL, W. J. and P. L. McCARTY. 1971. Aérobic decomposition of algae. Environmental Science and Technology 5:1023-1031. KAPLAN, D. R. and W. HAGEMANN. 1991. The relationship of cell and organism in vascular plants. BioSdence 41:693-703. KARL, D. M., G. A KNAUER, and J. H. MARTIN. 1988. Downward flux of particulate organic matter in the ocean: a particle decomposition paradox. Nature 332:438-441. KAUFFMAN, S. A. 1987. Developmental logic and its evolution. BioEssays 6:82-87. KAUFMAN, A. J., A. H. KNOLL, and S. M. AWRAMIK. 1992. Bio- and chemo- stratigraphic correlation of Neoproterozoic sedimentary successions: the upper Tindir Group, northwestern Canada, as a test case. Geology 20:181-185. KID WELL, S. M. and T. BAUMILLER. 1990. Experimental disintegration of regular echinoids: roles of temperature, oxygen, and decay thresholds. Paleobiology 16:247-271. 211

KIEBAR, R. J., X. ZHOU, and K. MOPPER. 1990. Formation of carbonyl compounds from UV-induced photodegradation of humic substances in natural waters: Fate of riverine carbon in the sea. Limnology and Oceanology 35:1503-1515. KIRCHNER, O. 1898. Oscillatoriaceae. Pp. 61-70. In A. Engler and K. Prantl (eds.). Die Natiirlichen Pflanzenfamilien, I, la. KNOLL, A. H. 1981. Paleoecology of late Precambrian microbial assemblages. Pp. 17-54. In K. J. Niklas (ed.). Paleobotany, Paleoecology, and Evolution; Volume 1. Praeger; New York. KNOLL, A. H. 1982. Microfossils from the late Precambrian Draken Conglomerate, Ny Friesland, Svalbard. Journal of Paleontology 56:755-790. KNOLL, A. H. 1984. Microbiotas of the late Precambrian Hunnberg Formation, Nordaustlandet, Svalbard. Journal of Paleontology 58:131-162. KNOLL, A. H. 1985. Exceptional preservation of photosynthetic organisms in silicified carbonates and silicified peats. Philosophical Transactions of the Royal Society of London B 311:111-122. KNOLL, A. H. 1991. End of the Proterozoic Eon. Scientific American 265(4):64-73. KNOLL, A. H. and N. J. BUTTERFIELD. 1989. N ew Window on Proterozoic Life. Nature 337:602-603. KNOLL, A. H. and S. CALDER. 1983. Microbiotas of the late Precambrian Rysso Formation, Nordaustlandet, Svalbard. Palaeontology 26:467-496. KNOLL, A. H. and S. GOLUBlC. 1979. Anatomy and taphonomy of a Precambrian algal stromatolite. Precambrian Research 10:115-151. 212

KNOLL, A. H., J. M. HAYES, A. J. KAUFMAN, K. SWETT, and I. B. LAMBERT. 1986. Secular variation in carbon isotope ratios from Upper Proterozoic successions of Svalbard and East Greenland. Nature 321:1-7. KNOLL, A. H. and Y. OHTA. 1988. Microfossils in metasediments from Prins Karls Borland, western Svalbard. Polar Research 6:59-67. KNOLL, A. H., K. SWETT, and E. BURKHARDT. 1989. Paleoenvironmental distribution of microfossils and stromatolites in the upper Proterozoic Blacklundtoppen Formation, Spitsbergen. Journal of Paleontology 63:129-145. KNOLL, A. H. and K. SWETT. 1990. Carbonate deposition during the Late Proterozoic Era: an example from Spitsbergen. American Journal of Science 290-A:104-132. KNOLL, A. H., K. SWETT, and J. MARK. 1991. Paleobiology of a Neoproterozoic tidal flat/lagoonal complex: the Draken Conglomerate Formation, Spitsbergen. Journal of Paleontology 65:531-570. KNOLL, A. H., P. K. STROTHER, and S. ROSSI. 1988. Distribution and diagenesis of microfossils from the lower Duck Creek Dolomite, Western Australia. Precambrian Research 38:257-279. KOLOSOV, P. N. 1982. Verkhnedokembriiskie paleoalgologicheskie ostatki Sibirskoi platformy. Nauka; Moscow. 94 pp. KOLOSOV, P. N. 1984. Pozdnedokembriiskie Mikroorganizmy Vostoka Sibirskoi platformy. Akademiia Nauk SSSR, Ordena Lenina Sibirskoe Otdelenie, Yakutskii Filial, Institut Geologii; Yakutsk. 84 pp. KUC, M. 1972. Fossil statoblasts of Cristatella mucedo Cuvier in the Beaufort Formation and in interglacial and postglacial deposits of the Canadian Arctic. Geological Survey of Canada, Paper 72-28:1-12. 213

KÜTZING, F. T. 1843. Phycologia Generalis. F. A. Brockhaus; Leipzig. 458 pp. KÜTZD^G, F. T. 1845. Phycologia Germanica. W. Kôhne; Nordhausen. 340 pp. LANIER, W. P. 1988. Structure and morphogenesis of microstromatolites from the Transvaal Supergroup, South Africa. Journal of Sedimentary Petrology 58:89-99. LEO, R. F. and E. S. BARGHOORN. 1976. Silicification of wood. Botanical Museum Leaflets, Harvard University 25(1). 47 pp. LICARI, G. R. 1978. Biogeology of the late pre-Phanerozoic Beck Spring Dolomite of eastern California. Journal of Paleontology 52:767-792. LIU C. 1982. Micro-fossil algal communities from the Wumishan Formation in Jixian, China, and their geological significance. Bulletin of Nanjing University - Special Paper on Algae. Pp. 121-166. LIU X., LIU Z., ZHANG L., and XU X. 1984. A study of late Precambrian microfossil algal community from Suining County, Jiangsu Province. Acta Micropalaeontologica Sinica 1:171-182. LIU Z and LI H. 1986. Fossil blue-green algal community from the upper Permian of Guangxi and its significant role in forming bottom of coal beds. Acta Micropalaeontologica Sinica 3:261-272. LO, S. C. 1980. Microbial fossils from the lower Yudoma Suite, earliest Phanerozoic, eastern Siberia. Precambrian Research 13:109-166. LOCHTE, K. and C. M. TURLEY. 1988. Bacteria and cyanobacteria associated with photodetritus in the deep sea. Nature 333:67-69. MÀDLER, K. 1964. Bemerkenswerte Sporenformen aus dem keuper und unteren Lias. Fortschritte in der Geologie von Rheinland und Westfalen 12:169-200. 214

MAITHY, P. K. 1975. Micro-organisms from the Bushimay System (late pre- Cambrian) of Kanshi, Zaire. The Palaeobotanist 22:133-149. MAITHY, P. K. and M. SHUKLA. 1977. Microbiota from the Suket Shales, Ramapura, Vindhyan System (late pre-Cambrian), Madhya Pradesh. The Palaeobotanist 23:176-188. MAITHY, P. K. and M. SHUKLA. 1984. Reappraisal of Fermoria and allied remains from the Suket Shale Formation, Ramapura. The Palaeobotanist 32:146-152. MALIVA, R. G., A. H. KNOLL, and R. SIEVER. 1989. Secular change in chert distribution: a reflection of evolving biological participation in the silica cycle. Palaios 4:519-532. MARCHANT, H. J. 1977. Cell division and colony formation in the green alga Coelastrum (Chlorococcales). Journal of Phycology 13:102-110. MATSUDA, Y. 1988. The Chlamvdomonas cell walls and their degrading enzymes. Japanese Journal of Phycology 36:246-264. MATTOX, K. R. and K. D. STEWART. 1984. Classification of the green algae. Pp. 29-72. In D. E. G. Irvine and D. M. John (eds.), Systematics of the Green Algae. Academic Press; London. McMENAMIN, D. S., S. KUMAR, and S. M. AWRAMIK. 1983. Microbial fossils from the Kheinjua Formation, Middle Proterozoic Semri Group (Lower Vindhyan), Son Valley area. Central India. Precambrian Research 21:247-271.

MENDELSON, C. V. and J. W. SCHOPF. 1982. Proterozoic microfossils from the Sukhaya Tunguska, Shorikha, and Yudoma formations of the Siberian Platform, USSR. Journal of Paleontology 56:42-83. 215

MIKHAILOVA, N. S. 1986. Novye nakhodki mikrofitofossilii iz otlozhenii verkhnego Rifeia Krasnoiarskogo kraia. Pp. 31-37. In B. S. Sokolov (ed.), Aktualnye Voprosy Sovremennoi Paleoalgologii. Naukova Dumka; Kiev. MUIR, M. D. 1976. Proterozoic microfossils from the Amelia Dolomite, McArthur Basin, Northern Territory. Alcheringa 1:143-158. NÀGELI, C. 1849. Gattungen Einzellinger Algen, Physiologisch und Systematisch Bearbeitet. F. Schulthess; Zürich. 139 pp. NAUTIYAL, A. C. 1980. Cyanophycean algal remains and paleoecology of the Precambrian Gangolihat Dolomites Formation of the Kumaun Himalaya. Indian Journal of Earth Sciences 7:1-11. NAUTIYAL, A. C. 1982. Algal remains from the Random Formation (late Precambrian) of Newfoundland, Canada. Indian Journal of Earth Sciences 9:174-177. NYBERG, A. V. and J. W. SCHOPF. 1984. Microfossils in stromatolitic cherts from the upper Proterozoic Min'yar Formation, southern Ural Mountains, USSR. Journal of Paleontology 58:738-772. OEHLER, J. H. 1977. Microflora of the H.Y.C. Pyritic Shale Member of the Barney Creek Formation (McArthur Group), middle Proterozoic of northern Australia. Alcheringa 1:315-349. OGURTSOVA, R. N. and V. N. SERGEEV. 1987. Mikrobiota Chichkanskoi svity verkhnego dokembriia Malogo karatau (iuzhnyi Kazakhstan). Paleontologicheskii Zhumal 1987(2):107-116. OGURTSOVA, R. N. and V. N. SERGEEV. 1989. Megasferomorfidy Chichkanskoi svity verkhnego dokembriia iuzhnogo Kazakhstana. Paleontologicheskii Zhum al 1989(2):119-122. 216

O'KELLY, C. J. and G. L. FLOYD. 1984. Correlations among patterns of sporangial structure and development, life histories, and ultrastructural features in the Ulvophyceae. Pp. 121-156. In D. E. G. Irvine and D. M. John (eds.), Systematics of the Green Algae. Academic Press; London. OLTMANNS, F. 1904. Morphologie und Biologie der Algen, Band 1. G. Fischer; Jena. 733 pp. OOSTHUYZEN, E. J. and A. J. BURGER. 1973. The suitability of apatite as an age indicator by the uranium-lead isotope method. Earth and Planetary Science Letters 18:29-36. GUYANG S., YIN L., and LI Z. 1974. Sinian and Cambrian spores. Pp. 72-80 and 114-123. In: A Handbook of the Stratigraphy and Paleontology of Southwestern China. Science Publishing House; Beijing. PASCHER, A. 1914. Über Flagellaten und Algen. Berichte der Deutschen Botanischen Gesellschaft 32:136-160. PERASSO, R., A. BAROIN, LIANG HU QU, J. P. BACHELLERIE, and A. ADOUTTE. 1989. Origin of the algae. Nature 339:142-144. PHILP, R- P. and M. CALVIN. 1976. Possible origin for insoluble organic (kerogen) debris in sediments from insoluble cell-wall materials of algae and bacteria. Nature 262:134-136. PJATILETOV, V. G. 1979. O nakhodkakh sinezelenykh vodoroslei v Yudomskikh otlozheniiakh Yakutii (vend). Doklady Akademii Nauk SSSR 249:714-716. PJATILETOV, V. G. 1980. Yudomskii kompleks mikrofossilii iuzhnoi Yakutii. Geologiia i GeoGzika 21(7):8-20. 217

PJATILETOV, V. G. 1988. Mikrofitofossilii pozdnego dokembriia Uchuro- Maiskogo raiona. Pp. 47-104. In V. V. Khomentovsky and V. Yu. Shenfil (eds.), Pozdnii Dokembrii i Rannii Paleozoi Sibiri Rifei i Vend. Akademiia Nauk SSSR, Sibirskoe Otdelenie, Institut Geologii i Geofiziki; Novosibirsk. PJATILETOV, V. G. and G. A. KARLOVA. 1980. Verkhnerifeiskii kompleks rastitelnykh mikrofossilii Yenisieskogo kriazha. Pp. 56-71. 7« V. V. Khomentovsky (éd.), Novye Dannye po Stratigrafii Pozdnego Dokembriia Zapada Sibirskoi Platformy i ee Skladchatogo Obramleniia. Akademiia Nauk SSSR, Sibirskoe Otdelenie, Institut Geologii i Geofiziki; Novosibirsk. PUEL, P., C. LARGEAU, and G. GIRAUD. 1987. Occurrence of a resistant biopolymer in the outer walls of the parasitic alga Prototheca wickerhamii (Chlorococcales): ultrastructural and chemical studies. Journal of Phycology 23:649-656. REITLINGER, E. A. 1948. Kembriiskie foraminifery Yakutii. Biulleten Moskovskogo Obshestva Ispytatelei Prirody (Novaia Seriia) 23(2):77-84. REVSBECH, N. P., B. B. JORGENSEN, and T. H. BLACKBURN. 1979. Oxygen in the sea bottom measured with a microelectrode. Science 207:1355- 1356. REX, G. 1983. The compression state of preservation of Carboniferous lepidodendrid leaves. Review of Palaeobotany and Palynology 39:65-85. REX, G. M. and W. G. CHALONER. 1983. The experimental formation of plant compression fossils. Palaeontology 26:231-252. 218

ROUND, F. E. 1963. The taxonomy of the Chlorophyta. British Phycological Bulletin 2:224-235. RUNNEGAR, B. 1982. A molecular-clock date for the origin of the animal phyla. Lethaia 15:199-205. SCHENFIL, V. Yu. 1980. Obruchevelly v Rifeiskikh otlozheniiakh Yeniseiskogo kriazha. Doklady Akademii Nauk SSSR 254:993-995. SCHENFIL, V. Yu. 1983. Vodorodi v Dokembriiskikh otlozheniiakh Vostochnoi Sibiri. Doklady Akademii Nauk SSSR 269:471-473. SCHEPELEVA, E. D. 1960. Nakhodki dnezelenykh vodoroslei v nizhnekembriiskikh otlozheniiakh Leningradskoi oblasti. Vsesoiuznyi Neftianoi Nauchno-Issledovatel'skii Geologo-Razvedochnyi Institut (VNIGRI), Trudy 27:170-172. SCHOPF, J. W. 1968. Microflora of the Bitter Springs Formation, late Precambrian, central Australia. Journal of Paleontology 42:651-688. SCHOPF, J. W. 1992. Proterozoic prokaryotes: affinities, geologic distribution, and evolutionary trends. In J. W. Schopf and C. Klein (eds.). The Proterozoic Biosphere: A Multidisdplinary Study. Cambridge University Press; Cambridge. SCHOPF, J. W. and E. S. BARGHOORN. 1969. Microorganisms from the late Precambrian of South Australia. Journal of Paleontology 43:111-118. SCHOPF, J. W. and J. M. BLACIC. 1971. New microorganisms from the Bitter Springs Formation (late Precambrian) of the north-central Amadeus Basin, Australia. Journal of Paleontology 45:925-960. SEARS, J. R. and S. H. BRAWLEY. 1982. Smithsoniella gen. nov., a possible evolutionary link between the multicellular and siphonous habits in the Ulvophyceae, Chlorophyta. American Journal of Botany 69:1450-1461. 219

SEILACHER, A., W.-E. REIF, and F. WESTPHAL. 1985. Sedimentological, ecological and temporal patterns of fossil Lagerstatten. Philosophical Transactions of the Royal Society of London B 311:5-23. SERGEEV, V. N. 1991. Okremnennye mikrofossilii dokembriia Urala i Kazakhstana i ikh biostratigraficheskie vozmozhnosti. Izvestiia Akademii Nauk SSSR, Seriia Geologicheskaia 1991(11):87-97. SERGEEV, V. N. and I. N. KRYLOV. 1986. Mikrofossilii Miniaarskoi svity Urala. Paleontologicheskii Zhumal 1986(l):84-95. SHARP, L. W. 1934. Introduction to Cytology, 3rd edition. McGraw-Hill; New York. 567 pp. SHUKOVSKY, E. S. and L. N. HALFEN. 1976. Cellular differentiation of terminal regions of trichomes of Oscillatoria princeps (Cyanophyceae). Journal of Phycology 12:336-342. SLANSKY, M. 1986. Geology of Sedimentary Phosphates. Elsevier; New York. 210 pp. SOGIN, M. L. 1989. Evolution of eukaryotic microorganisms and their small subunit ribosomal RNAs. American Zoologist 29:487-499. SOKOLOV, B. S. and IWANOWSKI (eds.). 1985. Vendskaia Sistema; Volume 1. Nauka; Moscow. 221 pp. SÏANIER, R. Y., W. R. SISTROM, T. A. HANSEN, B. A. WHITTON, R. W. CASTENHOLZ, N. PFENNIG, V. N. GORLENKO, E. N. KONDRATIEVA, K. E. EIMHJELLEN, R. WHTTTENBURY, R. L. GHERNA, and H. G. TRÜPER. 1978. Proposal to place nomenclature of the Cyanobacteria (blue-green algae) under the rules of the International Code of Nomenclature of Bacteria. International Journal of Systematic Bacteriology 28:335-336. 220

STAPLIN, F. L., J. JANSONIUS, and S. A. J. POCOCK. 1965. Evaluation of some acritarchous hystrichosphere genera. Neus Jahrbuch für Geologie und Palâontologie, Abhandlungen 123:167-201. STROTHER, P. K., A. H. KNOLL, and E. S. BARGHOORN. 1983. Micro­ organisms from the late Precambrian Narssârssuk Formation, north­ western Greenland. Palaeontology 26:1-32. SUN W. 1987a. Palaeontology and biostratigraphy of late Precambrian macroscopic colonial algae: Chuaria Walcott and Tawuia Hofmann. Palaeontographica Abteilung B 203:109-134. SUN W. 1987b. Discussions on the age of the Liulaobei Formation. Precambrian Research 36:349-352. SUN W., WANG G., and ZHOU B. 1986. Macroscopic worm-like body fossils from the upper Precambrian (900-700 Ma), Huainan district, Anhui, China and their stratigraphie and evolutionary significance. Precambrian Research 31:377-403. TAPP AN, H. 1980. The Paleobiology of Plant Protists. W. H. Freeman; San Francisco. 1028 pp. TEGELAAR, E. W., J. W. de LEEUW, S. DERENNE, and C. LARGEAU. 1989. A reappraisal of kerogen formation. Geochimica et Cosmochimica Acta 53:3103-3106. TEHLER, A. 1988. A cladistic outline of the eumycota. Cladistics 4:227-277. THENG, B. K. G. 1979. Formation and Properties of Clay-Polymer Complexes. Elsevier; Amsterdam. 362 pp. THURET, G. 1875. Essai de classification des Nostochinées. Annales Sciences Naturelle (Botanique) 6:372-382. TIMOFEEV, B. V. 1966. MikropaleoHtologicheskoe Issledovanie Drevnikh S vit. Nauka; Moscow-Leningrad. 147 pp. 221

TIMOFEEV, B. V. 1969. Sferomorfidy Proterozoia. Nauka; Leningrad. 145 pp. TIMOFEEV, B. V. 1970. Sphaeromorphida géants dans le Précambrien avancé. Review of Palaeobotany and Palynology 10:157-160. TIMOFEEV, B. V., T. N. HERMANN, and N. S. MIKHAILOVA. 1976. Mikrofitofossilii Dokembriia, Kembriia i Ordovika. Nauka; Leningrad. 106 pp. TIMOFEEV, B. V. and T. N. HERMANN. 1979. Dokembriiskaia mikrobiota Lakhandinskoi svity. Pp. 137-147. In B. S. Sokolov (ed.), Paleontologiia Dokembriia i Rannego Kembriia. Nauka; Leningrad. TYNNI, R. and J. DONNER. 1980. A microfossil and sedimentation study of the late Precambrian formation of Hailuoto, Finland. Geological Survey of Finland, Bulletin 311. 27 pp. VAN DEN HOEK. 1984. The systematics of the Cladophorales. Pp. 157-178. In D. E. G. Irvine and D. M. John (eds.), Systematics of the Green Algae. Academic Press; London. VIDAL, G. 1976. Late Precambrian microfossils from the Visingso Beds in southern Sweden. Fossils and Strata 9. 57 pp. VIDAL, G. and T. D. FORD. 1985. Microbiotas from the late Proterozoic Chuar Group (northern Arizona) and Uinta Mountain Group (Utah) and their chronostratigraphic implications. Precambrian Research 28:349-389. WALCOTT, C. D. 1899. Pre-Cambrian fossiliferous formations. Geological Society of America, Bulletin 10:199-244. WALCOTT, C. D. 1919. Middle Cambrian algae. Cambrian Geology and Paleontology IV. Smithsonian Miscellaneous Collections 67:217-260. 222

WALTER, M. R., DU RULIN, and R. J. HORODYSKI. 1990. Coiled carbonaceous megafossils from the Middle Proterozoic of Jixian (Tianjin) and Montana. American Journal of Science 290-A:133-148. WALTER, M. R., J. H. OEHLER, and D. Z. OEHLER. 1976. Megascopic algae 1300 million years old from the Belt Supergroup, Montana: a reinterpretation of Walcott's Helminthoidichnites. Journal of Paleontology 50:872-881. WANG P., ZHANG X., and GUO R. 1983. The Sinian microfossils from Jinning, Yunnan, south west China. Precambrian Research 23:133-175. WANG G. 1982. Late Precambrian Annelida and Pogonophora from the Huainan of Anhui Province. Tianjin Institute of Geology and Mineral Resources, Bulletin 6:9-22. WANG G. (ed.). 1984. Research on the Upper Precambrian of Northern Jiangsu and Anhui Provinces. Anhui Science and Technology Publishing House. 209 pp. WEISS, A. F. 1984. Mikrofossilii iz verkhnego rifeia Turukhanskogo raiona. Paleontologicheskii Zhurnal 1984(2):102-108. WETTSTEIN, R. 1924. Handbuch der Systematischen Botanik, Band 1. Franz Deuticke; Leipzig. 467 pp. WIEBE, W. J., W. M. SHELDON, Jr., and L. R. POMEROY. 1992. Bacterial growth in the cold: evidence for an enhanced substrate requirement. Applied and Environmental Microbiology 58:359-364. WILSON, C. B. 1961. The upper middle Hecla Hoek rocks of Ny Friesland, Spitsbergen. Geological Magazine 98:89-116. WIMAN, C. 1894. Paleontologische notizen. 1 und 2. Bulletin of the Geological Institution of the University of Upsala 2:109-117. 223

WOESE, C. and G. FOX. 1977. Phylogenetic structure of the prokaryotic domain. Proceedings, National Academy of Science, U.S.A. 74:5088-5090. WOESE, C., O. KANDLER, and M. L. WHEELIS. 1990. Towards a natural system of organisms: Proposal for the domains Archaea, Bacteria, and Eucarya. Proceedings, National Academy of Science, U.S.A. 87:4576^579. XING Y., DUAN C., LIANG Y. GAG R. et al. 1985. Late Precambrian Palaeontology of China. People's Republic of China, Ministry of Geology and Mineral Resources, Geological Memoirs, Series 2, Number 2. Geological Publishing House; Beijing. 243 pp. XU Z. 1984. Investigation on the procaryotic microfossils from the Gaoyuzhuang Formation, Jixian, north China. Acta Botanica Sinica 26:216-222 and 26:312-319. YAKSCHIN, M. S. and V. A. LUCHININA. 1981. Novye dannye po iskopaemym vodorosliam semeistva Oscillatoriaceae (Kirchn.) Elenkin. Pp. 28-34. In N. P. Meshkova and I. V. Nikolaeva (eds.), Pogranichnye Otlozheniia Dokembriia i Kembriia Sibirskoi Platformy. Akademiia Nauk SSSR, Sibirskoe Otdelenie, Trudy Institute Geologii i Geofiziki (Bypusk 475) Nauka; Novosibirsk. YIN C. 1985. Micropalaeoflora from the late Precambrian in Huainan region of Anhui Province and its stratigraphie significance. Professional Papers of Stratigraphy and Paleontology 12:97-119. YIN C. 1990. Spinose acritarchs from the Toushantou Formation in the Yangtze Gorges and its geological significance. Acta Micropalaeontologica Sinica 7:265-270 224

YIN L. 1979. Microflora from the Anshan Group and the Liaohe Group in E. Liaoning with its stratigraphie significance. Pp. 39-58. In: Selected Works For a Scientific Symposium on Iron-Geology and Palaeontology. Science Press; Beijing. YIN L. 1985a. Microfossils of the Doushantuo Formation in the Yangtze Gorge district, western Hubei. Palaeontologica Cathayana 2:229-249. YIN L. 1985b. A glimpse of Precambrian microfossils. In: Selected Papers from the 1th National Fossil Algal Symposium. Pp. 179-186. Geological Publishing House; Beijing. YIN L. 1987. Microbiota of latest Precambrian sequences in China. Stratigraphy and Palaeontology of Systemic Boundaries in China; Precambrian- Cambrian Boundary 1:415-494. YIN L. and LI Z. 1978. Pre-Cambrian microfloras of southwest China, with reference to their stratigraphical significance. Memoirs of the Nanjing Institute of Geology and Palaeontology 10:41-102. ZALESSKY, M. D. 1926. Premières observations microscopiques sur le schiste bitumineux du Vogien inférieur. Société Géologique du Nord, Annales 51:65-104. ZANG W. 1988. Late Proterozoic - Early Cambrian Microfossils and Biostratigraphy in China and Australia. Unpublished Ph.D thesis, Australian National University; Canberra, Australia. 432 pp. ZANG W. and M. R. WALTER. 1989. Latest Proterozoic plankton from the Amadeus Basin in central Australia. Nature 337:642-645. ZECHMAN, F. W., E. C. THERIOT, E. A. ZIMMER, and R. L. CHAPMAN. 1990. Phylogeny of the Ulvophyceae (Chlorophyta): cladistic analysis of nuclear encoded rRNA sequence data. Journal of Phycology 26:700-710. 225

ZELIBOR, J. L., Jr., L. ROMANKIW, P. G. HATCHER, and R. R. COLWELL. 1988. Comparative analysis of the chemical composition of mixed and pure cultures of green algae and their decomposed residues by ’^C nuclear magnetic resonance spectroscopy. Applied and Environmental Microbiology 54:1051-1060. ZHANG P. 1981. Microfossil blue-green algae from the Wumishan Formation of Jixian. Acta Geologica Sinica 55(4):253-257. ZHANG P. 1982. Microfossils from the Wumishan Formation of Jixian County. Acta Geologica Sinica 56(1):34-41. ZHANG P. 1987. Microfossils from the middle Proterozoic in Kuancheng County, eastern Hebei, China. Professional Papers of Stratigraphy and Paleontology 17:263-276. ZHANG P. and GU S. 1986. Microfossils from the Wumishan Formation of the Jixian System in the Ming Tombs, Beijing, China. Acta Geologica 60(4):13-22. ZHANG P. and Y AN X. 1984. Microfossils from the Gaoyuzhuang Formation in Laishui County, Hebei, China. Acta Geologica Sinica 58(3):196-204. ZHANG P., ZHU M., and SONG W. 1989. Middle Proterozoic (1200-1400 Ma) microfossils from the Western Hills near Beijing, China. Canadian Journal of Earth Sciences 26:322-328. ZHANG R., FENG S., MA G., XU G., and VAN D. 1991. Late Precambrian macroscopic fossil algae from Hainan Island. Acta Palaeontologica Sinica 30:115-125. ZHANG Y. 1981. Proterozoic stromatolite microfloras of the Gaoyuzhuang Formation (early Sinian: Riphean), Hebei, China. Journal of Paleontology 55:485-506. 226

ZHANG Y. 1985. Stromatolitic microbiota from the Middle Proterozoic Wumishan Formation (Jixian Group) of the Ming Tombs, Beijing, China. Precambrian Research 30:277-302. ZHANG Y. 1989. Multicellular thallophytes with differentiated tissues from Late Proterozoic phosphate rocks of South China. Lethaia 22:113-132. ZHANG Z. 1981. A new Osdllatoriaceae-like filamentous microfossil from the Sinian (late Precambrian) of western Hubei Province, China. Geological Magazine 118:201-206. ZHANG Z. 1982. Upper Proterozoic microfossils from the Summer Isles, N.W. Scotland. Palaeontology 25:443-460. ZHANG Z. 1984. A new microphytoplankton species from the Sinian of western Hubei Province. Acta Botanica Sinica 26:94-98. ZHANG Z. 1986. New material of filamentous fossil cyanophytes from the Doushantuo Formation (Late Sinian) in the eastern Yangtze Gorge. Scientia Geologica Sinica 1986(1):30-37. ZHENG W. 1980. A new occurrence of fossil group of Chuaria from the Sinian System in north Anhui and its geological meaning. Chinese Academy of Geological Sciences (Tianjin Institute of Geology and Mineral Resources), Series 6,1:49-69. ZHU S. 1982. A preliminary study of fossil micro-organisms from stromatolites in the lower part of Sinian suberathem, Yanshan Range. Bulletin of the Tianjin Institute of Geological and Mineralogical Resources 5:1-26. ZHU S., WANG Y., and ZHANG L. 1984. Formation of the Kaiyang phosphorites in China as related to ancient organisms. Symposium of 5th International Field Workshop and Seminar on Phosphorite. Pp. 165- 193. 227

APPENDIX A - LIST OF FOSSIL TAXA OCCURRING IN THE SVANBERGFJELLET FORMATION

Multicellulai Eukaryotes assignable to modem taxa PALAEASTRUM DYPTOCRANUM n. gen., n. sp ...... page 47 PROTEROCLADUS MAJOR n. gen., n. sp ...... 50 PROTEROCLADUS MINOR n. gen., n. sp ...... 51 SEGREGOCLADUS HERMANNAE n. gen., n. sp ...... 53

Multicellular Eukaryotes - Incertae Sedis PSEUDOTAWUIA BIRENIFERA n. gen., n. sp ...... 54 TAWUIA DALENSIS Hofmann, 1979 ...... 57 VALKYRIA BOREALIS n. gen., n. sp ...... 59

Unicellular Eukaryotes - Incertae Sedis (= Acritarcha) CEREBROSPHAERA BUICKE n. gen., n. sp ...... 63 CHUARIA CIRCULARIS Walcott, 1899 ...... 67 COMASPHAERIDIUM sp...... 70 CYMATIOSPHAEROIDES KULLINGn Knoll, 1984 ...... 70 DICTYOTIDIUM COLANDRUM n. sp ...... 72 GERMINOSPHAERA BISPINOSA Mikhailova, 1986 ...... 74 GERMINOSPHAERA FIBRILLA (Ouyang, Yin and Li, 1974), n. comb 75

GERMINOSPHAERA JANKAUSKASn n. sp ...... 76 GGNIOSFHAERIDIUivl sp...... 77 GORGONISPHAERIDIUM sp...... 78 LEIOSPHAERIDIA CRASSA (Naumova, 1949) Jankauskas, 1989 ...... 79 228

LEIOSPHAERIDIA JACUTICA (Timofeev, 1966) Mikhailova and

Jankauskas, 1989 ...... 80 LEIOSPHAERIDIA TENUISSIMA Eisenack, 1958 ...... 80 LEIOSPHAERIDIA WIMANII (Brotzen, 1941) n. comb ...... 81 LEIOSPHAERIDIA spp...... 83 OSCULOSPHAERA HYALINA n. gen., n. sp ...... 85 PTEROSPERMOPSIMORPHA PILEIFORMIS Timofeev, 1966 ...... 86 TRACHYHYSTRICHOSPHAERA AIMIKA Hermann, 1976 ...... 89 TRACHYHYSTRICHOSPHAERA POLARIS n. sp ...... 91

Unambiguous cyanobacteria, assignable to modem taxa GLOEODINIOPSIS LAMELLOSA Schopf, 1968 ...... 93 EOENTOPHYSALIS BELCHERENSIS Hofmann, 1976 ...... 94 EOENTOPHYSALIS CROXFORDH (Muir, 1976) n. comb ...... 94 POLYBESSURUS BIPARTTTUS Fairchild, 1975, ex Green et al., 1987 ...... 96 OBRUCHEVELLA BLANDHA Schenfil, 1980 ...... 97

Probable cyanobacteria, assignedC?) to modem taxa EOSYNECHOCOCCUS MOOREI Hofmann, 1976 ...... 99 SPHAEROPHYCUS PARVUM Schopf, 1968 ...... 100 SPHAEROPHYCUS spp ...... 100 CEPHALONYX GEMINATUS (Jankauskas, 1980) n. comb ...... 102 CYANONEMAsp...... 103 OSCILLATORIOPSIS OBTUSA Schopf, 1968 ...... 108 OSCILLATORIOPSIS AMADEA (Schopf and Blade, 1971), n. comb 110 OSCILLATORIOPSIS LONGA Timofeev and Hermann, 1979 ...... I l l PALAEOLYNGBYA CATENATA Hermann, 1974 ...... 114 229

PALAEOLYNGBYA HEBEIENSIS Zhang and Yan, 1984 ...... 114 RUCOSOOPSIS TENUIS Timofeev and Hermann, 1979 ...... 117 SIPHONOPHYCUS THULENEMUM n. sp ...... 119 SIPHONOPHYCUS SEPTATUM (Schopf, 1968) Knoll et al., 1991 ...... 119 SIPHONOPHYCUS ROBUSTUM (Schopf, 1968) Knoll et al., 1991 ...... 121 SIPHONOPHYCUS TYPICUM (Hermann, 1974) n. comb ...... 122 SIPHONOPHYCUS KESTRON Schopf, 1968 ...... 124 SIPHONOPHYCUS GRACILE (Pjatiletov, 1980) n. comb ...... 125 TORTUNEMA WERNADSKII (Schepeleva, 1960) n. comb ...... 128 VETERONOSTOCALE AMOENUM Schopf and Blade, 1971 ...... 130

Incertae Sedis BRACHYPLECANON KHANDANUM Lo, 1980 ...... 131 CHOLOROCLOEAOPSIS ZAIRENSIS Maithy, 1975 ...... 132 DIGITUS ADUMBRATUS n. sp...... 133 MYXOCOCCOIDES MINOR Schopf, 1968 ...... 133 MYXOCOCCOIDES CANTABRICIENSIS Knoll, 1982 ...... 134 MYXOCOCCOIDES spp...... 134 OSTIANA MICROCYSTIS Hermann, 1976 ...... 135 PSEUDODENDRON ANTERIDIUM n. gen., n. sp ...... 137 PSEUDODENDRON POLYTAENIUM n. sp ...... 139 Filament Bearing Body ...... 140 Sub-vertical branched tubes ...... 140 Sub-vertical branched(?) filaments in apatite ...... 141 Filaments mineralized by goethite(?) ...... 141 230

III. PALEOENVIRONMENTAL DISTRIBUTION OF PROTEROZOIC MICROFOSSILS, WITH AN EXAMPLE FROM THE AGU BAY FORMATION, BAFFIN ISLAND

ABSTRACT

A shale sample from the Black Shale Member of the ca. 1250 Ma old Agu Bay Formation, Fury and Hecla Group, northwest Baffin Island contains abundant, well preserved microfossils. The assemblage is dominated by small leiosphaerid acritarchs of which ca. 15% show median split release structures. Colonial unicells and larger spheroidal acritarchs are uncommon, and filamentous microfossils extremely rare. A single specimen of the highly distinctive acritarch Valeria lophostriata extends the geographic range of this taxon and, in concert with geochronologic and chemostratigraphic data, constrains the timing of Fury and Hecla deposition to the early Late Riphean. The overall habit of the Black Shale Member assemblage, including the even bedding plane distribution of fossils, supports the sedimentological and stratigraphie evidence for a mid to outer shelf depositional environment. Incorporating these findings into a review of Proterozoic shale-hosted microfossils reveals a distinct depth/diversity trend in assemblage structure such that five broad zones can be recognized extending from restricted nearshore to basinal environments. The depositional environments of the Neoproterozoic Mineral Fork Formation, Utah, and the terminal Proterozoic Pertatataka Formation, Australia, are reconsidered using this paleoecological measure of paleoenvironment. 231

INTRODUCTION

Like their Phanerozoic counterparts, Proterozoic microfossils are now being usefully applied to questions of biostratigraphy (Vidal and Knoll, 1983; Jankauskas, 1989; Knoll and Butterfield, 1989), paleoecology, and paleoenvironmental analysis (Ivanovskaya and Timofeev, 1971; Knoll, 1984; Vidal and Nystuen, 1990; Knoll et al., 1991), as well as illuminating patterns of biotic evolution. The degree to which this is possible depends on a number of factors, including the quantity and quality of fossil preservation and a detailed understanding of their distribution in time and space. As organic-walled microfossils in younger rocks show pronounced paleoenvironmental partitioning (Staplin, 1961; Jacobson, 1979; Dale, 1983; Doming, 1981) so too are environmental effects to be expected in the Proterozoic. These must of course be factored out of any biostratigraphic or evolutionary calculations, but a systematic documentation of such trends offers valuable insights into depositional environment, especially where other indicators are absent or ambiguous. A shale sample from the mid to outer shelf Black Shale Member of the Proterozoic Agu Bay Formation, Fury and Hecla Group, Baffin Island, Canada contains an abundant, well-preserved microfossil assemblage. Detailed analysis of both sedimentary facies and fossil diversity/distribution permits its incorporation into a broad characterization of depth-dependent trends in Proterozoic microfossils. Individual components of the Agu Bay biota also offer useful physiologic, taxonomic and, in one instance, biostratigraphic data. 232

GEOLOGICAL SETTING

The Fury and Hecla Group is a ca. 6000 m thick, little-altered Proterozoic sedimentary sequence overlying Hudsonian basement gneisses on northern Baffin Island (Fig. 1) (Chandler, 1988) and adjacent Melville Peninsula (Schau and Beckett, 1986). The predominantly siliciclastic succession is divided into six formations (Fig. 2) with changes in facies and unit thicknesses occurring along a strike length of ca. 115 km: the basal Nyeboe Formation (500 m) consists largely of alluvial red quartz arenite with minor marine stromatolitic dolomite and red amygdaloidal basalt flows. It is overlain by the shallow, tidal-dominated Sikosak Formation (150+ m) and the apparently subaerial basalts of the Hansen Formation (0-30 m thick); Schau (pers. comm., 1991), however, considers the Hansen Formation a sill. The succeeding Agu Bay Formation is up to ca. 600 m thick and comprises three members. The basal 0-10 m thick Dolomite Member includes stromatolitic reefs, oncolites, oolites and mudcracks, indicative of shallow- subtidal to supratidal environments. The overlying, and fossiliferous. Black Shale Member (Fig. 2) represents a marked shift to deeper water deposition. It is up to 75 meters thick and is dominated by friable, often rhythmically interbedded black shale and siltstone (see Chandler, 1988: fig. lid). Upward- coarsening and variously cross-bedded quartz arenite beds up to 2 m thick make up several per cent of the section and contain clasts that may have derived from the underlying Dolomite member. In the absence of any shallow water features, the overall thickness, textures and lateral continuity of the Black Shale Member point to its deposition on a marine shelf largely below storm wave base and across which sand-bars occasionally migrated (Chandler, 1988). The upper 500 m of the Agu Bay Formation, the Redbed Member, 233 marks a return to shallow water conditions - an abundantly mudcracked, shoaling coastal unit. Fury and Hecla deposition is concluded with the overlying Whyte Inlet (0-3000 m) and Autridge (2000 m) formations, together interpreted as an upward-deepening marine shelf succession (Chandler, 1988). The Fury and Hecla Group is considered broadly correlative to several other Proterozoic sequences in the eastern arctic of North America. Good lithostratigraphic comparisons have been drawn with the Eqalulik Group of the northernmost Baffin Island Bylot Supergroup, and -with the Wolstenholme and Dun das formations of the Thule Group, northwestern Greenland (Jackson and lannelli, 1981; Jackson, 1986; Chandler, 1988). Thick carbonate units overlie these siliciclastic rocks in both the Bylot (Uluksan Group) and Thule (Narrsârrsuk Formation) sections, but not in the Fury and Hecla.

Geochronology

The Fury and Hecla Group contains subaerial basalt flows in the Nyeboe and, probably, Hansen formations. The two igneous units are similar lithologically and occur stratigraphically within 200 m of one another, suggesting a broadly equivalent generation. The group is topped by the Dybbol Sill and cut by mafic dykes. Chandler (1988, fig. 17) correlated the Hansen Formation with the

Nauyat Formation basalts at the base of the Bylot Supergroup. Potassium- Argon dates from the Nauyat volcanics range from 762 ± 26 to 1221 ± 31 Ma (Jackson and lannelli, 1981) while the Hansen basalts have yielded K-Ar ages of 1089 ± 32,1117 ± 40 and 1121 ± 33 Ma (Chandler and Stevens, 1981). Fahrig et al. (1981) placed the paleomagnetic pole for the Nauyat Formation at 1220 Ma. The Dybbol Sill gives K-Ar ages of 746 ± 87 Ma and 716 ± 166 Ma, 234 while the dykes cutting the Fury and Hecla Group have K-Ar ages of 643 ± 27 Ma and 631 ± 43 Ma (Chandler and Stevens, 1981). The ages from the mafic igneous rocks in the Fury and Hecla Group coincide closely with the widespread Mackenzie and Franklin dyke swarms of the Canadian arctic (Chandler and Stevens, 1981). LeCheminant and Heaman (1989) have dated the Mackenzie event at 1267 ± 2 Ma, and Heaman et al. (1990) the Franklin event at 723 ± 3 Ma, both ages from U-Pb analyses on baddeleyite. Given a short duration for the Mackenzie intrusions (Lecheminant and Heaman, 1989) and assuming a Mackenzie age for the red flows in the Nyeboe Formation, the above data tie Fury and Hecla deposition more closely to the Mackenzie than to the Franklin event, and suggest that the Agu Bay fossils are close to 1250 Ma old. Support for a Mackenzie as opposed to a Franklin age also comes from chemostratigraphy. As the Fury and Hecla Group is dominantly siliciclastic, secular trends in carbon and strontium isotopes are not readily available. However, isotopic analysis of carbonate rocks from the broadly correlative Uluksan Group (Butterfield, Knoll, and Hayes, unpublished data), and Narrsârrsuk Formation (A. H. Knoll, pers. comm., 1991) yield Ô13C values from -1.0 to +3.2 %o PDB, values substantially lighter than those typical of 700- 800 Ma old (Franklin age) carbonates in the Canadian arctic and elsewhere (Knoll et al., 1986; Asmerom et al., 1991; Kaufman et al., 1992). While not definitive in themselves, these isotopic signatures are more in accord with those of older Proterozoic carbonates such as the unweathered dolomites of the 1100-1200 Ma old Mescal Limestone of Arizona (Beeunas and Knauth, 1985). 235

PALEONTOLOGY

Bedding-parallel thin sections of Fury and Hecla Group shales and siltstones were prepared for preliminary micropaleontological examination. A single green-brown shale (sample L161) from the Black Shale Member of the Agu Bay Formation revealed abundant and very well preserved populations of spheroidal microfossils. Most of the remainder of L161 was similarly prepared, yielding 67 thin-sectioned shale chips with a total area of ca. 40 cm^; ca. 1 cm^ of the sample was processed with hydrofluoric acid. The fossil assemblage is dominated overwhelmingly by small, thin- walled and unomamented spheroidal microfossils (leiosphaerids) which provide little other than size as a distinguishing taxonomic character (Fig. 3). A unimodal size frequency distribution (n = 2292) skewed moderately to the right (Fig. 4A) suggests that the population is monospecific, and a mean diameter of 10.6 pm (s.d. = 3.65 pm) places them within the form taxon Leiosphaeridia minutissima (Jankauskas, 1989). Approximately 15% of the population consistently exhibits medial split release structures (Figs. 3.5-3.10), of which ca. 12% are oriented so that they superficially appear as cellular diads (Figs. 3.11-3.14; note the specimen in 3.10 in which one hemisphere was oriented 'parallel' to bedding, the other 'lateral'); the edges of the separated hemispheres are commonly enrolled. These medially split forms fall entirely within the size range of the entire leiosphaerids, although their mean measured diameter is somewhat larger (% = 17.5 pm, s.d. = 3.03 pm, n = 354; Fig. 4B). As the planar dimensions of split compression fossils are typically exaggerated relative to unsplit material (Harris, 1974), the closely comparable wall structure, bedding-plane distribution (see below), and size of these two forms suggest that they belong to a single population. 236

Other taxa in the Black Shale Member are rare, together comprising less than 1% of the assemblage. These include larger spheroids measuring up to 200 pm diameter and ascribable to the acritarch species Leiosphaeridia crassa (Figs. 5.1, 5.3) and L. iacutica (Fig. 5.2), and infrequent colonial spheroids that resemble Svnsphaeridium (Fig. 5.9) and Satka (Fig. 5. 6). Only two filamentous fossils were encountered and their occurrence as short isolated fragments in thin-section (Fig. 5.8) identifies them as allochthons. Their preservation nevertheless suggests that the general absence of filaments in the Black Shale Member is a reflection of original distribution rather than a taphonomic bias. The most distinctive component of the Black Shale assemblage is a single specimen of the acritarch Valeria (Kildinosphaera) lophostriata (Figs. 5.4- 5.5), characterized by its unique wall sculpture of very fine (sub-micron­ spaced) concentric striae emanating from opposite 'poles'. It is 250 pm in diameter, well within the previously reported 60-700 pm size range for the taxon, but it differs from all other occurrences in having a thin outer envelope (275 pm) surrounding the sculptured spheroid. This, however, is considered an ontogenetic or taphonomic feature rather than a differentiating taxonomic one; its retention argues against the possibility that the specimen was reworked from older sediments. Valeria lophostriata has a worldwide distribution which, in combination with its readily identifiable surface sculpture, recommends it as an important biostratigraphic marker. To date Valeria has been reported from the southern Urals (Jankauskas, 1979), the Siberian Platform (Pjatiletov, 1980; Volkova, 1981), the Kola Peninsula (Liubtsov et al., 1988), both the Barents Sea (Vidal and Siedlecka, 1983) and Tanaljord-Varangerfjord (Vidal, 1981) regions of northern Norway, the Visingso Beds of southern Sweden (Vidal and Siedlecka, 1983), the Chuar and Uinta Mountain Groups of the southwestern United States 237

(Vidal and Ford, 1985), the Thule Group of northwestern Greenland (Dawes and Vidal, 1985), and southern Africa (A. H. Knoll, pers. comm., 1991). In all instances it is temporally restricted to the Late Riphean, and it occurs most characteristically during the first one-third of that period (Liubtsov et al., 1988: table 15; Jankauskas, 1989: table 4). The discovery of Valeria in the Agu Bay Formation corroborates the geochronological evidence for an early Late Riphean deposition of Fury and Hecla sediments (p. 234). Moreover, its similar occurrence in the Wolstenholme and Dundas formations on northwest Greenland (Dawes and Vidal, 1985) substantially reinforces the lithostratigraphic and chemostratigraphic data correlating the Thule Group with the Fury and Hecla Group (pp. 233, 234) (Jackson and lannelli, 1981; Jackson, 1986; Chandler, 1988).

Paleobiology

As organisms, the larger spheroidal microfossils of the Agu Bay Black Shale assemblage can be reasonably identified as some life cycle stage of protistan grade, probably photosynthetic eukaryotes. In contrast, the small simple leiosphaerids that comprise the majority of the Agu Bay fossils are not immediately identifiable at even the kingdom/superkingdom level. The occurrence of medial splits in spheroids has conventionally been taken as evidence of eukaryotic grade (e.g.. Green et al., 1989), but such structures are also found in baeocyte-releasing pleurocapsalean cyanobacteria (Waterbury and Stanier, 1978: fig. 4) and the germinating akinetes of various filamentous cyanobacteria (Nichols and Adams, 1982). The overall habit of these cyanophytes is nevertheless distinct from the Agu Bay fossils: pleurocapsaleans live as attached benthic organisms, often with distinctive 238 patterns of cell division, while akinetes are generally associated with filamentous growth. The medially split, filament-free, and fully planktic (see below) habit of the Agu Bay L. minutissima population supports their interpretation as eukaryotes. Finer biological analysis of these small spheroidal fossils is frustrated by their morphological simplicity. Nonetheless, additional paleobiological information can often be gleaned from the details of fossil distribution and orientation, i.e., preserved 'behavior' (e.g., Butterfield et al., 1988; Green et al., 1988; 1989). Unfortunately, the conventional procedure for studying shale- hosted microfossils, palynological maceration, necessarily destroys such potentially instructive information. The in situ (i.e., thin section) analysis of the Agu Bay fossils allowed a detailed assessment of their original bedding-plane distribution (Fig. 5.7). Five separate counts of ca. 500 leiosphaerids (from five separate thin-sections of sample L161 recording all fossils encountered during a systematic scan) were remarkably consistent (Figs. 4C-4G), and a Kolmogorov-Smimov analysis of the size-frequency distributions indicated that all five samples were drawn from the same population. The abundance of fossils varies somewhat between different laminae but averages ca. 63 mm'^ is.d. = 15 mm'^) (= ca. 1000 mm'^). Within any single lamina, however, fossil distribution is much more uniform. For example, fossils on five randomly chosen 1 mm^ plots of thin-section L161-16C showed an Index of Dispersion (s^/x: 6.2/78.8 = 0.078) significantly less than unity (> 99% confidence), indicating their very even distribution on individual bedding planes (cf., Odum, 1971; Rosenberg, 1974), a distribution that could only have derived from dispersed plankton settling out of a water column (Vidal and Knoll, 1983). The organisms preserved in the Black Shale Member were not only eukaryotic but planktic. 239

These small leiosphaerids are comparable to the degradation-resistant resting/dispersal stages of various algae which often show similar medial-split excystment structures (e.g.. Dale, 1983; Margulis et al., 1988). Alternatively, they might be interpreted simply as the cell walls and cast off cell walls of an actively growing unicellular alga. Cells of the modem green alga Chlorella, for example, fall within the morphological range of the fossil population, and some species develop highly recalcitrant sporopollenin-bearing walls (Atkinson et al., 1972; Rascio et al., 1979). The Chlorella life cycle also involves the production of intracellular autospores which, upon reaching maturity, are released from the mother cell via a medial split (Atkinson et al., 1972: fig. 21). Interestingly, the split edges of a vacated Chlorella mother cell tend to curl in on themselves very much in the manner of the medially split fossils; a small spheroid attached to a larger, medially split leiosphaerid (Fig. 3.11) is possibly interpreted as a daughter autospore. These Agu Bay leiosphaerids may thus as likely represent a population of physiologically active unicellular algae as one of dormant cysts or spores.

PALEOENVIRONMENTAL DISTRIBUTION

Fossil biotas play an important role in determining the paleoenvironment of Phanerozoic sediments (Boucot and Carney, 1981) but similar application in Proterozoic sequences has not been generally exploited. For carbonate facies with their diverse and relatively diagnostic sedimentary structures this may not present much of an impediment, and stromatolitic facies offer paleoenvironmental indicators down to the base of the photic zone (Hoffman, 1976). In contrast, fine-grained siliddastic sediments occur from terrestrial to abyssal environments and, in the absence of biological markers. 240

can often not be constrained to particular paleoenvironmental settings. Microfossils are commonly preserved in Proterozoic shales and a detailed understanding of their spatial distribution offers a potentially valuable measure of paleoenvironment. Just as all shales are not necessarily distal and deep water, all shale- hosted Proterozoic microfossils did not occupy a common ecological niche. Indeed, most such assemblages are reported from relatively shallow water facies (Vidal, 1981a; Vidal and Knoll, 1983). For example, shallow subtidal shales from the 750-850 Ma old Wynniatt Formation on Victoria Island, arctic Canada (Butterfield and Rainbird, 1988), and the marginally younger Svanbergfjellet Formation of Spitsbergen (Butterfield et al., 1988) preserve abundant and diverse microfossils. Unlike the Agu Bay Black Shale material, however, bedding-parallel thin sections of these rocks reveal localized populations and associations of microfossils and abundant, often mat-forming filaments. This patchy and otherwise autochthonous distribution indicates an important contribution by benthic (Rosenberg, 1974), probably photosynthetic organisms. Shallow-water shale biotas of Proterozoic age tend to be taxonomically diverse and typically include a variety of large (50-3000 pm diameter) spheroidal acritarchs, set amongst an almost ubiquitous background of small leiosphaerids. Strongly ornamented Proterozoic acritarchs were dominantly, if not exclusively shallow water inhabitants (Knoll and Butterfield, 1989; Vidal, 1990; Jenkins et al., 1992). Perhaps the most regularly recurring constituents of these shallow water assemblages, however, are filamentous microfossils comparable to modem Lyngbva /Phormidium/Plectonema-tvpe (LPP-type) cyanobacteria. These are best known as the unambiguously autochthonous and photosynthetic builders of stromatolites, but they also commonly occur in 241

fine-grained siliciclastic environments where they likewise constructed extensive microbial mats (Hermann, 1974; Peat et al., 1978; Schieber, 1986; Butterfield et al., 1988; Chapter II). That they were benthic and photosynthetic is supported by their usual occurrence in demonstrably shallow-water sediments, and a negative correlation between filament abundance and depth (Horodyski, 1980; Knoll and Swett, 1985). It is of course conceivable that some filaments were planktic, however the general absence of LPP-type filaments in offshore environments, both in the modem oceans (Fogg, 1982) and in Phanerozoic acritarch-preserving sediments, argues against such an interpretation. Alternatively, fossil filaments may simply be transported and redeposited shallow-water benthos, as appears likely for the very rare filament fragments in the Agu Bay shale. Nonetheless, their consistent appearance in shallow water sediments and conspicuous absence from deeper water environments reflects at least a broad paleoenvironmental partitioning. In contrast, distal shelf to slope microbiotas in the Proterozoic are characterized by taxonomically depauperate assemblages of small but often very abundant spheroidal microfossils, generally assignable to the genus Sphaerocongreeus (= Bavlinella): both filaments and larger acritarchs are typically absent (e.g., Moorman, 1974; Mansuy and Vidal, 1983; Vidal and Siedlecka, 1983; Palacios, 1989; Vidal and Nystuen, 1990). Vidal and Nystuen (1990) further suggest that probably no acritarch-produdng eukaryotic plankton occupied these distal, open ocean environments during the Proterozoic. The facies analysis of the Agu Bay Black Shale Member (Chandler, 1988) places it dearly on the mid to outer shelf, intermediate between the two environmental extremes mentioned above. Interestingly, the Black Shale Member fossils also show an intermediate character. Deep water aspects 242 include the low diversity, high abundance and entirely allochthonous nature of the assemblage, while shallower water influence is reflected in the dominance of eukaryotic organisms and the presence of at least a few larger acritarchs and (rare) filament fragments. Thus, the Black Shale Member fossil assemblage provides an independent, yet fully concurrent measure of the mid to outer shelf paleoenvironment. More importantly, it appears that the overall paleontological character of a Proterozoic shale, particularly as it may be corroborated by a number of distributional and dominance criteria, can be used in assessing depositional environments. The paleontological, and hence paleoenvironmental, onshore-offshore gradient in late Proterozoic proximal carbonate sequences has been discussed in detail by Knoll (1984) and Knoll et al. (1991). Here the highest diversity was met in the most 'distal' open-water conditions in apparent reversal of the trend observed in siliciclastic environments. This, however, derives simply from the differences of scale. Most shallow-water shales can be considered indicative of relatively unrestricted, usually subtidal settings and will therefore correspond broadly with the open-water, but nevertheless shallow-water facies of the carbonate studies; indeed, the two shallow open-water environments share a significant number of both benthic microbial and acritarch taxa (Knoll, 1984; Butterfield et al., 1988). Similarly, restricted intertidal to supratidal shales are likely to preserve low diversity assemblages of autochthonous (patchy distribution) prokaryotes, comparable to the microbial mat biotas of restricted carbonate environments. Thus, shale-hosted fossil assemblages can be applied to paleoenvironmental analysis in the same manner as those in carbonate facies. In addition, since fossiliferous shales are not limited to shallow-water, peri-stromatolitic facies, they offer an excellent opportunity to extend the paleontological characterization of Proterozoic basins out across the shelf and 243

slope. Systematic trends in Proterozoic microfossil occurrence and distribution can be usefully categorized into five broad environmental zones (Fig. 6):

(1) Near shore restricted, lagoonal to supratidal: fossils dominantly autochthonous (i.e., patchy bedding plane distribution); abundant filaments and/or small spheroids; dominantly prokaryotic; low diversity. The microfossils of this facies are best known from silicified microbial mat material of carbonate environments although they might well occur in intertidal or lagoonal shales.

(2) Near shore unrestricted, shallow subtidal: fossils both autochthonous and allochthonous (planktic); abundant filaments; moderate to high diversity of large (>50 pm diameter) acritarchs; if of appropriate age, large process-bearing acritarchs. Examples; McMinn Formation (Peat et al., 1978); Wynniatt Formation (Butterfield and Rainbird, 1988); Svanbergfjellet Formation (Butterfield et al., 1988); Rodda Beds - Murnaroo-1 drillhole (Jenkins et al., 1992). Comparable distribution of large complex acritarchs is found in open- water carbonate environments where they likewise co-occur with benthic

microbial mats (e.g.. Knoll, 1984; Awramik et al., 1985; Yin, 1990; Allison and Awramik, 1989; Knoll, 1992).

(3) Mid shelf, moderate depth (lower photic zone): fossils dominantly allochthonous (i.e., even bedding plane distribution); moderate diversity of spheroidal acritarchs but no substantially ornamented forms; sparse filaments. Examples: Chamberlain Shale (Horodyski, 1980); Clasgowbreen Formation (Knoll and Swett, 1985); Arcoona Quartzite Member of the Tent Hill Formation (Damassa and Knoll, 1986; note, this is a lateral equivalent of the Pertatataka Formation discussed below (p. 246)). 244

(4) Mid to outer shelf, deep water (sub photic zone): fossils entirely allochthonous; low taxonomic diversity, dominated by moderate to small (<50 pm) diameter spheroids; eukaryotes common; filaments rare (wash-ins?) to absent. Examples: Biri Formation (Vidal and Nystuen, 1990); Black Shale Member of the Agu Bay Formation (present study).

(5) Slope to basinal, turbidite dominated environments, deep water: fossils entirely allochthonous; often very abundant; very low taxonomic diversity; unambiguous eukaryotes absent; filaments absent. Assemblages are typically monospecific, composed exclusively of planktic prokaryotes with growth series similar to Sphaerocongregus Moorman. Examples: Hector Formation (Moorman, 1974); the Brioverian of France (Mansuy and Vidal, 1983); Bottrum Formation (Vidal and Nystuen, 1990).

Microfossil distribution as a means of deciphering paleoenvironment is subject to a number of caveats: 1) superior fossil preservation is necessary for any meaningful analysis; 2) local circumstances may impinge upon the generalized pattern, particularly in nearshore settings where sediment input and salinity are likely to be variable; 3) secular changes in fossil distribution must be accommodated; 4) as benthic organisms are not exclusively photosynthetic, bedding-plane distribution alone offers only a one-tailed test for allochthony. Clearly no one feature is sufficient for determining paleoecology (and thereby paleoenvironment); rather, it derives from an overall analysis of a fossil assemblage, incorporating all available paleontological and, of course, sedimentological data. Because the scheme presented here is based on the extremely conservative prokaryotic components of a biota and broad grades of eukaryotic organization (thus eliminating most 245 evolutionary effects), it should be applicable to most of the middle and late Proterozoic, limited on the one hand by the evolution of eukaryotic plankton, and on the other by the decline of organic-walled fossil preservation associated with the appearance of bioturbating animals. Despite its obvious simplification, the distribution shown in Figure 6 accords with a majority of Proterozoic microfossil assemblages. If the above characterization is generally applicable then it should help to resolve the paleoenvironmental setting of Proterozoic sequences in which the sedimentological context is ambiguous or in dispute. For example, the Neoproterozoic Mineral Fork Formation, Utah has been interpreted both as a terrestrial to shallow marine glacigene unit (Crittenden et al., 1952; Knoll et al., 1981), and as a distal shelf sequence involving submarine mud and debris flows (Condie, 1967; Schermerhom, 1974). Relatively deep-water glacial- marine conditions are now thought to have prevailed during most of Mineral Fork deposition, however a detailed accounting of the constituent facies remains wanting (Chistie-Blick, 1983). Knoll et al. (1981) reported a microfossil assemblage from fine-grained facies of the Mineral Fork Formation that appears to be entirely planktic (thin section study), includes no fossil filaments, and is dominated by abundant small spheroids. To a remarkable degree the biota replicates the form and distribution of Sphaerocongregus in the distal, deep-water Hector Formation (Moorman, 1974). Moreover, it differs conspicuously from other Neoproterozoic tillite-assodated microbiotas which typically include a number of larger acritarch taxa (e.g., Vidal, 1976; 1979; 1981b). Thus, the paleontological data would argue that at least the fine­ grained Mineral Fork facies were deposited in water sufficiently deep to preclude benthic photosynthesizers, and sufficiently distal to preclude wash-ins of filaments and acritarch-forming eukaryotes (zone 5). 246

At the other end of the scale, the very late (post-Varangian) Proterozoic Pertatataka Formation of central Australia has been described as a deep water distal turbidite (Korsch, 1986; Zang, 1988; Zang and Walter, 1989) yet, paleontologically, it contains one of the most diverse assemblages of large, morphologically complex acritarchs so far described (Zang and Walter, 1989). If this environmental interpretation is correct, it clearly contradicts the paleoecological pattern outlined above. It is worth noting, however, that the high acritarch diversity in the Pertatataka Formation occurs consistently at the top of the sequence where it appears to be conformably overlain by the oolitic and stromatolitic (photic zone) carbonates of the Julie Formation (Wells et al., 1970; Preiss et al., 1978; Lindsay and Korsch, 1991). The diverse acritarchs furthermore co-occur with a variety of well preserved filaments (Zang, 1988; Zang and Walter, 1989) which, in some cases, are recovered as entangled, laterally continuous mats (Zang, 1988: pi. 103, figs. b-d). Conversely, filaments are conspicuously absent in the middle and lower parts of the Pertatataka Formation, and the acritarch assemblage becomes systematically less diverse downsection (Zang, 1988: text-fig. 4-16). At least in some areas the lower Pertatataka Formation is dominated by small spheroids with growth sequences comparable to Sphaerocongregus (Zang and Walter, 1989), very similar to turbidite-hosted biotas elsewhere. Thus, the distribution of microfossils argues strongly for a more distal, deeper water environment at the base of the Pertatataka Formation (zone 5/4), but shallowing upwards to well within the photic zone in the upper parts where the diverse acritarchs and filamentous microfossils occur (zone 2) (see Lindsay and Korsch, 1991). Comparable ornamented acritarchs in correlative units of the Officer Basin, South Australia are limited to nearshore facies (Jenkins et al., 1992). Paleoenvironmental information is also essential for making accurate 247

biostratigraphic correlations. There is good potential for a rigorous, acritarch- based biostratigraphy of at least the late Proterozoic (Knoll and Butterfield, 1989). However, it is also clear that the morphologically complex forms upon which such a scheme must necessarily be based have relatively restricted paleoenvironmental distributions - most occur in shallow water (zone 2) environments. Before meaningful biostratigraphic correlations can be made it must be evident that similar paleoecosystems are being compared. For example, the ca. 1300 Ma old McMinn Formation of central Australia was clearly deposited under shallow marine conditions as is indicated by both its sediments (Peat et al., 1978) and the character of its preserved fossils, i.e., abundant filaments, entangled filamentous mats and large acritarchs. Thus, the absence of morphologically complex acritarchs typical of 900-600 Ma old shallow-water sequences indicates an evolutionary rather than an environmental control on taxonomic composition and is consequently of biostratigraphic significance. In contrast, the trend from low to intermediate to high diversities (and simple to complex morphologies) of acritarchs observed going upsection through the Pertatataka Formation (Zang, 1988; Zang and Walter, 1989) appears to be one of simple paleoenvironmental change, i.e., a shallowing upward sequence.

CONCLUSION

Proterozoic fossils clearly have distributions that vary systematically in both space and time. Their full potential as paleoenvironmental and biostratigraphic tools nevertheless rests in our ability to accurately distinguish the two. Thus, forms that have previously been applied biostratigraphically, such as Sphaerocongregus. appear to be more paleoenvironmentally than 248 temporally controlled, while large process-bearing acritarchs offer useful constraints in both dimensions. Leiosphaerid acritarchs, the signature plankton of the Proterozoic, may also reveal important stratigraphie and environmental information under appropriate analysis. Thus, zone 2 assemblages of middle Proterozoic age (prior to the evolution of ornamented acritarchs) may be distinguished from superficially similar late Proterozoic zone 3 assemblages on the basis of bedding-plane distribution (patchy vs. even) and the prevalence of filamentous microfossils. In the same way, prokaryote-dominated zones 1 and 5 may be distinguished through indications of benthic growth. Paleoenvironmental and biostratigraphic analysis involve different approaches to the fossil record. The latter demands morphologically distinctive forms but a single specimen can often provide significant temporal resolution (e.g., the Agu Bay Valeria specimen). Conversely, paleoecology, as a proxy for paleoenvironment, is determined from large populations and overall 'community' structure. The depth-dependent characterization of Proterozoic microfossil assemblages outlined in this paper are in this sense ' community- based, using aspects such as diversity, abundance/dominance, and spatial distribution, as well as direct, moderately reliable indicators of depth such as filaments. As with their Phanerozoic counterparts, it is becoming increasingly dear that detailed investigation of Proterozoic microfossils can provide significant constraints on age, paleoenvironment, and evolutionary history. 249

REFERENCES

ALLISON, G. W. and S. M. AWRAMIK. 1989. Organic-walled microfossils from earliest Cambrian or latest Proterozoic Tindir Group rocks, northwest Canada. Precambrian Research 43:253-294. ATKINSON, A. W., Jr., B. E. S. GUNNING, and P. C. L. JOHN. 1972. Sporopollenin in the cell wall of Chlorella and other algae: ultrastructure, chemistry, and incorporation of 14C-acetate, studied in synchronous cultures. Planta (Berlin) 107:1-32. AWRAMIK, S. M., D. S. McMENAMIN, YIN C., ZHAO Z., DING Q., and ZHANG S. 1985. Prokaryotic and eukaryotic microfossils from a Proterozoic/Phanerozoic transition in China. Nature 315:655-658. BEEUNAS, M. A. and L. P. KNAUTH. 1985. Preserved stable isotopic signature of subaerial diagenesis in the 1.2-b.y. Mescal Limestone, central Arizona: implications for the timing and development of a terrestrial plant cover. Geological Society of America, Bulletin 96:737-745. BOUCOT, A. J. and R. S. CARNEY. 1981. Principles of Benthic Marine Paleontology. Academic Press; New York. 463 pp. BUTTERFIELD, N. J., A. H. KNOLL, and K. SWETT. 1988. Exceptional preservation of fossils in an Upper Proterozoic shale. Nature 334:424427. BUTTERFIELD, N. J. and R. H. RAINBIRD. 1988. The paleobiology of two Proterozoic shales. Geological Society of America, Abstracts with Programs 20:103. 250

CHANDLER, F. W. 1988. Geology of the late Precambrian Fury and Hecla Group, northwest Baffin Island, District of Franklin. Geological Survey of Canada, Bulletin 370:1-30. CHANDLER, F. W. and R. D. STEVENS. 1981. Potassium-Argon age of the Late Proterozoic Fury and Hecla Formation, Northwest Baffin Island, District of Franklin. Geological Survey of Canada, Paper 81-1 A:37-40. CHRISTIE-BLICK, N. 1983. Glacial-marine and subgladal sedimentation. Upper Proterozoic Mineral Fork Formation, Utah. Pp. 703-775. In B. F. Molnia (ed.), Glacial-Marine Sedimentation. Plenum Press; New York. CONDIE, K. C. 1967. Petrology of the Late Precambrian tillite (?) association in northern Utah. Geological Society of America, Bulletin 78:1317-1344. CRHTTNDEN, M. D., Jr., B. J. SHARP, and F. C. CALKINS. 1952. Geology of the Wasatch Mountains east of Salt Lake City, Parleys Canyon to Traverse Range. Utah Geological and Mineralogical Survey Guidebook to the Geology of Utah 8:1-37. DALE, B. 1983. Dinoflagellate resting cysts: "benthic plankton". Pp. 69-136. In G. A. Fryxell (ed.). Survival Strategies of the Algae. Cambridge University Press; Cambridge. DAMASSA, S. P. and A. H. KNOLL. 1986. Micropalaeontology of the Late Proterozoic Arcoona Quartzite Member of the Tent Hill Formation, Stuart Shelf, South Australia. Alcheringa 10:417-430. DAWES, P. R. and G. VIDAL. 1985. Proterozoic age of the Thule Group: new evidence from microfossils. Gronlands Geologiske Undersogelse Rapport 125:22-28. 251

DORNING, K. J. 1981. Silurian acritarch distribution in the Ludlovian shelf sea of South Wales and the Welsh Borderland. Pp. 31-36. In J. W. Neale and M. D. Brasier (eds.), Microfossils From Recent and Fossil Shelf Seas. Ellis Horwood Limited; Chichester. FAHRIG, W. F., K. W. CHRISTIE, and D. L. JONES. 1981. Paleomagnetism of the Bylot Basins: evidence for MacKenzie Continental tensional tectonics. In F. H. A. Campbell (ed.), Proterozoic Basins of Canada. Geological Survey of Canada, Paper 81-11:301-312. FOGG, G. E. 1982. Marine plankton. Pp. 491-513. In N. G. Carr and B. A. Whitton (eds.). The Biology of Cyanobacteria. University of California Press; Berkley. GREEN, J. W., A. H. KNOLL, and K. SWETT. 1988. Microfossils from oolites and pisolites of the Upper Proterozoic Eleonore Bay Group, central East Greenland. Journal of Paleontology 62:835-852. GREEN, J. W., A. H. KNOLL, and K. SWETT. 1989. Microfossils from silicified stromatolitic carbonates of the Upper Proterozoic Limestone-Dolomite 'Series', central East Greenland. Geological Magazine 126:567-585. HARRIS, T. M. 1974. Williamsoniella lienieri: its pollen and the compression of spherical pollen grains. Palaeontology 17:125-148. HEAMAN, L. M., A. N. LeCHEMINANT, and R. H. RAINBIRD, R. H. 1990. A U-Pb baddeleyite study of Franklin igneous events, Canada. Geological Association of Canada, Program with Abstracts 15:A55. HERMANN, T. N. 1974. Nakhodki massovykh skoplenii trikhomov v Rifee. Pp. 6-10. In B. V. Timofeev (ed.), Mikrofitofossilii Proterozoia i Rannego Paleozoia SSSR. Nauka; Leningrad. 252

HOFFMAN, P. 1976. Environmental diversity of middle Precambrian stromatolites. Pp. 599-611. In M. R. Walter (ed.). Stromatolites. Elsevier; Amsterdam. HORODYSKI, R. J. 1980. Middle Proterozoic shale-facies microbiota from the lower Belt Supergroup, Little Belt Mountains, Montana. Journal of Paleontology 54:649-663. IVANOVSKAYA, A. V. and B. V. TIMOFEEV. 1971. Relationship between salinity and distribution of phytoplankton. Geologoiia i Geofizika 8:113-117. JACKSON, G. D. 1986. Notes on the Proterozoic Thule Group, northern Baffin Bay. Geological Survey of Canada, Paper 86-1 A:541-552. JACKSON, G. D. and T. R. lANNELLI. 1981. Rift-related cyclic sedimentation in the Neohelikian Borden Basin, Northern Baffin Island. In F. H. A. Campbell (ed.), Proterozoic Basins of Canada. Geological Survey of Canada, Paper 81-10:269-302. JACOBSON, S. R. 1979. Acritarchs as paleoenvironmental indicators in Middle and Upper Ordovician rocks from Kentucky, Ohio, and New York. Journal of Paleontology 53:1197-1212. JANKAUSKAS. T. V. 1979. Srednerifeiskaia mikrobiota luzhnovo Urala i Bashkirskovo Priuralia. Doklady Akademii Nauk SSSR 248:190-193. JANKAUSKAS. T. V. (ed.). 1989. Mikrofossilii Dokembriia SSSR. Nauka; Leningrad. 188 pp. JENKINS, R. J. P., D. M. McKIRDY, C. B. FOSTER, T. O'LEARY, and S. D. PELL. 1992. The record and stratigraphie implications of organic-walled microfossils from the Ediacaran (terminal Proterozoic) of South Australia. Geological Magazine (in press). 253

KAUFMAN, A. J., A. H. KNOLL, and S. M. AWRAMIK. 1992. Biostratigraphic and chemostratigraphic correlation of Neoproterozoic sedimentary successions; Upper Tindir Group, northwestern Canada, as a test case. Geology 20:181-185. KNOLL, A. H. 1984. Microbiotas of the late Precambrian Hunnberg Formation, Nordaustlandet, Svalbard. Journal of Paleontology 58:131-162. KNOLL, A. H. 1992. Vendian microfossils in metasedimentary cherts of the Scotia Group, Prins Karls Forland, Svalbard. Palaeontology (in press). KNOLL, A. H., N. BLICK, and S. M. AWRAMIK. 1981. Stratigraphie and écologie implications of late Precambrian microfossils from Utah. American Journal of Science 281:247-263. KNOLL, A. H. and N. J. BUTTERFIELD. 1989. New window on Proterozoic life. Nature 337:602-603. KNOLL, A. H., J. M. HAYES, A. J. KAUFMAN, K. SWETT, and 1. B. LAMBERT. 1986. Secular variation in carbon isotope ratios from Upper Proterozoic successions of Svalbard and East Greenland. Nature 321:1-7. KNOLL, A. H. and K. SWETT. 1985. Micropalaeontology of the late Proterozoic Veteranen Group, Spitsbergen. Palaeontolgy 28:451-473. KNOLL, A. H., K. SWETT, and J. MARK. 1991. Paleobiology of a Neoproterozoic tidal flat/lagoonal complex: the Draken Conglomerate Formation, Spitsbergen. Journal of Paleontology 65:531-570. KORSCH, R. 1986. Pertatataka Formation. Pp. 103-107. In J. M. Kennard, R. S. Nicoll, and M. Owen (eds.). Field Excursion 25B, 12th International Sedimentological Congress. Bureau of Mineral Resources, Geology and Geophysics, Canberra. 254

LeCHEMINANT, A. N. and L. M. HEAMAN. 1989. Mackenzie igneous events, Canada: Middle Proterozoic hotspot magmatism associated with ocean opening. Earth and Planetary Science Letters 96:38-48. LINDSAY, J. F. and R. J. KORSCH. 1991. The evolution of the Amadeus Basin, central Australia. Bureau of Mineral Resources, Geology and Geophysics Bulletin 236:7-32. LIUBTSOV, V. V., N. S. MIKHAILOVA, and A. A. PREDOVSKD. 1989. Vyvody. Pp. 80-82. In F. P. Mitrofanov and T. V. Jankauskas (eds.), Litostratigragiia 1 Mikrofossilii Pozdnego Dokembria Kolskogo Poluostrova. Kolskii Nauchnyi Tsentr; Apatity. MANSUY, C. and G. VIDAL. 1983. Late Proterozoic Brioverian microfossils from France: taxonomic affinity and implications of plankton productivity. Nature 302:606-607. MARGULIS, L., G. HINKLE, H. McKHANN, B. MOYNIHAN, and S. W. BROWN. 1988. Mychonastes desiccatus Brown sp. nova (Chlorococcales, Chlorophyta) - an intertidal alga forming achlorophyllous desiccation- resistant cysts. Archiv fur Hydrobiologie, Supplement 78:425-446. MOORMAN, M. 1974. Microbiota of the late Proterozoic Hector Formation, southwestern Alberta, Canada. Journal of Paleontology 48:524-539. NICHOLS, J. M. and D. G. ADAMS. 1982. Akinetes. Pp. 387-412. In N. G. Carr and B. A. Whitton (eds.). The Biology of Cyanobacteria. University of California Press; Berkley. ODUM, E. P. 1971. Fundamentals of Ecology, 3rd edition. W. B. Saunders; Philadelphia. 574 pp. PALACIOS, T. 1989. Microfosiles de pared organica del Proterozoico Superior (regiôn central de la peninsula Ibérica). Memorias del Museo Paléontologie© de la Universidad de Zaragoza 3(2):1-91. 255

PEAT, C. J., M. D. MUIR, K. A. PLUMB, D. M. McKIRDY, and M. S. NORVICK. 1978. Proterozoic microfossils from the Roper Group, Northern Territory, Australia. BMR Journal of Australian Geology and Geophysics 3:1-17. PREISS, W. V., M. R. WALTER, R. P. COATS, and A. T. WELLS. 1978. Lithological correlations of Adelaidean glaciogenic rocks in parts of the Amadeus, Ngalia, and Georgina Basins. BMR Journal of Australian Geology and Geophysics 3:43-53. PJATILETOV, V. G. 1980. Mikrofossilii dashkinskoi svity Eniseickogo kriazha. Doklady Akademii Nauk SSSR 255:453-454. RASCIO, N., G. CASADORO, and C. ANDREOLI. 1979. Ultrastructural features of Chlorella nana sp. nov. Botanica Marina 22:223-227. ROSENBERG, R. 1974. Spatial dispersion of an estuarine benthic faunal community. Journal of Experimental Marine Biology and Ecology 15:69-80. SCHAU, M. and M. BECKETT. 1986. High grade metamorphic rocks of northwestern Melville Peninsula, District of Franklin. Geological Survey of Canada, Paper 86-lA:667-674. SCHERMERHORN, L. J. G. 1974. Late Precambrian mixités: glacial and/or nongladal?. American Journal of Science 274:673-824. SCHIEBER, J. 1986. The possible role of benthic microbial mats during the formation of carbonaceous shales in shallow Mid-Proterozoic basins. Sedimentology 33:521-536. STAPLIN, P. L. 1961. Reef controlled disruption of Devonian microplankton in Alberta. Palaeontology 4:392-424. 256

VIDAL, G. 1976. Late Precambrian acritarchs from the Eleonore Bay Group and Tillite Group in East Greenland. Gronlands Geologiske Undersogelse Rapport 78:1-19. VIDAL, G. 1979. Acritarchs from the Upper Proterozoic and Lower Cambrian of East Greenland. Gronlands Geologiske Undersogelse Bulletin 134:1-40. VIDAL, G. 1981a. Aspects of problematic acid-resistant, organic-walled microfossils (acritarchs) in the Upper Proterozoic of the North Atlantic region. Precambrian Research 15:9-23. VIDAL, G. 1981b. Micropaleontology and biostratigraphy of the upper Proterozoic and lower Cambrian sequence in East Finnmark, northern Norway. Norges Geologiske Undersokelse 362:1-53. VIDAL, G. 1990. Giant acanthomorph acritarchs from the upper Proterozoic in southern Norway. Palaeontology 33:287-298. VIDAL, G. and P. R. DAWES. 1980. Acritarchs from the Proterozoic Thule Group, north-west Greenland. Gronlands Geologiske Undersogelse Rapport 100:24-29. VIDAL, G. and T. D. FORD. 1985. Microbioteis from the late Proterozoic Chuar Group (northern Arizona) and Uinta Mountain Group (Utah) and their chronostratigraphic implications. Precambrian Research 28:349-389. VIDAL, G. and A. H. KNOLL. 1983. Proterozoic plankton. Geological Society of America, Memoir 161:265-277. VIDAL, G. and J. P. NYSTUEN. 1990. Micropaleontology, depositional environment, and biostratigraphy of the upper Proterozoic Hedmark Group, southern Norway. American Journal of Science 290-A:170-211. 257

VIDAL, G. and A. SIEDLECKA. 1983. Planktonic, acid-resistant microfossils from the upper Proterozoic strata of the Barents Sea region of Varanger Peninsula, East Finnmark, northern Norway. Norges Geologiske Undersokelse 382:45-79. VOLKOVA, N. A. 1981. Akritarkh verkhnego Dokembriia iugo-vostochnoi Sibiri (Ustkirbinskaia svita). Biulleten Moskovskogo Obshchestva Ispytatelei Prirody Otdel Geologicheskii 56(4):66-75. WATERBURY, J. B. and R. Y. STANIER. 1978. Patterns of growth and development in pleurocapsalean cyanobacteria. Microbiological Reviews 42:2-44. WELLS, A. T., D. J. FORMAN, L. C. RANFORD, and P. J. COOK. 1970. Geology of the Amadeus Basin, central Australia. BMR Geology and Geophysics, Bulletin 100:1-222. YIN, C. 1990. Spinose acritarchs from the Toushantou Formation in the Yangtze Gorges and its geological significance. Acta Micropalaeontologica Sinica 7:265-270. ZANG, W. 1988. Late Proterozoic - Early Cambrian Microfossils and Biostratigraphy in China and Australia. Unpublished PhD thesis. Australian National University. 432 pp. ZANG, W. L. and M. R. WALTER. 1989. Latest Proterozoic plankton from the Amadeus Basin in central Australia. Nature 337:642-645. 258

Figure 1. Geological map of the Fury and Hecla Group, northwestern Baffin

Island, Canada showing microfossil locality "S". BAFFIN ISLAND

+ + + 70 15 + + + +

DRIFT

25 Km FURY and HECLA STRAIT

HADRYNIAN NYBOE FM Dybbol Sill, dykes Redbeds AGU BAY FM Mainly Redbeds ARCHEAN-APHEBIAN NEOHELIKIAN-HADRYNIAN + +1 HUDSONIAN BASEMENT FURY AND HECLA GROUP HANSEN FM + + GNEISS AUTRIDGE FM Basalt Quartzarenlte, Shale Fault SIKOSAK BAY FM WHYTE INLET FM Quartzarenlte Site discussed in text Quartzarenlte © (L161) 260

Figure 2. Stratigraphie column of the Fury and Hecla Group and an expanded view of the Black Shale Member of the Agu Bay Formation. Fossiliferous sample L161 was collected from an incomplete section of the Black Shale

Member 23 km to the east of the depicted column, thus its exact level in the

Member is unknown. DYBBOL SILL

AUTRIDGE 500 FORMATION (2 km preserved)

|~—I Fissile shale liS'l Synoresis cracks WHYTE INLET FORMATION BLACK Crossbeds (3 km preserved) Desiccation cracks SHALE \A//.I Stromatolites

MEMBER ^ Basait, diabase AGU BAY Quartz conglomerate FORMATION + + + + + + Granitic rocks HANSEN FM 10 SIKOSAK FM NYEBOE FORMATION

Shale Dolomite _|Medium/coarse Sandstone iStiale I ^Grlt I ^Conglomerate Sandstone Siltstone 262

Figure 3. Leiosphaeridia minutissima isolated from sample L161, Black Shale

Member, Agu Bay Formation. Figure captions include Harvard University

Paleobotanical Collection (HUPC) number, slide identification, and England-

Finder co-ordinates (in parentheses). Magnification for all figures, x 715.

Entire leiosphaerids. 1. HUPC 62450; L161-35 (D-23-2). 2. HUPC 62451; L161-35 (M-27-3). 3. HUPC 62452; L161-35 (N-37-3). 4. HUPC 62453; L161-24 (H-20-0).

Medially split leiosphaerids in 'lateral' orientation 5. HUPC 62454; L161-30 (H-18-3). 6. HUPC 62455; L161-23 (L^O-1). 7. HUPC 62456; L161-28 (J-21-0). 8. HUPC 62457; L161-35 (M-38-0). 9. HUPC 62458; L161-35 (N-39-0). 10. HUPC 62459; LI61-26 (J-24-0); with one hemisphere oriented 'parallel' to bedding, the other 'lateral'.

Medially split leiosphaerids with both hemispheres 'parallel' to bedding. 11. HUPC 62460; L161-36 (L-37-3); note the small (ca. 3 pm diameter) attached leiosphaerid. 12. HUPC 62461; LI 61-30 (L-49-4). 13. HUPC 62462; L161-35 (M-34-0). 14. HUPC 62463; L161-31 (P-49-2). 8 10

11 12 13 14 264

Figure 4. Size-frequency distribution of Black Shale Member leiosphaerids

(counts from bedding-parallel thin sections).

A. Distribution of entire (unsplit) leiosphaerids (cumulative of histograms

C through G).

B. Distribution of medially split leiosphaerids.

C-G. Distributions of five separate counts of entire leiosphaerids from five

different thin sections of sample L161: C, L161-15-C; D, L161-22-C;

E, L161-16-C; F, L161-7-R; G, L161-1-L. 700 150 •

600 100 n = 497 n = 436 n = 440

500 5 0 -

n=2292

400 ED^ B 300 150 Medial Splits n = 354 200 100 n = 461 n = 458

100 50

±L 0 ~ | I .k a 5 7 9 11 13 15 17 19 21 23 5 7 9 11 13 15 17 19 21 23 5 7 9 11 13 15 17 19 21 23 5 7 9 11 13 15 17 19 21 23 266

Figure 5. Microfossils from sample LI 61, Black Shale Member, Agu Bay

Formation. Figure captions include Harvard University Paleobotanical

Collection (HUPC) number, slide identification, England-Finder co-ordinates

(in parentheses), and magnification.

Larger leiosphaerid acritarchs. 1. HUPC 62464; L161-29 (N-24-3); x 320; L. crassa. 2. HUPC 62465; L161-29 (H-27-2); x 320; L. iacutica. 3. HUPC 62466; L161-33 (K-20-4); x 320; L. crassa.

Valeria lophostriata 4. HUPC 62467; L161-31 (M-30-3); x 130; whole specimen preserved within an unomamented envelope. 5. HUPC 62467; L161-31 (M-30-3); x 755; detail of Fig. 5.4.

Satka(?) 6. HUPC 62468; L161-25 (N-26-0); x 755.

Leiosphaeridia minutissima 7. Entire and medially split leiosphaerids in bedding-parallel thin section L161-16 (M-19-1); x 320.

Filamentous microfossil - Siphonophycus 8. Allochthonous filament fi-agment in bedding-parallel thin section LI61-1 (L-19-4); X 320.

Svnsphaeridium(?) 9. HUPC 62469; in bedding-parallel thin section L161-10 (K-42-3); x 755. t- A * t ? • • ;2 p . ' • ' f- &

•. ♦

l é " ' . 8 268

Figure 6. Schematic diagram illustrating the paleoenvironmental distribution of Proterozoic microfossils from restricted near-shore (zone 1) through distal shelf and slope (zone 5) environments. Figures below the base-line represent non-randomly distributed and/or filamentous benthos limited to the photic zone. The large ornamented plankton in zone 2 are limited to the pre-

Ediacaran late Proterozoic. The Black Shale Member of the Aug Bay Formation was deposited in a zone 4 environment.

270

IV. THE PALEOBIOLOGY OF THE FROTEROZOIC HUNTING FORMATION, SOMERSET ISLAND, ARCTIC CANADA

ABSTRACT

Silidfied peritidal carbonates of the 1270 to 725 million year old Hunting Formation, Somerset Island, arctic Canada, contain fossils of a well-preserved bangiophyte red alga, Baneiomorpha antigua n. gen, n. sp. Morphological details, including evidence of transverse intercalary cell division in uniseriate filaments and longitudinal intercalary division to yield multiseriate filaments of radially-arranged wedge-shaped cells, indicate that the fossils are closely related to extant Baneia. Such taxonomic resolution distinguishes Baneiomorpha from most other pre-Ediacaran eukaryotes and contributes to growing evidence that multicellular algae diversified well before the Ediacaran radiation of large animals.

INTRODUCTION

Multicellular organisms characterized by cellular integration and differentiation occur in all four eukaryotic kingdoms, including at least eight separate phyla in the Protista (Buss, 1987). Most of these experiments in complex multicellularity must have made their first appearance well before the Ediacaran radiation of large animals (ca. 600 million years ago), but fossil evidence is limited and in many cases equivocal. Interpretational problems include the lack of preserved cellular structure in many fossils, the non­ diagnostic nature of such structures as may be preserved, and the absence of 271

biochemical and ultrastructural information. Rocks from the ca. 750 to 1250 million year old (Ma) Hunting Formation, Somerset Island contain fossils in which the habit and patterns of cell division are sufficiently distinctive to permit their assignment to the bangiophyte red algae (Bangiales, Rhodophyta). They are here ascribed to a new genus and species, Bangiomorpha antigua.

GEOLOGICAL SETTING

Somerset Island is a narrowly separated extension of the Boothia Peninsula in the southeastern portion of the Canadian arctic archipelago (District of Franklin, Northwest Territories) (Fig. 1). In the Aston Bay area on its northwest coast, little altered middle to upper Proterozoic sedimentary rocks sit with ca. 10-25° of easterly dip on crystalline basement of the Canadian Shield (Fig. 2) (Blackadar, 1963; 1967; Tuke et al., 1966; Dixon et al., 1971; Dixon, 1974; Stewart, 1987). The sedimentary sequence comprises two unconformity-bound units. The basal Aston Formation (also cropping out on adjacent Prince of Wales, Prescott, and Pandora islands) consists largely of quartz arenite, but with conspicuous horizons of red chert and red to white intradastic dolostone and a striking ca. 10-30 m thick red stromatolitic biostrome; diabase sills account for approximately one third of the 800-1100 m thick formation. Aston rocks are disconformably to unconformably overlain by the predominately dolomitic Hunting Formation. A measured section along the northernmost west tributary of the Hunting River (94° 52'W, 73° 36'N) and an abbreviated basal section further to the south (94° 42'W, 73° 34'N) document ca. 625 meters of Hunting Formation strata (Figs. 2, 3). Uppermost Hunting rocks were not measured, but a reconnaissance survey along Aston Bay (94° 52'W, 73° 39^N) points to a later influx of silidclastic sediments. 272

Three relatively distinct units can be identified in the measured Hunting strata (Fig. 3). A lowermost 'basal member' (20+ m) consists of conglomeratic cross-bedded sandstone with quartz pebbles up to 3 cm long, grading upwards into interbedded purple siltstone, sandy dolostone, and dolosiltite. An undetermined thickness separates this basal sequence from the fossiliferous 'cherty member', ca. 300 meters of buff to grey dolomite with abundant, sometimes pervasive, diagenetic chert. Primary textures in this unit include abundant microdigitate and domal stromatoloids (cf.. Grey and Thome, 1985; Hofmann and Jackson, 1987), substantial thicknesses (up to 1 m) of 'giant' fibro-radiate aragonite(?) (cf., Peryt et al., 1990; Aitken, 1991: figs. 7-9), laminae- disruptive fenestrae, and coated grains; localized horizons feature tepee structures, concentrically-laminated speleothem-like domes (cf., Pelechaty and James, 1991: fig. 8C), and Baicalia-like columnar stromatolites. The prevalence of precipitated carbonate here is comparable to numerous sequences of early Proterozoic age (cf., Grotzinger, 1989), and the overall character is remarkably reminiscent of the Duck Creek Dolomite, Western Australia (cf.. Grey and Thorne, 1985). Restricted supratidal or high intertidal conditions clearly dominated most depositon. Fossiliferous cherts occur at the 80 and 81 meter levels of this 'member' (F in Figure 3) and, in the case of the new bangiophyte fossil, at a broadly correlative horizon ca. 3 km to the south (Fig. 2: arrow). The 'cherty member' is succeeded by at least 300 meters of buff to red dolomite, conspicuously distorted by peritidal tepee structures (cf., Assereto and Kendall, 1977) and occasional mud-cracked horizons. Chert is relatively uncommon, and the only distinctly laminated structures are rare columnar stromatolites; no microfossils were encountered. The dearth of diagenetic silica (cf., Mali va et al., 1989) and tufa-like carbonate fabrics suggests that deposition of this 'tepee member' took place under somewhat less restricted conditions 273

than that which attended the underlying unit. Hunting deposition is only broadly constrained in time, bracketed by two generations of mafic intrusions, one that cuts only the Aston Formation, the other that intrudes both Aston and Hunting strata (Kerr and deVries, 1976). Lower Paleozoic rocks on Somerset Island and elsewhere in the Canadian arctic show no evidence of such intrusions, and the two events are accepted as belonging to the 1267 ± 2 Ma Mackenzie (LeCheminant and Heaman, 1989) and 723 ± 3 Ma Franklin (Heaman et al., 1990) dyke swarms (both U-Pb ages on baddeleyite); a diabase sill in the upper Aston Formation has yielded a K-Ar age of 702 ± 25 Ma (in Dixon, 1974). The restricted intertidal to supratidal microfossils that dominate the Hunting assemblage are of minimal biostratigraphic utility. However, secular trends in carbon isotopes, as recorded in marine carbonates (chemostratigraphy), may further constrain the time of deposition (Knoll et al., 1986). Preliminary isotopic analyses of Hunting carbonates indicate that they have values between +2.5 and +3.8 PDB (Butterfield and Hayes, unpublished data), values lighter than those typical of 700 to 800 Ma carbonates in the Canadian arctic and elsewhere (Knoll et al., 1986; Asmerom et al., 1991). Few isotopic data have been reported for well dated 800 Ma to 1200 Ma carbonate sequences, but of those available, the Hunting signature most closely approximates that of un weathered dolomites from the 1100-1200 Ma Mescal Limestone, Arizona (Beeunas and Knauth, 1985), or the ca. 1250 Ma Society Cliffs Formation on adjacent Baffin Island (Jackson and lannelli, 1981; Butterfield and Hayes, unpublished data). Although not definitive, the carbon isotopic evidence suggests that Hunting sedimentation probably occurred closer to the older (Mackenzie) rather than the younger (Franklin) dyke- defined chronological boundary. 274

PALEOBIOLOGY

Apart from the relatively infrequent Baicalia-tvpe columnar stromatolites (and rare branched forms northwest of the measured section), the most obvious record of biological activity on the Hunting tidal flats comes from microfossils preserved in early diagenetic chert. Of the two fossiliferous samples from the measured section, one (HU-249) is dominated by 10-50 pm diameter spheroidal vesicles, usually with multiple envelopes and commonly containing unsheathed or multiple-sheathed internal vesicles (Figs. 4.3, 4.4). As a population, most of these fossils would be ascribed to Gloeodiniopsis Schopf, 1968, although some of the more colonial forms approach the habit of Eoentophvsalis Hofmann, 1976. The other sample (HU-245) consists largely of laminated and densely packed Siphonophvcus tvpicum (Hermann. 1974) microbial mat, with localized occurrences of Siphonophvcus (?) gracile (Pjatiletov, 1980) (Fig. 4.6), Gloeodiniopsis(?). Pterospermopsimorpha Timofeev, 1966 (Fig. 4.5), Polvbessurus bipartitus Fairchild, 1975 ex Green et al., 1987, and possible Bangiomorpha n. gen. (alternatively, Palaeolvngbva Schopf, 1968). Dense microbial mats (S. tvpicum. S. kestron Schopf, 1968, and S. gracile) and Polvbessurus also occur in the partial southern section of the 'cherty member' (Fig. 3: arrow; Figs. 4.1, 4.2), pointing to their closely comparable paleoenvironmental settings (cf.. Knoll et al., 1991). A specimen of Polvbessurus here shows unequivocal branching of its 'cone-in-cone' mucilaginous stalk (Fig. 4.1), a feature thought to be likely for the taxon but not actually observed in the type material (Green et al., 1987). 275

Bangiomorpha antigua n. gen., n. sp.

A laminated chert from the southern fossil locality (Fig. 2) preserves a spectacular population of multicellular fossils, here described as Bangiomorpha antigua n. gen., n. sp. Twelve perpendicular to bedding thin sections have so far exposed over 1000 filaments in various orientations; it regularly co-occurs with Polvbessurus. In its most simple (and most common) state, Bangiomorpha consists of stacked disc-shaped cells enclosed in a relatively transparent enveloping sheath, and forms unbranched uniseriate filaments 15- 45 pm in diameter and up to 2 mm long (Figs. 4, 6.1). Almost all specimens are oriented vertically and they commonly occur in groups attached to locally stabilized surfaces (Fig. 5.1); the basal ends of most of the filaments are elaborated into multicellular structures that appear to have served as attachment rhizoids (Fig. 4.9). Larger diameter specimens (> 45 pm across) of these same populations are generally multiseriate, with a number of cells occupying any given level along the filament (Figs. 5.2b-c, 6.2, 4.8). Transverse sections of these more complex forms reveal that the component cells are usually wedge-shaped and arranged axially around a central core (Fig. 6.3). The specimen in Figure 6.3 preserves three of what were originally eight such cells, and, in different focal planes of the same specimen, four adjacent layers show a similar pattern of cell division. Other multiseriate filaments have a four-fold radial cleavage (e.g., b in Fig. 5.2), while some appear to be constructed of more numerous spheroidal cells (Fig. 4.8). In most details of its morphology, and particularly in its distinctive patterns of cell division, Baneiomorpha n. gen. compares closely to the macroscopic, gametophytic (IN) generation of species belonging to the modern red algal genus Baneia. Bangia is a cosmopolitan seaweed that today ranges 276 from boreal to subtropical oceans and from normal marine waters to freshwater lakes and streams (Carbary et al., 1980; Sheath and Cole, 1980; Sheath and Cole, 1984; Sheath et al., 1985). It tolerates extended emersion and desiccation, and is usually found in the upper intertidal zone of marine coastlines and at or above the waterline in freshwater; it commonly occurs with filamentous cyanobacteria above the zone occupied by green algae (Sheath and Cole, 1980; Sheath et al., 1985). Perhaps not insignificantly, this habit places it in the zone where, in the Proterozoic, it would have been particularly subject to early diagenetic silica permineralization (Maliva et al., 1989). In its most simple form, gametophytic Baneia is characterized by unbranched uniseriate filaments composed of stacked, disc-shaped cells. The polysaccharide cell walls are distinctly bi-phasic with an inner wall defining individual cells and an outer wall enveloping the whole organism (Figs. 6.4, 6.5). In more mature, multiseriate forms the 'outer v/all' also occupies the central core of the filament (Fig. 6.6); a sub-micron thick 'cuticle' gives the filament a sharply delineated outer wall (Cole et al., 1985: fig. 4). Unlike the apical cell division typical of florideophytes, vegetative growth in bangiophyte red algae generally occurs by transverse intercalary cell division (Carbary et al. 1980; Cole and Conway, 1975); consequently, Bangia filaments often show a hierarchical packaging of successive cell generations (Fig. 6.4). In fossil Bangiomorpha n. gen., both a differentiated inner and outer cell wall and a pattern of transverse intercalary cell divisions are clearly evident (Figs. 6.1,

6 .2).

Mature filaments of gametophytic Bangia generally become multiseriate through longitudinal intercalary division (Figs. 6.5, 4.7). They do not, however, become simple constructions of randomly arranged, equidimensional 277 cells. Not only is the position of each parent cell broadly retained and reflected in the multiseriate filament, but the initial cleavage of the disc-shaped cells is also radially oriented (Carbary et al. 1980; Cole et al., 1985). This orientation yields a distinctive pattern of 4, 8, or 16 wedge-shaped cells arranged axially around a central core and is particularly apparent in transverse cross sections of vegetative Bangia filaments (Fig. 6.6) (Carbary et al., 1980: figs. 9.f-9.h); subsequent spore differentiation (asexual, male, or female) results in multiseriate filaments constructed of numerous spheroidal cells (Sommerfeld and Nichols, 1970: figs. 11-12; Carbary et al., 1980: figs. 9.i- 9.n; Cole et al., 1985: figs. 9-11). Because many of the gross morphological characteristics of sub-tissue-grade organisms are convergent or ecophenotypic, these underlying cellular patterns are likely to be of considerable taxonomic and phylogenetic significance. The distinctive cell division patterns of Bangiomorpha n. gen., along with its overall habit and such details as the localized constrictions of multiseriate filaments (Figs. 6.2, 4.7, 4.8) or abrupt transitions from a uniseriate to multiseriate condition (Fig. 4.8; cf. Lamb et al., 1977: fig. 45), strongly support its assignment to the bangiophyte red algae, probably to the level of family (Bangiaceae). The presence of spheroid- composite multiseriate filaments (Fig. 4.8) further resolves a spore- differentiating, possibly sexual phase (cf.. Cole et al., 1985). Bangiomorpha antigua n. sp. is not identical to extant species of Bangia. The modem alga varies widely in filament diameter but averages ca. 100 pm (Sheath and Cole, 1984). The Proterozoic filaments fall within the size range of modem Bangia but are generally smaller, having a mean diameter of 25.4 pm and a minor mode of 51.3 pm in the case of multiseriate forms (see systematic section); maximum measured diameter is 72 pm. The fossils also differ in possessing multicellular rhizoids or holdfast structures (Fig. 4.9), whereas 278

Bangia has long, non-septate rhizoids descending from a number of basal vegetative cells (Sommerfeld and Nichols, 1970: fig. 3). Multicellular rhizoids are nevertheless common in the red algae, and holdfasts essentially identical to those of B. antigua are found in the filamentous bangiophytes Ervthrotrichia (Erythropeltidaceae) and Porphvra (Bangiaceae) (Carbary et al., 1980). Are there other possible interpretations of these fossils? Stigonematacean cyanobacteria often form a comparably thick external sheath and are commonly multiseriate. However, these prokaryotes are also characterized by true branching and several types of differentiated cells (e,g., heterocysts and akinetes), which are not observed in the fossils. More importantly, the multiseriate thallus of stigonemataceans appears to be a relatively loosely knit association of cells and filaments that lack any particular geometrical or developmental regularity (Martin and Wyatt, 1974); the distinctive pattern of radial cell division characteristic of both Bangia and Bangiomorpha n. gen. is unknown in any cyanobacterium (S. Golubic, pers. comm.). Other extant eukaryotes, such as species of the green algal family Schizomeridaceae, likewise form uniseriate and multi-seriate filaments but these also lack radially oriented cells. Bangiomorpha n. gen. is neither the only nor the oldest Proterozoic metaphyte. Undoubted multicellular algae are known from several Late Riphean (1000-700 Ma) shale biotas and include coenobial colonies (Butterfield and Knoll, 1989; Chapter II) and branching filamentous forms, both septate (Butterfield et al., 1988; Chapter II) and coenocytic (Hermann, 1981; Chapter II). Macroscopic tawuiids and lonfengshaniids (ca. 900-700 Ma) have a worldwide distribution and represent at least probable metaphytes (Hofmann, 1985; Chapter II), as do filamentous macrofossils in ca. 1400 million year old deposits from Montana and China (Walter et al., 1990). Problematic bedding 279

plane markings in middle Proterozoic sandstones of Western Australia are conceivably the imprints of a relatively large filamentous seaweed (Grey and Williams, 1990). In latest Proterozoic rocks a calcareous, encrusting phylloid alga (Grant et al., 1991) and a variety of structured multicellular fossils (Zhang, 1989) have been documented, and filamentous organic-walled vendotaenids are generally, although not universally (Vidal, 1989), thought to be of algal origin (Gnilovskaya, 1988). Most of these fossils, however, cannot be placed confidently into any higher-order taxon. The same can be said for most of the organic-walled fossil algae of Phanerozoic age (e.g., Walcott, 1919; Fry, 1983). In contrast, the clearly interpretable morphology of Bangiomorpha provides an unambiguous datum point for the reconstruction of early red algal and protistan phylogeny. The two classes of the Rhodophyta, Bangiophyceae and Florideophyceae, are distinguished on the basis of thallus form, growth habit, and various intracellular characteristics (Cole and Conway, 1975; Garbary et al., 1980). They appear to be linked phylogenetically by the microscopic chonchocelis phase (2N) sporophyte) observed in some species of Bangia and Porphvra (Bangiaceae). The chonchocelis exhibits a number of otherwise florideophyte characteristics, including apical cell division and plugged pit connections between cells (Cole and Conway, 1975). The Bangiophyceae are generally considered to be less advanced, although not necessarily ancestral to, the Florideophyceae (Fritsch, 1965; Bold and Wynne, 1985). As a gametophytic bangiophyte, Bangiomorpha n. gen. predates the earliest record of probable florideophytes (Walcott, 1919) by at least 200 million years and thus provides at least circumstantial support for a bangiophyte to florideophyte phylogenesis. Even so, the paucity of the non-mineralized fossil record in general, and of non-mineralized rhodophytes in particular, rules out any conclusive 280 evolutionary statement based solely on the fossil record. Whether the bangiophyte life cycle included an alternation of generations by Hunting time is speculative at this point as there is no evidence of a preserved chonchocelis phase in these rocks. The oldest reported chonchocelis comes from Silurian rocks in Poland (Campbell, 1980). Molecular data from both small subunit (Bhattacharya et al., 1990) and large subunit (Perasso et al., 1989) ribosomal RNA indicate that the three principal clades of multicellular algae (Rhodophyta, Chlorophyta, and Chromophyta) all diverged during a brief but marked radiation relatively late in eukaryotic history. This pattern is broadly supported by the moderate diversity of mid to late Proterozoic fossil metaphytes (Chapter H). Multicellularity, however, is almost certainly a derived condition, and Tappan's (1976) interpretation of the ca. 2000 million year old Gunflint microfossil Eosphaera as a Porphvridium-like rhodophyte, although controversial, underscores the probability that unicellular red, green, and brown algae antedated their multicellular descendants by a significant interval. Although the timing of early diversification remains incompletely known, the Hunting Formation bangiophytes contribute to a growing paleontological record which indicates that several and perhaps all major groups of multicellular algae were well established long before the Ediacaran and Cambrian radiations. 281

SYSTEMATIC PALEONTOLOGY

All type and illustrated specimens are housed in the Paleobotanical Collections of Harvard University and assigned Harvard University Paleobotanical Collection (HUPC) numbers.

Domain EUCARYA Woese, Kandler, and Wheelis, 1990 Division RHODOPHYTA Wettstein, 1924 Class BANGIOPHYCEAE Melchior, 1954 Order BANGIALES Schmitz, 1892 Family BANGIACEAE Nageli, 1847 Genus BANGIOMORPHA n. gen.

Type spgczes.—Bangiomorpha antigua n. sp. DzagMoszs.—Unbranched, filamentous bangiaceans, either uniseriate or multiseriate. Uniseriate filaments constructed of stacked, disc-shaped cells with evidence of transverse intercalary cleavage. Multiseriate filaments constructed of stacked units of 4 to 8 radially arranged wedge-shaped cells; occasionally of more numerous spheroidal cells. All filaments enclosed in a thick, relatively translucent outer wall. Basal, multicellular holdfast structures. Discussion.—A short uniseriate filament fragment with evidence of transverse intercalary cell division has been described from the Narssârssuk Formation, northwestern Greenland (Enzien, 1990). This sequence is likely correlative with the Hunting Formation (Jackson and lannelli, 1981) and the two units both record similar high intertidal to supratidal carbonate environments. The 282

possibility that the Narssârssuk fossil is a specimen of Bangiomorpha n. gen. is intriguing, but, in the absence of the diagnostic multiseriate filaments, cannot be unambiguously distinguished from a Tohannesbaptistia-like cyanobacterium. Etymology.—With reference to the marked similarity to modern Bangia.

BANGIOMORPHA ANTIQUA n. sp. Figures 4.8, 4.9, 5.1, 5.2, 6.1-6.3 Diagnosis.—A species of Baneiomorpha with uniseriate filaments less than 45 pm wide and multiseriate filaments equal to or greater than 45 pm wide. DescriptioTi.—Vertically oriented uniseriate and multiseriate unbranched filaments up to 2 mm long; isolated or gregarious. Uniseriate forms 15-45 pm wide (x = 25.4 pm; s.d. = 6.0 pm; n = 500) with disc-shaped cells showing evidence of transverse intercalary cleavage. Multiseriate forms 45-72 pm wide ix = 51.3 pm; s.d. = 9.1 pm; n = 13) with 4 to 8 radially arranged wedge-shaped cells occupying any given level along the filament, or occasionally with more numerous spheroidal cells; a single filament may include intervals of both the uniseriate and multiseriate habit. All filaments are surrounded by a prominent outer wall; many differentiate a basal, multicellular holdfast structure. Etymology.—With reference to its Proterozoic age. Material.—Several hundred specimens recorded from chert sample HU-ST-1; possible occurrences in HU-ST-5, HU-245. Holotype.—HUPC 62910; Figure 6.2; Slide HU-ST-IA; England-Finder co-ordinates: 0-37-3. 283

REFERENCES

AITKEN, J. D. 1991. The Ice Brook Formation and post-Rapitan, late Proterozoic glaciation, Mackenzie Mountains, Northwest Territories, Geological Survey of Canada, Bulletin 404:1-43. ASMEROM, Y., S. B. JACOBSEN, A. H. KNOLL, N. J. BUTTERFIELD, and K. SWETT. 1991. Strontium isotopic variations of Neoproterozoic seawater: Implications for crustal evolution. Geochimica et Cosmochimica Acta 55:2883-2894. ASSERETO, R. L. and C. G. St. C. KENDALL. 1977. Nature, origin and classification of peritidal teepee structures and related breccias. Sedimentology 24:153-210. BEEUNAS, M. A and L. P. KNAUTH. 1985. Preserved stable isotopic signature of subaerial diagenesis in the 1.2-b.y. Mescal Limestone, central Arizona: Implications for the timing and development of a terrestrial plant cover. Geological Society of America, Bulletin 96:737-745. BHATTACHARYA, D., H. J. ELWOOD, L. J. GOFF, and M. L. SOGIN. 1990. Phylogeny of Gracilaria lemaneiformis (Rhodophyta) based on sequence analysis of its small subunit ribosomal RNA coding region. Journal of Phycology 26:181-186. BLACKADAR, R. G. 1963. Somerset Island between Aston Bay and Creswell Bay. Pp. 143-150. In Y. O. FORTIER, R. G. BLACKADAR, B. F. GLENISTER, H. R. GREINER, D. J. McLAREN, N. J. McMILLAN, A. W. NORRIS, E. F. ROOTS, J. G. SOUTHER, R. THORSTEINSSON, and E. T. TOZER. Geology of the north-central part of the Arctic Archipelago, 284

Northwest Territories (Operation Franklin). Geological Survey of Canada, Memoir 320. 671 pp. BLACKADAR, R. G. 1967. Precambrian geology of Boothia Peninsula, Somerset Island, and Prince of Wales Island, District of Franklin. Geological Survey of Canada, Bulletin 151. 62 pp. BOLD, H. C and M. J. WYNNE. 1985. Introduction to the Algae, 2nd edition. Prentice-Hall; Englewood Cliffs, NJ. 720 pp. BUSS, L. W. 1987. The Evolution of Individuality. Princeton University Press; Princeton. 201 pp. BUTTERFIELD, N. J., A. H. KNOLL, and K. SWETT. 1988. Exceptional preservation of fossils in an Upper Proterozoic shale. Nature 334:424-427. BUTTERFIELD, N. J. and F. W. CHANDLER. 1992. Palaeoenvironmental distribution of Proterozoic microfossils, with an example from the Agu Bay Formation, Baffin Island. Palaeontology (in press). CAMPBELL, S. E. 1980. Palaeoconchocelis starmachii, a carbonate boring microfossil from the Upper Silurian of Poland (425 million years old): implications for the evolution of the Bangiaceae (Rhodophyta). Phycologia 19:25-36. COLE, K. M. and E. CONWAY. 1975. Phenetic implication of structural features of the perennating phase in the life history of Porphvra and Bangia (Bangiophyceae, Rhodophyta). Phycologia 14:239-245. COLE, K. M., C. M. PARK, P. E. REID, and R. G. SHEATH. 1985. Comparative studies on the cell walls of sexual and asexual Baneia atropurpurea (Rhodophyta). I. Histochemistry of polysaccharides. Journal of Phycology 21:585-592. 285

DIXON, J. 1974. Revised stratigraphy of the Hunting Formation (Proterozoic), Somerset Island, Northwest Territories. Canadian Journal of Earth Sciences 11:635-642. DIXON, O. A., S. R. WILLIAMS, and J. DIXON. 1971. The Aston Formation (?Proterozoic) on Prince of Wales Island, arctic Canada. Canadian Journal of Earth Sciences 8:732-742. DIXON, P. S. 1973. Biology of the Rhodophyta. Oliver and Boyd; Edinburgh. 285 pp. ENZIEN, M. V. 1990. Cyanobacteria or Rhodophyta? Interpretation of a Precambrian microfossil. Biosystems 24:245-251. FRITSCH, F. E. 1965. The Structure and Reproduction of the Algae. The University Press; Cambridge. 2 Volumes. FRY, W. L. 1983. An algal flora from the Upper Ordovician of the Lake Winnipeg region, Manitoba, Canada. Review of Palaeobotany and Palynology 39:313-341. GARBARY, D. J., G. I. HANSEN, and R. F. SCAGEL. 1980. The marine algae of British Columbia and northern Washington: Division Rhodophyta (Red Algae), Class Bangiophyceae. Syesis 13:137-195. GNILOVSKAYA, M. B. (ed.). 1988. Vendotaenids of the East European Platform. Nauka; Leningrad. 144 pp. GRANT, S. W. F., A. H. KNOLL, and G. J. B. GERMS. 1991. Probable calcified metaphytes in the latest Proterozoic Nama Group, Namibia: origin, diagenesis, and implications. Journal of Paleontology 65:1-18. GREEN, J. W., A. H. KNOLL, S. GOLUBIC, and K. SWETT. 1987. Paleobiology of distinctive benthic microfossils from the upper Proterozoic Limestone- Dolomite "Series", central East Greenland. American Journal of Botany 74:928-940. 286

GREY, K. and A. M. THORNE. 1985. Biostratigraphic significance of stromatolites in upward shallowing sequences of the early Proterozoic Duck Creek Dolomite, Western Australia. Precambrian Research 29:183-206. GREY, K. and I. R. WILLIAMS. 1990. Problematic bedding-plane markings from the Middle Proterozoic Manganese Subgroup, Bangemall Basin, Western Australia. Precambrian Research 46:307-328. GROTZINGER, J. P. 1989. Facies and evolution of Precambrian carbonate depositional systems: emergence of the modem platform archetype. Pp. 79-106. In P. D. Crevello, J. L. Wilson, J. F. Sarg, and J. F. Read (eds.). Controls on Carbonate Platform and Basin Development. Society of Economic Paleontologists and Mineralogists, Special Publication 44. HEAMAN, L. M., A. N. LeCHEMINANT, and R. H. RAESfBlRD, R. H. 1990. A U-Pb baddeleyite study of Franklin igneous events, Canada. Geological Association of Canada, Program with Abstracts 15:A55. HERMANN, T. N. 1981. Nitchatye mikroorganizmy Lakhandinskoi svity reki Mai. Paleontologicheskii Zhurnal 1981(2):126-131. HOFMANN, H. J. 1985. Precambrian carbonaceous megafossils. Pp. 20-33. In D. F. Toomey and M. H. Nitecki (eds.), Paleoalgology: Contemporary Research and Applications. Springer Verlag; Berlin. HOFMANN, H. J. and G. D. JACKSON. 1987. Proterozoic ministromatolites with radial-fibrous structure. Sedimentology 34:963-971. HORODYSKI, R. J. and C. MANKIEWICZ. 1990. Possible Late Proterozoic skeletal algae from the Pahrump Group, Kingston Range, southeastern California. American Journal of Science 290-A:149-169. 287

JACKSON, G. D. and T. R. lANNELLI. 1981. Rift-related cyclic sedimentation in the Neohelikian Borden Basin, northern Baffin Island. In F. H. A. Campbell (ed.), Proterozoic Basins of Canada. Geological Survey of Canada Paper 81-10:269-302. KERR, J. W. and C. D. S. de VRIES. 1976. Structural geology of Somerset Island, District of Franklin. Geological Survey of Canada, Paper 76-lA:493-495. KNOLL, A. H., J. M. HAYES, A. J. KAUFMAN, K. SWETT, and I. B. LAMBERT, I. B. 1986. Secular variation in carbon isotope ratios from Upper Proterozoic successions of Svalbard and East Greenland. Nature 321:1-7. KNOLL, A. H., K. SWETT, and J. MARK. 1991. Paleobiology of a Neoproterozoic tidal flat/Iagoonal complex: the Draken Conglomerate Formation, Spitsbergen. Journal of Paleontology 65:531-570. LAMB, I. M., M. H. ZIMMERMANN, and E. E. WEBBER. 1977. Artificial Key to the Common Marine Algae of New England North of Cape Cod. Farlow Herbarium, Harvard University; Cambridge. 53 pp. LeCHEMINANT, A. N. and L. M. HEAMAN. 1989. Mackenzie igneous events, Canada: Middle Proterozoic hotspot magmatism associated with ocean opening. Earth and Planetary Science Letters 96:38-48. MALIVA, R. G, A. H. KNOLL, and R. SIEVER. 1989. Secular change in chert distribution: a reflection of evolving biological participation in the silica cycle. Palios 4:519-532. MARTIN, T. C. and J. T. WYATT. 1974. Comparative physiology and morphology of six strains of stigonematacean blue-green algae. Journal of Phycology 10:57-65. 288

MELCHIOR, H. 1954. Rhodophyta. Pp. 123-138. In H. Melchior and E. Werdermann (eds.), A. Engler's Syllabus der Pflanzenfamilien, 12th Edition, Volume 1. Borntraeger; Berlin. NÀGELI, C. 1847. Die neueren Algensysteme. F. Schulthess; Zürich. 275 pp. PELECHATY, S. M. and N. P. JAMES. 1991. Dolomitized Middle Proterozoic calcretes, Bathurst Inlet, Northwest Territories, Canada. Journal of Sedimentary Petrology 61:988-1001. PERASSO, R., A. BAROIN, L. H. QU, J. P. BACHELLERIE, and A. ADOUTTE. 1989. Origin of the algae. Nature 339:142-144. PERYT, I. M., A. HOPPE, T. BECHSTÀDT, J. KÔSTER, C. PIERRE, and D. K. RICHTER. 1990. Late Proterozoic aragonite cement crusts, Bambui Group, Minas Gerais, Brazil. Sedimentology 37:279-286. SCHMITZ, F. 1892. Rhodophyceae. Pp. 15-23. In A. Engler, Syllabus der Vorlesungen. G. Borntraeger; Berlin. SHEATH, R. G. and K. M. COLE. 1980. Distribution and salinity adaptations of Bangia atropurpurea (Rhodophyta), a putative migrant into the Laurentian Great Lakes. Journal of Phycology 16:412-420. SHEATH, R. G. and K. M. COLE. 1984. Systematics of Baneia (Rhodophyta) in North America. I. Biogeographic trends in morphology. Phycologia 23:383-396. SHEATH, R. G., K. L. VANALSTYNE, and K. M. COLE. 1985. Journal of Phycology 21:297-303. SOMMERFELD, M. R. and H. W. NICHOLS. 1970. Developmental and cytological studies of Bangia fuscopurpurea in culture. American Journal of Botany 57:640-648. 289

STEWART, W. D. 1987. Late Proterozoic to early Tertiary stratigraphy of Somerset Island and northern Boothia Peninsula, District of Franklin, N.W.T. Geological Survey of Canada, Paper 83-26. 78 pp. STEWART, W. D. and J. W. KERR. 1984. Geology of Somerset Island North, District of Franklin. Geological Survey of Canada, Map 1595A. TAPP AN, H. 1976. Possible eucaryotic algae (Bangiophycidae) among early Proterozoic microfossils. Geological Society of America, Bulletin 87:633-639. TUKE, M. F., D. L. DINELEY, and B. R. RUST. 1966. The basal sedimentary rocks in Somerset Island, N.W.T. Canadian Journal of Earth Sciences 3:697-711. VIDAL, G. 1989. Are late Proterozoic carbonaceous megafossils metaphytic algae or bacteria? Lethaia 22:375-379. WALCOTT, C. D. 1919. Middle Cambrian algae. Cambrian Geology and Paleontology IV. Smithsonian Miscellaneous Collections 67:217-260. WALTER, M. R., DU RULIN, and R. J. HORODYSKI. 1990. Coiled carbonaceous megafossils from the Middle Proterozoic of Jixian (Tianjin) and Montana. American Journal of Science 290-A;133-148. WETTSTEIN, R. 1924. Handbuch der Systematischen Botanik, Band 1. Franz Deuticke; Leipzig. 467 pp. WOESE, C., O. KANDLER, and M. L. WHEELIS. 1990. Towards a natural system of organisms: Proposal for the domains Archaea, Bacteria, and Eucarya. Proceedings, National Academy of Science, U.S.A. 87:45764579. ZHANG Y. 1989. Multicellular thallophytes with differentiated tissues from Late Proterozoic phosphate rocks of South China. Lethaia 22:113-132. 290

Figure 1. Schematic geological map of Somerset Island showing the location of

Proterozoic strata. ASTON' BAY SOMERSET

PHANEROZOIC SEDIMENTS

ISLAND HUNTING and ASTON FM’s

CRYSTALLINE BASEMENT

50 km

m 292

Figure 2. Geological map of the Aston Bay area showing the relationship of the Proterozoic Aston and Hunting Formations to the underlying crystalline basement of the Canadian shield and the overlying Phanerozoic sediments; the

Hunting River drainage system is shown for geographic reference. The black dot marks the location of the 'Basal Member' outcrop and its contact with the underlying Aston Formation (= locality D of Tuke et al., 1966). After Stewart and Kerr (1984). ASTON

BAY

5 km

Fossil locality

Measured section

Phanerozoic sediments

Hunting Formation

Aston Formation

Crystaline basement 294

Figure 3. Partial stratigraphie column of the Hunting Formation, as recorded at the 'measured section' locality of Figure 2. The 'Basal Member' sequence was measured from an abbreviated section several kilometers to the south.

The three informally named members are delineated on the basis of gross lithological character; except for differences in measured thickness, they compare closely with those of Tuke et al. (1966), who studied the same section.

Fossiliferous chert samples HU-245 and HU-249 were collected from the 80 and 81 meter levels, respectively (marked with an F). 600 -

□C LU 500 CQ 2 S LU

LU LU — / / CL 400 - UJ H- I Tepee structures

g 1 ^ I Stromatolites

300 - s l-»-| Mud-cracks » / ^ I Chert, nodular and bedded

I Chert, fibro-radiate CC LU 1X1 Covered section 2 0 0 - m Z 2 LU

OC 100 - LU % ü

(m) 0 [■—■ L7-T./.‘-ir ‘ BASAL MEMBER

ASTON FORMATION 296

Figure 4. Hunting Formation fossils and modem Bangia. Scale bar in 5 equals

50 pm for 1, 2, 6, 7, 9; 20 jim for 3-5; 75 pm for 8. Figure captions include

HUPC number, slide identification and "England Finder" location coordinates

(in parentheses).

Polvbessurus bipartitus 1. HUPC 62913; HU-ST-IA (N-39-0); specimen showing upward bifurcation of a single stalk. 2. HUPC 62914; HU-ST-IA (N-41-0).

GloeodmiopsisC?) sp. 3. HUPC 62915; HU-249-1A (L-12-1); unicell with multiple sheaths. 4. HUPC 62916; HU-249-1A (K-12-4); tetrad with multiple sheaths.

Pterospermopsimorphaf?) sp. 5. HUPC 62917; HU-245-1 (D-59-1).

(?)Siphonophvcus gracile (with longitudinal striation) 6. HUPC 62918; HU-245-1 (F-68T).

Bangia atropurpurea (collected at Marblehead, Massachusetts; January, 1990) 7. Longitudinal cross-section of a multiseriate filament with marked transverse constrictions (compare with Figs. 6.2, 4.8).

Bangiomorpha antigua n. gen., n. sp. 8. HUPC 62919; HU-ST-IB (M-42-2); fossil filament showing the abrupt change from a uniseriate to multiseriate condition; multiseriate region composed of multiple spheroidal cells. 9. HUPC 62920; HU-ST-IA (N-36-4); basal portion of a filament with a multi-cellular holdfast structure.

298

Figure 5. Bangiomorpha antigua n. gen., n. sp. In pétrographie thin sections of chert cut perpendicular to bedding. Scale bar in 1 equals 250 pm for 1; 150 pm for 2.

1. Low magnification photomicrograph illustrating the vertical orientation

and substrate attachment of the fossils (Slide HU-ST-IC; England Finder

coordinates 0-41-0).

2. Fossil filaments showing a number of characteristic bangiophyte

morphological features: (a) uniseriate filament consisting of ensheathed

paired cells; (b) transverse cross section of a filament showing four-fold

radial cleavage; (c) transverse, oblique, and longitudinal cross sections of

multiseriate filaments with ca. eight-fold radial cleavage (Slide HU-ST-

IA; England Finder coordinates 0-35-0). / H Iv

W jT:

% 300

Figure 6. Comparison of Bangiomorpha antigua n. gen., n. sp. with modern

Bangia atropurpurea. Figure captions include HUPC number, slide identification and "England Finder" location coordinates (in parentheses). Scale bar in 6 equals 25 pm for 1; 33 pm for 3; 40 pm for 4; 50 pm for 2, 6; 85 pm for

5.

Bangiomorpha antigua n. gen., n. sp. 1. HUPC 62911; HU-ST-IA (0-35-0); uniseriate filament showing the hierarchical packaging of cells into pairs and quartets, indicative of intercalary cell division; all cells are enclosed in a sharply delineated outer wall.

2. FIUPC 62910; HU-ST-IA (0-37-3); Holotype. Multiseriate filament with a number of constrictions (similar constrictions are also found in m odem Bangia: Fig. 4.7).

3. HUPC 62912; HU-ST-IA (0-35-1); transverse cross section of a multiseriate filament with an eight-fold radial cleavage. Three complete wedge-shaped cells are visible and the bases of an additional five cells can be discerned.

Bangia atropurpurea (collected at Marblehead, Massachusetts; January, 1990) 4. Longitudinal cross section of a uniseriate Bangia filament showing the inner and outer cell walls and the hierarchical packaging of cell pairs.

5. Longitudinal cross section of multiseriate Bangia.

6. Transverse cross section of a multiseriate Bangia filament showing an approximate eight-fold radial cleavage and a centra) core of 'outer wall'. A

ZT' CO CO

CM in 302

PUBLISHED ABSTRACTS, ETC. (1988-1992)

BUTTERFIELD, N. J. 1992a. Pre-Ediacaran multicellular life: harbinger of a Phanerozoic radiation. North American Paleontology Convention V; Chicago. BUTTERFIELD, N. J. 1992b. Taphonomy and secular distribution of organic- walled fossils. International Geological Congress; Kyoto. BUTTERFIELD, N. J. 1991. Proterozoic plankton and its paleoenvironmental distribution. Geological Association of Canada, Program with Abstracts 16:A18. BUTTERFIELD, N. J. 1990a. A Carboniferous analogue of 'Pevtoia' and a possible taxonomic affiliation for 'Anomalocaris'. Geological Society of America, Abstracts with Programs 22:A35. BUTTERFIELD, N. J. 1990b. Taphonomy and some revised taxonomy of the Burgess Shale fossil Lagerstatte (Middle Cambrian, British Columbia).

Canadian Paleontology and Biostratigraphy Seminar, Kingston, Ontario. BUTTERFIELD, N. J. and A. H. KNOLL. 1989a. Morphologically complex fossils through the Late Proterozoic: Their application to biostratigraphy and paleobiology. Geological Association of Canada, Program with Abstracts 14:A100. BUTTERFIELD, N. J. and A. H. KNOLL. 1989b. Metaphytes and multi- cellularity in the Proterozoic: Examples from Svalbard and Arctic Canada. Geological Society of America, Abstracts with Programs 21:A146. 303

BUTTERFIELD, N. J., A. H. KNULL, and K. SWETT. 1988. Exceptional preservation of a Late Riphean shale biota, Svanbergfjellet Formation, Spitsbergen. Geological Association of Canada, Program with Abstracts 13:A16. BUTTERFIELD, N. J. and R. H. RAINBIRD. 1988. The paleobiology of two Proterozoic shales. Geological Society of America, Abstracts with Programs 20:A103. KAUFMAN, A. J., A. H. KNOLL, N. J. BUTTERFIELD, and J. M. HAYES. 1990. The Neoproterozoic carbon isotope record: its validity in chemostratigraphic correlation. Geological Society of America, Abstracts with Programs 22:A114. KNOLL, A. H., and N. J. BUTTERFIELD. 1989. Large, morphologically complex phytoplankton from Proterozoic cherts and shales. American Journal of Botany, supplement to Volume 76. p.l68. KNOLL, A. H., N. J. BUTTERFIELD, and A. J. KAUFMAN. 1990. The Neoproterozoic Akademikerbreen Group, Spitsbergen: characteristics and correlation. Geological Association of Canada Nuna Research Conference. Pp. 20-21.

ASMERMOM V., S. B. JACOBSEN, A. H. KNOLL, N. BUTTERFIELD, and K. SWETT. 1991. Sr isotopic variations of Neoproterozoic seawater: implications for crustal evolution. Geochimica et Cosmochimica Acta 55:2883-2894. KNOLL, A. H. and N. J. BUTTERFIELD. 1989. New Window on Proterozoic Life. Nature 337:602-603. Reprinted from Nature. Vol. 334. No. 6181. pp. 424-427. 4 August 1988 Mdanillan Müi’uziiu's Lt(l.. I9SS

bergfjellet carbonates are similar to those reported for Upper Exceptional preservation of fossils in an Riphean rocks elsewhere, but distinctly different from reported Upper Proterozoic shale Vendian or Palaeozoic values (refs 8, 9 and unpublished data). Metaphytes (‘seaweeds') are a conspicuous component of the Svanbergfjellet shale benthos. One population exhibits exten­ N. J. Butterfield, A. H. Knoll & K. Swett* sive, branching filamentous thalli composed of large (6-50 x Botanical Museum, Harvard University, Cambridge, 50-800 p.m) cylindrical cells. Adjacent cells are attached via Massachusetts 02138, USA distinctively thickened septal plates (Fig. 2a, b). B ranching is • Department of Geology, University of Iowa, Iowa City, intercalary, with branch insertion near the apices of axial cells Iowa 52242, USA (Fig. 2a). It is impossible to ascertain the complete height of the algae from which these fossils were derived, but preserved fragments are up to 1 cm long. This population has close Late Proterozoic organisms must have been diverse and widely morphological analogues among extant cladophoralean green distributed, but in general their fossil record Is both taxonomically algae (Ulvophyceae), including species that today form ‘algal and environmentally limited. Exceptional preservation of Pro­ carpets' below wave base in tropical reef areas'®. terozoic fossils is not unknown, but it is usually associated with More enigmatic are robust, complex branched structures 10- siliciRed carbonates from restricted peritidal or playa lake environ­ 60|im in diameter and up to 1 mm long found in the lower ments'"^, We report here an exceptionally well preserved and ' shale horizon (Fig. 2c). These show no obvious cellularity, distinctive assemblage of Late Proterozoic fossils from subtidal consisting instead of robust fibrils that branch repeatedly. marine shales. In addition to the sphaeromorphic acritarchs and Among living algae, the rhizoids or hold-fasts of certain of the cyanobacterial sheaths routinely preserved In Proterozoic rocks, hetcrotrichous Chaeiophorales (Cholophyceae, jensu srricro) this assemblage includes multicellular algae ( seaweeds*), a diverse provide a close structural analogue. assortment o f morphologically complex protistan vesicles, and Solitary filaments (**5xl00p.m) that terminate in bulbous probable heterotrophic bacteria. Thus, it provides one of the clear­ (*=35)xm) swellings occur sporadically in thin sections (Fig. est and most taxonomically varied views of Proterozoic life yet 2d); these may be related developmentally, or at least be com­ reported. parable functionally, to large (30-110 pm) vesicles bearing 2-7 The fossils occur in the Svanbergfjellet Formation, a tubular extensions (Fig. 3d). Given their marked asymmetry SSO-650-m succession of carbonates and subordinate shales and pliant walls, these structures resemble, and may well rep­ exposed in the glaciated mountains and coastal cliffs of north­ resent, germinating zoospores of filamentous protists—modern eastern Spitsbergen’ (Fig. 1). Fossiliferous horizons occur within analogues include the xanthophycean alga Vaucheria. a distinctive lOO-200-m sequence of massive stromatolitic bio- Oscillaiorian and chroococcalean cyanobacteria in the shales stromes interbedded with subequal thicknesses of shale and, compare closely with those reported from silicificd Proterozoic lower in the formation, within a 5-m shale bed immediately carbonates. Isolated, short trichomcs (Fig. 3/) may be alloch- beneath a laterally extensive stromatolitic bench. The thonous, but dense tangles of filaments that cover bedding planes fossiliferous shales are green to black, entirely siiiciclastic, and indicate that in siiu benthic microbial mats were common in extremely fine grained, with most mineral grains significantly muddy environments, just as they were in coeval carbonates. finer than 1 n m . Sub-millimetric lamination indicates that they Geochemical signatures indicate that heterotrophic bacteria were deposited in quiet subtidal waters protected from strong were present in Proterozoic communities, but their remains have wave activity. Thin clay laminae provided rapid and efficient seldom been observed directly". Svanbergfjellet shales contain barriers to continued physical and biological degradation. Regu­ populations of rod-shaped and filamentous fossils that simply lar lamination also facilitates study by means of bedding-parallel on the basis of size (<2 pm) might be considered as possible thin sections^, which retain not only delicate and otherwise heterotrophs. More persuasive is the preserved distribution unrecoverable fossils, but also biologically significant details of (inferred behaviour) of these populations within individual fossil orientation and bedding-plane distribution. laminae. For example, small (1-1.5 x 5-10 pm) rods occur Although the age of the Svanbergfjellet fossils is poorly con­ repeatedly as dense colonies within local patches of organic strained by radiometric data, a Late Riphean age of approxi­ detritus (Fig. 3a, 6), with individuals often aligned toward or mately 700-800 Myr BP is indicated on several grounds. The along discrete organic particles. Other possible heterotrophic fossiliferous sediments sit some 2,200 m below lowermost Cam­ bacteria include clusters and chains of smaller rods (0.4 x brian strata and 1,100-1,300 m beneath the base of Vendian 0.7 pm), occasionally branching filaments that resemble extant glacial beds; both stromatolites and acritarchs in this and associ­ actinomycetes. ated formations are distinctly Late (but not latest) Riphean in Superimposed on these benthic associations are a wide variety aspect’"^; and carbon and strontium isotopic ratios for Svan­ of presumed allochthonous remains. Megascopic forms include

a Cm) 8000 500

7000 Fig. 1 Map of Spitsbergen showing locations 400 g # G - G 3 of fossiliferous shales in coastal exposures at 6000 m ^♦G -61, 62 Geerabukta (G) and in nunatuk walls along P* M - O - 3 3 Polarisbrccn glacier (P). Also shown are geo­ 5000 logical columns illlustrating the general 300 stratigraphy of the Upper Proterozoic Akademikerbreen Group and the stratigraphy 4000 of the Svanbergfjellet Formation in a section 200 measured at Geerabukta. Stratigraphie posi­ 3000 tions of fossiliferous shales are indicated by • P - 2 9 4 5 asterisks followed by sample numbers. 2000 100 1000 Fig. 2 a, h, Multicellular algae comparable to some extant cladophoralean green algal species (bar in c = 60 p,m for a, ICO p.m for b ). c. Robust, branching structure comparable to the prostrate holdfasts of certain extant chactophcralcan green algae; for example, Sfigeo- clonium pusilium (b ar= ISO^im). d. Spheroidal vesicle with filamentous extension, tentatively identified as the germinating zoospore of a filamentous protist (bar in c = 60 p.m). fl. h and d are from bedding- parallel thin sections; c is from an acid maceration.

a

Fig. 3 a. Cluster of probable heterotrophic bacterial rods. 6, Higher power photomicrograph showing the morphology of individual rods, c, g. Large process-bearing acritarch, which in g is preserved within a larger unomamenled vesicle, d. Vesicle with three asymmetrically placed filamentous extensions, comparable to the germinating zoospores of some extant protists, for example. Vaucheria setsilis. e. Ornamented acritarch, comparable in general morphology to Paleozoic fossils assigned to the genus A c r u m .f, Well preserved oscillatorian cyanobacterial trichome. The fossils in o -/are from bedding-parallel thin sections; g is from an acid maceration. Bar in c = 50 jtm for o. c, d and e; =20 nm for 6; =110^m for/; and = iTO^m for g.

Chuaria, Tawuia, and thin-walled spheroidal vesicles 200- previously unreported forms. Particularly striking are large ( 150- 1,500 |xm in diameter. Smaller unornamented acritarchs occur 250 uni), opaque, bristle-covered vesicles that bear long (up as isolated individuals, small clusters, or large populations of to 60p,m), terminally flared processes (Fig. 3c). Occasionally, medially split spheroids that cover entire bedding planes, sug­ these are preserved within still larger (300 p.m) unornamented gesting the episodic occurrence of plankton blooms or mass vesicles, strongly supporting their interpretation as cncystmeni cncysimcnis. More distinctive arc the at least eight taxa of structures (Fig. 3g). Their phylogenetic relationships remain morphologically complex acritarchs. In addition to the Late un clear. Riphean genus Trachyhysirichosphaera, the Svanbergfjellet In concert with equally distinctive, but specifically different assemblage includes forms comparable to the predominantly acritarch assemblages in latest Proterozoic shales and cherts Palaeozoic genera A crum (Fig. 3d) and Micrhyslridium, large from China'^’^ and Australia (W. Zang and M. R. Walter, process-bearing vesicles resembling late Vendian acritarchs personal communication), Svanbergfjellet acritarchs suggest the assigned to Cotnasphaeridium a n d Baltisphaeridium, and several potential for a much enhanced Upper Proterozoic microfossil zonation. Coupled with emerging data on isotropic chemo- ved in the shales indicates that metazoans larger than SO-100 pm stratigraphy". magnetostratigraphy'’, and sequence stratigra­ in diameter were not present in at least the local benthos. Perhaps phy'*, complex acritarchs hold promise for a major improvement most important, the Svanbergfjellet fossils demonstrate that a in our ability to resolve Late Proterozoic time. much wider sampling of Proterozoic life is available than has Palaeobiologically, the Svanbergfjellet biota provide evidence hitherto been appreciated. fora substantially pre-Ediacaran radiation of at least two classes This research was supported in part by grants from the NSF of metaphytic green algae, as well as morphologically complex and NASA, an NSERC pre-doctoral fellowship to NJ.B., and unicellular protists. At the same time, the fine lamination preser­ a John Simon Guggenheim Foundation fellowship to A.H.K.

Received 17 March; accepted 16 June 1988. 8. Knoll. A. H. tta l Sature 321. 832-838 (1986). 1. K n o ll. A . H. P*tiL Trcni. A Joe 31 IB, 111-1^(19851. 9. Veizcr. J. et aL Ceochim. eestmyhim. Aeia 47. 295-302 (1983). 1 Southgate. P. N. C e o lo ty 14, 683-6S6 (19861. 10. Round, P. E. 77ir Eeohgy of Algae (Cambridge Univ. Press. Cambridge. 1981). 3. Wilaofi. C. B. C t o l M ag. n. 8 9 -1 1 6 ( 1961 ). 11. Lanier. W, P. J. led . fV frot S I , 89-99 (1981). 4. Hofodyiki. R. J. A B lo e tie r. 8. Jele^ce 199. 682-684 (19781. 12. Awramik. S. M. r i aL S a tu r e 315, 655-659 5. Ja n k a u fk a a . T . V. in Stnioiypt of the Riphean: Paleoniotogy and M eomagneiKS (ed. Keller. 13. Yin, L PalaeonL Cathayana 2, 229-249 (1984), B. M.) 84-120 (USSR Acad. Sci. Ceol. Inal. Nauka. Moscow. 19821. 14. Magaritz, M. Holser, W, T. A Kirschvink, J. L S a tu re 320. 258-259 (1986). 6. V idal, G . t Knoll, A. H. Ceol Soc Am. Mem. 161, 265-277 (1983). 15. Kirschvink, J. L Geoi M ag. IIS. 139-150(1971), 7. R a a b e n . M . E . A Z a b ro d in . V. E. Tr. CeoL Inst. Akad Nmik S.S.S.R. 217. 1-130 (19721. 16. Chhttie-Blick. N.. Crotzinger, J. P. A van der Borsch, C. C. C e o h g y 16, 100-104 (1988). A Bangiophyte Red Alga from the Proterozoic of Arctic Canada

N i c h o l a s J. B utterfield, Andrew H. K noll, Keene Swett

Stlicified peritidal carbonate rocks o f the 1250- to 750-million-year-old Hunting Formation, Somerset Island, arctic Canada, contain fossils o f well-preserved bangio­ phyte red algae. Morphological details, especially the presence o f multiseriate filaments composed o f radially arranged wedge-shaped cells derived by longitudinal divisions from disc-shaped cells in uniseriate filaments, indicate that the fossils are related to extant species in the genus Bangia. Such taxonomic resolution distinguishes these fossils from other pre-Edicaran eukaryotes and contributes to growing evidence that multicellular algae diversified well before the Ediacaran radiation of large animals.

ulticellular o r g a n i s m s northwest coast, litdc altered middle to up­ characterized by cellular integra­ per Proterozoic sedimentary rocks sic on tion and differentiation occur in crystalline basement of the Canadian Shield M all four eukaryotic kingdoms, including (2). at The sedimentary sequence comprises least eight separate phyla in the Protista ( /). two unconformity-bounded units. The low­ Most of these experiments in complex multi- er unit, the Aston Formation, consists large­ cellularity must have made their first appear­ ly of quartz arenite, whereas the overlying ance well before the Edicaran radiation o f Himdng Formation is dominated by inter­ large animals (600 million years ago. Ma), tidal to supra tidal carbonate rocks. The fos­ but fossil evidence is limited and in many siliferous horizons are thin (1 cm thick) cases equivocal. Interprctational problems chert layers in stratiform laminated shallow- result from the lack of preserved cellular subtidal to intertidal carbonate rocks of the structure in many fossils, the nondiagnostic Hunting Formation, approximately 150 m nature of such structures as may be pre­ above the Aston-Hunting contact. served, and the absence of biochemical and Him ting deposition is only broadly con­ ultrastructural information. In this report, strained in time, bracketed by two genera­ we describe well-preserved fossils in 1250- tions of mafic dikes (2) that are dated else­ to 750-million-year-old rocks from arctic where at 1267 ± 2 M a (3) and 723 ± 3 Ma Canada in which the habit and patterns of (4). Constituent stromatolites (stratiform to cell division are sufiiciendy distinctive to simple Baicalia-xypc columns) and associated permit their assignment to the bangiophyte microfossils (dense filamentous mats, spher­ red algae. oids, and the stalk-forming cyanobacterium The fossiliferous rocks were collected ott Polybessurus) are useful as paleoenvironmen­ Somerset Island, a narrowly separated exten- ‘ tal indicators but cannot provide finer time sion of the Boothia Peninsula in the south­ resolution. On the other hand, several analy­ eastern part of the Canadian arctic archipela­ ses of Hunting carbonate rocks indicate that go (District of Franklin, Northwest Terri­ they have values between -H2.5 and tories) (Fig. I). In the Aston Bay area on its -t-3.8 per mil (5), values lighter than those typical of 700- to 800-million-year-old car­ bonate rocks in the Canadian arctic and N. J. Butterfield and A. H. Knoll, Botanical Museum, elsewhere (5, 6). The Hunting isotopic sig­ Harvard University, Cambridge, MA 02138. K Swett, Department of Geology, University of Iowa, nature more closely approximates that of Iowa City, LA 52242. un weathered dolomite from the 1100- to

104 SCIENCE, VOL. 250 1200*million*ycar*old Mescal Limestone, istics of sub-rissuc grade organisms arc con­ Arizona (7), and 1150- to 1300-million- vergent or ccophcnot)’pic (9, 10, 13, 16), year-old marble of the Upper Marble unit, such underlying cellular patterns arc likely to Grenville Series, New York (S). Thus, the be of considerable phylogenetic significance. carbon isotopic evidence, although not de­ Pfianetozcc Along with its other cellular and extracellu­ □ seOimeniary rocKs finitive, suggests that Hunting sedimenta­ [— I Hunting and lar details, the radial pattern of cells in the tion might have occurred closer to the older Aston locmaions fossil material strongly supports its assign­ r n Crystalline rather than the younger dike-defined chro­ ' basement ment to the bangiophyte red algae, possibly nological boundar)'. to the level of family (Bangiaceac). The fossils are moderately to extraordi­ The fossils arc not identical to extant narily well preserved, having been pcrminer- species of Bati(;ia. The modem alga varies alizcd by ven* early diagenctic chert. Pétro­ widely in filament width but averages 100 graphie thin sections cut perpendicular to p,m diameter (10). The Proterozoic fila­ bedding have so far exposed over 500 fossil ments fall in the size range of modem filaments. In their most simple (and most but are generally smaller, having a mean common) state, the fossils consist o f stacked diameter o f 26 ^im (« = 519; SD = 7.4 p.m) disc-shaped cells enclosed in a relatively Fig. 1. Schematic geological map of Somerset and a minor mode of 53 p.m in the case of transparent enveloping sheath and form un- Island showing the location of fossiliferous Hunt­ multiseriate forms (« = 13; SD = 9.1 jxm). branchcd uniseriate filaments 15 to 45 jxm ing Formation strata. Maximum measured diameter in our sam­ in diameter and up to 2 mm long (Fig. 2). ples is 72 fim. The fossils also differ in Almost all o f the fossils are oriented vertical­ 4D). In more mature, multiseriate forms possessing multicellular rhizoids or holdfast ly and the)’ commonly occur in groups this ‘outer’ wall also occupies the central structures (Fig. 3), whereas B ai^ia has long, attached to locally stabilized surfaces (Fig. core of the filament (Fig. 4F); a sub-micron nonseptate rhizoids descending from a num ­ 2A); the basal ends of the filaments arc thick “cuticle” gives the filament a sharply ber o f basal vegetative cells (12). Multicellu­ elaborated into multicellular structures that delineated boundar)' layer {14). Unlike the lar rhizoids are nevertheless common in the appear to ser\c as attachment rhizoids (Fig. apical cell division t)'pical of florideophvte red algae, and holdfasts essentially identical 3). Larger diameter specimens (>45 p.m red algae, vegetative growth in bangio* to those of the fossils are found in the across) of these same populations are gener­ ph\'tes generally occurs by transverse inter- filamentous bangioph)tes Brythroirichûi ally multiseriate with a number of cells calar>’ cell division {/2, 13, /5); consequent­ (Enthropcltidaccac) and P c rp h y ra (Bangia- occupying any given level along the filament ly, Banfjia filaments often show a hierarchical ccae) (12, 13). (Figs. 2B and 4B). Transverse cross sections packaging of successive cell generations Arc there other possible interpretations of of these more complex forms (Fig. 4C) (Fig. 4D). In the fossil filaments both a these fossils? Stigonematacean c\anobacteria reveal that the component cells are wedge- differentiated inner and outer cell wall and a often form a comparably thick external shaped and arranged axially around a central pattern of transverse intercalai")' cell divi­ sheath and are commonly multiseriate. core. The specimen in Fig. 4C prcser\es sions are clearly evident (Figs. 2B and 4A). However, these prokaryotes are also charac­ three of what were originally eight such Mature filaments of Banf^ia generally be­ terized by true branching and several t)'pcs cells, and in different focal planes of the come multiseriate through longitudinal in­ of differentiated cells (for example, hetero- same specimen, four adjacent layers show a tercalai")' division {12, 14) (Fig. 4E). They cysts and akinetes), which are not obserxed similar pattern of cell division. Other fila­ do not, however, become simple construc­ in the fossils. More importantly, the multi­ ments have a fourfold radial cleavage (b in tions of randomly arranged, equidimen- seriate thallus of stigonemataccans appears Fig. 2B). sional cells. N ot only is the position o f each to be a relatively loose knit association of In most details of their morpholog)', and parent cell broadly retained in the multiscr- cells and filaments lacking any particular particularly in their distinctive patterns of iate filament, but initial cleavage of the disc- cell division, these fossil filaments compare shaped cells is also radially oriented. This closely to the gametoph)'tic ( IN) generation orientation yields a distinctive pattern of of species belonging to the modem red algal wedge-shaped cells arranged axially around genus Bangia (Bangiaceac, Bangiales, Bangio- a central core (12, 14) and is particularly phyceac, Rhodophyta). Bangia is a cosm opoli­ apparent in transverse cross sections of ma­ tan seaweed that today ranges from boreal to tu re Bangia filaments (Fig. 4F). Because subtropical oceans and from marine waters to many of the gross morphological character­ freshwater lakes and streams {9 -Î3 ). It toler­ ates extended emersion and desiccation and is usually found in the upper intertidal zone of Fig. 2. Populations of fossil bangiophyte fila­ marine coastlines and at or above the water­ ments in pétrographie thin sections of Hunting line in fresh water; it commonly occurs with Formation chert cut perpendicular to bedding. filamentous cyanobacteria above the zone (A) Low magnification photomicrograph illus­ occupied by green algae (9, ÎÎ). trating the vertical orientation and substrate at­ tachment of the fossils; scale bar equals 0.25 mm. In its most simple form, gamctophv’tic (B) Fossil filaments showing a number o f m or­ is characterized by unbranched uni­ phological features characteristic of bangiophvtcs; seriate filaments composed of stacked, disc- a, uniseriate filament consisting of ensheathed shaped cells. The polysaccharide cell walls paired cells; b. transverse cross section o f ' fila­ ment showing fourfold radial cleavage; c, trans­ arc distinctly biphasic with an inner wall verse. oblique, and longitudinal cross sections of defining individual cells and an outer wall multiseriate filaments with approximate eight­ enveloping the whole organism (N) (Fig. fold radial cleavage; scale bar equals 0.15 mm.

5 OCTOBER 1990 REPORTS 105 branching filamentous forms, both septate (75). The Bangiophyccac arc generally con­ (20) and cocnocytic (21). Macroscopic sidered to be less advanced than, although tawuiids and lonfengshaniids (1100 to 700 not necessarily ancestral to, the Floridco- Ma) have a worldwide distribution and rep­ phyccae (73). As gamctophytic bangio- resent at least probable metaphytes (22), as phncs, the Hunting Formation fossils pre­ do filamentous macrofossils in 1400 mil- date the earliest record o f probable florideo- lion-year-old deposits from Montana and phncs (76) by at least 200 million years and China (23). Problematic bedding plane thus provide at least circumstantial support markings in middle Proterozoic sandstones for a bangiophnc to floridcophne phylo­ of Western Australia may be imprints of a genesis. V^iether the bangiophne life q'cle relatively large filamentous seaweed (24). In included an alternation of generations by latest Proterozoic rocks a calcareous, en­ Hunting time is speculative at this point as crusting phylloid alga (25) and a variety of we have found no evidence of a chonchocc- structured multicellular fossils (26) have lis phase in our samples. The oldest reported been documented, and filamentous organic- chonchocelis comes from Silurian rocks in walled vendotaenids are generally, although Poland (29). not universally (27), thought to be of algal Molecular data from both small subunit origin (2&). None of these fossils, however, (30) and large subunit (37) ribosonial RNA Fig. 3. Basal portion o f a fossil filament showing can be placed confidently into any extant suggest that the three principal cladcs con­ the multicellular rhizoid structure; scale bar higher-ordcrtaxon. The same can be said for taining multicellular algae (Rhodophna, equals 30 ^m. most of the organic-walled fossil algae of Chlorophna, Phaeophvta) diverged during Phanerozoic age (16). In contrast, the clearly a brief but marked radiation relatively late in geometrical or developmental regularity' interprétable morphology of the Somerset cukaiyotic histoiy*. This interpretation is (/7); the distinctive partem of radial cell Island fossils provides a detailed datum broadly supported by the moderate diversity' division characteristic of both Batiffia and the point for the reconstruction of early red of mid to late Proterozoic fossil metaphytes. fossil material is unknown in any cyanobac­ algal and protistan phylogcny. Multiccllularit)', however, is almost certainly terium (JH). Other extant eukaryotes, such The two classes of the Rhodophyta, Ban- a derived condition, and Tappan's (32) in­ as species of the green algal family Schizo- giophyccac and Florideophyccac, are distin­ terpretation of the 2000-million-ycar-old meridaceae, likewise form uniseriate and guished on the basis o f thallus form, growth Gunflint microfossil Losphavra as a Pury^hyri- multiseriate filaments (fJ), but these also habit, and various intracellular characteris­ Jium-like rhodophyte, although controver­ lack radially oriented cells. tics (12. 73, 75). They appear to be linked sial, underscores the probability that unicel­ These fossils are neither the only nor the phylogenetically by the microscopic chon- lular red, green, and chromophyte algae oldest Proterozoic metaphytes. Undoubted chocclis phase (2N sporophyte) observed in antedated their multicellular descendants by multicellular algae arc known from seseral some species of Bani^ia and Porphyra (Ban- a significant interval. Although the timing late Riphean (900 to 700 Ma) shale biotas glaceae); the conchocelis cxliibits a number of early algal diversification remains incom­ and include coenobial colonics (19) and of otherwise horideophyte characteristics pletely known, the Hunting Formation fos­ sils contribute to a growing paleontological record which indicates that sc\’cral and per­ Flg. 4. Comparison o f fossil filaments (A to C) with haps all major groups of multicellular algae modem Bangia atropurpurea were well established long before the Edia­ (collected at Marblehead, caran and Cambrian radiations. Massachusetts, January', 1990) (D to F). (A) Unisc'f. iate filament showing the hi­ REFERENCES AND NOTES erarchical packaging o f cells 1. L. \V. Buu, The hi'oluiioti o f InJii'iJujlity (Pnncctcjn into pairs and quartets in­ Un«v. Prcsj, Pnnccton, 1987). dicative o f intercalary cell di­ 2. W. D. StwiTT, Ceci Sun'. Can Vap ( 1987), vision; all cells are enclosed 3. A. N. LeChcmiiunt and L, M. Hcaman. lianU in a sharply delineated outer Pljftet Sa. U it 96. 38 (1989). wall. (B) Multiseriate fila­ ♦. L. M. Hcaman and R. H. Rainbird. Ceci .Asia Can. Profi Ahiir. IS. A55 (1990). ment with a numbci of con­ r .’:r T - 5. N. I Burtcfficid, A. H. Knoll, |. M, Mayes, unpub­ strictions (similar constric­ lished data. &"C is defined as tions are also found in mod­ em Bançia). (C) Transscrse cross section o f a multiscr* 10' iate filament with an eight­ fold radial cleavage. Three and reported in parts per mil relative to the Pee Dec complete, wedge-shaped Bclcmnitc (PDB) standard cells arc visible, and the 6- A M Knoll, f M, Mayes. A J Kaufman. K Swett. bases of an additional five I- B. lumbcrt. .Varnre 32X. 832 (1986) cells can be discerned. (D) 7. ,M A. Beeunas and I.. P. Knauth. (,rcl S a .4m Longitudinal cross section li„H 96, 737 (1985) of a uniseriate Ban^iia fila­ 8 I F Whelan. R. C) Rve. W dcI.orraine, 11, Oh ment showing the inner and moto. .4m y .Su 290. 396 ( 1990) outer cell walls and the hier­ 9 R C Sheath and K, M Ode. I n,y,X 16. 412 (1980) archical packaging o f cell pairs. (E) Ixingirudinal cross section of multiseriate (F) Transverse 1 0 lH,y,chy,a 23. 383 (1984) cross section o f a multiseriate filament showing an approximate eight-fold radial cleavage and a 11. R G. Sheath. K L VanAisrsnc. K .M C'xilc, / central core o f “outer wall.” Scale bar [shown in (F)j equals 20 vim for (A); 50 vim for (R); 30 vvm I V '’/ 21. 297 (198.4), for (Cl; 35 p.m for (D); 80 jim for (E); and 50 p.m for (F). 12 D 7 Garban. G I Man sen. R. F .Stagcl. .Synu 13.

106 SCIENCE, VOL. 250 137(1980). 24. K. Grey and I. R. Williams, Precambr. Res. 46, 307 13. H. C. Bold and M. J. Wynne» Intwduaion to the Algae (1990). (Prcndcc-Hall, Englewood Cliffs, NJ, ed. 2, 1986); 25. S. W. F. Grant and A. H . Knoll, Ceol. Soc. Am. F. E. Fritsch, The Structure and Reproduction of the Abstr. Prog. 19, 681 (1987). Algae (Cambridge Univ. Press, Cambridge, 1965). 26. Zhang Y , Lethaia 22, 113 (1989). 14. K. M. Cole, C. M. Park, P. E. Reid, R. G. Sheath, J. 27. G. Vidal, ibid., p. 375. Phycol. 21, 585 (1985). 28. M. B. Gnilovskaya, Ed., Vendotaenids o f the East 15. K. Cole and E. Conway, Phycologia 14,239 (1975). European Platform (Nauka, Leningrad, 1988). 16. C. D. Walcott, Smithsonian Misc. Coll. 67, 217 29. S. CampbcU, J . Phycol. 19, 25 (1980). (1919); W. L, Fry, Rev. Paiaeobot. Palynol. 39, 313 30. D. Bhattacharva, H. J. Elwood, L. J. Goff, M. L. (1983). Sogin, ibid. 26, 181 (1987). 17. T. C. Martin and J. T. Wyatt, J . Phycol. 10, 57 31. R. Perasso, A. Baroin, L. H.

5 OCTOBER 1990 REPORTS 107 PLEASE NOTE

Copyrighted materials in this document have not been filmed at the request of the author. They are available for consultation, however, in the author’s university library.

Appendix B - E, 311-342

University Microfilms International