University of Alberta

Lower Palaeozoic arthropod traces: Clarity in the mud

by

Stacey Gibb

A thesis submitted to the Faculty of Graduate Studies and Research in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

Earth and Atmospheric Sciences

©Stacey Gibb Spring 2011 Edmonton, Alberta

Permission is hereby granted to the University of Alberta Libranes to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms

The author reserves all other publication and other nghts in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be pnnted or otherwise reproduced in any material form whatsoever without the author's prior wntten penmssion Library and Archives Bibliotheque et 1*1 Canada Archives Canada Published Heritage Direction du Branch Patrimoine de I'edition

395 Wellington Street 395, rue Wellington OttawaONK1A0N4 OttawaONK1A0N4 Canada Canada

Your Tile Votre reference ISBN: 978-0-494-80955-6 Our file Notre r6f6rence ISBN: 978-0-494-80955-6

NOTICE: AVIS:

The author has granted a non­ L'auteur a accorde une licence non exclusive exclusive license allowing Library and permettant a la Bibliotheque et Archives Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par telecommunication ou par I'lnternet, preter, telecommunication or on the Internet, distribuer et vendre des theses partout dans le loan, distribute and sell theses monde, a des fins commerciales ou autres, sur worldwide, for commercial or non­ support microforme, papier, electronique et/ou commercial purposes, in microform, autres formats. paper, electronic and/or any other formats.

The author retains copyright L'auteur conserve la propriete du droit d'auteur ownership and moral rights in this et des droits moraux qui protege cette these. Ni thesis. Neither the thesis nor la these ni des extraits substantiels de celle-ci substantial extracts from it may be ne doivent etre imprimes ou autrement printed or otherwise reproduced reproduits sans son autorisation. without the author's permission.

In compliance with the Canadian Conformement a la loi canadienne sur la Privacy Act some supporting forms protection de la vie privee, quelques may have been removed from this formulaires secondaires ont ete enleves de thesis. cette these.

While these forms may be included Bien que ces formulaires aient inclus dans in the document page count, their la pagination, il n'y aura aucun contenu removal does not represent any loss manquant. of content from the thesis.

1*1 Canada This dissertation is dedicated to my Mom...

... who passed too young and never saw this to completion.

May she forever rest in peace. ABSTRACT

Ichnology is the study of trace fossils and therefore the inferred behaviour of organisms that may have created the ichnofossils. Through analyses of traces from numerous different localities in the southern Rocky Mountains of Canada, the coast of Wales, and the deserts of Morocco and central Australia, from the lower Cambrian to the Upper Devonian, numerous conclusions were made. From the lower Cambrian Gog Group of Canada, it is evident that the 'Cambrian

Explosion' was indeed an explosion of epic portions, not only in the advent of hard exo- and endoskeletons, but also of novel behaviours. The Cambrian of

Morocco provided evidence for a relationship between sea floor current directions and the orientations of the trace Selenichnites in a shallow clastic palaeoenvironment. The Ordovician of Morocco provided direct evidence of a putative tracemaker, Asaphellus aff.fezouataensis, in very close physical association with the trace that it made Rusophycus carleyi, and allowed me to speculate on the form of the, yet to be discovered, ventral appendages of that trilobite. This locality also provided further insights on the importance of

Microbially Induced Sedimentary Structures in the formation and preservation of ichnofossils. The Devonian of Wales showed that trace fossils can provide the first evidence of the migration of an organism into a new environment

(xiphosurans into fresh water); it provides additional evidence of preferential orientations of Selenichnites in relation to currents occurring at the surface of the substrate; and shows that turbulent hydrodynamic influences assisted in the creation of the traces and perhaps caused the trace maker to burrow into the substrate.

The synergy of these apparently quite different chapters is that the ichnofossils were all plausibly made by arthropods. Preliminary neoichnological research provided a behavioural reason to justify the morphological distinctions between the ichnogenera Cruziana and Rusophycus. Further research is only going to provide more insight into the 'other dimension' of palaeontology - ichnology: therefore, the ethology of organisms that cannot be as definitely acquired from a static body fossil. ACKNOWLEDGEMENTS

The academic exchange, support, and friendship of my supervisors, Brian Chatterton and George Pemberton, was invaluable. I also acknowledge Murray Gingras who was a valuable member of my supervisory committee. My examiners, both in the final and candidacy examinations, provided challenging, though entertaining and thought provoking experiences, for this I thank: Nigel Hughes, Rich Palmer, Lindsey Leighton, Alex Wolfe and Heather Proctor. Funding from the University of Alberta and Department of Earth & Atmospheric Sciences, in the form of scholarships, teaching assistantships, research assistantships, office space, and administrative assistance, and financial support from the NSERC Discovery Grants of B. Chatterton and G. Pemberton. The supervisory assistance of Derek Siveter while I was a 'visiting scholar' at the University of Oxford is greatly appreciated. I thank the University of Oxford, the Oxford University Museum of Natural History and Wolfson College for acceptance and hospitality. The following individuals, whose help time and effort was invaluable in the field: B. Chatterton, H. ait H'ssaine, R. McKellar, D. Molinaro, D. Nordheimer, and A. Madden. Institutions which either allowed and/or provided assistance for research include: Parks Canada (granting Research permits JNP- 2005-2178 and LL-2010-6392); Royaume du Maroc, Ministere de I'Energie, des Mines, de l'Eau et de l'Environnement (collection and export permits); in Australia, the Australian Museum, the Maryvale Station, and the Idracowra Station; Oxford University Museum of Natural History; and the University of Alberta and more specifically the Department of Earth & Atmospheric Sciences. I would like to thank the following for their support and assistance through most, if not all of my dissertation: my Mom and Dad, my sister and her family, my grandma, grandpa and other extended family members (you know who you are), R. McKellar, B. Frizzell, R. Madden, A. Lindoe, J. Bohun, J. Duffy, S. Hesse, K. Hibbard, and the many sports teams/clubs I have been involved with. Finally, it would not have been possible without the love and support of Art Madden. TABLE OF CONTENTS

CHAPTER 1: INTRODUCTION Introduction 1 Localities 2 Cambrian southern Rocky Mountains of Canada 2 Cambrian of Morocco 3 Ordovician of Australia 3 Ordovician of Morocco 4 Devonian of Wales 4 Previous Research 5 Methodology 6 Format 7 References 9

CHAPTER 2: CAMBRIAN OF CANADA

Introduction 11 Geographical Location 12 Geological Background 13 Systematic Palaeontology (Ichnology) 16 Cruziana d'Orbigny, 1842 17 Cruziana billingsi Fillion & Pickerill, 1990 18 Cruziana dispar (Linnarsson, 1869) 19 Cruziana irregularis Fenton & Fenton, 1937 20 Cruziana Jenningsi Fenton & Fenton, 1937 21 Cruziana mesodelta isp. nov. 23 Cruziana navicella Fenton & Fenton, 1937 24 Cruziana omanica Seilacher, 1970 26 Cruziana penicillata Gibb, Chatterton & Pemberton, 2009 26 Cruzianaplicata Crimes, Legg, Marcos & Arboleya, 1977 27 Cruzianaproblematica (Schindewolf, 1928) 29 Cruziana ramellensis (Legg, 1985) 30 Cruziana rugosa d'Orbigny, 1842 31 Diplichnites Dawson, 1873 32 Diplichnites obliquus isp. nov. 33 Diplichnites twelvetreesi (Chapman, 1928) 34 Monomorphichnus Crimes, 1970b 36 Monomorphichnus bilinearis Crimes, 1970b 36

Marcos & Arboleya, 1977 37 Monomorphichnus trilinearis isp. nov. 37 Rusophycus Hall, 1852 38 Rusophycus arizonensis (Seilacher, 1970) 39 Rusophycus eutendorfensis (Linck, 1942) 41 Rusophycus latus Webby, 1983 43 Rusophycus mesodeltus isp. nov. 45 Rusophycus radwanskii Alpert, 1976 47 Rusophycus subnotous isp. nov. 48 Rusophycus unilobus (Seilacher, 1970) 49 Rusophycus vie torus isp. nov. 50 Conclusions 52 References 57

CHAPTER 3: CAMBRIAN OF MOROCCO

Introduction 98 Geological and Geographical Setting 99 Material 100 Systematic Ichnology 101 Selenichnites (Romano and Whyte, 1987) 101 Selenichnites tacfihtus n. isp. 101 Interpretation 105 References 110

CHAPTER 4: ORDOVICIAN OF MOROCCO

Introduction 122 Regional Setting 123 Geographical Location 124 Geological Setting 124 Palaeoenvironmental Reconstruction 124 Smoking Gun Scenario 127 Hard Evidence 128 Soft Evidence 130 Systematic Palaeontology 131 Ichnogenus Rusophycus Hall, 1852 131 Rusophycus carleyi (James, 1885) 131 Family Asaphidae Burmeister, 1843 134 Genus Asaphellus Callaway, 1877 134 Asaphellus aff. fezouataensis Vidal, 1998b 134 Conclusions 137 References 138 CHAPTER 5: ORDOVICIAN OF AUSTRALIA Introduction 156 Geological Background 157 Systematic Palaeontology 160 Cruziana d'Orbigny, 1842 160 Cruziana barriosi Baldwin, 1977 161 Cruziana furc if era d'Orbigny, 1842 163 Cruziana goldfussi (Rouault, 1850) 164 Cruziana omanica Seilacher, 1970 166 Cruziana penicilatta isp. nov. 170 Diplichnites Dawson, 1873 172 Diplichnites arboreus isp. nov. 172 Monomorphichnus Crimes, 1970b 173 Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya, 1977 174 Monomorphichnus lineatus Alpert, 1976 175 Monomorphichnus sinus isp. nov. 176 Monomorphichnus spp. 177 Rusophycus Hall, 1885 177 Rusophycus unilobus (Seilacher, 1970) 177 Conclusions 179 References 180

CHAPTER 6: DEVONIAN OF WALES Introduction 204 Geological Setting and Stratigraphy 205 Systematic Palaeoichnology 205 Ichnogenus Selenichnites (Romano & Whyte, 1987) 206 Selenichnites rossendalensis (Hardy, 1970) 206 Possible trace-makers 210 Associated ichnofauna 215 Palaeoecology 215 Conclusions 218 References 219

CHAPTER 7: CONCLUSION Summary of Work 234 Future Work 236 Lower Cambrian Gog Group 237 Cambrian of Morocco 238 Ordovician of Morocco 238 Ordovician of Australia 239 Devonian of Wales 239 References 241 LIST OF TABLES

5-1. GPS coordinates for the localities collected, with an ichnospecies list of traces collected from these localities, including some traces not mentioned within the paper. 191 LIST OF FIGURES

2-1. National Parks along the British Columbia and Alberta border, with inset illustrating the location of the parks in relation to the two provinces. 68

2-2. Banff and Yoho National Parks with the three localities marked as dashed boxes: A, Lake O'Hara. B, Mount Babel. C, Redoubt Mountain (modified from Nelson et ah, 2005). 69

2-3. Images of the localities. A, Lake O'Hara. B, Mount Babel. C, Redoubt Mountain. 70

2-4. Generalized stratigraphy of the localities, not to scale nor representing any unconformities. 71

2-5. A-B. Ichnofossils illustrating the intrastratal mode of formation of digging down to the muddy substrate, followed by the retreat of the organism, whereupon the sand immediately infills, leaving a convex hyporelief sandstone trace. A, Cruziana dispar (Linnarsson, 1869) at tip of finger with shale encompassing the sandstone trace. B, Cruziana isp., finger on the ventral surface of a lobe with shale at the base of the sandstone trace. C-E. Microbially Induced Sedimentary Structures (MISS) from the Fort Mountain Formation, Gog Group. C, Wrinkle structures verging into Kinneyia structures with Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya, 1977 (in more detail on Fig. 2-18 D) from Mount Babel, Alberta (T80). D, T90 displaying Kinneyia structures from Redoubt Mountain, Alberta. E, T89 intricate Kinneyia structures from Redoubt Mountain, Alberta. Scale bar = 1 cm. 72

2-6. Cruziana billingsi Fillion & Pickerill, 1990, from the Fort Mountain Fonnation, Gog Group. A, T55 from Redoubt Mountain, Alberta. B, T54 from Lake O'Hara, British Columbia. C, T56 from Redoubt Mountain, Alberta. D, T53 from Mount Babel, Alberta. Scale bar = 1 cm. 73

2-7. Abundant Cruziana billingsi Fillion & Pickerill, 1990 and Diplichnites twelvetreesi (Chapman, 1928) transforming from one to another from the Fort Mountain Formation, Gog Group, Lower Cambrian, Mount Babel, Alberta. Image is of the actual rock surface on the talus slope and a P1nctnr\mf> fTS7 Rr T7QA ic in rnllertinnc "sc^lp har = S CTTI 75 2-8. All specimens from the Fort Mountain Formation, Gog Group, Lower Cambrian. A, Cruziana dispar (Linnarsson, 1869), T44, from Lake O'Hara, British Columbia. B-C, Cruziana jenningsi Fenton & Fenton, 1937. B, T38, from Lake O'Hara, British Columbia. C, T61,from Redoubt Mountain, Alberta. D, Cruziana ramellensis (Legg, 1985), T73, from Lake O'Hara, British Columbia. Scale bar = 1 cm. 76

2-9. Cruziana jenningsi Fenton & Fenton, 1937, from the Fort Mountain Formation, Gog Group, Lower Cambrian of Lake O'Hara, British Columbia, T35-T37. A, Ventral view of convex hyporelief trace, T35. B, Ventral view of trace, T36. C, Anterior view of A (T35), emphasizing the anterior ridges. D, Ventrolateral view of trace, T37. E, The rock in which A, B and D are situated upon and labeled. Scale bar = 1 cm. 77

2-10. Cruziana irregularis Fenton & Fenton, 1937. All specimens from the Fort Mountain Formation, Gog Group, Lower Cambrian. A, T43 from Lake O'Hara, British Columbia. B-D, from Redoubt Mountain, Alberta. B, T60. C, T59. D, T58. Scale bar = 1 cm. 78

2—11. Cruziana mesodelta isp. nov. from the Fort Mountain Formation, Gog Group, Lower Cambrian, Redoubt Mountain, Alberta. A, Holotype, T62. B, Paratype, T63. Scale bar = 1 cm. 79

2-12. Cruziana navicella Fenton & Fenton, 1937, from the Fort Mountain Formation, Gog Group, Lower Cambrian of Alberta. A, T42 from Mount Babel. B, T41 from Redoubt Mountain, with Cruziana penicillata Gibb, Chatterton & Pemberton, 2009, terminal trace (T40), indicated by the brackets. Scale bar = 1 cm. 80

2-13. All specimens are from the Fort Mountain Formation, Gog Group, Lower Cambrian. A-B, Cruziana omanica Seilacher, 1970. A, T64 from Mount Babel, Alberta (arrow pointing to the trifid ridge). B, T65 from Redoubt Mountain, Alberta. C, Cruziana penicillata Gibb, Chatterton & Pemberton, 2009, T39, from Redoubt Mountain, Alberta. D, F-G, Cruzianaproblematica (Schindewolf, 1928), T69-T71, from Redoubt Mountain, Alberta. E, Cruziana plicata Crimes, Legg, Marcos & Arboleya, 1977 (T68). F, Transitions from C. plicata to C. problematica (T70) upon meeting perpendicular trace of C. problematica (D: T69). Scale bar = 1 cm. 81

2-14. All specimens are from the Fort Mountain Formation, Gog Group, Lower Cambrian. A, C, Cruziana plicata Crimes, Legg, Marcos & Arboleya, 1977, from Lake O'Hara, British Columbia. A, T67. C, T66. B, Cruziana ramellensis (Legg, 1985), T72, from Redoubt Mountain, Alberta. Scale bar = 1 cm. 82

2-15. Cruziana rugosa d'Orbigny, 1842, Fort Mountain Formation, Gog Group, Lower Cambrian. A, C-D, Redoubt Mountain, Alberta. A, T45. C, T74 (arrow pointing to cephalon impression). D, T75. B, T76 from Lake O'Hara, British Columbia. Scale bar = 1 cm. 83

2—16. A, C-D. Diplichnites twelvetreesi (Chapman, 1928) Gog Group, Lower Cambrian. A, T77, Fort Mountain Formation, Lake O'Hara, British Columbia. C, T46, Fort Mountain Formation, Redoubt Mountain, Alberta. D, T78, Lake Louise Formation, Lake O'Hara, British Columbia. B, Diplichnites obliquus isp. nov. Holotype, T47, Fort Mountain Formation, Lake O'Hara. Scale bar = 1 cm. 84

2-17. Specimens from the Fort Mountain Formation, Gog Group, Lower Cambrian, Mount Babel, Alberta. A, C-D. Diplichnites twelvetreesi (Chapman, 1928). A, T48. C, T91. D, Diplichnites twelvetreesi transitional form to Cruziana billingsi Fillion & Pickerill, 1990 (Plastotype: T79 - T57). B, Monomorphichnus bilinearis Crimes, 1970b, T49. Scale bar = 1 cm. 85

2-18. All specimens from the Fort Mountain Formation, Gog Group. Lower Cambrian. A, Monomorphichnus bilinearis Crimes, 1970b, T50, Lake O'Hara, British Columbia. B, Monomorphichnus trilinearis isp. nov. (Holotype) T51, Redoubt Mountain, Alberta. C-D. Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya, 1977. C, T52, Mount Babel, Alberta. D, T80, Mount Babel, Alberta. Scale bar = 1 cm. 86

2-19. A, Coffee bean viewed laterally to visual the idea of two juxtaposed lobes of a 'single' coffee bean, equivalent to the lobes on Rusophycus and space between lobes is equal to the median line or mesial opening. B-D. Rusophycus eutendorfensis (Linck, 1942) from the Fort Mountain Formation, Gog Group, Lower Cambrian. B, T82 from Mount Babel, Alberta. C, T84 from Redoubt Mountain, Alberta. D, T83 from Redoubt Mountain, Alberta. Scale bar = 1 cm. 87

2-20. All specimens from the Fort Mountain Formation, Gog Group, Lower PamV»rion A J?i/vs\T-tli-\>r"iiv /Ti-iVnH/jncTp ^Q<=>ilar»1-»^r- 1 Q7fA HPS21 -frr*m Redoubt Mountain, Alberta. B-C. Rusophycus latus Webby, 1983. B, T29 from Mount Babel, Alberta. C, T28 from Lake O'Hara, British Columbia. Scale bar = 1 cm. 88

2-21. Rock surface with traces from the Fort Mountain Formation, Gog Group, Lower Cambrian, Lake O'Hara, British Columbia. A, Rusophycus latus Webby, 1983, T31. B-D. Rusophycus subnotous isp. nov. B, Paratype, T32. C, Paratype, T33. D, Holotype, T34. E, Whole rock with the individual traces labeled with responding letter to those figured on plate. Scale bar = 1 cm. 90

2-22. Rusophycus mesodeltus isp. nov. from the Fort Mountain Formation, Gog Group, Lower Cambrian, Redoubt Mountain, Alberta. A, Holotype, T85. B, Paratype, T86. Scale bar = 1 cm. 92

2-23. All specimens from the Fort Mountain Formation, Gog Group, Lower Cambrian, Redoubt Mountain, Alberta. A, Rusophycus unilobus (Seilacher, 1970), T88. Yl, Rusophycus radwanskii Alpert, 1976, T27. C, Rusophycus victorus isp. nov., Holotype: T30. Scale bar = 1 cm. 93

2-24. Arthropod ichnofaunal distribution from Lake O'Hara. 94

2-25. Mount Babel arthropod ichnofaunal distribution. 95

2-26. Redoubt Mountain distribution of arthropod ichnofauna. 96

2-27. Arthropod ichnofaunal distribution across the three localities: Lake O'Hara, Mount Babel and Redoubt Mountain (from the collections of S. Gibb and J. Magwood). 97

3-1. Schematic of area where Cambrian Moroccan traces are located. A. Overview of Morocco, with boxed region enlarged to highlight the geology. B. The geological overview of the region where Selenichnites tacfihtus n. isp. occurs, labeled as 'Cambrian locality' (Saadi et al, 1974- 1977). 115

3-2. Stratigraphic section with occurrence of trilobites at the base and Selenichnites tacfihtus n. isp. at the top of the section. 116

3-3. The Cambrian rock face with Selenichnites tacfihtus n. isp. A. Schematic of the traces upon the bedding surface, with arrows indicating direction of interference ripples. The thin mudstone bed overlying the ichnofossil bedding surface, and upper right corner is the next stratigraphically higher sandstone horizon. B. Image of the actual surface with boxed numbers corresponding to illustrated traces in subsequent figures (numbering corresponds to figure). 118

3—4. Selenichnites tacfihtus n. isp. all from the Azlag Formation, Tabanite Group, Middle Cambrian, near Tansikht, south-central Morocco. A. Two traces with anterior directed to the top left corner, which is relatively consistent with the original orientation on the bedding surface. These traces are approximately 90° to the majority of the other traces, though aligned with one of the ripple trains (T21c). B. Three distinct traces, with two in the right portion of the image, and the upper trace is only the left half of the trace, with the third trace being the deeply incised anterior marking in the left of the image. The traces are not oriented according to original positioning on the bedding surface, but rotated 135° counter clockwise (T22c). C. Single trace rotated 55° counter clockwise from position on bedding surface (T24c). D. Single trace in approximately original position on bedding surface. Another occurrence that is 90° to the majority of traces, though aligned with a ripple train (T23c). E. Three traces, with two overlapping each on the bottom of the image, and the third trace is approximately 90° to the traces below. The image is rotated roughly 50° counter clockwise from position on the bedding surface (T25c). F. One prominent trace, and two less prominent with the image being rotated 150° clockwise from the bedding surface. * is the holotype (T20c). All scale bars are 1 centimeter in length. 119

3-5. Selenichnites tacfihtus n. isp. all from the Azlag Formation, Tabanite Group, Middle Cambrian, near Tansikht, south-central Morocco. Oblique angle image of two 'trackways' and one single trace, white arrows indicate hypothetical direction of movement of the tracemakers. The single trace is to the far left of the image. The longest 'trackway' is just right of the single trace and the last trace of the six clear traces (eight traces in total to the 'trackway') are a series of possible appendage markings (bottom left corner of image). The second 'trackway' is on the right of the image, with two visible traces in the track, and three traces creating the total of the 'trackway'. Scale bar image is segmented into 1 centimeter increments. 121

4-1. Locality maps. 1, Map of Morocco and surrounding area (modified from Chatterton et al, 2006; McKellar and Chatterton, 2009). 2, Ordovician (shaded regions) locality map for trilobite/trace locality (marked with %) within the Tafilalt basin (modified from Fetah et al., 1986). 145

4-2. Microbial mat "cracking" and preferential iron staining along the crack lines from the Upper Fezouata Formation, Ouzina, southern Morocco. 1, Asaphellus aff'. fezouataensis in bottom right corner (UA13664). 2, Asaphellus aff'. fezouataensis in middle (UA13667n). Scale bar is 1 cm. 146

4—3. The proximity ofAsaphellus aff'. fezouataensis to Rusophycus car ley i. I, A. fezouataensis with white arrow directed at a lobe of R. carleyi under the right pleural region of the trilobite (UA13664). The trilobite would have become trapped within the sediment. The left lobe is not visible because perhaps the trilobite had dug alongside another R. carleyi and only the right lobe of the most recent behavior is recorded. 2, R. carleyi, convex hyporelief, with white arrow pointing to the right lateral section of the cephalon of a trilobite (UA 13657). 3, R. carleyi, convex hyporelief, with white arrow directed towards the left genal spine of a trilobite (UA 13662). 4, R. carleyi, convex hyporelief, with the white arrow at the right ventral lateral section of the cephalon and the black arrow pointing at the transected portion of the hypostome of the trilobite, within the trace (UA 13661). Specimens UA 13662 & UA13661 were submerged in ethyl alcohol to enhance the exoskeleton within the trace, while UA13664 & UA13657 were coated in ammonium chloride. Specimens are from the Upper Fezouata Formation, Ouzina, southern Morocco. Scale bar is 1 cm. 147

4-4. Rusophycus carleyi (UA13658), convex hyporelief with left lateral border of pygidium (white arrow), and thoracic segment impressions (white arrows outlined by black), from the Upper Fezouata Formation, Ouzina, southern Morocco. Scale bar is 1 cm. 149

4-5. Scatter plot (length vs. width) of Rusophycus carleyi and the trilobites from the Ouzina locality. The length and width of the trilobite, when only fragments were available were inferred by extrapolating proportions from parts, with ratios taken from four Asaphellus species from the Ordovician of Morocco. The black diamonds are the trilobites and the grey squares are Rusophycus carleyi. All specimens are from the Upper Fezouata Formation, Ouzina, southern Morocco. 150

4-6. Rusophycus carleyi (UA13655), convex hyporelief, from the Upper Fezouata Formation, Ouzina, southern Morocco. 1, Lateral oblique close- up of hypostome impression (black arrow). 2, Oblique lateral view of trace. 3, Lower posterior view of endopodal spine scratches within the medial opening. 4, Close-up of coxal impressions, with the left being the anterior end of the trace. Scale bar is 1 cm. 151

A—7 Pi/cnnJiv/'tit mrlpvi 1 TTAI^fiSfi Hicnlavino enHrmnrlal Qninp QrratchpQ within the posterior lateral medial opening. 2, UA13660 displaying endopodal spine scratches within the lateral medial opening. 3, 4, UA 13659. 3, Entire view of convex hyporelief of trace. 4, Close-up of thoracic segment impressions on trace. All specimens are from the Upper Fezouata Formation, Ouzina, southern Morocco. Scale bar is 1 cm. 152

4-8. Schematic of the tracemaker (trilobite), Asaphellus aff. fezouataensis, removing itself from the substrate, resulting in the trace Rusophycus carleyi. From the striations/ridges within the mesial opening, it is assumed that A aff'. fezouataensis would have endopodomeral spines. Sketch created by Darrin Molinaro. 154

4-9. Asaphellus aff. fezouataensis, all dorsal views. 1, UA13664, cephalon and majority of thorax (submerged in ethyl alcohol to enhance trilobite morphology). 2, UA13665, cephalon, minority of thorax and taphonomically altered hypostome (coated with ammonium chloride). Important note, with regard to this specimen is the pits observed on the posterior end of the cephalon, associated with the alimentary structures (diverticulae). The pits are similar in outline to trilobite alimentary diverticulae illustrated by Chatterton et al. (1994, Figs. 1.1-1.3, 2.1, 2.4 and 3.4). 3, UA13663, pygidium (coated with ammonium chloride). All specimens are from the Upper Fezouata Formation, Ouzina, southern Morocco. Scale bar is 1 cm. 155

5-1. A. Generalised map of mainland Australia. B. Locality map indicating positions of Charlotte Range and Mount Watt. 190

5-2. Specimens are all convex hyporelief. A—B, Cruziana barriosi Baldwin, 1977 from the Charlotte Range. A, F133884 from CR2. B, F133880 from CR7. C-G, Cruzianafurcifera d'Orbigny, 1842. C-D, F133871 from CRR from the Charlotte Range. C, anteroventral oblique view. D, ventral view. E-F,F133938 from CR7 from the Charlotte Range. G,F135852 from Mount Watt. Scale bar = 1 cm. 192

5-3. Cruziana goldfussi (Rouault, 1850). Specimens are all convex hyporelief. A-D, F133937 from CR7. A, lateral view, arrow depicts scratch marks up the trace. B, oblique lateral view and arrow pointing to lateral ridge marking. C, opposite oblique lateral view of B. D, view of ventral surface of hyporelief. E, ventral surface of F133887 from CRR. F, ventrolateral view of F133936 from CR7. All from the Charlotte Range. Scale bar= 1 cm. 194

5-4. Cruziana omanica Seilacher, 1970. Specimens are all convex hyporelief. A, F133893 (CR.7). B, F133886 (CR.7). C, F133895 with fragment of possible external mould of trilobite pygidium (CR7). D, F133007 (CR7). E, portion of sample of F133877(CR7). F, F133 883 (CR7). All from the Charlotte Range. Scale bar = 1 cm. 196

Disarticulated and fragmented trilobite sclerites. A, impression of thoracic segment (F133950) from the Charlotte Range (CR3). B, external mould of axial region and portion of pleural region of thoracic segment, pleural furrow, anterior and posterior band have similar morphology to Lycophron howchini (Etheridge, 1894), indicated by the white arrow; black on white arrow is pointing at the typical gastropod found in the region (F133949 - CR3). C, external mould of left pleural region of fragmented pygidium, similar morphology to asaphid pygidia (F133895 - CR7). Scale bar = 1 cm. 198

A-H, Cruziana penicillata isp. nov. All convex hyporelief. A-B, paratype (F133018) from the Charlotte Range. A, lateral view. B, ventral view of convex hyporelief C, paratype: anterior portion of trace (F133871) from the Charlotte Range. D, paratype: F135855 from Mount Watt. E, holotype: F133868 from the Charlotte Range. F, paratype: F135853 from Mount Watt. G, paratype: F135856 from Mount Watt. H, paratype: F133021 from Mount Watt. I, Monomorphichnus spp. F133890 (CR7) from the Charlotte Range. Scale bar = 1 cm. 199

A-C, Diplichnites arboreus isp. nov. Specimens are all convex hyporelief. A, holotype indicated by white arrow (F58984) from the Charlotte Range (R.d. CI) (coin for scale is 23.6 mm diameter). B, paratype underprint, highlighted by white arrows (F133014) from Mount Watt. C, paratype (F133022) from Mount Watt. D, Monomorphichnus lineatus Crimes, Legg, Marcos & Arboyleya, 1977 (F133022) from Mount Watt. Scale bar =1 cm. 201

All convex hyporelief. A, Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya, 1977 (F133 896) from CR7 of the Charlotte Range. B, Monomorphichnus sinus isp. nov. (white, and black-on-white arrows) and Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya, 1977 (F133931) from CR7 in the Charlotte Range. C, Same as B: Monomorphichnus lineatus depicted as white straight lines and Monomorphichnus sinus traced by curved grey lines. D, Monomorphichnus multilineatus Alpert, 1976 (F133943) from CR7 in the Charlotte Range. E, Monomorphichnus sinus isp. nov. (F133919) from CR4 in the Charlotte Range. F, Rusophycus unilobus (Seilacher, 1970) (F133920) from CR4 in the Charlotte Range. Scale bar = 1 cm. 202

(l) Geological man of the region (modified from Marshall. 2000a. b): (2) 'Freshwater West' locality in relation to the west of Wales; (3) Section image of the 'Freshwater West' locality, with geologist of approximate height of 170 centimetres, with hand on location in which rock was removed with the five traces of Selenichnites rossendalensis. The division of the Stackpole Sandstone Member, within the Gupton Fonnation, into the older 'Heterolithic Association' and younger 'Quartzarenite Association' (Marshall, 2000b); (4) Strati graphic columns of the Freshwater West locality (modified from Marshall, 2000a, b). 226

6-2. (1) Image of the complete slab replica with the 6 traces Selenichnites rossendalensis of; (2) Trace la' from corresponding image. Note scour marks (small arrows); (3) Trace 'c'; (4) Trace '/, with all scale bars = 1 centimetre; (5) Euproops depicted in stage one (Eldredge, 1970) of burrowing into substrate (stippling is substrate covering the anterior of the prosoma). 228

6-3. Measurements of each ichnofossil, and letters correspond to Figure 6-2.1. 230

6^1. Sketch of the rock with Selenichnites rossendalensis and physical sedimentary structures depicted as stippled. The arrows delineate the hypothesized palaeocurrent direction (i: is a horseshoe vortex with missing obstacle; iii, iv and v: tool marks). There is only a slight variance in current direction observed from the physical sedimentary structures. Inset: Rose diagram depicting the long axes orientation of the traces (shaded blocks) and the mean palaeocurrent direction (black arrow), orientation is based purely on the illustrated surface, an arbitrarily defined direction for the purposes of the description. Scale bar = 1 centimetre. 231

6-5. Horseshoe vortex system. (1) Instantaneous flow ribbons showing strong circulation around the obstacle and weaker, more random flow to the lee of the obstacle; (2) Time averaged shear velocity associated with a horseshoe vortex. The shear velocity gives a realistic estimate of the zones of erosion (dark). Note the coincidence of the zones of erosion and the scours associated with Selenichnites rossendalensis. 233 LIST OF APPENDICES

Appendix: Specimen details 186 CHAPTER 1: INTRODUCTION

'Which animal produced that trace fossil?' remains typical of the first question put to an ichnological sample. This is not the right way of addressing trace fossils; they tell us much more interesting things than that, while rarely divulging the nature of their architects. (Bromley, 1990, p. 137)

INTRODUCTION The purpose of this dissertation is to improve our knowledge and understanding of Lower Palaeozoic arthropod traces. The ichnofossils were collected, photographed, identified and classified. For each of the localities, data that could be used to deduce the palaeoenvironment are considered, particularly with regard to their effects on preservation of the traces and/or to deduce (speculate on) the animals that made the trace(s). It was found, in some instances, that prevailing current directions affected the orientation of the traces. Another variable considered is the diversity and disparity of traces that are generally thought to have been made by arthropods, in each of the localities. Surprisingly, the most diverse traces that would generally be assigned to arthropods considered herein are found in the lower Cambrian strata of the Gog Group. This suggests that arthropods radiated and evolved the ability to perform a wide variety of behaviours that produced trace fossils very early in their history. They were able to move a remarkable amount of sediment and dig quite deeply in the Early Cambrian. It is often considered that, apart from the appearance of many organisms with hard shells, the greatest change from the Proterozoic to the Cambrian is the ability of organisms to dig downward into the sediment and produce trace fossils with a vertical component. The diverse assemblage of arthropod traces found in the lower Cambrian Gog Group amply supports this behaviour.

1 The synergy between the apparent disparity of the chapters from the lower Cambrian to the Devonian, the southern Rocky Mountains, across to Wales, through Morocco and down under to Australia, is interesting. Different types of arthropod behaviour are revealed and regaled to paint a bigger picture, part of a masterpiece of epic portions, the history of life. Leonardo da Vinci, in the 15l - 16th centuries already recognized the impressive nature of ichnology in his writings: "Between one layer and the other there remains traces of the worms that crept between them when they had not yet dried. All the sea mud still contains shells, and the shells are petrified together with the mud" (from the Leicester Code, folio lOv of Leonardo da Vinci in Baucon, 2010); and paintings Virgin of the Rocks and Madonna of the Yarnwinder (Baucon, 2010). This chapter is not intended to be a comprehensive literature review of arthropod ichnology, but an overview of the chapters contained herein.

LOCALITIES The localities studied are a progression through geological time from the lower Cambrian to the Devonian.

Cambrian southern Rocky Mountains of Canada: 'Arthropod traces of the lower Cambrian Gog Group, southern Rocky Mountains of Canada' This chapter illuminates the 'phylogenetic lawn' theory to the evolution of organisms, though in this chapter, of ichnofossils. The diversity and abundance of arthropod traces that occur in the lower Cambrian Gog Group in the southern Rocky Mountains of Canada is epic. Four ichnogenera, Cruziana, Diplichnites, Monomorphichnus, and Rusophycus are discussed with a total of 25 different ichnospecies. The distribution of the ichnogenera from the three localities (Lake O'Hara, Mount Babel, and Redoubt Mountain) demonstrate a larger portion of Cruziana and thus confirming a clastic palaeoenvironmental setting within the Cruziana ichnofacies, thus within the region between fairweather wave base and stormweather wave base.

2 Cambrian of Morocco: 'Moroccan prod traces {Selenichnites) from the middle Cambrian and the implication of the tracemakers behaviour' A single bedding surface was identified along the roadside close to an important road junction at Tansikht and north of Zagora in southern Morocco. This bedding plane has concomitant interference ripples and distinct horseshoe shaped concave epirelief ichnofossils dispersed across the surface. The sandstone bed belongs to the upper middle Cambrian Azlag Formation. The Azlag Formation occurs above the Bailiella Formation, with its distinctive trilobite fauna, in this section. The traces provide evidence of the ethology of an organism that inhabited the shallow waters during this time. Body fossils are rarely preserved in a clastic setting. The traces are assigned to the ichnogenus Selenichnites, and a new ichnospecies, Selenichnites tacfihtus, is proposed, identified, diagnosed, described and illustrated. Possible trace-makers and behaviours are considered.

Ordovician of Australia: 'Arthropod ichnofossils from the Ordovician Stairway Sandstone of central Australia' The arthropod ichnofossils from the Middle Ordovician (Darriwilian Stage) Stairway Sandstone are, surprisingly, lower in diversity than those found in the lower Cambrian Gog Group. Though the preservation of the traces is exceptional, and the palaeoenvironmental setting was quite similar to that of the Gog Group, the diversity of the arthropod traces is much lower. The ichnogenera illustrated and discussed from this unit are Cruziana, Diplichnites, Monomorphichnus and Rusophycus. Three new ichnospecies were proposed, and eight previously described and one indeterminate ichnospecies were described from the Stairway Sandstone. The arthropod traces co-occur with other ichnofauna in a fine to medium grained sandstone with minor interbeds of mudstone. This unit of the upper part of the Stairway Sandstone can be classified as belonging to the Cruziana ichnofacies, based upon the ichnofauna and the sediments.

3 Ordovician of Morocco: 'Rusophycus carleyi (James, 1885), trace fossils from the Lower Ordovician of southern Morocco, and the trilobites that made them' "Thin-bedded, pyrite-rich, fine sandstones and mudstones of the Floian- Dapingian Upper Fezouata Formation contain abundant trace fossils, Rusophycus carleyi, in close association with a species of the asaphid trilobite, Asaphellus. The sizes and shapes of this trilobite and the traces match closely. Five specimens have even been found where an articulated specimen of Asaphellus appears to be directly located over a specimen of Rusophycus carleyi within a thin bed of sandstone, suggesting that the trilobite animal may have been trapped on (op of a trace that it had just made. Such intimate associations between a putative trace maker and a trace are very rare in the fossil record and particularly rare for i rkibita. The number of coxal impressions that form part of R. carleyi, eleven, matches the number expected for an asaphid trilobite (one for each of eight thoracic segments and one for each of three post-oral cephalic appendages). impressions of the hypostome, thoracic tip impressions, cephalic margin and oy^idial margin in a few of the traces also match those of this asaphid trilobite. R. ar:eyi has been found in Ordovician strata of other parts of the world in association with asaphid trilobites" (Gibb et al., 2010, p. 271).

Devonian of Wales: 'Devonian Selenichnites (ichnofossil) from Pembrokeshire (Hates, UK)' Well-preserved ichnofossils, Selenichnites rossendalensis, are described Trcm the Famennian Gupton Formation of Wales. Possible trace-makers are hypothesized. The behaviours that produced these traces are discussed or considered. Research in neoichnology and the documentation of sedimentary structures associated with these traces shed light on the likely palaeoenvironmental setting. This palaeoenvironment is posited to have been a deep lacustrine setting that has been associated with the Mermia ichnofacies.

4 PREVIOUS RESEARCH Leonardo da Vinci, the archetypal Renaissance Man, is now referred to as the 'Father of Ichnology' (Baucon, 2010, p. 361). It was not only his paintings, of Virgin of the Rocks and Madonna of the Yarnwinder, but his sketches, in particular those of Paleodictyon, and writings on ichnology (Baucon, 2010). Baucon (2010, p. 362-364, fig. IB) discussed and illustrated the four stages of ichnological history: 'Age of Naturalists', 'Age of Fucoids', 'Period of Reaction' and 'Development of the Modern Approach'. The 'Age of Naturalists' was during the Renaissance Period, or approximately between the years of 1500 - 1650, when a number of individuals (i.e. Fracastoro, Gesner, Bauhin, and Aldrovandi) were illustrating and/or commenting on ichnofossils, though none as accurately as da Vinci (Baucon, 2010). It was not until Brongniart erected the genus Fucoides in 1828, though then classified the 'trace' as algal that we observe the 'Age of Fucoids' (Baucon, 2010; Osgood, 1970, 1975). Osgood (1970; 1975) expanded on this time in ichnological history, citing numerous authors who wrote on the 'algal remains': Hall, Billings, Miller, Dyer, Schimper, Heer, Saporto, Delgado, Barrande, and Agassiz. It should be noted that some authors were actually recognizing the fact that they were ichnofossils, and were described as such, these were: Dawson, Logan, and Miller (Osgood, 1970, 1975). The 'Age of Reaction' was lead by Nathorst (1881), in which Osgood stated that it was "the first comprehensive attack against 'fucoids,' and in the final analysis the most devastating" (1970, p. 287). Nathorst (1881), also used neoichnology to refute some of the 'fucoid' interpretations. Osgood (1970, p. 289) continued with his 'Development of the Ethological and Paleoecological Approaches' or as Baucon (2010, fig. IB) referred to it as 'Development of the Modern Approach' with the work of Rudolf Richter and his associates with the establishment of the Forschungsanstalt fur Meeresgeolgie und Meerespalaontologie der Senckenbergischen Naturforschenden Gesellschaft zu Frankfurt am Main, or commonly referred to as 'Senckenberg am Meer' (Osgood, 1970, p. 289). As Osgood (1970, p. 289) stated, "the purpose [of Senckenberg am Meer] was to exploit the vast Wattenmeer mudflats of the North Sea for what they might teach

5 with respect to geologic phenomena. This was the first careful systematic study of Recent ecology, with the principal emphasis on invertebrate traces". Further advancement of ichnology was accelerated by Seilacher (to cite just a few: 1953; 1955; 1964; 1970; 1983; 1985; 1990; 1991), especially in the area of arthropod lebensspuren research. Finally the work of compiling the multitude of ichnogenera that had been proposed up to that date was done by Hantzschel (1975) for the Treatise on Invertebrate Paleontology. Obviously since then many researchers have contributed to the advancement of knowledge, of particular interest to this dissertation, of arthropod trace fossils. Though they are far too numerous to cite in this short introductory chapter, many of their works are cited in the following chapters.

METHODOLOGY The collection of trace fossils was based on quality of preservation, diversity, disparity, and some relationship to the overall composition of the ichnofauna. Where possible, sections were measured and traces were collected in place. However, because many ichnofossils are most visible on bedding surfaces, many of the best specimens were collected as talus specimens, therefore they are not in situ. The photographic techniques are discussed in each chapter and when different techniques were required, they are explained. The systematic palaeoichnology is always focused on the importance of nomenclature that is based purely on the morphological character states and not on behaviour and/or the organisms that created the ichnofossil (Bertling et al., 2006; Bromley, 1990; Gibb et al., 2009; Jensen, 2003; Magwood, 1992; Minter et al., 2007; Pemberton and Frey, 1984; Pickerill, 1994; Ride et al., 2001). This is a point that is reiterated throughout the chapters. When appropriate and enough information was available, be it sediment composition, physical sedimentary structures, previous work/publications and/or co-occurring fossil(s)/ichnofossil(s), a palaeoenvironmental reconstruction was

6 suggested. Not all localities were in need of such a reconstruction because appropriate environmental reconstruction(s) had been proposed in previous publications. Some localities had insufficient defining features available to me to allow me to provide a satisfactory interpretation. However, palaeoenvironmental requirements for the requisite presence and preservation of the ichnofossils described in each chapter are considered.

The main objectives of this body of work are: (a) to establish the systematics of the arthropod traces from various localities throughout the world during different geological times within the Lower Palaeozoic; (b) where appropriate to propose possible tracemakers, based upon fossils within close proximity and/or based on literature reviews; (c) to put forward plausible palaeoenvironmental reconstructions based on the traces, strata, physical sedimentary structures, and previously published work from the same region; (d) to incorporate neoichnology in deducing the morphology of the animals that made the traces, and constrain explanations of the behaviour(s) of the organism(s) that made the traces; (e) to use hydrodynamics to establish relationships between sedimentary structures, preservation of and orientation of traces and movement in fluids; and (f) to provide useful additions to the vast amount of knowledge that can be gained from the ichnofossils, including their diversity, disparity, and abundances throughout the world.

In all of the chapters, I am the primary investigator/writer/photographer. My co-authors provided assistance in the form of field assistance (or provider of specimen(s)), photographic assistance, scholarly advice, advancement of knowledge and/or editorial guidance.

FORMAT This thesis adheres to paper/published format. The editorial format for each of the chapters are as follows: Cambrian Gog Group of Canada submitted and in

7 review with Memoirs of the Association of Australasian Palaeontologists; Cambrian of Morocco submitted and in review with Ichnos; Ordovician of Australia was published in Memoirs of the Association of Australasian Palaeontologists 37, 695-716 (ISSN 0810-8889); Ordovician of Morocco was published michnos 17, 271-283 (DOI: 10.1080/10420940.2010.535452) with an Erratum, for figures, michnos 18:1-9 (DOI: 10.1080/10420940.2011.559873); and the Devonian of Wales will be submitted to the Geological Journal. The introduction and concluding chapters are in the editorial format of the Journal of Paleontology; and the information on localities and repositories is provided in each distinct chapter.

8 LITERATURE CITED

BAUCON, A. 2010. Leonardo da Vinci, the founding Father of ichnology. Palaios, 25:361-367. BERTLING, M., S. J. BRADDY, R. G. BROMLEY, G. R. DEMATHIEU, J. GENISE, R. MIKULAS, J. K. NIELSEN, K. S. S. NIELSEN, A. K. RINDSBERG, M. SCHLIRF, AND A. UCHMAN. 2006. Names for trace fossils: a uniform approach. Lethaia, 39:265-286. BROMLEY, R. G. 1990. Trace fossils: Biology and taphonomy. Unwin Hyman, London, 280 p. GIBB, S., B. D. E. CHATTERTON, AND M. K. GINGRAS. 2010. Rusophycus carleyi (James, 1885), trace fossils from the Lower Ordovician of Southern Morocco, and the trilobites that made them. Ichnos, 17:271-283. GIBB, S., B. D. E. CHATTERTON, AND S. G. PEMBERTON. 2009. Arthropod ichnofossils from the Ordovician Stairway Sandstone of central Australia. Memoirs of the Association of Australasian Palaeontologists, 37:695-716. HANTZSCHEL, W. 1975. Treatise on Invertebrate Paleontology: Part W, Miscellanea, Supplement 1, Trace Fossils and Problematica. The Geological Society of America, Inc. and The University of Kansas, Boulder, Colorado, and Lawrence, Kansas, W269 p. JAMES, J. F. 1885. The fucoids of the Cincinnati Group, Pt. 2. Journal of the Cincinnati Society of Natural History, 7:151-166. JENSEN, S. 2003. The Proterozoic and earliest Cambrian trace fossil record; patterns, problems and perspectives. Integrative and Comparative Biology, 43:219-228. MAGWOOD, J. P. A. 1992. Ichnotaxonomy: a burrow by any other name...?, p. 15- 33. In C. G. Maples and R. R. West (eds.), Trace fossils. Short courses in paleontology. No. 5. University of Tennessee, Knoxville. MINTER, N. J., S. J. BRADDY, AND R. B. DAVIS. 2007. Between a rock and a hard place: arthropod trackways and ichnotaxonomy. Lethaia, 40:365-375. NATHORST, A. G. 1881. Om spar af nagra evertebrerade djur m. m. och deras palaeontologiska betydelse. Svenska Vetenskaps Akademiens Handlingar, 18(7):1-104. OSGOOD, R. G. J. 1970. Trace fossils of the Cincinnati area. Palaeontographica Americana, 6(41):281-439. OSGOOD, R. G. J. 1975. The history of invertebrate ichnology, p. 3-12. In R. W. Frey (ed.), The study of trace fossils: A synthesis of principles, problems, and procedures in ichnology. Springer-Verlag, New York. PEMBERTON, S. G., AND R. W. FREY. 1984. Quantitative methods in ichnology: spatial distribution among populations. Lethaia, 17:33-49. PICKERILL, R. K. 1994. Nomenclature and taxonomy of invertebrate trace fossils, p. 3-42. In S. K. Donovan (ed.), The Paleobiology of trace fossils. The Tnhn HnnVinQ TTnivprcitv PrpQQ Raltimnrp RIDE, W. D. L., H. G. COGGER, C. DUPUIS, O. KRAUS, A. MINNELLI, F. C. THOMPSON, AND P. K. TUBBS. 2001. International Code of Zoological

9 Nomenclature. International Trust of Zoological Nomenclature, London, 306 p. SEILACHER, A. 1953. Uber die Methoden der Palichnologie, Studien zur Palichnologie, no. 1. Neues Jahrbuch fur Geologie und Palaontologie, Abhandlungen, 96:421-452. SEILACHER, A. 1955. Spuren und Lebenweise der Trilobiten, p. 342-372. In O. H. Schindewolfe and A. Seilacher (eds.), Beitrage zur Kenntnis des Kambriums in der Salt Range (Pakistan). Akademie der Wissenschafter und der Literatur Abhandlungen der Mafhematisch- Naturwissenschaftlichen Klasse, 10. SEILACHER, A. 1964. Sedimentological classification and nomenclature of trace fossils. Sedimentology, 3:253-256. SEILACHER, A. 1970. Cruziana stratigraphy of "non-fossiliferous" Palaeozoic sandstones, p. 447-476. In T. P. Crimes and J. C. Harper (eds.), Trace fossils. Seel House Press, Liverpool. SEILACHER, A. 1983. Upper Paleozoic trace fossils from the Gilf Kebir-Abu Ras area in southwestern Egypt. Journal of African Earth Sciences, l(l):21-34. SEILACHER, A. 1985. Trilobite paleobiology and substrate relationships. Transactions of the Royal Society of Edinburgh: Earth Sciences, 76:231- 237. SEILACHER, A. 1990. Paleozoic trace fossils, p. 649-670. In R. Said (ed.), Geology of Egypt. Rotterdam, Balkema. SEILACHER, A. 1991. An updated Cruziana stratigraphy of Gondwanan Palaeozoic Sandstones, p. 1565-1580. In M. J. Salem, O. S. Hammuda, and B. A. Eliagoubi (eds.), The geology of Libya. Elsevier, Amsterdam.

10 CHAPTER 2: CAMBRIAN OF CANADA

ARTHROPOD TRACES OF THE LOWER CAMBRIAN GOG GROUP, SOUTHERN ROCKY MOUNTAINS OF CANADA1

INTRODUCTION THE Gog Group, found throughout the Rocky Mountains of Alberta and British Columbia contains a surprising diversity and disparity of arthropod ichnofossils from the lower Cambrian. This 'group' provides evidence of the ethology of'older' organisms. Due to the siliciclastic nature of most of the sediments very few body fossils are ever preserved in the Gog Group, and though arthropod fossils of this age have been documented, speculations on the creators of the traces are not addressed in this paper. However, the preservation of the traces is exceptional. The diversity of ichnospecies is so extraordinary that one can conclude that many of the initial stages of the 'Cambrian Explosion', in the Rocky Mountains, can be observed in trace fossils before we see many body fossils preserved in abundant diversity in the famous Middle Cambrian Burgess Shale fauna. This moves beyond the concept of the Early Cambrian arthropods not being classified "to any of the major groups of living arthropods, or to the trilobites" (Briggs 1990, p. 24). This should not be a surprise since the Arthropoda is very diverse in the Chengjiang biota that is at least partially equivalent in age to the trace-bearing units of the Gog Group (Hou et al. 2004). A wide range of behaviours developed in arthropods early in geological time and similar, if not identical, behaviours can still be observed among Modern arthropods (Gibb et al. in prep). Thus the ichnofossil diversity is comparable to the arthropod diversity observed in the Burgess Shale echoed by Briggs (1990, p.

A version of this chapter has been submitted for review as: GIBB, S., PEMBERTON, S.G. & CHATTERTON, B.D.E.. Arthropod traces of the lower Cambrian Gog Group, Southern Rocky Mountains of Canada. Memoirs of the Association of Australasian Palaeontologists. 11 29) from Manton & Anderson (1979, p. 281) in the analogy of arthropod phylogeny being comparable to a lawn of grass ("phylogenetic lawn") with many blades of grass, instead of the more 'typical' image of evolutionary biology shown in introductory texts that look like a tree with many limbs splitting off to numerous terminal branches ("evolutionary tree"). The focus herein is on the traces believed by most workers to have been made by arthropods (due to the presence of scratch marks produced by jointed appendages): Cruziana d'Orbigny, 1842, Diplichnites Dawson 1873, Monomorphichnus Crimes 1970, and Rusophycus Hall, 1852. These were collected from three localities: Lake O'Hara, Mount Babel, and Redoubt Mountain, representing the Fort Mountain, Lake Louise and St. Piran Formations of the Gog Group. These arthropod ichnofossils occur with numerous other traces not usually considered to have been made by arthropods, to name a few of the more abundant ichnogenera: Bergaueria Prantl, 1946, Chondrites von Sternberg, 1833, Phycodes Richter, 1850, Planolites Nicholson, 1873, Skolithos Haldeman, 1840, Teichichnus Seilacher, 1955a, and Trichophycus Miller & Dyer, 1878. Thus, the lower Cambrian ichnocoenosis for the Gog Group is very diverse. Previous research on the lower Cambrian ichnology was published as early as 1937 by Fenton & Fenton on three ichnospecies of Cruziana, which have been identified in our own research. Publications by Arai & McGugan (1968) and Pemberton & Magwood (1990) focused on the ichnogenus Bergaueria, while others (Young 1972; Magwood 1988; Magwood & Pemberton 1988, 1990; Desjardins et al. 2010) addressed an array of ichnogenera.

GEOGRAPHICAL LOCATION While the Gog Group is widespread, many sections of that rock unit do not contain abundant or well-preserved trace fossils. We have concentrated in localities where the traces are more abundant and better preserved. Three localities (Figs. 2-1-2-3) concentrated on in this work are located in the Main Ranges of the Rocky Mountains, in the Banff and Yoho National Parks. These

12 parks straddle the British Columbia and Alberta provincial border. The first locality at Lake O'Hara in Yoho National Park (Figs. 2-2, 2-3.A) is located at approximately 51° 21' 27.7" N and 116° 19' 56.7" W. The second locality is located on the east shore of Moraine Lake, from the scree slopes of Mount Babel (Fig. 2-2, 2-3.B), 51° 19' 29.2"N and 116° 10' 42.8" W. Redoubt Mountain (Figs. 2-2, 2-3.C) is the third locality, situated at 51° 27' 25.9" N and 116° 06' 01.5" W. The Gog Group outcrops extensively along the British Columbia- Alberta border from Jasper National Park to Mount Assiniboine. The formations in the Gog Group change between the Jasper and Banff-Lake Louise-Yoho areas (Aitken 1983). Some sampling collection of traces was done at the following localities that contain similar ichnofossils to those described herein in Jasper National Park: Pyramid Mountain (52° 57' 35.4" N and 118° 08' 11.3" W), Whistler Mountain (52° 49' 38.6" N and 118° 07' 40.8" W) and Mount Kerkeslin (52° 38' 39.3" N and 117° 50' 32.4" W). Unfortunately, these localities are beyond the scope of this research and will be the focus of further research. In general, the traces obtained from these more northern localities to date are not as abundant, well preserved and diverse as those that are the focus of the present work.

GEOLOGICAL BACKGROUND The Gog Group consists of four formations, in sequence: the Fort Mountain Formation is a cliff-forming resistant unit, that predominantly consists of quartzite with thin, interbedded mudstones; the Lake Louise Formation is a thinner, less resistant unit comprised of shales and quartzites; the St. Piran Formation is another resistant quartzite unit with occasional thin, interbedded mudstones; and the Peyto Formation is a carbonate unit. Traces may be found throughout the Gog Group from the Fort Mountain Formation at the base through the Lake Louise Formation, the St. Piran Formation and to where it is capped by the Peyto Formation, of the lower Cambrian (Fig. 2-4). Only the first three formations are addressed in this paper, as traces are more abundant in these units.

13 As may be expected the traces are more abundant and better preserved where these largely sandstone or quartzite units contain some interbedded finer grained siltstone or mudstone units. Sections where these units consist entirely of coarser sands contain few if any well-preserved arthropod traces. A "profound truncation of Miette Group strata of Windermere age at the base of the Gog Group of Early Cambrian age" (Aitken 1969, p. 193) defines the beginning of the Palaeozoic in the southern Canadian Rocky Mountains. This "profound regional unconformity" (Aitken 1969, p. 197) occurs at the base of the Gog Group. It is located at the base of the Sauk Sequence of Sloss (1963; 1976) and as was noted by Aitken (1989, p. 106, fig. 7.2), marks a worldwide transgression, also noted by Vail et al. (1977). In this region, it occurred on a subsiding passive continental margin (Aitken 1983; Hein & Arnott 1983; Hein 1987; Hein et al. 1991; Desjardins, et al. 2010) as calculated by Bond & Kominz (1984). The three localities that produced the ichnofossils considered herein were situated on the western margin of Laurentia on a shallow shelf in the proto- Pacific. Laurentia, in the Early Cambrian, was rotated approximately 90° clockwise from the orientation it presently occupies (in regard to lines of latitude and longitude). The geographic location was slightly south of the equator, which is a tropical setting (McKerrow et al. 1992, p. 601; Scotese 2002). The defining sediments of the Gog Group are quartzites. Hein (1987) stated that "the Gog Group deposits are part of the proto-Pacific miogeocline, which is marked by a westward thickening wedge of basal coarse clastic units" (p. 157). Hein & Arnott (1983) noted that "the source of sediment was mainly to the east, consisting of cratonic crystalline rocks and overlying sedimentary cover. The maturity of the sediment suggests reduction of relief in the source terrain" (p. 8). The Gog Group sediments were deposited on a subsiding margin, in a shallow marine setting. The Fort Mountain Formation was deposited in an environment that was moderately deep, therefore an offshore setting above storm weather wave base that was "driven by a tidally dominated current system" (Hein,

14 et al. 1991, p. 426). The Lake Louise Formation is greatly reduced in the percentage of sand present, and is comprised of intercalated mudstone and/or very fine-grained sandstone. This would be indicative of a deeper water setting, which Desjardins et al. (2010) considered to be a maximum flooding surface. The environment is interpreted as a distal offshore shelf setting. The St. Piran Formation has been interpreted as "a deep water, offshore shelf system, generally below wave-base, and driven by a mainly unidirectional, littoral current system. Some flows were modulated by tidal currents" (Hein, et al. 1991, p. 426). Desjardins et al. (2010) further divided the formation into two units: "The lower part of the St. Piran Formation records the migration and overall progradation of a sand-sheet complex in an outer-shelf environment. Lowstand tidal-flat complexes developed... These are followed by sand-ridge complexes in an outer-shelf setting" (p. 512). To analyze the palaeoenvironmental interactions of the organisms with the substrate, it was observed that the organisms producing the traces, in a number of cases were digging through the sand into a mud substrate. This was particularly evident in the Fort Mountain Formation at Lake O'Hara (Fig. 2-5A-B). Therefore, many of the traces illustrated were produced in a muddy substrate at an interface with the overlying sand in the manner discussed by Fenton & Fenton (1937), Seilacher (1955b; 1970; 1983; 1985), Birkenmajer & Bruton (1971) and Goldring(1985). A number of bedding planes occur which display Microbially Induced Sedimentary Structures (MISS) (Noffke et al. 2001; Noffke 2010) (Fig. 2-5 C-E). The presence of MISS provides a reason why the organisms would dig into the muddy substrate: to ingest biofilms and/or microbial mats or other organisms feeding on those materials. The hypothesis of digging into the substrate was suggested by MacNaughton & Narbonne (1999, p. 105): "that infaunal burrowing evolved first as a strategy for utilizing unexploited nutrients". The MISS observed in Figure 2-5C, is that of 'wrinkle structures' verging into complete Kinneyia structures. Examples of Kinneyia wrinkle structures (Fig. 2-5 D-E) have been determined to be those of microbial mats (Hagadorn & Bottjer 1997,

15 1999; Pfliiger 1999; Porada & Bouougri 2007; Porada et al. 2008) occurring in a siliciclastic environment. The flat-topped crests present, display both bifurcation and a honey comb structure (Porada & Bouougri 2007; Porada, et al. 2008). Crests are typically 1 millimetre high, with an average width of 2.5 millimetres, and an average length of 30 millimetres. Kinneyia structures have been cited to occur in an intertidal to subtidal environment (Porada & Bouougri 2007; Porada, et al. 2008). The structures identified in the Fort Mountain Formation from Redoubt Mountain, were more likely in a subtidal setting, due to the reasoning of Hein et al. (1991) that this was a deeper water setting, though above the storm weather wave base, thus conforming with the suggestion of Porada et al. (2008, p. 67) that they are "interpreted as event deposits, resulting from either storms or floods". Further indications of microbial activity enhancing the preservation of ichnofossils are the limited occurrences, at Mount Babel, that have the actual trace(s) (concave epirelief) occurring in a micaceous mudstone overlain by the sandstone (quartzite) bed, with convex hyporelief traces. The micaceous mudstone is cohesive enough to be able to peal off of the quartzite beds to produce the part and counterpart of the ichnofossil(s). The cohesive nature of the mudstone is likely the result of biofilms/microbial mats, as was described by Schieber (1999, p. 3, 7) as having a 'flypaper effect' for a mudstone, enriching it with mica that makes it more cohesive as a rock. Obviously, further research is required as to the setting of these microbially induced sedimentary structures, though, at present, it is only a point of observation.

SYSTEMATIC PALAEONTOLOGY (ICHNOLOGY) The specimens are housed in the University of Alberta Trace Fossil Collection, assigned the numbers T27 - T89. The specimens were painted black, coated with a sublimate of ammonium chloride and, using low-angle light, were photographed at several incremental focal planes to increase depth of field. 'Helicon Focus' and 'PhotoShop CS' are used to 'stitch' these images together in order to enhance the specimen illustration

16 by producing a greater 'depth of field' (showing more of the specimen in sharp focus). The traces, if necessary, were scrubbed using hydrogen peroxide and a medium bristle toothbrush to remove lichen before photography. Terminology used herein has no relation to any organism that would have created the trace(s), and the classification and nomenclature of the trace is based solely on morphological character states (Bromley 1990; Magwood 1992; Pickerill 1994; Ride et al. 2001; Jensen 2003; Bertling et al. 2006; Minter et al. 2007; Gibb et al. 2009).

Cruziana d'Orbigny, 1842

Type ichnospecies. Cruziana rugosa d'Orbigny, 1842 by subsequent designation (Miller 1889, p. 115).

Diagnosis (sensu Gibb et al. (2009, p. 698)). An elongate, bilobate furrow that can vary in depth with respect to the bedding plane, may be straight and/or gently curved (not tightly or sharply) within or upon the bedding surface, and may be composed of repetitive sets (or series) of imprints (ridges) along its length (though in most cases these ridges cannot be differentiated into distinctive set patterns); ridges in central part of furrow usually aligned in a herringbone shape and more or less continuous along length of furrow; also, sometimes, outside herringbone- aligned ridges, the trace may exhibit pair of narrow, comparatively smooth outer zones with or without fine brush-like impressions, and additionally there may be presence of lateral ridges.

Remarks. This was recently commented on by Gibb et al. (2009, p. 698) as: Hantzschel (1975, p. W55) stressed, Rusophycus is not at all equivalent to Cruziana. Cruziana is characterized by the following morphological traits: "elongate bandlike furrows covered by herringbone-shaped ridges" (Hantzschel 1975, p. W55) and not "short bilobate bucklelike fonns, resembling [the] shape of coffee beans" (Hantzschel 1975, p. Wl01) as are characteristic in Rusophycus this

17 practice is accepted by many other authors (e.g. Radwanski & Roniewicz 1963; Crimes 1968; Crimes 1970b, a; Orlowski et al. 1970; Osgood 1970; Birkenmajer & Bruton 1971; Bergstrom 1973; Crimes 1975c, a; Crimes et al. 1977; Fillion & Pickerill 1990; Uchman et al. 2004). Seilacher (1990, p. 651) had a contrary view: "They are united under the ichnogeneric name Cruziana d'Orbigny, whether made in a stationary (coffee bean-shaped 'rusophyciform' expression) or in a bulldozing manner (band-shaped 'cruzianaeform' expression)." Consequently, the two ichnogenera are clearly distinguishable by their distinctive morphological traits, therefore the ichnogenera Cruziana and Rusophycus should not be synonymized, but acknowledged as unique. Fillion & Pickerill (1990, p. 24), also stated that: "Although Seilacher (1970, p. 455) united both long furrows and short excavations (=Rusophycus) under Cruziana because similar scratch marks made it possible to 'attribute burrows of very different outline to the same animals', most subsequent authors considered, as we do, their morphologies to differ significantly and preferred to retain the two distinctive ichnogenera because knowledge of the tracemaker is not taxonomically significant in ichnology." Thus, any forward/backward/declined morphology defines Cruziana and the singular vertical morphology, which creates a near image of the ventral morphology of the tracemaker with little to no lateral movement characterizes Rusophycus.

Cruziana billingsi Fillion & Pickerill, 1990 (Figs. 2-6 A-D, 2-7, 2-17 D)

1972 Diplichnites sp.; Young p. 13 - 14, fig. 6.

Material and locality. All plesiotypes and plastotype from the Fort Mountain Formation, Gog Group, lower Cambrian of the southern Rocky Mountains: T53 shallow representative of ichnospecies from Mount Babel, Alberta; T54 from Lake O'Hara, British Columbia; T55 & T56 from Redoubt Mountain, Alberta. Plastotype: T57 from Mount Babel, Alberta.

18 Diagnosis. See Fillion & Pickerill (1990, p. 25).

Description. Refer to Fillion & Pickerill (1990, p. 25).

Remarks. The specimens both illustrated and others collected for size dimensions, provide the following data: maximum length is essentially indeterminate due to the fact that most specimens are truncated by the edges of the specimen, and the plastotype displays Cruziana billingsi, which transforms into Diplichnites twelvetreesi (Chapman 1928), therefore determining a minimum and maximum length is a frivolous endeavour; maximum width = 27.2 - 49.5 mm; average ridge width = 1.4 - 2.8 mm; maximum depth =1.9- 12.3 mm; and a V-angle = 140° - 180°. Occasionally a bifid ridge morphology is identified, but due to taphonomic variables, the ridge morphology is indeterminate. Cruziana billingsi differs from C. kufraensis Seilacher et ah, 2002, purely in the orientation of the ridges, for in C. kufraensis the ridges are transverse, forming a 180° V-angle, making it impossible to assign definitively anterior and posterior directions to the traces (Seilacher et al. 2002, p. 262). Also, C. kufraensis consistently displays bifid ridges. Another transverse ridged ichnospecies is C. transversa Landing & Brett, 1987. Cruziana transversa was also stated have 'straight furrows'. The traces from the Gog Group have a straight to slightly sigmoidal pattern to the furrows, with a deviation of upwards of 45° from original direction. As stated by Fillion & Pickerill (1990, p. 25): "Cruziana liujingensis Yang, 1983 differs from C. billingsi... in having more angular scratch marks and a V-angle that may be as small as 130 degrees". Due to the transverse nature of the ridges on the paired lobes, Cruziana billingsi falls within the 'Pudica Group' of Seilacher (1970; 1991).

Cruziana dispar (Linnarsson, 1869) (Fig. 2-10 A)

19 Material and locality. Plesiotype: T44 from the Fort Mountain Formation, Gog Group, lower Cambrian, Lake O'Hara, Canada.

Diagnosis. Bi-directional ridges. Anterior ridges are proverse with ridges pointing to anterior, while posterior ridges are retroverse. At front of trace, paired lobes form overall heart-shape (based on Jensen 1990, p. 31).

Description. Front of trace is heart-shaped. Anterior ridges fonn tight V-angles, with apices pointing forward. Posterior ridges are opposite to anterior ridges in direction. Trace is vertical and/or inclined.

Remarks. The specimen was found in situ, therefore providing definitive evidence of the direct downwards direction of motion of the organism, in a head first manner, not a whole body vertically downward direction, as would be observed for Rusophycus. The vertical orientation of the trace is such that one may argue that this would not be an ethology that any organism would engage in, though it has been observed in modern notostracans, which will be elaborated on in a forthcoming publication (Gibb in prep). This 'style' of burrowing was referred to by Seilacher (2007, p. 38, pi. 39) as 'divergent scratching'. The organism is in a head-down position, and the head is flexed back dorsally, allowing it to scratch the distal front of the trace with the anterior appendages. As stated by Seilacher (1970, p. 459), Cruziana dispar displays a similar overall morphology to C. grenvillensis (Dawson 1864), though the angle of inclination of the trace into the substrate is much steeper in C. dispar than in C. grenvillensis. There is no evidence of pleural markings. Seilacher (1970, p. 459), considered such markings help to define C. grenvillensis. Following the Cruziana classification established by Seilacher (1970; 1991), Cruziana dispar is the nominate form of the 'Dispar Group'.

Cruziana irregularis Fenton & Fenton, 1937 (Fig. 2-8 A - D)

20 Material and locality. Plesiotypes: from the Fort Mountain Formation, Gog Group, lower Cambrian: T43 from Lake O'Hara, British Columbia; T58, T59, and T60 from Redoubt Mountain, Alberta.

Diagnosis. Bifid ridges adorn paired lobes. Ridges randomly cross thin median line. Median line is not straight, but has distinct shift of 30° - 50°. Trace may vary between horizontal and vertical orientations.

Description. Near circular overall shape of convex hyporelief trace. Paired lobes do not possess any lateral markings. Lobes are adorned with bifid ridges. Ridges are orientated near to transverse. Thin median line is occasionally criss-crossed by ridges. Median line has well-defined change in direction of 30° - 50°. Trace is vertically declined, 40° - 50° from horizontal.

Remarks. Fenton & Fenton (1937, p. 448-450) first described these traces with the 30° change of direction of digging, therefore implying that this is a locomotive trace with forward direction, as one would observe in Cruziana. They suggested the trace served as an 'egg-receptacle', as discussed further below under C. jenningsi. While the production of this trace as the result of an egg-laying behaviour cannot be discounted, there are a number of possible behavioural alternatives, the most likely being that offodinichnia. Due to the distinct kink in the median line, and the combination of horizontal and declined directions to the trace, no other traces are closely comparable.

Cruziana jenningsi Fenton & Fenton, 1937 (Figs. 2-9 A-E, 2-10 B-C)

21 Material and locality. Plesiotypes from the Fort Mountain Formation, Gog Group, lower Cambrian: 3 specimens on one surface (T35-T37) and T38 from Lake O'Hara, British Columbia; and T61 from Redoubt Mountain, Alberta.

Diagnosis. Oval shaped. Inclined trace with possible transverse ridges on anterior edge. Well-defined ridges on lobes with distinct medial line. Typically truncated posteriorly.

Description. Overall egg-shaped trace with truncated posterior. Anterior is defined by two inflated lobes that meet at thin medial line. Anterior has one to several defined, hypichnial transverse ridges. Thin hypichnial striations occur on two lobes, creating 120° - 160° V-angle at anterior and 100° - 140° V-angle at posterior. Occasionally ridges criss-cross medial line. Trace declined 10° - 50° from horizontal at posterior of trace to near horizontal at anterior.

Remarks. Fenton and Fenton (1937, p. 447-448) described their specimens from the St. Piran Formation of the Gog Group, though remarked that they were 'nests' for egg deposition. Though this statement cannot be definitively \ cfu.ed, it is more likely that the organism was digging into the sand-mud interface for feeding. Both the Lake O'Hara and Redoubt Mountain, Fort Mountain Formation specimens, illustrated herein and the specimens of Fenton at id Fenton (1937) are noted to be the result of digging down into a mud substrate and then the trace was infilled with sand. This explains the exquisite preservation of morphological features in these traces. Seilacher (2007, p. 38, pi. 39) referred to this type of burrowing as a "heads-down (prosocline)... position". Seilacher (1993) erected the ichnosubspecies Cruziana acacensis sandalina from the Silurian of Libya. Seilacher (1993, p. 526) stated that "the steep front part is equally scooped by the cephalon, while the scratches on the rear surface reveal a very different endopodal structure" to that of C. jenningsi. Cruziana acacensis sandalina can be differentiated by the repeated pattern of

22 ridges, while C. jenningsi has a variation in the ridge pattern between the anterior and posterior. The Lake O'Hara and Redoubt Mountain specimens range in sizes from a width of 27.9 - 72.0 mm, length of 27.7 - 91.0 mm and depth of 21.9 - 41.1 mm. The specimens described by Fenton and Fenton (1937) have a width = 54-57 mm, length = 72-78 mm, and the depth = 45-50 mm. Cruziana jenningsi differs from C. irregularis by having a straight median line, not the meandering median line observed in C irregularis.

Cruziana mesodelta isp. nov. (Fig. 2-11 A - B)

Material and locality. Holotype: T62 from the Fort Mountain Formation, Gog Group, lower Cambrian, Redoubt Mountain, Alberta; Paratype: T63 from the Fort Mountain Formation, Gog Group, lower Cambrian, Redoubt Mountain, Alberta (both collected by J. Magwood).

Etymology. Greek meso, middle, to the mesial opening and Greek delt-, triangle, distinctive V-shape within the mesial opening. Accounting for the cruzianaeform ichnofossil.

Diagnosis. Elongate convex hyporelief trace. Paired lobes separated by mesial opening. Defined V-shaped ridges within mesial opening.

Description. Elongated paired lobes with no anterior or posterior closure. Lobes are independent, with mesial opening filled by strongly defined bifid ridges with 100° - 110° V-angle. Strong medial ridges are weaker, thinner and more oblique laterally. No lateral morphology.

Remarks. A distinction was made between the cruzianaeform, Cruziana mesodelta, and the rusophyciform, Rusophycus mesodeltus, to account for the

23 distinct anterior rounded lobes observed on R. mesodeltus, and the elongate, lacking a true beginning and/or end to the trace in C. mesodelta. Obviously, these two ichnospecies are very similar to each other in all regards other than R. mesodeltus has a distinct anterior end. The argument can obviously be made that R. mesodeltus was produced by an organism that 'dwelled' within the substrate, hence the rusophyciform trace, and C. mesodelta does not have an 'ending', therefore cruzianaeform. The fact that we do not have extremely long traces of C. mesodelta does not allow us to conclude that the rusophyciform and cruzianaeform are conspecific and should thus be classified as C. mesodelta. Until that time, the distinction is made that these two forms are unique ichnospecies within separate genera. The dimensions of the specimens are: maximum length = 96.8 - 142.4 mm; maximum width = 42.7 - 88.1 mm; maximum depth = 14.8 - 18.6 mm; width of mesial opening = 9.1 mm; and average ridge width =1.3 — 1.6 mm. Due to the distinct morphology of the paired lobes being separated by a mesial opening and the mesial opening adorned with strong, obliquely oriented bifid ridges, no other ichnospecies are comparable.

Cruziana navicella Fenton & Fenton, 1937 (Fig. 2-12 A - B)

Material and locality. Plesiotypes from the Fort Mountain Formation, Gog Group, lower Cambrian: T42, Mount Babel, Alberta; and T41 (collected by J. Magwood) from Redoubt Mountain, Alberta.

Diagnosis. Complex, elongate, and entwined convex hyporelief trace. Obtuse angled ridges on paired lobes, flanked by lateral ridges.

Description. Twisted on itself, elongate convex hyporelief trace. Turns of trace vary from minute adjustments to as great as 310°, if not larger. Clearly defined bifid ridges adorn paired lobes at V-angle of 140° - 160°. Average ridge width of

24 ~1.7 mm. Median line is not clearly defined, for ridges randomly cross midline. Dimensional ranges of specimens (incomplete or partially encapsulated in rock): maximum length = 210.7 - 272.8 mm; maximum width = 37.6 - 48.6 mm; and maximum trace depth = 21.7 - 24.9 mm.

Remarks. Cruziana navicella is almost a conundrum due to the fact that it is a complex and intricate trace. Not following the standard protocol of Cruziana, in the sense of a straight and/or slightly meandering trace, but one in which turns sharper than a meander and may cross over itself at greater depth. This ichnospecies is morphing into a number of different realms within the substrate, though it is thought that it should maintain the status of Cruziana due to a number of features: two equivalent lobes with striations angled into the midline of the trace, whereupon they are divided by the median line; continuous nature of the trace, therefore the organism which created this trace was engaging in a 'burrowing' as well as travelling behaviour, but not that of simple burrowing; and the presence of cephalon markings on the anterior in some instances and/or distinct transformation into a different ichnospecies. Seilacher (1970, p. 461) stated that Cruziana irregularis, C. jenningsi and C navicella should be classified as a single ichnospecies: C. jenningsi. We do not agree with Seilacher for he stated that since it is only a behavioural change observed that one could not assign different ichnospecies. The morphological characteristics of the three ichnospecies definitively warrant their assignment into separate ichnospecies. Morphologically, the only ichnospecies that is similar to Cruziana navicella is C. irregularis, which has a short, single turn in its trackway. It is obvious that C. irregularis could have developed into C. navicella with continued digging into the substrate. The organisms, in the lower Cambrian, were clearly capable of digging deeply into the substrate and taking advantage of the food sources and/or protection of the substrate. No other ichnospecies of Cruziana are as complex as C. navicella.

25 Cruziana omanica Seilacher, 1970 (Fig. 2-13 A-B)

Material and locality. Plesiotypes from the Fort Mountain Formation, Gog Group, lower Cambrian: T64 from Mount Babel, Alberta; T65 from Redoubt Mountain, Alberta.

Diagnosis. See Gibb et al. (2009, p. 702-704).

Description. Refer to Gibb et al. (2009, p. 704).

Remarks. The specimens from the Gog Group extend the geological range of Cruziana omanica from the lower Cambrian to the Lower Devonian (Uchman, et al. 2004, p. 76; Gibb, et al. 2009, p. 704). The maximum width and length of the Gog representatives of this species are slightly less than those from the Ordovician Stairway Sandstone of Australia (Gibb, et al. 2009), though the morphological characteristics are those of C. omanica. Size should not be a variable as to assignment to ichnospecies, for organisms of varying dimensions are capable of producing similar traces. Further discussion of C. omanica occurs in Gibb et al. (2009, p. 704-705).

Cruziana penicillata Gibb, Chatterton and Pemberton, 2009 (Fig. 2-12 B, 2-13 C)

Material and locality. Plesiotypes: T39 and T40 {Cruziana navicella ending in C. penicillata), collected by J. Magwood from the Fort Mountain Formation, Gog Group, lower Cambrian, Redoubt Mountain, Alberta.

Diagnosis. See Gibb et al. (2009, p. 707).

26 Description. See Gibb et al. (2009, p. 707).

Remarks: Cruziana penicillata was first reported from the Ordovician of central Australia, and it was thought that the organism could possibly have been restricted geographically and/or ecologically. The occurrence of this ichnospecies in the lower Cambrian of Western Canada, extends both the time and geographic ranges of this ichnospecies. The dimensions of the specimen are all within the ranges observed for the Ordovician specimens of the Stairway Sandstone: maximum length = 40.2 - 51.1 mm; maximum width = 26.9 - 27.2 mm; maximum depth = 22.7 - 26.8 mm; and angle of trace from horizontal = 20° - 30°. Seilacher (1970, pi. lh) figured similar specimens, which he identified as Cruziana fasciculata, Seilacher, 1970. Seilacher's (1970) specimens are from the lower Cambrian, and if they are conspecific, this would support chronoichnology. His specimens have bundle sets of 5 or more striations, as is characteristic of C. penicillata. However, C. penicillata lacks a 'turning' of approximately 30° of the ridges to the median line, and C. fasciculata is a horizontal trace. Cruziana penicillata is a symmetrically aligned trace, not taking on an irregular furrow appearance. Cruziana penicillata was discussed thoroughly in Gibb et al. (2009, p. 707).

Cruziana plicata Crimes, Legg, Marcos & Arboleya, 1977 (Fig. 2-13 F, 2-14 A, C)

Material and locality. Collected by J. Magwood, plesiotypes from the Fort Mountain Formation, Gog Group, lower Cambrian: T66 and T67 from Lake O'Hara, British Columbia; and T68 from Redoubt Mountain, Alberta.

Diagnosis. See Crimes et al. (1977, p. 100).

27 Description. Refer to Crimes et al. (1977, p. 100).

Remarks. The three specimens of this species found in the Gog Group present extremes of Cruziana plicata, though their assignment to this taxon is based upon strong evidence, and conviction that the specimens belong within the ichnospecies. The specimens are within the range stated and illustrated by Crimes et al. (1977, p. 100) for V-angle of the ridges on the elongate paired lobes. The ridges, on the paired lobes, occur as pairs and there is no evidence of lateral ridges, thus suggesting their placement within C. plicata and distinguishing them from C. semiplicata Salter, 1853. Our traces are on the small side for the ichnospecies, based on the dimensions presented by Crimes et al. (1977, p. 100). Our specimens possess the following ranges in size: maximum length = 20.7 - 124.8 mm; maximum width = 8.7 - 22.1 mm; average ridge width = 0.3 - 0.7 mm; and maximum depth = 2.5 - 5.7 mm. One of the specimens (T68: Fig. 2-12 F) appears to change from Cruziana plicata to C. problematica upon 'crossing' over a perpendicular trackway of C. problematica (Fig. 2-12 D). Specimen T67 is perhaps atypical for this ichnospecies, based on its small size and short length. However, we believe that these two traces that some workers might identify as a Rusophycus are definitively Cruziana, and more specifically, C. plicata. These traces were formed by forward and slightly declined locomotion, so are classified within Cruziana, though obviously they display a terminal end with rusophyciform morphology. This is just the end of a locomotory Cruziana trace. The fact that at least one of the traces has a forward and declined nature is just an extreme of the ichnospecies. These two small versions, one on top of the other, demonstrate a small size for the ichnospecies, and possibly the oldest ichnological evidence of arthropod sexual behaviour preserved. Assuming that the 'female' dug into the substrate and the 'male' also dug into the substrate, though not as deeply and in a position so as to fertilize the eggs in the semi-protection of a sand veneer.

28 Cruziana problematica (Schindewolf 1928) (Fig. 2-13 D, F-G)

Material and locality. Plesiotype: 3 traces on surface (T69 -T71) from the Fort Mountain Formation, Gog Group, lower Cambrian, of Redoubt Mountain, Alberta.

Diagnosis. Refer to Fillion & Pickerill (1990, p. 26).

Description. See Fillion & Pickerill (1990, p. 26).

Remarks. It has already been suggested that Cruziana problematica can be transitional with C. plicata, as a result of change of behaviour of the organism (probably changing depth of burrow). Further speculation on this phenomenon will be left for future research. The size dimensions portrayed in the specimens from the Gog Group fall close to the dimensions of the specimens described by Fillion & Pickerill (1990, p. 26) for their eastern Canadian specimens of ?upper Cambrian to Lower Ordovician ichnofossils, though the width is slightly greater in the Gog specimens. The dimensions of the lower Cambrian traces are: maximum length = 27.4 - 58.2 mm; maximum width = 7.6 - 8.7 mm; average ridge width = 0.6 - 0.8 mm; maximum depth =1.3-1.8 mm; and V-angle = 160° - 180°. Cruziana problematica differs from C billingsi in the following: it is difficult, due to the size of the specimens of C. problematica to ascertain whether or not the ridges are bifid as observed on C. billingsi; and the ridges are more tightly spaced together on C. problematica. Specimens with better preservation may demonstrate that the C. billingsi and C. problematica from the Gog Group are conspecific, but until that time, C. problematica remains a 'problem'. Cruziana stromnessi (Trewin 1976) is another small trace. The ridges on C. stromnessi begin as transverse ridges but then are deflected at the median line

29 to result in an obtuse V-angle (Fillion & Pickerill 1990, p. 27). It has also been noted (Trewin 1976; Fillion & Pickerill 1990) that C. stromnessi sometimes exhibits lateral ridges. Lateral ridges are clearly not evident in the Gog specimens referred to C. problematica. Seilacher's (1970; 1991) Cruziana strati graphic classification system would place Cruziana problematica, based upon the morphological trait of transverse ridges, within the 'Pudica Group'.

Cruziana ramellensis (Legg 1985) (Fig. 2-10 D, 2-14 B)

1992 Cruziana barbata; Orlowski, p. 24, fig. 7(2).

Material and locality. Plesiotypes: T72 from the Fort Mountain Formation, Gog Group, lower Cambrian, Redoubt Mountain, Alberta; T73 from the Fort Mountain Formation, Gog Group, lower Cambrian, Lake O'Hara, British Columbia.

Diagnosis. Anterior ridges are transverse and posterior ridges curve from steep lateral angle to approximately 100-120° V angle at median line; distinct transverse furrow demarks anterior and posterior ridges.

Description. See Legg (1985, p. 159).

Remarks. Seilacher (2007, p. 192), when he commented on Cruziana barbata, stated that "its resemblance [is] to a moustache and a goat's beard in the rusophyciform version". Though from the diagnosis of Cruziana as having either horizontal movement and/or both vertical and horizontal movement, the specimens demonstrate a cruzianaeform (plesiotype T72 has both posterior movement down into the substrate and plesiotype T73 of depth and forward movement). As stated, T72 depicts a movement both backwards and downwards, thus confirming assignment within Cruziana. Legg's (1985) figures of specimens

30 (pi. 3A-C) also appear to have posterior movement, especially in pi. 3A & B, that have similar anterior cephalic impressions on the anterior that progress deeper into the substrate with a posterior directional movement. The final specimen figured by Legg (1985), on pi. 3C, is perhaps similar to T73 with an apparent anterior direction of movement. The dimensions observed from the two specimens are: maximum length (though truncated at posterior for trace continues into the rock) = 23.9 - 47.6 mm; maximum width = 21.1-65.6 mm; maximum length anterior to transverse furrow = 8.2 - 22.6; maximum depth = 5.7 - 30.4 mm; anterior V-angle = 160° - 180°; and posterior V-angle ~ 100°. The specimens have absolutely no resemblance to the types of Cruziana barbata, and from previous authors (Seilacher 1970; Legg 1985; Seilacher 2007), they are now classified as Cruziana ramellensis, though Legg (1985, p. 159) stated that he thought that they are rusophyciform. As stated above, we do not concur. Rusophycus bonnarensis Crimes, Legg, Marcos & Arboleya, 1977 is only similar with regard to the anterior portion of Cruziana ramellensis looking like a 'moustache'. It lacks the transverse furrow and the posterior 'beard'. Cruziana ramellensis, in accordance with Seilacher (1970; 1991), is classified in the 'Dispar Group' with the division between the directions of the anterior and posterior ridges.

Cruziana rugosa d'Orbigny, 1842 (Fig. 2-15 A - D)

Material and locality. Plesiotypes are all from the Fort Mountain Formation, Gog Group, lower Cambrian: T45 and T74 from Redoubt Mountain, Alberta (collected by J. Magwood); T75 from Redoubt Mountain, Alberta; T76 from Lake O'Hara, British Columbia (collected by J. Magwood).

Diagnosis. See Fillion & Pickenll (1990, p. 26).

31 Description. Convex hyporelief trace. Shallow declined angle or horizontal trace. Defined, deep median furrow. Paired corrugated lobes have at least 9 ridges on each. Median line is defined. Individual ridges vary from 90° - 100° V-angle. Lateral ridges are absent.

Remarks. Cruziana rugosa can be recognized by its corrugated bilobate form. The only other ichnospecies that displays a similar corrugated-like lobe structure is C. imbricata Seilacher, 1970. Cruziana imbricata was diagnosed to have no scratches on the trace. Fillion & Pickerill (1990, p. 27) found it necessary to differentiate C. rugosa from Cfurcifera d'Orbigny, 1842, and stated that it is "the deep, comblike scratch marks that extend from the central furrow to the lateral margin of C. rugosa separate[s] it from C. furcifera". We obviously disagree as to the geological range of C. rugosa as stated by Fillion & Pickerill (1990, p. 27) to be "from the Arenig to the Llandeilo [sic] (Crimes 1975b) and possibly into the Devonian (Bradshaw 1981)". Fillion & Pickerill (1990, p. 27) stated that they did not agree to the assignment of a Gog Group specimen to C. rugosa by Magwood & Pemberton (1988). Fillion & Pickerill (1990, p. 27) proposed that that specimen be assigned to Rusophycus marginatus Bergstrom & Peel, 1988. Rusophycus marginatus does display a corrugated bilobate morphology, but the fine ridges are not confined to each individual lobe, as is observed on C. rugosa, but sweep across the transverse ridges. We believe that the illustrations clearly depict C. rugosa, therefore the range of the ichnospecies is lower Cambrian to "possibly into the Devonian" (Fillion & Pickerill 1990, p. 27). Cruziana rugosa was assigned by Seilacher (1970; 1991) as the nominate ichnospecies for his 'Rugosa Group', and he stated that that group shows little in the way of being constrained strati graphically.

Diplichnites Dawson, 1873

Type ichnospecies. Diplichnites aenigma Dawson, 1873.

32 Diagnosis. See Fillion & Pickerill (1990, p. 31).

Remarks. Seilacher (1985, p. 234) clearly stated that Diplichnites is an undertrack of Cruziana. This would occur when the organism, digging into a mud-sand interface (Seilacher 1955b, 1970; Birkenmajer & Bruton 1971; Goldring & Seilacher 1971; Seilacher 1983, 1985, 1986, 2007), is not digging as 'deeply' as occurred in Cruziana trackways. While Crimes (1970b, p. 64, fig. 6) suggested that whether an organism produced Cruziana or Diplichnites is a function of the speed of locomotion, with Diplichnites resulting from a faster speed of travel for the organism than that of an organism making Cruziana. Crimes (1970b, p. 57, pi. 9f) also considered that some Cruziana to Diplichnites trackways are transitional. This transitional nature was also illustrated by Young (1972, p. 13- 14, figs. 6,7), though he identified them as Diplichnites from the Gog Group. The very same trackways are illustrated here (Figs. 2-7 & 2-17 D) and it is clear that these traces transform from Cruziana to Diplichnites, depending upon the depth to which the organism 'dug' into the substrate. We agree with Donovan's (2010, p. 283) statement that "Cruziana and Rusophycus are trace fossils that were produced by bilaterally symmetrical organisms", and add Diplichnites to that list. These traces are ascribed to arthropods, due to the scratches thought to have been made by paired jointed appendages.

Diplichnites obliquus isp. nov. (Fig. 2-16 B)

Material and locality. Holotype: T47 from the Fort Mountain Formation, Gog Group, lower Cambrian, Lake O'Hara, British Columbia.

Etymology. Due to the angle of the ndges to the midline = oblique.

33 Diagnosis. Bifid ridges with an oblique V-angle and defined mesial gap.

Description. Trackway in convex hyporelief of bifid ridges oriented to create 110° V-angle. Obvious mesial gap between right and left series.

Remarks. Diplichnites obliquus, as can be observed in Fig. 2-16 B, is a trackway that morphs into D. twelvetreesi (Fig. 2-16 A). This obviously was caused by a change in the morphological behaviour of the organism that produced this trackway. The distinct morphological change in ridge direction requires differentiation of these two ichnospecies. It differs fvomD. arboreus, as the ridges part forward (open anterior), whereas the ridges converge forward creating a closure to the anterior in D. arboreus. The trackway has the following size dimensions: length of ridges = 3.2 - 6.2 mm; width between right and left series (mesial gap) = 4.2 mm; and ridge width =1.1 mm. The ridge width and mesial gap width difference between D. obliquus and D. twelvetreesi can be accounted for by a rotation of the appendages to create the trackway. This would alter the angle in which the appendage enters the substrate and the extent to which the leg was flexed. Speculation as to the cause of the morphological change from angled to transverse ridges could be that the organism was originally propelling itself in a forward direction with little regard as to the substrate content. More transverse ridges would have been caused by the legs flexing directly medially under the body with no backwards rotation. This could indicate that the organism was possibly ingesting substrate as it travelled slowly forward.

Diplichnites twelvetreesi (Chapman 1928) (Figs. 2-7, 2-16 A, C-D, 2-17 A, C- D)

1957 Tasmanadia twelvetreesi; Glaessner, p. 103-104, pi. 11, fig. 4. 1972 Diplichnites sp.; Young, p. 13 - 14, fig. 7.

34 1976 Diplichnites sp.; Alpert, p. 234, pi. 1, fig. 3. 1983 Diplichnites binatus; Webby, p. 65 - 68, fig. 3B.

Material and locality. Plesiotypes and Plastotype are from the Gog Group, lower Cambrian. Plesiotypes: T77 from the Fort Mountain Formation at Lake O'Hara, British Columbia; T46 from the Fort Mountain Formation at Redoubt Mountain, Alberta; T48, T91 from the Fort Mountain Formation, Mount Babel, Alberta; T78 from the Lake Louise Formation, Lake O'Hara, British Columbia. Plastotype: T79 from the Fort Mountain Formation, Mount Babel, Alberta.

Diagnosis. Paired transverse bifid ridges or grooves; trackways straight to slightly sigmoidal; and distinct gap between paired series of ridges.

Description. Refer to Webby (1983, p. 68).

Remarks. The first conclusion that must be made is that the range of size of the traces of Diplichnites twelvetreesi from the lower Cambrian of the southern Rocky Mountains has expanded from Webby's (1983, p. 68) numbers: the minute trackways observed on T78 have the following dimensions: length of ridge =1.1 - 2.6 mm; width of ridge = 0.4 mm; and width between series = 1.2 - 2.1 mm. While the largest trackways from the Gog Group, T48, possess the following size ranges: length of ridge = 5.5 - 12.8 mm; width of ridge =1.5 mm; and width between series = 10.6 - 15.8 mm. All of the Gog specimens have a prominent five ridges per series sequence, which was not noted by Webby (1983, p. 68) in his Ordovician specimens from Australia and differs from the four ridges recorded for the Cambrian specimen from Tasmania (Glaessner 1957, p. 103). Due to the near transverse V-angle of the traces, they are easily distinguishable from Diplichnites arboreus Gibb et al., 2009, and D. obliquus where the ridges meet at an angle of approximately 90° - 110°, respectively. Specimens also differ from D. arboreus in that there are no ridges pinching in at

35 the anterior. The transverse orientation of the ridges in D. twelvetreesi sets it apart from D. arboreus, which has ridges oriented in an oblique angle between the right and left series in the trackway.

Monomorphichnus Crimes, 1970b

Type ichnospecies. Monomorphichnus bilinearis Crimes, 1970b.

Monomorphichnus bilinearis Crimes, 1970b (Fig. 2-18 A)

Material and locality. Plesiotypes from the Fort Mountain Formation, Gog Group, lower Cambrian: T50 is from Lake O'Hara, British Columbia; T49 is from Mount Babel, Alberta.

Diagnosis. Refer to Crimes (1970b, p. 57) and Fillion & Pickerill (1990, p. 41).

Description. See Crimes (1970b, p. 57-58).

Remarks. The bilinear ridges occur in sets of four or five, which are much lower than the numbers cited by Crimes (1970b, p. 57-58) of six or seven, but are within the broader range of two to ten ridges of Fillion & Pickerill (1990, p. 41). The ridges have a slight curve distally. The specimens from the lower Cambrian Gog Group have the following morphological dimensions, though the ridge length is truncated in the specimen from Lake O'Hara: length = 13.7 - 65.9 mm; and average ridge width = 1.3 mm. The nature of the ichnospecies, to have two ridges, occurring in pairs allows limited confusion with any other ichnospecies within the ichnogenus, Monomorphichnus.

36 Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya, 1977 (Fig. 2-18 C-D)

Material and locality. Plesiotypes from the Gog Group, lower Cambrian of Mount Babel, Alberta. T52 from the St. Piran Formation and T80 from the Fort Mountain Formation.

Diagnosis. See Crimes et al. (1977, p. 103) and Fillion & Pickerill (1990, p. 42).

Description. See Fillion & Pickerill (1990, p. 42).

Remarks. Both specimens have five ridges per series, which splits the difference cited by Crimes et al. (1990, p. 41) of either four or six ridges per set, while Fillion & Pickerill (1990, p. 42) have a range of two to seven ridges. There does not appear to be a limit as to the number of ridges per set. Some of the sets have a wider proximal ridge that tapers distally, the ridges occasionally display a slight curve distally, and the distance between the ridges (per set) is not consistent. The dimensions observed from the two specimens are: Length = 6.6 - 29.6 mm; width between ridges = 2.2 - 8.6 mm; and average width of ridge is = 1.5 mm. Monomorphichnus lineatus has a single ridge morphology, and therefore it is easily distinguished from M. bilinearis. Desai et al. (2010, p. 239) noted that "Monomorphichnus is a common and abundant trace fossil of the Cambrian sediments of the Tethys (Bhargava & Bassi 1988)". Huges et al. (2005, p. 64) also addressed this issue and stated that the "concentration of trace fossils may be related to palaeoenvironmental or taphonomic factors".

Monomorphichnus trilinearis isp. nov. (Fig. 2-18 B)

37 Material and locality. Holotype: T51 from the Fort Mountain Formation, Gog Group, lower Cambrian on Redoubt Mountain, Alberta.

Etymology. Trilinearis for the three subridges on each ridge.

Diagnosis. Monomorphichnus with three defined markings (subridges) on each individual convex hyporelief ridge.

Description. Convex hyporelief trace. Each ridge is marked with three subridges with one being most prominent and cascading down to second and third subridge. Straight and/or slightly recurved ridges. Tapers distally to single line.

Remarks. The holotype has a series of 3 - 4 ridges per set, though a limit on the number of ridges to define the ichnospecies is counterproductive. The length of the ridges is truncated on both sides of the specimen, therefore the size is limited in this dimension: 34.3 - 104.7 mm. The width of the ridges is from 2.1 - 5.1 mm proximally and 0.8 mm distally. The obvious three-subridge cascading ridge defines the ichnospecies and therefore is dissimilar to the single ridge Monomorphichnus lineatus and the paired ridges of M. bilinearis. Monomorphichnus trilinearis is highly unlikely to be confused with any other ichnospecies within the ichnogenus. It was presumably made by an organism with a subdivided tip to the appendage that scraped the trace in the substrate.

Rusophycus Hall, 1852

Type species. Fucoides biloba Vanuxem, 1842.

Diagnosis. See Fillion & Pickerill (1990, p. 52).

38 Remarks. As stated by Gibb et al. (2009, p. 698 & 711) Cruziana and Rusophycus have caused, and will probably continue to cause, disagreement and controversy in the literature. Trace fossil systematics are based on the morphological characteristics of the trace, and not the inferred behaviour(s) of the trace(s). Osgood (1970, p. 306), when he discussed the justification for naming a number of his specimens as Rusophycus pudicum Hall, 1852 was mistaken in the following statement: "the trilobite burrowed, moved forward a short distance, and then burrowed again. Specimens such as this are similar to Cruziana but there is a basic behavioral difference. The surface of the long bandlike Cruziana is always even, demonstrating that the organism was moving horizontally across the substrate. In contrast the specimen of Rusophycus under discussion shows an alteration of vertical and horizontal movement". Unfortunately, Osgood (1970) not only incorporated behaviour into this discussion, but also the commented that it was made by a trilobite. He stated that Cruziana must be purely horizontal. We do not agree with this. We consider that the extended form of Cruziana may be the result of purely horizontal movement or a combination vertical and horizontal movement. We consider that the Rusophycus form resulted from a purely vertical movement, with very little to no evidence of horizontal and/or lateral movement. The overall morphological shape of a Rusophycus trace has been compared to a coffee bean, and to two coffee beans juxtaposed (Fillion & Pickerill 1990). However, we consider that the best analogy is to that of a single coffee bean with the two lobes slightly separated (Fig. 2-19 A). Crimes (1975a) recognized this issue. This concept is used for all Rusophycus ichnospecies in this work.

Rusophycus arizonensis (Seilacher 1970) (Fig. 2-20 A)

Material and locality. Unfortunately, Seilacher (1970, p. 460) did not designate a holotype for the ichnospecies Cruziana arizonensis, and therefore a lectotype for Rusophycus arizonensis is designated herein as fig. 7.8 (Seilacher 1970, p. 458)

39 from the Flathead Sandstone, Middle Cambrian, in the Lewis and Clark Reserve, Montana, which is housed in the U.S. National Museum (Seilacher 1970, p. 459). Plesiotype: T81 from the Fort Mountain Formation, Gog Group, lower Cambrian, Redoubt Mountain, Alberta.

Diagnosis. Ovoid shaped trace. Transverse trifid ridges across paired lobes. Ridges randomly criss-cross tight median line. Posterolateral thinner ridges apparent.

Description. Convex hyporelief ovoid paired lobe ichnofossil. Lobes are adorned with trifid transverse ridges with V-angle of 150°. Median line defined and tight with ridges criss-crossing randomly. Thin ridges flank posterolateral edges of lobes with V-angle of 90°. No lateral marginal morphological character states are observed. Dimensions of specimen are: maximum length = 90 mm; maximum width = 50 mm; and average width of ridges = 2 mm.

Remarks. The lectotype as illustrated by Seilacher (1970, fig. 7.8) is only a sketch, therefore a degree of interpretative license must be accounted for, along with the diagnosis (Seilacher 1970, p. 460) provided: "Mainly resting tracks. Towards the rear end the sharp endopodal scratches are bordered, and eventually covered, by 'brushed' exopodal lobes". This diagnosis infers far too much with regard to behaviour of the trace-producer rather than the purely morphological traits to define the trace, especially when one consults the remarks section, which contradicts a portion of the diagnosis, with regard to bordering of the trace: "this species... differs from C. semiplicata in the prevalence of short resting tracks and in the rarity of pleural impressions" (Seilacher 1970, p. 460). His remarks section makes a clear statement that the ichnospecies does not have pleural impressions, which is a defining character trait of Cruziana semiplicata. Rusophycus arizonensis has an overall ovoid shape, while C. semiplicata is an elongate trace defined by 'continuous' lobes, flanked by lateral ridges.

40 The only other ichnospecies that displays similar morphological character states to Rusophycus arizonensis is R. mesodeltus. These species are compared and contrasted below under the latter ichnospecies. The reassignment of this species to Rusophycus from Cruziana is purely due to the shape of the trace suggesting that the direction of movement of the organism was in a purely vertical direction, thus placing it within Rusophycus. Seilacher (1970; 1991) classified the cruzianaeform traces within the 'Semiplicata Group', as having "finer exopodal 'brushings'" on the lateral lobes (Seilacher 1970, p. 460). The 'Pudica Group' was defined as: "mainly resting tracks with sharp and multiple endopodal scratches that run mainly transverse" (Seilacher 1970, p. 471). Therefore, there is an issue as to whether Rusophycus arizonensis should remain within the 'Semiplicata Group' with its posterolateral fine ridges adorning the transverse ridges, or to follow the trend that some would refer to the trace as a 'resting trace'. The transverse ridges would key the trace into the 'Pudica Group', but we have issues with classifying any trace using an inferred generalized behaviour for the overall morphological shape. Further research of arthropod traces and revisiting the classification system established by Seilacher (1970; 1991) are required.

Rusophycus eutendorfensis (Linck 1942) (Fig. 2-19 B-D)

Material and locality. All plesiotypes from the Fort Mountain Formation, Gog Group, lower Cambrian, Alberta: T82 from Mount Babel, T3 and T84 collected by James Magwood on Redoubt Mountain.

Diagnosis. Refer to Fillion & Pickerill (1990, p. 54).

Description. Specimens are preserved as convex hyporelief ichnofossils. Two defined lobes with oblique ridges with anterior V-angle of approximately 100° and posterior V-angle of 80°-120°. Median furrow defined and becomes more

41 mesial opening at anterior. Size variable: maximum length = 4.5 - 33.1 mm; and maximum width = 4.2 - 21.4 mm.

Remarks. Fillion & Pickerill (1990, p. 54) diagnosed Rusophycus didymus (Salter 1856) as "resembling two small juxtaposed coffee beans". This point needs be clarified for the traces do display morphology similarity to Fig. 2-19 A, but this is not of two coffee beans, but two halves of a coffee bean, equivalent to the whole bean with the two halves slightly parted to resemble a median furrow and the bean halves as the lobes (Crimes 1975a). In the case of Rusophycus eutendorfensis (Linck 1942), we agree with Fillion & Pickerill (1990, p. 55) that Bromley & Asgaard (1979, p. 66) are exercising "an excessive Tumping' of many different fonns" (Fillion & Pickerill 1990, p. 55) when they referred to R. didymus as nomen dubium, thus assigning it to R. eutendorfensis. Fillion & Pickerill's (1990, p. 55) comments as to the difference between the two are worth reiterating: "Rusophycus eutendorfensis differs from R. didymus in the presence of closely spaced, rarely bifid striellae, which make a V-angle of 70 -160 degrees and appear almost as if they are arranged radially from the center of the coffee-bean". This correlates with our traces from the lower Cambrian Gog Group to the assignment of R. eutendorfensis. The traces on T83 present an interesting contradiction to a hypothesis presented by Crimes (1970a, p. 123), for he stated that "it appears that while some of the smaller trilobites may have preferred a quiet water environment, the absence of traces that would be produced by trilobites less than 0.5 cm wide can only be satisfactorily explained by assuming that after hatching these trilobites behaved planktonically and rarely if ever touched the bottom". The maximum width observed in these small Rusophycus traces is no more than 4.5 mm, thus demonstrating that small multi-limbed organisms inhabited the sediment-water interface, occasionally, if not permanently, therefore exhibiting benthic behaviour and not the planktonic behaviour proposed by Crimes (1970a). Quite possibly the localities that Crimes (1970a) investigated (areas of Wales, the Welsh Borderlands and Leinster), the palaeoenvironmental settings were not conducive

42 for the preservation of traces of this size. We also have a difficult problem assigning an organism to the trace, as Crimes (1970a) has stated that they would have been trilobites. Once again, there is no definitive evidence that the traces were made by trilobites, in the lower Cambrian Gog Group, though Crimes (1970a) negative evidence, of the lack of small traces cannot lend itself to stating that the protaspids and/or freshly hatched arthropods were exclusively planktonic. Chatterton & Speyer (1997) posited that while some trilobite early growth stages are planktonic, the majority are benthic. Also, the setting in which the Gog specimens lived was not a 'quiet water environment', for the sand grains and previously stated ichnofacies for this formation are within the Cruziana ichnofacies, therefore below the 'fair weather wave base' and above the 'storm weather wave base' (Pemberton et al. 1992, p. 54). Rusophycus eutendorfensis is compared and contrasted with the most similar ichnofossil, R. subnotous isp. nov., in the remarks section of the latter ichnospecies. The classification system established by Seilacher (1970; 1991) does not allow for Rusophycus eutendorfensis? We cannot find a 'Group' in which this ichnospecies can be classified. Therefore, it may simply be that more groups are required and/or only Cruziana can be classified using his system.

Rusophycus latus Webby, 1983 (Figs. 2-20 B - C, 2-21 A, E)

1983 Rusophycus latus Webby, fig. 4D, 5C.

Material and locality. Plesiotypes: T31, T28 from the Fort Mountain Formation, Gog Group, Lake O'Hara, Canada; T29 from the Fort Mountain Formation, Gog Group, Mount Babel, Canada.

Diagnosis. See Fillion and Pickerill (1990, p. 55).

43 Description. Refer to Webby (1983, p. 69-71).

Remarks. The specimens assigned to Rusophycus latus herein display a bifid ridge morphology and a V-angle of 140° - 160°, therefore near transverse. The maximum length is 27.8 - 47.1 mm and maximum width of 26.6 - 53.9 mm. The length versus width ratio is equivalent to Webby's (1983) ratio of 0.85 in only one specimen, though we do not consider that this is a defining characteristic of the trace. As Webby (1983, p. 72) discussed, in regard to a length/width ratio for ichnofossils, referring to Crimes (1970a), of Rusophycus having a 'shape factor' and this ratio increased with time. We agree with Webby (1983) that Crimes' (1970a) shape factor increase from 1.0 in the Upper Cambrian to 2.0 in the Lower Ordovician must have been localized. A length/width ratio within the lower Cambrian of the southern Rocky Mountains is from 0.9 - 1.2, therefore not conforming to the 'shape factor', unless one argues that the hypothesis during the Early Cambrian is displaying a number of different morphologies and this was then trimmed down with time. We do not believe the latter to be the situation, for the lower Cambrian trace of R. latus is similar to the same ichnospecies of the Lower Ordovician from Australia of Webby (1983). Other ichnospecies that display transverse ridges and therefore similarity to Rusophycus latus are Rusophycus unilobus, Cruziana salomonis (Seilacher 1990) and C. balsa (Seilacher 1991). Due to the unique morphology demonstrated by R. unilobus and the interfingering of the two lobes to obscure a median line any correlation between the two ichnospecies, aside from the presence of transverse ridges is eliminated. One could argue that there are a number of similarities to the rusophyciform Cruziana salomonis illustrations of Seilacher (1990, fig. 32.5; 1991, fig. 3; 2007, pi. 67), though the exopodite introvert brushing, mentioned in the discussion of C. salomonis (Seilacher 1990, p. 662) are not present in R. latus, thus providing an important distinction between the two ichnospecies. The only other ichnospecies that should be considered is C. balsa (this ichnospecies should be reassigned to Rusophycus, as Seilacher (1991, p. 1572) even stated, it is "two exopodal lobes of rusophycoid burrows").

44 Cruziana balsa, as described by Seilacher (1991, p. 1572) has the overall morphological shape of "an overturned 'balsa' reed boat...[with] transverse endopodal scratches [and] longitudinal exopodal brushings". Therefore, from the overall shape of C balsa and the longitudinal brushings that occur at the anterior of the trace, only the transverse ridges on the lobes connect Rusophycus latus and Cruziana balsa to one another. Referring to the classification by Seilacher (1970; 1991), Rusophycus latus can be either placed within the 'Petraea Group' or the 'Pudica Group'. Placement within the 'Petraea Group' would be based purely on the similarity of R. latus to Cruziana omanica. It is probably more aptly placed within the 'Pudica Group' if one infers that the traces are ovoid and have near transverse ridges upon the lobes to the median line.

Rusophycus mesodeltus isp. nov. (Figs. 2-22 A-B)

Material and locality. All specimens are from the Fort Mountain Formation, Gog Group, lower Cambrian, Redoubt Mountain, Alberta. Holotype: T85 (almost complete biogenic sedimentary structure, minus posterior end); Paratypes: T86 (near complete biogenic sedimentary structure, minus posterior of trace), and T87 (this specimen was collected by J. Magwood).

Etymology. Greek meso, middle, to the mesial opening and Greek delt-, triangle, distinctive V-shape within the mesial opening. Accounting for the rusophyciform ichnofossil.

Diagnosis. Overall oval shape with anterior being slightly more lobate than posterior. Thin ridges on paired convex hyporelief lateral lobes, that sweep medially to form distinct bifid ridges, criss-crossing thin, defined, shallow median line. Median ridges form distinct V-shaped angle at midsagittal line.

45 Description. Convex hyporelief oval shaped trace. Anterior of trace typically more defined in form of two distinct lobes. Two paired lobes. Thin ridges adorn lobes at more acute angle (60° - 80°) then sweep to median lobe to become defined bifid ridges at V-angle of 120°. Median ridges can be paired. Lateral adjustments into sediment possible. Cephalic impressions noted; no other lateral ridges are present.

Remarks. The ichnospecies Cruziana mesodelta was erected to address the cruzianaeform trace, while Rusophycus mesodeltus represents the rusophyciform trace. Both ichnospecies are quite similar with equivalent median V-angle of 120° and a unique bundle of five ridges occurring on the C. mesodelta specimen and on T85 of R. mesodeltus (though 2 bundles of 5 ridges are observed) that are perpendicular to the prominent ridges on both convex left lobes. The C. mesodelta specimen displays an angle of 45° to the median line, while the R. mesodeltus specimen has an angle of 65°. Why the uncharacteristic anomaly of bundles of 5 ridges, perpendicular to the prominent ridge direction, occurs is unclear, though it provides an intriguing morphological characteristic to tie the cruzianaeform and rusophyciform species to each other. Another two ichnospecies that have similar overall morphological characteristics are Rusophycus arizonensis and Cruziana salomonis Seilacher, 1990. Differences between R. mesodeltus and R. arizonensis are that R. arizonensis has the following character states: trifid ridges that are transverse and the most posterior lateral fine ridges occur at an oblique angle, as if produced by lateral appendages (exites) sweeping the sediment. The other ichnospecies, C. salomonis, may actually prove to be Rusophycus and then R. mesodeltus would be a synonym of C. salomonis. The sketches of Seilacher (1990, fig. 32.5; 1991, fig. 3; 2007, pi. 67) do not appear to demonstrate strong similarities to R. mesodeltus, aside from transverse ridges. However, Seilacher (1990, fig. 32.4; 1994, fig. 5B; 2007, pi. 12) sketched quite different illustrations of C. salomonis which confuses the true morphological characters of this species. If one then refers to the images of the ichnospecies in Seilacher (1990, pi. 32.1c-g), at least one image (pi. 32. Id)

46 suggests that C. salomonis is close to R. mesodeltus, though the other images are of specimens that are not similar to R. mesodeltus, with once again the only similarity being the presence of transverse ridges. The fact that Seilacher (1990, p. 662) stated in his diagnosis that there are "rarely preserved exopodite brushings", indicates that the two ichnospecies are morphologically different. Seilacher's (1970; 1991) classification system for Cruziana, lacks a category for Rusophycus mesodeltus. Seilacher (1990, p. 662) stated that Cruziana salomonis falls within the 'Barbata Group' though there is no further discussion as to this group in any subsequent publications.

Rusophycus radwanskii Alpert, 1976 (Fig. 2-23 B)

Material and locality. Plesiotype: T27 collected by J. Magwood from the Fort Mountain Formation, Gog Group, lower Cambrian, Redoubt Mountain, Alberta.

Diagnosis. See Alpert (1976, p. 233-234).

Description. Refer to Alpert (1976, p. 233).

Remarks. Admittedly, the incomplete description and diagnosis of Alpert (1976, p. 233-234) leaves much open to creative imagination. The presence of bidirectional scratch marks is key to the assignment of the ichnofossil from Redoubt Mountain to Rusophycus radwanskii. It is difficult to see in the figured holotype (Alpert, 1976, pi. 3, fig.7) the bidirectional ridges mentioned in his work. However, if the ridges are bidirectional and the shape of the trace is elliptical, our specimen meets his criteria. The specimen from Redoubt Mountain is slightly weathered but still portrays an overall elliptical shape, with the maximum length of 126.3 mm and maximum width of 74.6 mm. The bidirectional ridges are either parallel to the midline near the anterior of the trace and the second direction is located at the posterior of the trace, with a V-angle of

47 70°. The anterior ridges occur in bundles, though due to weathering, only 6 ridges per bundle can be detected. The posterior ridges on the left convex hyporelief lobe have at least 11 defined ridges per bundle. The maximum depth of the trace is 37.3 mm. The trace-producer appears to have made adjustments slightly to the posterior and laterally in digging into the substrate. This ichnofossil differs substantially from any other Rusophycus, therefore it cannot, unfortunately, be compared and/or contrasted with any other ichnospecies. Furthermore, it cannot be classified into any of the groups put forward by Seilacher (1970; 1991).

Rusophycus subnotous isp. nov. (Figs. 2-21 C - E)

Material and locality. All specimens are from the Fort Mountain Formation, Gog Group, lower Cambrian, Lake O'Hara, British Columbia. Holotype: T34 (complete specimen); Paratypes: T32 (truncated at anterior); T33 (complete specimen).

Etymology. Latin subnoto for 'to mark underneath', as the organism marked the substrate under the body.

Diagnosis. Oval, coffee-bean shaped. Distinct median line. Lobes tight to median line at anterior and open at posterior. Ridges on lateral lobes create 80° V angle.

Description. Overall oval shape, with two clearly defined lobes separated by distinct median line at anterior that opens at posterior to equate to median opening. Lobes marked with distinct, single lined ridges that form 80° V angle between lobes. No lateral ridges. Size range of the specimens (accounting for truncated specimen): maximum length is 11.6-20.2 mm; maximum width of

48 10.0-13.5 mm; maximum depth of 2.7-5.4 mm; and maximum ridge width is 1.1 mm.

Remarks. Rusophycus subnotous differs from the most similar ichnospecies, R. didymus, for R. subnotous has a posterior opening and tight to median line anterior lobes, and R. didymus is the exact opposite with an anterior opening and closed posterior lobes. This point can also be used as an example as to why it is not assigned to R. eutendorfensis, with its anterior gape, while R. subnotous has a posterior gape. The only other ichnospecies that displays splaying lobes was described and illustrated by Seilacher (1970, p. 462, fig. 7.5) as Cruziana carinata. Seilacher's (1970) C. carinata, which should probably be reassigned to Rusophycus, also has an anterior mesial opening and tight posterior median line, similar to R. didymus and opposite to R. subnotous. Seilacher (1970, p. 462) noted that the medial ridges, on C. carinata, are angled acutely to the median line, and from his illustration (Seilacher, 1970, fig. 7.5) the V angle appears to range from 60°-70°, thus marginally tighter to the medial line than R. subnotous. The ridges that define C. carinata do not run continuously to the edge of the lobe, but change to a finer striation pattern laterally and have prominent lateral crests flanking the paired lobes. Seilacher's (1970; 1991) classification system does not accommodate the Rusophycus subnotous readily in any of his categories.

Rusophycus unilobus (Seilacher 1970) (Fig. 2-23 A)

Material and locality. Plesiotype: T88 complete specimen from the Fort Mountain Formation, Gog Group, lower Cambrian, Redoubt Mountain, Alberta.

Diagnosis. See Seilacher (1970, p. 473).

Description. Refer to Gibb et al. (2009, p. 711).

49 Remarks. The front of the specimen can almost be considered to display a median furrow, though farther back the interfingering of the left and right striations obscure the remainder of any trace of a median furrow. The size of the specimen, at least in width, is closer to Seilacher's (1970, p. 473) range of 20-25 mm, with a maximum width of 29 mm. Thus the Gog specimen is smaller than the Stairway Sandstone specimen of 54.9 mm, from the Middle Ordovician of Australia (Gibb, et al. 2009, p. 712), though slightly larger than the Lower Devonian specimens from Algeria (Seilacher 1970, p. 473). The geological range for this ichnospecies has now been extended from the lower Cambrian to the Lower Devonian, once again, throwing doubt in the utility of trace fossils for biostratigraphy in contrast to Seilacher's claims in various publications (Seilacher 1970; 1990; 1991; 1994; Seilacher et al. 2002; Seilacher 2007). The specimen appears to have a bifid distal claw morphology, though due to imperfect preservation, this is obscured in the majority of the striations. The maximum length is 43.7 mm; maximum width of 29.2 mm; and maximum depth of 6.5 mm. No lateral morphology is evident. Due to the unique morphology of this ichnospecies, it does not require further distinction from other possible ichnospecies. It should be noted that Seilacher (1970; 1991) classified Cruziana uniloba (Seilacher 1970, p. 473) within the 'Pudica Group', we do not foresee any issue with this placement, even with the classification of this ichnospecies as Rusophycus unilobus.

Rusophycus victorus isp. nov. (Fig. 2-23 C)

1999 Rusophycus avalonensis, MacNaughton & Narbonne, p. 108, fig. 9.A.

Material and locality. Holotype: T30 from the Fort Mountain Formation, Gog Group, lower Cambrian, Redoubt Mountain, Alberta.

50 Etymology. The 'North American Treaty Organization' phonetic alphabet name for the letter 'V is Victor, which represents the shape of the angle between the ridges on the right and left lobes.

Diagnosis. Distinct right and left lobes adorned with clearly defined and widely spaced ridges. Right and left lobes do not meet, producing shallow mesial opening. Ridges uni- or bi-fid. No lateral markings.

Description. Convex hyporelief. Defined ridges produce shallow right and left lobes, maximum depth is 5.21 mm. Right and left ridges do not meet. Ridges are oblique to near transverse: posterior V-angle is 100° and anterior V-angle is 160°. Number of ridges is variable, left lobe with 14 ridges. Longest ridge is 14.67 mm (84 and 9( ridge) of left lobe. Average width between ridges is 1.1 mm. Ridges uni- and/or bi-fid. Mesial opening shallow. Maximum length is 26.21 mm and maximum width is 33.9 mm.

Remarks. Unfortunately not all lobes are created and/or preserved equal, hence there is a disparity between the two lobes on the specimen. MacNaughton & Narbonne (1999, p. 108) figured a specimen (fig. 9.A) which they assigned to Rusophycus avalonensis from the Backbone Ranges Formation, lower Cambrian from the Northwest Territories of Canada. The specimen displays a closer affinity to Rusophycus victorus morphologically than it does to R. avalonensis. MacNaughton & Narbonne's (1999) specimen has shallow though defined right and left lobes with distinct uni- and bi-fid ridges that do not meet, thus creating a shallow mesial opening. The anterior and posterior V-angle is quite similar, 140° and 120° respectively. On their image (fig. 9.A), the right lobe displays at least 10 ridges, while the left is not as well preserved, thus appears to have one less ridge. Rusophycus avalonensis was described by Crimes & Anderson (1985, p. 331-333, figs. 5-2, 5-3, 12-2) as possessing the following morphological character states that are different to those ofR. victorus: bundles of ridges occurring in

51 groups of 5; median line (not a mesial opening); and the individual ridges are much closer together. The only other ichnospecies that is close to Rusophycus victorus was described by Fillion & Pickerill (1990, p. 53-54, pi. 13, fig. 14), R. crimes! Rusophycus crimesi has similar characteristics: with undisturbed region separating the two lobes, hence a mesial opening and clearly defined ridges producing the two lobes that have a maximum depth of 5mm. The morphological differences between the two ichnospecies is that R. crimesi has a V-angle of 90°, the ridges occur as pairs and will occasionally criss-cross each other thus creating a rhombic appearance, and has definitive lateral bifid markings. Yet again, this ichnospecies does not allow for assignment within Seilacher's (1970; 1991) classification.

CONCLUSIONS The Gog Group contains a surprisingly diverse arthropod trace assemblage. Many of the previously described taxa were originally described from strata that are not lower Cambrian in age. Indeed, Seilacher 2007 (p. 66) and Ekdale et al. (1984, p. 63, table V-2) provided range charts for these ichnospecies that suggest they may be useful for biostratigraphy, but some ichnospecies were stated to have a younger geological age than the lower Cambrian of the Gog Group. Clearly, these range charts need to be revised. Some of these traces have been noted by one of us (Gibb in prep) in the Recent. Indeed, the notostracan, Triops, in lab experiments, was able to produce several Cruziana and Rusophycus species found in the Gog Group and other Palaeozoic strata. Seilacher (2007, p. 190), in order to support his use of Cruziana ichnospecies for the biostratigraphy of Gondwana, noted that some of his zonal taxa occur in strata of different ages in Laurentia, suggested that "a case of homeomorphy caused by the convergent evolution of comblike appendages in two unrelated clades of burrowing trilobites". If such is the case, then we are observing numerous cases of'convergent evolution' throughout geological time

52 to account for the numerous ichnospecies that occur in the lower Cambrian Gog Group in Western Canada but also occur in younger strata elsewhere. Most authors readily agree that different organisms may make identical traces and one organism is often able to make a variety of different traces. Clearly to invoke convergent evolution every time a trace taxon is found in strata of a different age in a different region in order to prop up a biostratigraphic scheme based on ichnofossils is awkward, if not absurd. While we acknowledge that some organisms may have had behaviours that produced traces that are unique in the geologic record (in form and time range), or have been of such a large size that no other organism living at another time could have been able to make the same trace, in general that was not the case. Another type of problem associated with using trace fossils in biostratigraphy is the possibility for burrowing and boring organisms to produce traces in substrates that may be much older than the traces (possible even hundreds of millions of years older). This is, however, extremely unlikely to be have been the case for the traces described from the Gog Group. Thus biostratigraphic schemes based on trace fossils are inherently unreliable. It is much easier for similar behaviours to evolve than for identical organisms to evolve. Biostratigraphy based upon body fossils is much more reliable than biostratigraphy based upon ichnofossils. The following ichnospecies that are found in the Gog Group are listed with, in brackets, the geological range stated by Seilacher (2007, pi. 66) or Ekdale et al. (1984, p. 63, table V-2): C. dispar (lower Cambrian); C. barbata [now C ramellensis] (middle Cambrian); C. arizonensis [now R. arizonensis for rusophyciform] (middle Cambrian); C jenningsi (lower to upper Cambrian); C. omanica (upper Cambrian, though this ichnospecies has already been demonstrated to occur in Middle Ordovician strata in Australia and Lower Devonian rocks in Ukraine (Gibb, et al. 2009)); C. rugosa (Middle to Upper Ordovician); and C. uniloba [now R. unilobus] (Lower Devonian). The palaeoenvironmental signals provided by trace fossils are perhaps by contrast even more reliable than those provided by body fossils. Traces are seldom transported from the environment where they were made. Body fossils are more often found in a different environment from that where they lived (e.g.

53 planktonic and pelagic organisms are usually preserved in sea or lake bottom sediments; and body fossils may be winnowed out of one environment and transported to another or even carried into deeper water environments by mass wastage events). In this context, it is important to note that few body fossils have been found in the Gog Group, and most of those found have been collected from the Peyto Formation, a formation not noted for its trace fossils. Thus the traces obtained from the Gog Group provide us with a plenitude of data, not available from body fossils, most importantly, that there were numerous organisms living in the shallow seas of western Canada. We know from lower Cambrian Konservat Lagerstdtten like that obtained from Chengjiang in China (Hou, et al. 2004) that numerous diverse and disparate organisms, including arthropods, were living in quieter, deeper water marine environments during the lower Cambrian. The Gog Group shows us that there was a wealth of life, including arthropods, in shallow, more energetic near shore environments. It also shows us that these organisms had evolved a wide variety of behaviours. The lower Cambrian of the southern Rocky Mountains of Canada provides us with a clearer window into the shallow marine environments in the geological past, with the preservation of a diverse number of arthropod ichnospecies. Crimes (1970b, fig. 7, p. 65) presented distribution pie charts for two different environments from the Upper Cambrian of Wales. The pie charts we have produced for the different localities (Figs. 2-24-2-26), where most of the traces described herein were collected, do not match in proportion to either of his charts. Indeed, a chart produced from the data from all three of these Gog localities (Fig. 2-27) does not match the distributions of either of his charts. There is some correspondence between Crimes (1970b, fig. 7a, p. 65), which is his higher energy setting in Wales, to Redoubt Mountain and Lake O'Hara. There is little correspondence between our data and those he provided for a lower energy setting in southern Caernarvonshire (1970b, fig. 7b, p. 65). This makes sense, since our samples were collected from strata that were deposited on a continental shelf, in a moderately high-energy setting. These strata include thick sandstone (quartzite) beds as well as much thinner fine-grained clastic beds.

54 Crimes (1970b, p. 65-66), stated his reasoning for the different distributions with regard to the different environments (higher energy in north Caernarvonshire and lower energy in southern Caernarvonshire) were: "these traces therefore presumably represent the different life-activities of the trilobites in a given environment, rather than any response to a change in environment... it appears that the rareness of furrow [Cruziana] relative to surface traces [in southern Caernarvonshire] is the result of a response of the animals to the lower energy environment. Thus, unless predators were fewer in the lower energy environment, the furrowing technique must be for physical protection or food- hunting in the higher energy environment, rather than to keep out of sight of predators". We concur with this statement and add that in the higher energy environments, the organisms would be required to dig into the sand to reach the possibly more nutrient-rich mud or organisms living in the sands, thus requiring 'furrowing' and the subsequent preservation of Cruziana. We have observed several types of evidence for the existence of microbial mats at some levels in the Gog Group. These include the distinctive wrinkled appearance on some bedding plane surfaces, and concentrations of mica crystals parallel to bedding in some of the finer clastic beds clinging to the surfaces of thicker quartzite units. Microbial mats have been shown to be able to trap and concentrate mica crystals that are being carried across the sea floor by currents. These mica-rich fine sands and shale tend to be more flexible and often remain adhered to sandstone slabs that have fallen from cliffs, for some time after they enter the talus slope. We observe this intrastratal, mud-sand interface in outcrop at Lake O'Hara (Figs. 2-5 A-B) and at Mount Babel with micaceous mudstones peeling off of the thick quartzite blocks, thus preserving the concave hyporelief trace in the mudstone and convex hyporelief trace on the undersurface of the quartzite (Fenton & Fenton 1937; Martinsson 1970; Seilacher 1970; Birkenmajer & Bruton 1971; Seilacher 1983, 1985). The pie charts that we provide are based on the specimens collected and present at the University of Alberta. They are influenced by collecting and preservation biases. Most of them are on the surfaces of sandstone beds that are

55 thick enough to be competent and thin enough not to be too heavy to transport. Clearly, the collections were based upon selecting the best-preserved samples that could be transported (often involving being carried a considerable distance down a steep mountainside). Large slabs covered in traces (often Cruziana and/or Diplichnites trackways and sometimes groups of Rusophycus) were left in the field and not counted. It is interesting to note the higher percentage of recorded Diplichnites at Mount Babel in comparison to the other two localities. This phenomenon is partly explained when one observes the slab shown in Figure 2-7, where numerous trackways on a single bedding surface show Cruziana and Diplichnites transforming into one another, back and forth. We saw numerous examples of this on large slabs at Mount Babel, though we only recorded the numbers from a single large specimen. We plan to return to Mount Babel to study these slabs to see if it is possible to relate current direction data (obtainable from cross-bedding within the quartzite blocks) to the directions of these trackways of Cruziana and Diplichnites. This may be able to provide additional interesting information on arthropod behaviour in these shallow lower Cambrian seas. The concept of an ichnofacies, a body of strata defined by a distinctive set of trace fossils, is well established in the literature. One of the best-known ichnofacies is the Cruziana ichnofacies, which shows a rich trace fossil diversity, with horizontal repichnia, cubichnia, and subvertical burrows. It is considered to represent mid to distal continental shelf locations below normal wave base, but the sediments may be affected by storms (Pemberton, et al. 1992). The predominance of Cruziana, high diversity of the ichnofauna, and nature of the sediments from the localities we have described with the Gog Group would suggest that they should all be assigned to the Cruziana ichnofacies.

56 REFERENCES

AITKEN, J.D., 1969. Documentation of the sub-Cambrian unconformity, Rocky Mountains Main Ranges, Alberta. Canadian Journal of Earth Sciences 6, 193-200.

AITKEN, J.D., 1983. Cambrian and Lower Ordovician - Sauk Sequence. 96-124 in Stott, D.F. & Aitken, J.D. (eds), Sedimentary cover of the craton in Canada. Geology ofCanda, no. 5. Geological Survey of Canada, Ottawa.

AITKEN, J.D., 1989. The Sauk Sequence — Cambrian to Lower Ordovician miogeocline and platform. 105-119 in Ricketts, B.D. (ed), Western Canada sedimentary basin: A case history. Canadian Society of Petroleum Geologists, Calgary.

ALPERT, S.P., 1976. Trilobite and star-like trace fossils from the White-Inyo Mountains, California. Journal of Paleontology 50, 226-239.

ARAI, M.N. & MCGUGAN, A., 1968. A problematical coelenterate(?) from the Lower Cambrian, near Moraine Lake, Banff area, Alberta. Journal of Paleontology 42, 205-209.

BERGSTROM, J., 1973. Organization, life, and systematics of trilobites. Fossils and Strata 2, 1-69.

BERGSTROM, J. & PEEL, J.S., 1988. Lower Cambrian trace fossils from northern Greenland. Rapp. Gronlands geol. Unders. 13 7, 43-53.

BERTLING, M., BRADDY, S.J., BROMLEY, R.G., DEMATHIEU, G.R., GENISE, J.,

MIKULAS, R., NIELSEN, J.K., NIELSEN, K.S.S., RINDSBERG, A.K., SCHLIRF,

M. & UCHMAN, A., 2006. Names for trace fossils: a uniform approach. Lethaia 39, 265-286.

BHARGAVA, O.N. & BASSI, U.K., 1988. Trace fossils from the Palaeozoic- Mesozoic sequences of Spiti-kinnaur (Himachal Pradesh) with comments on paleoenvironment control of their frequency. Journal Geological Society of India 32, 47-68.

BIRKENMAJER, K. & BRUTON, D.L., 1971. Some trilobite resting and crawling traces. Lethaia 4, 303-319.

57 BOND, G.C. & KOMINZ, M.A., 1984. Construction of tectonic subsidence curves for the early Paleozoic miogeocline, southern Canadian Rocky Mountains: Implications for subsidence mechanisms, age breakup, and crustal thinning. Geological Society of America Bulletin 95, 155-173.

BRADSHAW, M.A., 1981. Paleoenvironmental interpretations and systematics of Devonian trace fossils from the Taylor Group (lower Beacon Supergroup), Antarctica. New Zealand Journal of Geology and Geophysics 24, 615-652. BRIGGS, D.E.G., 1990. Early arthropods: Dampening the Cambrian Explosion. 24-43 in Mikulic, D.G. (ed), Arthropod Paleobiology. Short courses in paleontology. Number 3. The Paleontological Society,

BROMLEY, R.G., 1990. Trace fossils: Biology and taphonomy. Unwin Hyman, London, 280 p.

BROMLEY, R.G. & ASGAARD, U., 1979. Triassic freshwater ichnocoenoses from Carlsberg Fjord, East Greenland. Palaeogeography, Palaeoclimatology, Palaeoecology 28, 39-80.

CHAPMAN, F., 1928. On some remarkable annelid remains from the Arthur River, N.W. Tasmania. Papers and Proceedings ofthe Royal Society of Tasmania 1-5.

CHATTERTON, B.D.E. & SPEYER, S.E., 1997. Ontogeny. 173-247 in Kaesler, R.L. (ed), Treatise on Invertebrate Paleontology, Part O, Arthropoda I, Trilobita, Revised. The Geological Society of America, Inc. and The University of Kansas, Boulder, Colorado and Lawrence, Kansas.

CRIMES, T.P., 1968. Cruziana: A stratigraphically useful trace fossil. Geological Magazine 105, 360-364.

CRIMES, T.P., 1970a. The significance of trace fossils in sedimentology, stratigraphy, and palaeoecology with examples from Lower Palaeozoic strata. 101-126 in Crimes, T.P. & Harper, J.C. (eds), Trace fossils. Seel House Press, Liverpool.

CRIMES, T.P., 1970b. Trilobite tracks and other trace fossils from the Upper Cambrian of North Wales. Geological Journal 7, 47-68.

CRIMES, T.P., 1975a. The production and preservation of trilobite resting and

58 furrowing traces. Lethaia 8, 35-48.

CRIMES, T.P., 1975b. The stratigraphical significance of trace fossils. 109-130 in Frey, R.W. (ed), The study of trace fossils. Springer-Verlag, New York.

CRIMES, T.P., 1975c. Trilobite traces from the Lower Tremadoc of Tortworth. Geological Magazine 112, 33-46.

CRIMES, T.P. & ANDERSON, M.M., 1985. Trace fossils from late Precambrian- Early Cambrian strata of southeastern Newfoundland (Canada): Temporal and environmental implications. Journal of Paleontology 59, 310-343.

CRIMES, T.P., LEGG, I., MARCOS, A. & ARBOLEYA, M., 1977. ?Late Precambrian- low Lower Cambrian trace fossils from Spain. 91-138 in Crimes, T.P. & Harper, J.C. (eds), Trace fossils 2: Geological Journal, Special issue 9. D'ORBIGNY, A., 1842. Voyage dans TAmerique meridionale (le Bresil, la Republique oriental de T'uruguay, la Republique Argentine, la Patagonie, la Republique du Chili, la Republique de Bolivia, la Republique du Peron) execute pendant les annees 1826, 1827, 1829, 1830, 1831, 1832, et 1833. Pitois-Levrault, Levrault, Paris, Strasbourg, 188 p.

DAWSON, J.W., 1864. On the fossils of the genus Rusophycus. Canadian Naturalist and Geologist 1, 363-367, 458.

DAWSON, J.W., 1873. Impressions and footprints of aquatic animals and imitative markings on Carboniferous rocks. American Journal of Science 105, 16- 24.

DESAI, B.G., SHUKLA, R. & SAKLANI, R.D., 2010. Ichnology of the Early Cambrian Tal Group, Nigalidhar Syncline, Lesser Himalaya, India. Ichnos 77,233-245.

DESJARDINS, P.R., MANGANO, M.G., BUATOIS, L.A. & PRATT, B.R., 2010. Skolithos pipe rock and associated ichnofabrics from the southern Rocky Mountains, Canada: colonization trends and environmental controls in an early Cambrian sand-sheet complex. Lethaia 43, 507-528. DONOVAN, S.K., 2010. Cruziana and Rusophycus: trace fossils produced by trilobites... in some cases? Lethaia 43, 283-284.

EKDALE, A.A., BROMLEY, R.G. & PEMBERTON, S.G., 1984. Ichnology: Trace

59 fossils in sedimentology and stratigraphy. Society of Economic Paleontologists and Mineralogists, Tulsa, Oklahoma, 317 p.

FENTON, C.L. & FENTON, M.A., 1937. Trilobite "nests" and feeding burrows. The American Midland Naturalist 18, 446-451.

FILLION, D. & PICKERILL, R.K., 1990. Ichnology of the Upper Cambrian? to Lower Ordovician Bell Island and Wabana groups of eastern Newfoundland, Canada. Palaeontolographica Canadiana 17, 1-119.

GIBB, S., CHATTERTON, B.D.E. & PEMBERTON, S.G., 2009. Arthropod ichnofossils from the Ordovician Stairway Sandstone of central Australia. Memoirs of the Association of Australasian Palaeontologists 37, 695-716.

GLAESSNER, M.F., 1957. Palaeozoic arthropod trails from Australia. Palaontologische Zeitschrift 31, 103-109.

GOLDRING, R., 1985. The formation of the trace fossil Cruziana. Geological Magazine 122, 65-72.

GOLDRING, R. & SEILACHER, A., 1971. Limulid undertracks and their sedimentologic implications. Neues Jahrbuch fur Geologie und Paldontologie. Abhandlungen 137, 422-442.

HAGADORN, J.W. & BOTTJER, D.J., 1997. Wrinkle structures: Microbially mediated sedimentary structures common in subtidal siliciclastic settings at the Proterozoic-Phanerozoic transition. Geology 25, 1047-1050.

HAGADORN, J.W. & BOTTJER, D.J., 1999. Restriction of a Late Neoproterozoic biotope: Suspect-Microbial structures and trace fossils at the Vendian- Cambrian Transition. Palaios 14, 73-85. HALDEMAN, S.S., 1840. Supplement to Number One of "A monograph of the Limniades, or fresh-water univalve shells of North America," containing descriptions of apparently new animals in different classes, and the names and characters of the subgenera in Paludina and Anculosa. J. Dobson, Philadelphia.

HALL, J., 1852. Palaeontology of New York. State of New York, Albany, NY, 362

P- HANTZSCHEL, W., 1975. Treatise on Invertebrate Paleontology: Part W,

60 Miscellanea, Supplement 1, Trace Fossils and Problematica. The Geological Society of America, Inc. and The University of Kansas, Boulder, Colorado, and Lawrence, Kansas, W269 p.

HEIN, F.J., 1987. Tidal/Littoral offshore shelf deposits — Lower Cambrian Gog Group, southern Rocky Mountains, Canada. Sedimentary Geology 52, 155-182. HEIN, F.J. & ARNOTT, R.W., 1983. Precambrian Miette conglomerates, Lower Cambrian Gog quartzites and modern braided outwash deposits Kicking Horse Pass Area. Canadian Society of Petroleum Geologists field trip guidebook, 46 p.

HEIN, F.J., ROBB, G.A., WOLBERG, A.C. & LONGSTAFFE, F.J., 1991. Facies descriptions and associations in ancient reworked (?trangressive) shelf sandstones: Cambrian and Cretaceous examples. Sedimentology 38, 405- 431.

Hou, X.-G., ALDRIDGE, R.J., BERGSTROM, J., SIVETER, D.J., SIVETER, D., J. & FENG, X.-H. (eds), 2004. The Cambrian fossils of Chengjiang, China : The flowering of early animal life. Blackwell Science Ltd., Oxford, 233 p.

HUGHES, N.C., PENG, S., BHARGAVA, O.N., AHLUWALIA, A.D., WALIA, S.,

MYROW, P.M. & PARCHA, S.K., 2005. Cambrian biostratigraphy of the Tal Group, Lesser Himalaya, India, and early Tsanglangpuan (late early Cambrian) trilobites from the Nigali Dhar syncline. Geological Magazine 142, 57-80.

JENSEN, S., 1990. Predation by early Cambrian trilobites on infaunal worms — evidence from the Swedish Mick witzia Sandstone. Lethaia 23, 29-42.

JENSEN, S., 2003. The Proterozoic and earliest Cambrian trace fossil record; patterns, problems and perspectives. Integrative and Comparative Biology 43,2X9-22%.

LANDING, E. & BRETT, C.E., 1987. Trace fossils and regional significance of a Middle Devonian (Givetian) disconformity in southwestern Ontario. Journal of Paleontology 61, 205-230. LEGG, I., C, 1985. Trace fossils from a Middle Cambrian deltaic sequence, north

61 Spain. 151-165 in Curran, H.A. (ed), Biogenic structures: Their use in interpreting depositional environments. Special Publication No. 35. Society of Economic Paleontologists and Mineralogists, LlNCK, O., 1942. Die spur Isopodichnus. Senckenbergiana 25, 232-255.

LINNARSSON, J.G.O., 1869. Om nagra forsteningar fran Vestergotlands sandstenslager. Ofversikt av Kongliga Vetenskaps-Akademiens Handlingar 337-357.

MACNAUGHTON, R.B. & NARBONNE, G.M., 1999. Evolution and ecology of Neoproterozoic - lower Cambrian trace fossils, NW Canada. Palaios 14, 97-115. MAGWOOD, J.P.A., 1988. The ichnology of the Lower Cambrian Gog Group, Lake Louise, Alberta, Master of Science (unpublished). University of Alberta, 306 p.

MAGWOOD, J.P.A., 1992. Ichnotaxonomy: a burrow by any other name...? 15-33 in Maples, C.G. & West, R.R. (eds), Trace fossils. Short courses in paleontology. No. 5. University of Tennessee, Knoxville.

MAGWOOD, J.P.A. & PEMBERTON, S.G., 1988. Trace fossils of the Gog Group, a lower Cambrian tidal sand body, Lake Louise, Alberta. 14 in Landing, E., Narbonne, G.M. & Myrow, P.M. (eds), Trace fossils, small shelly fossils, and the Precambrian-Cambrian boundary; proceedings. Bulletin - New York State Museum, Albany, NY.

MAGWOOD, J.P.A. & PEMBERTON, S.G., 1990. Stratigraphic significance of Cruziana: New data concerning the Cambrian-Ordovician ichnostratigraphic paradigm. Geology 18, 729-732.

MANTON, S.M. & ANDERSON, D.T., 1979. Polyphyly and the evolution of arthropods. 269-321 in House, M.R. (ed), The origin of major invertebrate groups. Academic Press, London.

MARTINSSON, A., 1970. Toponomy of trace fossils. 323-330 in Crimes, T.P. & Harper, D. A.T. (eds), Trace Fossils. Geological Journal, Special Issue No. 3. Seel House Press, Liverpool.

MCKERROW, W.S., SCOTESE, C.R. & BRASIER, M.D., 1992. Early Cambrian

62 continental reconstructions. Journal of Geological Society, London 149, 599-606. MILLER, S.A., 1889. North American geology and palaeontology for the use of amateurs, students, and scientists. Cincinnati, 718 p.

MILLER, S.A. & DYER, C.B., 1878. Contributions to paleontology, no. 1. The Journal of the Cincinnati Society of Natural History 1, 24-39.

MINTER, N.J., BRADDY, S.J. & DAVIS, R.B., 2007. Between a rock and a hard place: arthropod trackways and ichnotaxonomy. Lethaia 40, 365-375.

NICHOLSON, H.A., 1873. Contributions to the study of the errant annelides of the older Paleozoic rocks. Proceedings of the Royal Society of London 21, 288-290. NOFFKE, N., 2010. Geobiology: Microbial mats in sandy deposits from the Archean Era to Today. Springer-Verlag, Berlin, 194 p.

NOFFKE, N., GERDES, G., KLENKE, T. & KRUMBEIN, W.E., 2001. Microbially induced sedimentary structures; a new category within the classification of primary sedimentary structures. Journal of Sedimentary Research 71, 649- 656.

ORLOWSKI, S., 1992. Trilobite trace fossils and their stratigraphical significance in the Cambrian sequence of the Holy Cross Mountains, Poland. Geological Journal 27, 15-34.

ORLOWSKI, S., RADWANSKI, A. & RONIEWICZ, P., 1970. The trilobite ichnocoenoses in the Cambrian sequence of the Holy Cross Mountains. 345-360 in Crimes, T.P. & Harper, J.C. (eds), Trace fossils. Seel House Press, Liverpool. OSGOOD, R.G.J., 1970. Trace fossils of the Cincinnati area. Palaeontographica Americana 6, 281-439.

PEMBERTON, S.G., MACEACHERN, J.A. & FREY, R.W., 1992. Trace fossil facies models: Environmental and allostratigraphic significance. 47-72 in Walker, R.G. & James, N.P. (eds), Facies models: Response to sea level change. Geological Association of Canada,

PEMBERTON, S.G. & MAGWOOD, J.P.A., 1990. A unique occurrence of

63 Bergaueria in the Lower Cambrian Gog Group near Lake Louise, Alberta. Journal of Paleontology 64, 436-440.

PFLUGER, F., 1999. Matground structures and redox facies. Palaios 14, 25-39.

PICKERILL, R.K., 1994. Nomenclature and taxonomy of invertebrate trace fossils. 3-42 in Donovan, S.K. (ed), The Paleobiology of trace fossils. The John Hopkins University Press, Baltimore.

PORADA, H. & BOUOUGRI, E.H., 2007. Wrinkle structures - a critical review. Earth-Science Reviews 81, 199-215.

PORADA, H., GHERGUT, J. & BOUOUGRI, E.H., 2008. Kinneyia-type wrinkle structures — criteria review and model of formation. Palaios 23, 65-11.

PRANTL, F., 1946. Two new problematic trails from the Ordovician of Bohemia. Bulletin International Science Academie Tcheque 46, 49-59.

RADWANSKI, A. & RONIEWICZ, P., 1963. Upper Cambrian trilobite ichnocoenosis from Wielka Wisniowka (Holy Cross Mountains, Poland). Acta Palaeontologica Polonica 8, 259-279, 210 pi.

RICHTER, R., 1850. Aus der thuringischen Grauwacke. Deutsche Geologische Gesellschaft, Zeitschrift 2, 198-206.

RIDE, W.D.L., COGGER, H.G., DUPUIS, C, KRAUS, O., MINNELLI, A., THOMPSON,

F.C. & TUBBS, P.K., 2001. International Code of Zoological Nomenclature. International Trust of Zoological Nomenclature, London, 306 p.

SALTER, J.W., 1853. On the lowest fossihferous beds of North Wales. Report of the British Association for the Advancement of Science (Belfast) for 1852. Geology and Physical Geography section 56-58.

SALTER, J.W., 1856. On fossil remains in the Cambrian rocks of the Longmynd and north Wales. Geological Society of London Quarterly Journal 12, 246-252. SCHIEBER, J., 1999. Microbial mats in terrigenous elastics: The challenge of identification in the rock record. Palaios 14, 3-12.

SCHINDEWOLF, O.H., i928. Studien aus dem Marburger Buntsandstein III-VTI.

64 IV. Isopodichnus problematicus (Schdwf.) im Unteren und Mittleren Buntsandstein. Senckenbergiana 10, 27-37.

SCOTESE, C.R., 2002. Early Cambrian Climate (540 million years ago), in Scotese, C.R. (ed), Paleomap Project.

SEILACHER, A., 1955a. Spuren und Fazies im Unterkambrium. in Schindewolfe, O.H. & Seilacher, A. (eds), Beitrdge zur Kenntnis des Kambriums in der Salt Range (Pakistan). Akademie der Wissenschafter und der Literatur Abhandlungen der Mathematisch-Naturwissenschaftlichen Klasse, 10.

SEILACHER, A., 1955b. Spuren und Lebenweise der Trilobiten. 342-372 in Schindewolfe, O.H. & Seilacher, A. (eds), Beitrdge zur Kenntnis des Kambriums in der Salt Range (Pakistan). Akademie der Wissenschafter und der Literatur Abhandlungen der Mathematisch- Naturwissenschaftlichen Klasse, 10.

SEILACHER, A., 1970. Cruziana stratigraphy of "non-fossiliferous" Palaeozoic sandstones. 447-476 in Crimes, T.P. & Harper, J.C. (eds), Trace fossils. Seel House Press, Liverpool.

SEILACHER, A., 1983. Upper Paleozoic trace fossils from the Gilf Kebir-Abu Ras area in southwestern Egypt. Journal of African Earth Sciences 1, 21-34.

SEILACHER, A., 1985. Trilobite paleobiology and substrate relationships. Transactions of the Royal Society of Edinburgh: Earth Sciences 76, 231- 237.

SEILACHER, A., 1986. Evolution of Behavior as Expressed in Marine Trace Fossils. 62-87 in Nitecki, M.H. (ed), Evolution of animal behavior: Paleontological and field approaches. Oxford University Press, New York.

SEILACHER, A., 1990. Paleozoic trace fossils. 649-670 in Said, R. (ed), Geology of Egypt. Rotterdam, Balkema.

SEILACHER, A., 1991. An updated Cruziana stratigraphy of Gondwanan Palaeozoic Sandstones. 1565-1580 in Salem, M.J., Hammuda, O.S. & Eliagoubi, B.A. (eds), The geology of Libya. Elsevier, Amsterdam.

SEILACHER, A., 1993. Evolution of burrowing behaviour in Silurian trilobites:

65 Ichnosubspecies of Cruziana acacensis. 523-530 in Salem, M.J., Mouzughi, A.J. & Hammuda, O.S. (eds), The geology of the Sirt Basin. Elsevier, Amsterdam.

SEILACHER, A., 1994. How valid is Cruziana stratigraphy? Geologische Rundschau 83, 752-758.

SEILACHER, A., 2007. Trace fossil analysis. Springer-Verlag, Berlin, 226 p.

SEILACHER, A., LUNING, S., MARTIN, M.A., KLITZSCH, E., KHOJA, A. & CRAIG, J., 2002. Ichnostratigraphic correlation of Lower Palaeozoic elastics in the Kufra Basin (SE Libya). Lethaia 35, 257-262.

SLOSS, L.L., 1963. Sequences in the cratonic interior of North America. Geological Society of America Bulletin 74, 93-113.

SLOSS, L.L., 1976. Areas and volumes of cratonic sediments, western North America and eastern Europe. Geology 4, 272-276.

TREWIN, N.H., 1976. Isopodichnus in a trace fossil assemblage from the Old Red Sandstone. Lethaia 9, 29-37'.

UCHMAN, A., DRYGANT, D., PASZKOWSKI, M., POREBSKI, S.J. & TURNAU, E., 2004. Early Devonian trace fossils in marine to non-marine redbeds in Podolia, Ukraine: palaeoenvironmental implications. Palaeogeography, Palaeoclimatology, Palaeoecology 214, 67-83.

VAIL, P.R., MITCHUM, R.M. & THOMPSON, S., Ill, 1977. Seismic stratigraphy and global changes of sea level; Part 4, Global cycles of relative changes of sea level. Memoir - American Association of Petroleum Geologists 26, 83-97. VANUXEM, L., 1842. Geology of New York, pt. Ill, comprising the survey of the 2d geological district. W. & A. White and J. Visscher, Albany, 3-6 p. VON STERNBERG, K.M.G., 1833. Versuch einer geognostich-botanischen Dartstellung der Flora der Vorwelt 5, 6. Johann Spurny, Prague, 80 p.

WEBBY, B.D., 1983. Lower Ordovician arthropod trace fossils from Western New South Wales. Proceedings of the Linnean Society of New South Wales 107, 59-74.

YANG, S., 1983. Early Devonian trace fossils from Liujing, Hengxian, Guangxi,

66 and their paleoecological significance. Regional Geology of China 5, 11- 22. [In Chinese, with English summary].

YOUNG, F.G., 1972. Early Cambrian and older trace fossils from the southern Cordillera of Canada. Canadian Journal of Earth Sciences 9, 1-17.

67 To Edmonton N

JASPER

Miette *- Hot v ALBERTA Jasper/ Springs

^NATIONAL)

PARK Rocky Mountain House

Columbia) Ice Fields^ BRITISH COLUMBIA

YOHO BANFF NATIONAL ^ PARK" NATIONAL ^Lake Louise PARK*

To Calgary KOOTENAY> NATIONAL v PARK

Radium^

Invermere I

Fig. 2-1. National Parks along the British Columbia and Alberta border, with inset illustrating the location of the parks in relation to the two provinces.

68 BractJioooa * fc

4- • •^ AiiftKtioan J c ^V J^P Redoubt

Kilometres

I Rock or Alpine | Forest or Grassland

j Glacier CJ Locality River British Columbia/ I Lake Alberta border

Fig. 2-2. Banff and Yoho National Parks with the three localities marked as dashed boxes: A, Lake O'Hara. B, Mount Babel. C, Redoubt Mountain (modified from Nelson et al., 2005).

69 Fig. 2-3. Images of the localities. A, Lake O'Hara. B, Mount Babel. C, Redoubt Mountain.

70 Peyto Formation

St. Piran Formation c Q. CD O -Q O E Lake Louise CD o o CD Formation

Fort Mountain Formation

e Q. CD O Hector Formation -Q O E CD ,+—» CD O CD CD s_ CL

Fig. 2-4. Generalized stratigraphy of the localities, not to scale nor representing any unconformities.

71 Fig. 2-5. A-B. Ichnofossils illustrating the intrastratal mode of formation of digging down to the muddy substrate, followed by the retreat of the organism, whereupon the sand immediately infils, leaving a convex hyporelief sandstone trace. A, Cruziana dispar (Linnarsson, 1869) at tip of finger with shale encompassing the sandstone trace. B, Cruziana isp., finger on the ventral surface of a lobe with shale at the base of the sandstone trace. C-E. Microbially Induced Sedimentary Structures (MISS) from the Fort Mountain Formation, Gog Group. C, Wrinkle structures verging into Kinneyia structures with Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya, 1977 (in more detail on Fig. 2-18 D) from Mount Babel, Alberta (T80). D, T90 displaying Kinneyia structures from Redoubt Mountain, Alberta. E, T89 intricate Kinneyia structures from Redoubt Mountain, Alberta. Scale bar = 1 cm.

72 Fig. 2-6. Cruziana billingsi Fillion & Pickerill, 1990, from the Fort Mountain Formation, Gog Group. A, T55 from Redoubt Mountain, Alberta. B, T54 from Lake O'Hara, British Columbia. C, T56 from Redoubt Mountain, Alberta. D, T53 from Mount Babel, Alberta. Scale bar = 1 cm.

73 74 Fig. 2-7. Abundant Cruziana billingsi Fillion & Pickerill, 1990 and Diplichnites twelvetreesi (Chapman, 1928) transforming from one to another from the Fort Mountain Formation, Gog Group, lower Cambrian, Mount Babel, Alberta. Image is of the actual rock surface on the talus slope and a plastotype (T57 &T 79) is in collections. Scale bar = 5 cm.

75 Fig. 2-8. All specimens from the Fort Mountain Formation, Gog Group, lower Cambrian. A, Cruziana dispar (Linnarsson, 1869), T44, from Lake O'Hara, British Columbia. B-C, Cruziana jenningsi Fenton & Fenton, 1937. B, T38, from Lake O'Hara, British Columbia. C,T61, from Redoubt Mountain, Alberta. D, Cruziana ramellensis (Legg, 1985), T73, from Lake O'Hara, British Columbia. Scale bar = 1 cm.

76 Fig. 2-9. Cruziana jenningsi Fenton & Fenton, 1937, from the Fort Mountain Formation, Gog Group, lower Cambnan of Lake O'Hara, British Columbia, T35-T37. A, Ventral view of convex hyporelief trace, T35. B, Ventral view of trace, T36. C, Anterior view of A (T35), emphasizing the anterior ridges. D, Ventrolateral view of trace, T37. E, The rock in which A, B and D are situated upon and labeled. Scale bar = 1 cm.

77 Fig. 2-10. Cruziana irregularis Fenton & Fenton, 1937. All specimens from the Fort Mountain Formation, Gog Group, lower Cambrian. A, T43 from Lake O'Hara, British Columbia. B-D, from Redoubt Mountain, Alberta. B, T60. C, T59. D, T58. Scale bar = 1 cm.

78 Fig. 2-11. Cruziana mesodelta isp. nov. from the Fort Mountain Formation, Gog Group, lower Cambrian, Redoubt Mountain, Alberta. A, Holotype, T62. B, Paratype, T63. Scale bar = 1 cm.

79 Fig. 2-12. Cruziana navicella Fenton & Fenton, 1937, from the Fort Mountain Formation, Gog Group, lower Cambrian of Alberta. A, T42 from Mount Babel. B, T41 from Redoubt Mountain, with Cruziana penicillata Gibb, Chatterton & Pemberton, 2009, terminal trace (T40), indicated by the brackets. Scale bar = 1 cm.

80 Fig. 2-13. All specimens are from the Fort Mountain Formation, Gog Group, lower Cambrian. A-B, Cruziana omanica Seilacher, 1970. A, T64 from Mount Babel, Alberta (arrow pointing to the trifid ridge). B, T65 from Redoubt Mountain, Alberta. C, Cruziana penicillata Gibb, Chatterton & Pemberton, 2009, T39, from Redoubt Mountain, Alberta. D, F-G, Cruziana problematica (Schindewolf, 1928), T69-T71, from Redoubt Mountain, Alberta. E, Cruziana plicata Crimes, Legg, Marcos & Arboleya, 1977 (T68). F, Transitions from C. plicata to C. problematica (T70) upon meeting perpendicular trace of C. problematica (D: T69). Scale bar = 1 cm.

81 Fig. 2-14. All specimens are from the Fort Mountain Formation, Gog Group, lower Cambrian. A, C, Cruziana plicata Crimes, Legg, Marcos & Arboleya, 1977, from Lake O'Hara, British Columbia. A,T67. C,T66. B, Cruziana ramellensis (Legg, 1985), T72, from Redoubt Mountain, Alberta. Scale bar = 1 cm.

82 Fig. 2-15. Cruziana rugosa d'Orbigny, 1842, Fort Mountain Formation, Gog Group, lower Cambrian. A, C-D, Redoubt Mountain, Alberta. A, T45. C, T74 (arrow pointing to cephalon impression). D, T75. B, T76 from Lake O'Hara, British Columbia. Scale bar = 1 cm.

83 v ,--*- -.

Fig. 2-16. A, C-D. Diplichnites twelvetreesi (Chapman, 1928) Gog Group, lower Cambrian. A, T77, Fort Mountain Formation, Lake O'Hara, British Columbia. C, T46, Fort Mountain Formation, Redoubt Mountain, Alberta. D, T78, Lake Louise Formation, Lake O'Hara, British Columbia. B, Diplichnites obliquus isp. nov. Holotype, T47, Fort Mountain Formation, Lake O'Hara. Scale bar = 1 cm.

84 *<•:

1 D mi-':.,',, ?l 4. ".^-ii1

Fig. 2-17. Specimens from the Fort Mountain Formation, Gog Group, lower Cambrian, Mount Babel, Alberta. A, C-D. Diplichnites twelvetreesi (Chapman, 1928). A, T48. C, T91. D, Diplichnites twelvetreesi transitional form to Cruziana billingsi Fillion & Pickerill, 1990, T79-T57 (Plastotype). B, Monomorphichnus bilinearis Crimes, 1970b, T49. Scale bar = 1 cm.

85 Fig. 2-18. All speimens from the Fort Mountain Formation, Gog Group, lower Cambrian. A, Monomorphichnus bilinearis Crimes, 1970b, T50, Lake O'Hara, British Columbia. B, Monomorphichnus trilinearis isp. nov. (Holotype) T51, Redoubt Mountain, Alberta. C-D. Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya, 1977. C, T52, Mount Babel, Alberta. D, T80, Mount Babel, Alberta. Scale bar = 1 cm.

86 Fig. 2-19. A, Coffee bean viewed laterally to visual the idea of two juxtaposed lobes of a 'single' coffee bean, equivalent to the lobes on Rusophycus and space between lobes is equal to the median line or mesial opening. B-D. Rusophycus eutendorfensis (Linck, 1942) from the Fort Mountain Formation, Gog Group, lower Cambrian. B, T82 from Mount Babel, Alberta. C, T52 from Redoubt Mountain, Alberta. D, T80 from Redoubt Mountain, Alberta. Scale bar = 1 cm.

87 Fig. 2-20. All specimens from the Fort Mountain Formation, Gog Group, Lower Cambrian. A, Rusophycus arizonensis (Seilacher, 1970), T81 from Redoubt Mountain, Alberta. B-C. Rusophycus latus Webby, 1983. B, T29 from Mount Babel, Alberta. C, T28 from Lake O'Hara, British Columbia. Scale bar = 1 cm.

88 89 Fig. 2-21. Rock surface .with traces from the Fort Mountain Formation, Gog Group, lower Cambrian, Lake O'Hara, British Columbia. A, Rusophycus latus Webby, 1983, T31. B-D. Rusophycus subnotous isp. nov. B, Paratype, T32. C, Paratype, T33. D, Holotype, T34. E, Whole rock with the individual traces labeled with responding letter to those figured on plate. Scale bar = 1 cm.

90 91 Fig. 2-22. Rusophycus mesodeltus isp. nov. from the Fort Mountain Formation, Gog Group, lower Cambrian, Redoubt Mountain, Alberta. A, Holotype, T85. B, Paratype, T86. Scale bar = 1 <

92 Fig. 2-23. All specimens from the Fort Mountain Formation, Gog Group, lower Cambrian, Redoubt Mountain, Alberta. A, Rusophycus unilobus (Seilacher, 1970), T88. B, Rusophycus radwanskii Alpert, 1976, T27. C, Rusophycus victorus isp. nov., Holotype: T30. Scale bar = 1 cm.

93 Lake O'Hara

Monomorphichnus 9%

Fig. 2-24. Arthropod ichnofaunal distribution from Lake O'Hara.

94 Mount Babel

Diplichnites 29%

' Cruziana 54%

Monomorphichnus 10%

Rusophycus 7%

Fig. 2-25. Mount Babel arthropod ichnofaunal distribution.

95 Redoubt Mountain

Monomorphichnus 7%

Fig. 2-26. Redoubt Mountain distribution of arthropod ichnofauna.

96 Gog Group

Monomorphichnus 9%

Fig. 2-27. Arthropod ichnofaunal distribution across the three localities: Lake O'Hara, Mount Babel and Redoubt Mountain (from the collections of S. Gibb and J. Magwood).

97 CHAPTER 3: CAMBRIAN OF MOROCCO

PROD TRACES {SELENICHNITES) FROM THE MIDDLE CAMBRIAN OF MOROCCO, WITH HYPOTHESES ON THE ETHOLOGY OF THE TRACEMAKER(S)1

INTRODUCTION

Trace fossils provide a unique window into the palaeontological behaviour that a simple body fossil is incapable of providing. The Azlag Formation, from the upper middle Cambrian of southern Morocco, on the roadside close to Tansikht, provides a spectacular glimpse into a shallow marine setting. Interference ripples are evident on the bedding surface, though it is the traces of Selenichnites tacfihtus n. isp. that provides an abundance of behavioural information about an unidentified organism with a rounded cephalon. The traces occur both isolated and in series in which they are analogous to trackways. Of particular note is that the traces are oriented in such a fashion that they appear to have been controlled by the current that produced one of the ripple trains in the interference ripples. A thin layer of mudstone (fine sand and mud) overlies the sandstone bed, indicating a shallow marine setting that experienced a storm event that resulted in disturbed finer particles settling out of suspension. It is posited that the organism that produced the traces was prodding through the mud to the sandstone in search of small organisms or microbial detritus, thus creating undertracks.

A version of this chapter has been submitted for review. GIBB, S., CHATTERTON, B.D.E. and PEMBERTON, S.G.. Prod traces (Selenichnites) from the middle Cambrian of Morocco, with hypotheses on the ethology of the tracemaker(s). Ichnos. GICH-2010-0009. 98 GEOLOGICAL AND GEOGRAPHIC SETTING

The trace fossil locality is conveniently located along the highway from Agdz to Zagora, immediately before the junction at Tansikht. The GPS coordinates are N 30° 41' 05.3" and W 06° 11' 44.2" (See Figure 3-1). A section was examined and measured at this locality, with the trace-bearing bed at the top of the section (above this level, the outcrop disappears under the highway). The section was dated using the trilobite Planolimbus hessikouanus Geyer, 1990 found lower in the section (Figure 3-2). Geyer (1990) described a new genus, Planolimbus, and species, P. hessikouanus from Jbel Hessiken (equivalent to Jbel Assiken) of the Bailiella Formation (Destombes, 1985) within the Tabanite Group (Destombes, 1985) of the middle Cambrian from the Souss Basin (Geyer, 1989; Geyer and Landing, 2006; Landing et al., 2006; Alvaro and Debrenne, 2010). Planolimbus hessikouanus occurs in the Bailiella Formation. We also found fragments (free cheeks and thoracic segments) of Paradoxides sp., also known from the Bailiella Formation (Destombes, 1985; Geyer and Landing, 1995; 2006). The Azlag Formation consists of coarser elastics and occurs above the Bailiella Formation. After identifying these two units in the Tansikht area, we measured a section from the trilobite-bearing part of the Bailiella Formation through into the Azlag Formation (Ja'idi et al., 1970; Saadi et al., 1974-1977; Destombes, 1985; Destombes and Feist, 1987; Fetah et al., 1989; Geyer, 1990; Geyer and Landing, 1995; 2006). The Bailiella Formation was considered by Geyer and Landing (1995; 2006) to be the result of a depositional event of higher energy, between the fair- weather and storm-wave base, due to the presence of fossil hash beds in which trilobites and crinoids are found. The Bailiella Formation conformably underlies the Azlag Formation (Ja'idi et al., 1970; Saadi et al., 1974-1977; Destombes, 1985; Destombes and Feist, 1987; Fetah et al., 1989; Geyer, 1990; Geyer and Landing, 1995; 2006). The exact boundary between the Bailiella Formation and the Azlag Formation is still open for debate for this locality as the two units are conformable and the boundary is gradational. Geyer and Landing (1995; 2006) described the

99 depositional environment of the Azlag Formation as being a tidally influenced, shallow marine setting. This description is confirmed by the moderate to thick- bedded clean sandstones, with concomitant physical sedimentary structures of symmetrical interference ripple marks and symmetric ripple marks. The Bailiella Formation and the Azlag Formation are both regarded as being upper middle Cambrian strata (Ja'idi et al., 1970; Saadi et al., 1974-1977; Destombes, 1985; Destombes and Feist, 1987; Fetah et al., 1989; Geyer, 1990; Geyer and Landing, 1995; 2006). The bedding surface bearing Selenichnites seems to have only a single type of trace preserved in clean, well-sorted sandstone, with distinct interference ripples. The currents that caused the interference ripples can only be speculated on. Selley (2000) defined interference ripples as "two obliquely intersecting sets of ripple crests" (p. 156) and discussed the formation of interference ripples as the "result from the modification of one ripple train due to one set of conditions by a later train, generated by waves or currents with a different orientation" (p. 156). A thin layer of 'mudstone' (fine-grained sand and mud) overlies the sandstone bed (Figure 3-3) and that is overlain by another sandstone bed. It is believed that this setting is within the distal intertidal zone to proximal upper shoreface, for no evidence of subaerial exposure, in the form of desiccation cracks, and/or rain imprints is present. Alvaro and Debrenne (2010) discussed the palaeoenvironment within the Souss Basin from the Ediacaran to the lower Cambrian as a carbonate belt. The subsequent carbonate collapse is referred to the Toyonian regression, which is noted to have occurred in the latest Early Cambrian. This regression led to a siliciclastic depositional environment and thus this is consistent with the Azlag Formation.

MATERIAL

The trace fossils discussed in this paper are located upon a rock face adjacent to the highway (Figure 3-3B), and removal is impossible. Latex peels

100 were made of the traces. Plaster casts were then produced from these peels to provide the concave epirelief ichnofossils that were studied and photographed. These types are lodged in the University of Alberta Trace Fossil Collection. The plaster casts were painted black, ammonium chloride coated and photographed under low-angle light at incremental focal planes to increase depth of field with the use of the programs 'Helicon Focus' and 'PhotoShop'.

SYSTEMATIC ICHNOLOGY

The nomenclature used herein has no bearing on any organism that could have created the trace(s), and is based solely on the morphological characteristics of the trace(s) (Bromley, 1990; Magwood, 1992; Jensen, 2003; Bertling et al., 2006).

Selenichnites (Romano and Whyte, 1987)

Herradurichnus Poire and Del Valle, 1996, p. 93-94.

Discussion: Poire and Del Valle (1996) erected a new ichnogenus for the ichnospecies Corophioides scagliai of Borrello (1966). We believe this ichnogenus, Herradurichnus, to be synonymous with Selenichnites (Romano and Whyte, 1987) due to the U-shaped concave epirelief morphological characteristics provided by Poire and Del Valle (1996) within the diagnosis (p. 93) and illustrations (fig. 3, pi. I, fig. 3). Even in the etymology of the name, which was provided, 'horseshoe shape', indicates that Herradurichnus is a junior synonym of Selenichnites.

Selenichnites tacfihtus n. isp. Figures 3-3 - 3-5

101 Selenichnites isp. Draganits, Braddy and Briggs, 2001, p. 135, fig. 8.

Type Material and locality: Sixteen latex peels and plaster casts from Morocco were made, though only the following are figured: Holotype: T20c and T20m (bottom left trace on figure 4f); Paratypes: T21c-T26c and T21m-T26m, made from a sandstone bedding surface in the Azlag Formation, Tabanite Group, middle Cambrian, near Tansikht, south-central Morocco. Specimens with T type numbers are stored in the University of Alberta Trace Fossil Collection in the Department of Earth & Atmospheric Sciences, University of Alberta, Edmonton, Alberta T6G 2E3, Canada.

Etymology: Due to the horseshoe-shape of the trace, the Berber term for horseshoe - tacfih 't - was chosen.

Diagnosis: Horseshoe-shaped trace of various length to width ratios. Maximum depth of trace anywhere within constraints of lunate shape. No other morphological characteristics are present.

Description: Crescent-shaped concave epirelief traces. Width of traces is 41.8 - 93.4 mm and lengths of 23.4 - 68.1 mm. Depth of traces vary from shallow to moderately incised into strata, and deepest region of trace is highly variable: anterior edge, anterolateral and posterolateral. Traces can be isolated and/or a series of traces arranged in a row analogous to a trackway.

Discussion: The traces found in the Tansikht locality of the Azlag Formation bear few distinctive morphological characteristics. Numerous examples of this trace are present on a single bedding plane. None of these individual traces show some of the features used to identify other ichnospecies assigned to Selenichnites. The traces do conform with the description for Selenichnites provided by Romano and Whyte (1987; 1990). A possible inconsistency in the depth of the posterolateral ridges may be accounted for by collapse of the leading ridge of the

102 sediment during or shortly after the formation of the trace. Trewin and McNamara (1995) posited that the setting of Tumblagooda Sandstone had a possible binding agent within the sand to create an overhanging edge on the anterior margin of S. langridgei. We have not had any indication of a binding agent within the strata, which could result from the fact that the substrate was better sorted and therefore more consolidated, and the angle of repose was such that there was a slight collapse of sediment at the leading edge. Nevertheless, it is possible that a binding agent was present between the sandstone and overlying mudstone. The bedding surface does possess interference ripples in a shallow marine setting (Geyer and Landing, 1995; 2006), allowing for a firm, though unlithified substrate. The individual traces display a prominent direction of movement in accordance with one of the two ripple trains (Figure 3-3 A-B). It is unlikely that traces could be assigned to a different ichnospecies due to the sheer lack of morphological characteristics of the traces. Other ichnospecies within the ichnogenus Selenichnites are: S. rossendalensis (Hardy, 1970) (Carboniferous of the U.K.); S. cordiformis (Fischer, 1978) (Ordovician of Colorado); S. bradfordensis (Chisholm, 1985) (Carboniferous of the U.K.); S. hundalensis Romano and Whyte, 1987 (Jurassic of the U.K.); S. langridgei Trewin and McNamara, 1995 (?Late Silurian of Western Australia); S. scagliai (Poire and Del Valle, 1996) (Cambrian/Ordovician of Argentina); S. antarcticus Weber and Braddy, 2004 (?Lower Ordovician of Antarctica); S. isp. Wang, 1993 (Upper Triassic of the U.K.); S. isp. Draganits et al., 2001 (Lower Devonian of Northern India); S. isp. Morrissey and Braddy, 2004 (Early Devonian of southwest Wales); and S. isp. Lucas and Lerner, 2005 (Lower Pennsylvanian of Alabama). Reasons for eliminating other ichnospecies from the ichnogenus Selenichnites are varied. S. rossendalensis is a lunate-shaped trace, but it also exhibits markings from appendages and a posterior morphological feature. Hardy (1970) had proposed that this was made by a 'telson', though a specific morphological term for a posited tracemaker should not be assigned in a diagnosis of an ichnofossil. Neither appendage nor 'telson' markings were observed in the

103 traces described herein. S. cordiformis was diagnosed by Fischer (1978) as a "ventral heart-shaped doublure imprint" of equal length and width, along with a telson marking. The traces from Morocco are not of equal length to width, with a ratio of length to width that displays a degree of variation within the traces from 0.54-1.19. This variation could be due to the traces being undertracks, a theory that is discussed further in the 'Interpretation'. The Moroccan traces do not have a heart-shape and, as previously stated, have no telson drag mark. Chisholm (1985) diagnosed S. bradfordensis. While Romano and Whyte (1987) disagree that the illustrated traces of Chisholm (1985) were Selenichnites, Draganits et al. (2001) did classify them within Selenichnites. Based on observation of the bedding traces of Chisholm (1985), it is clear that they conform to the diagnosis of Romano and Whyte (1987) for Selenichnites. Chisholm's (1987) S. bradfordensis is not comparable to S. tacfihtus, for the Moroccan traces lack a median furrow, and no transverse or chevron markings have been observed. Romano and Whyte (1987) described S. hundalensis as having a "relatively flat, subtriangular region between and in front of paired crescentic lobes". The distinguishing subtriangular feature set between two lateral lobes immediately separates S. hundalensis from the Moroccan traces, for the latter only displays a simple lunate depression, with no other distinguishing ventral features of the trace-maker. Selenichnites isp. of Draganits et al. (2001) is synonymized with S. tacfihtus due to the lack of morphological characteristics other than the simple horse-shoe shape of the trace that is the defining feature of this ichnospecies. Draganits et al. (2001) also stated that their traces "may indicate an orientation of the apices against depositional currents", therefore suggesting a current alignment. Trewin and McNamara (1995) erected the ichnospecies S. langridgei where the defining morphological trait is a median trefoil along with the lunate- shaped trace. None of the specimens of S. tacfihtus show a median trefoil. Selenichnites scagliai (Poire and Del Valle, 1996) does display similar limited morphological features, as identified with S. tacfihtus, though the overall

104 concave epirelief illustrated appears to be symmetrical in length and width, therefore a 1:1 ratio, thus not having the ratios characteristic of S. tacfihtus. It may be found that S. scagliai and S. tacfihtus are synonymous, though at this time, the traces appear diverse enough to warrant distinct ichnospecies. A trace from Antarctica described by Weber and Braddy (2004), S. antarcticus, is relatively similar to the traces found in Morocco, displaying an analogous lunate structure. However, S. tacfihtus lacks the transverse lineations found with S. antarcticus. Another behavioural similarity between 5". antarcticus and the traces of Morocco is their alignment with the current. Wang (1993) illustrated three forms of Selenichnites, though never assigned them to a specific ichnospecies. Of the three forms illustrated and discussed, all display more than just a lunate-shaped trace, with either paired lobes and/or scratch markings, none of which are observed with the Cambrian traces of Morocco. Lucas and Lerner (2005) also erected an unassigned ichnospecies that is now assigned to Selenichnites. Their traces are described and illustrated as having the following morphological traits that eradicate their synonymization with S. tacfihtus: a medially placed shallow furrow dividing the crescent into pairs; and lateral scratch markings on either side of the medial furrow (Lucas and Lerner, 2005). No other ichnogenera were considered in this analysis due to the simple nature of S. tacfihtus with only the depression in the shape of a horseshoe.

INTERPRETATION The stratum in which Selenichnites tacfihtus is found is a massive, clean, well-sorted sandstone with symmetrical interference ripples. As derived from the physical sedimentary structures, the palaeoenvironmental depositional setting is likely that of distal intertidal to proximal shoreface. Geyer and Landing (1995; 2006), as previously stated, interpreted the Azlag Formation to be "shallow, tidally influenced environments... [with] massive quartzites"; they also noted the occurrence of the ichnofossil Skolithos at the top of the Azlag Formation. Pemberton et al. (1992) described the features of the Skolithos ichnofacies that

105 correlate with the trace fossil bed in question: "a slightly muddy to clean, well- sorted, loose shifting particulate substrates". The mud is observed overlying the trace surface and the traces are found in a sandstone. "Abrupt changes in rates of deposition, erosion, and physical reworking of sediments are frequent. Such conditions commonly occur on the foreshore and shoreface" (Pemberton et al., 1992). This pattern is observed in the strata with interference ripples and leads to the conclusion that the traces were made in a distal intertidal or proximal upper shoreface setting. Distinguishing trace fossil evidence for the Skolithos ichnofacies that are applicable to the Moroccan locality, stated by Pemberton et al. (1990) are: "few horizontal structures, ...few structures produced by mobile organisms, ... [and] low diversity, although individual forms may be abundant". Selenichnites tacfihtus is a horizontal structure implied to be made by a mobile organism (to be discussed herein) and they are the only traces found on the surface (though they occur both as single indentations and series of traces that are analogous to a trackway). Thus, the locality can be established as one within the Skolithos ichnofacies both from the physical and biogenic sedimentary structures. The concept of a binding agent that Trewin and McNamara (1995) proposed for the preservation of their ichnospecies Selenichnites langridgei, will be more closely examined for possible explanations as to the preservation of S. tacfihtus. Seilacher (2008) proposed a number of different preservational scenarios of traces, involving microbes — for arthropods in particular. The most applicable to the Cambrian traces of Morocco is that of the traces being classified as 'undertracks'. Seilacher (2008) stated that "undertracks are more distinct than surface tracks and have much higher fossilisation potential (Goldring and Seilacher, 1971); they do, however, become visible only on mud/sand interfaces or if the sediment contained suitable and closely-spaced films that developed into parting upon lithification. Marine storm sands are commonly too thick for post- event undertracks to reach the base of the bed. Their rippled tops, however, favoured the preservation of surface tracks or shallow undertracks, because they became coated shortly afterwards by the mud veneer settling from suspension." These are near perfect examples as to how the traces could have been preserved in

106 the upper middle Cambrian of Morocco. As a result, one can conclude that our traces may be undertracks formed in a storm setting, where interference ripples could be observed, and the settled fine sand and mud overlies the sandstone deposit upon which the surface traces (possible undertracks) are found. For this to be the case, the prevailing direction of the traces could imply the predominant current, i.e. tidal current. The original ripple current direction could have been associated with the dying storm waves. Local tidal currents then modified this? The trace-maker of Selenichnites tacfihtus must have possessed a few morphological features; any other character traits would be absolute speculation. The two most obvious morphological character states are: rounded cephalon (or a structure that could also have this morphology: a pygidium, though it is highly unlikely that the traces are pygidial traces for the ethological and fluid dynamics do not fit this scenario) and the musculature to drive the cephalon into the sandy substrate. One trace does display some scratch marks (Figure 3-5), which have not been assigned to another ichnospecies due to the shallow nature of the scratches and the inability to tease out the number of appendages that could have created the scratches. From the scratch marks, one can conclude that the organism did have at least one pair of appendages. Due to the nature of the traces, Eldredge's (1970) first stage of burrowing of a horseshoe crab is relevant. The initial actions involved, in burrowing, do correspond to the traces of Selenichnites tacfihtus. Eldredge (1970) observed the burrowing behaviour of modern Limulus polyphemus and its implications for trilobites' functional anatomy. Eldredge (1970) explained four stages of limulid burrowing; the first phase of the first stage is most relevant to S. tacfihtus. As stated by Eldredge (1970): "the process is begun (Stage 1) by a downward flexion of the prosoma on the opisthosoma... The opisthosoma is kept in a more or less horizontal plane throughout the process. The symmetrical wall of sand is pushed up the steep slope of the anterior margins of the prosoma, reaching the eyes and the genal angle at the same time; the prosoma is flexed back to the horizontal position at this time, the animal moves forward, and the wall of sand is spread back along the horizontal central portion of the prosoma." The prosoma

107 can be generalized as a lunate cephalon and the opisthosoma as 'the body' of an organism. The initial driving of the cephalon into the substrate is the behaviour thought to be indicative of the traces observed. It is quite possible that the organism had driven its head into the substrate after a storm deposit, when the sand grains possibly had microbial matter as a veneer (Seilacher, 2008), searching for food through the mud, on top of the sandstone, with chemoreceptors on the cephalon. This would then account for the traces being undertracks and thus the reason only the horseshoe-shaped indentation remains. In Figure 5, the scratch marks found at the end of the trackway point to the possibility that food was found and then ingested by the organism by the process of a cephalon prodding technique. It should also be noted that the traces never deviate from one or the other ripple train directions, though predominantly with only the one. Therefore the organism was either using the current to assist in their movement within the water to prod the substrate, or it was working against the current. It is unlikely that an organism would expend extra energy in search of food, thus it is presumed that it was moving with one of the two currents. One group of organisms from the Cambrian that would have had the cephalon size to create Selenichnites tacfihtus are the anomalocaridids, though none to date have been found in Morocco. Though the cephalon is typically of a crescent shape (aside from Hurdia Daley, Budd, Caron, Edgecombe and Collins 2009, which has a triangular shape to the anterior aspect of the cephalon), its anterior/frontal appendages would hinder the creation of Selenichnites. The anomalocaridid would not be able to tuck the enormous front appendages back and under the cephalon without some trace evidence on the inside edge of the horseshoe-shaped trace. Xiphosurans have been recorded as possible trace-makers of Selenichnites by Hardy (1970), Fischer (1978), Chisholm (1985), Romano and Whyte (1987; 1990), Wang (1993), Draganits et al. (2001), and Lucas and Lerner (2005). The first geological occurrence of a horseshoe crab is from the Late Ordovician strata of Manitoba, Canada (Rudkin et al., 2008), thus eliminating that organism from

108 the equation. Middle Cambrian aglaspidids discussed to date are apparently not large enough to have produced the traces found in the Azlag Formation from the Cambrian (Stormer, 1956; Hesselbo, 1992) and/or the Ordovician (Van Roy, 2006). Briggs et al. (1979) illustrated Aglaspis spinifer from the upper Cambrian of Wisconsin that is large enough to have made these traces. Hesselbo (1988) also published on traces attributed to aglaspidids from the Cambrian of Wisconsin, but these are much smaller in size than the Moroccan traces and are rusophyciform, Raaschichnus gunersoni Hesselbo, 1988. Trilobites are the last obvious choice as a maker of these traces, and a number of different trilobites do occur in the middle Cambrian of southern Morocco. Size is obviously a parameter that must be considered. Another fact that should be addressed is the lack of trilobite fossils in upper middle Cambrian strata. This could be due to the siliciclastic nature of the sediments during this time, and the typical lack of body fossils to be found in a clastic setting. Therefore, one must look to trilobites of appropriate size occurring before and/or after this time period for a cephalon match. The lower middle Cambrian of Morocco strata has the following trilobites: Cambropallas telesto and Acadoparadoxides briareus Geyer 1993 (Geyer and Palmer, 1995) of the Order Redlichiida Richter, 1932. These two trilobites have a cephalon that is large enough to have created Selenichnites tacfihtus, though they are both from the slightly older lower middle Cambrian of Morocco. Yet another candidate, based on cephalon size, is Asphiscus Meek 1873, also occurring in middle Cambrian strata, but from the western United States of America (Harrington et al., 1959; McNamara and Rudkin, 1984). The upper Cambrian provided a number of trilobites within the approximate size range, though Family Dikelocephalidae Miller 1889 appears to be the closest candidate: more specifically Dikelocephalus Owen 1852 (Westrop et al., 2005), once again a genus from North America. There is little doubt, however, that an organism prodding the substrate with or against a current produced cephalon impressions, either isolated or arranged in a sequence analogous to a trackway: Selenichnites tacfihtus.

109 REFERENCES

Alvaro, J.J. and Debrenne, F. 2010. The Great Atlasian Reef Complex: An early Cambrian subtropical fringing belt that bordered West Gondwana. Palaeogeography, Palaeoclimatology, Palaeoecology, 294: 120-132. Bertling, M., Braddy, S.J., Bromley, R.G., Demathieu, G.R., Genise, J., Mikulas, R., Nielsen, J.K., Nielsen, K.S.S., Rindsberg, A.K., Schlirf, M. and Uchman, A. 2006. Names for trace fossils: a uniform approach. Lethaia, 39: 265-286. Borrello, A.V. 1966. Paleontografia Bonaerense. Fasciculo V. Trazas, restos tubiformes y cuerpos fosiles problematicos de la Formacion La Tinta, Sierras Septentrionales, provincia de Buenos Aires. Comision de Investigaciones Cientificas de la Provincia de Buenos Aires: 1-42. Briggs, D.E.G., Bruton, D.L. and Whittington, H.B. 1979. Appendages of the arthropod Aglaspis spinifer (Upper Cambrian, Wisconsin) and their significance. Palaeontology, 22: 167-180. Bromley, R.G. 1990. Trace fossils: Biology and taphonomy. Unwin Hyman, London, 280 pp. Chisholm, J.I. 1985. Xiphosurid burrows from the Lower Coal Measures (Westphalian A) of West Yorkshire. Palaeontology, 28: 619-628. Daley, A.C., Budd, G.E., Caron, J.-B., Edgecombe, G.D. and Collins, D.H. 2009. The Burgess Shale anomalocaridid Hurdia and its significance for early Euarthropod evolution. Science, 323: 1597-1600. Destombes, J. 1985. Middle Cambrian. In C.H. Holland (ed.) Lower Palaeozoic of north-western and west-central Africa, vol. 4. John Wiley & Sons, Chichester, 157-184. Destombes, J. and Feist, R. 1987. Decouverte du Cambrien superieur en Afrique (Anti-Atlas central, Maroc). Comptes Rendus de TAcademie des Sciences, Serie 2, Mecanique, Physique, Chimie, Sciences de TUnivers, Sciences de la Terre, 304: 719-724. Draganits, E., Braddy, S.J. and Briggs, D.E.G. 2001. A Gondwanan coastal

110 arthropod ichnofauna from the Muth Formation (Lower Devonian, Northern India): Paleoenvironment and tracemaker behavior. Palaios, 16: 126-147. Eldredge, N. 1970. Observations on burrowing behaviour in Limulus polyphemus (Chelicerata, Merostomata), with implications on the functional anatomy of trilobites. American Museum Novitates, 2436: 1-17. Fetah, M., Bensai'd, M. and Dahmani, M. 1989. Carte Geologique du Maroc: Zagora-Coude du Dra-Hamada du Dra (p.p.), 1/200 000. Editions du Service Geologique du Maroc. Notes et Memoires, No. 273. Royaume du Maroc. Ministere de I'Energie et des Mines. Direction de la Geologie, Rabat. Fischer, W.A. 1978. The habitat of the early vertebrates: trace and body fossil evidence from the Harding Formation (Middle Ordovician), Colorado. The Mountain Geologist, 15: 1-26. Geyer, G. 1989. Late Precambrian to early Middle Cambrian lithostratigraphy of southern Morocco. Beringeria, 1: 115-143. Geyer, G. 1990. Die marokkanischen Ellipsocephalidae (Trilobita: Redlichiida). Beringeria, 3: 3-363. Geyer, G. 1993. The giant Cambrian trilobites of Morocco. Beringeria, 8: 71-107. Geyer, G. and Landing, E. 1995. The Cambrian of the Moroccan Atlas regions. Beringeria Special Issue 2: 7-46. Geyer, G. and Landing, E. 2006. Latest Ediacaran and Cambrian of the Moroccan Atlas regions. Beringeria Special Issue 6: 7-46. Geyer, G. and Palmer, A.R. 1995. Neltneriidae and Holmiidae (Trilobita) from Morocco and the problem of Early Cambrian Intercontinental correlation. Journal of Paleontology, 69: 459-474. Goldring, R. and Seilacher, A. 1971. Limulid undertracks and their sedimentologic implications. Neues Jahrbuch fur Geologie und Palaontologie. Abhandlungen, 137: 422-442. Hardy, P.G. 1970. New xiphosurid trails from the Upper Carboniferous of Northern England. Palaeontology, 13: 188-190.

Ill Harrington, H.J., Henningsmoen, G., Howell, B.F., Jaanusson, V., Lochman-Balk, C, Moore, R.C., Poulsen, C, Rasetti, F., Richter, E., Richter, R., Schmidt, H., Sdzuy, K., Struve, W., Tripp, R., Weller, J.M. and Whittington, H.B. 1959. Classification. In R.C. Moore (ed.) Treatise on Invertebrate Paleontology. Part O. Arthropoda 1. Geological Society of America and University of Kansas, Boulder and Lawrence, 0170-0540. Hesselbo, S.P. 1992. Aglaspidida (Arthropoda) from the Upper Cambrian of Wisconsin. Journal of Paleontology, 66: 885-923. Hesselbro, S.P. 1988. Trace fossils of Cambrian aglaspidid arthropods. Lethaia, 21: 139-146. Ja'idi, M., Bencheqroun, Diouri, M. and Ennadifi, Y. 1970. Carte Geologique de l'Anti-Atlas Central et de la Zone Synclinale de Ouarzazate: Feuilles Ouarzazate, Alougoum et Telouet Sud, 1/200 000. Editions du Service Geologique du Maroc. Notes et Memoires, No. 138. Royaume du Maroc. Ministere de I'Energie et des Mines. Division de la Geologie, Rabat. Jensen, S. 2003. The Proterozoic and earliest Cambrian trace fossil record; patterns, problems and perspectives. Integrative and Comparative Biology, 43:219-228. Landing, E., Geyer, G. and Heldmaier, W. 2006. Distinguishing eustatic and epeirogenic controls on Lower-Middle Cambrian boundary successions in West Gondwana (Morocco and Iberia). Sedimentology, 53: 899-918. Lucus, S.G. and Lerner, A.J. 2005. Lower Pennsylvanian invertebrate ichnofossils from the Union Chapel Mine, Alabama: A preliminary assessment. In R.J. Buta, A.K. Rindsberg and D.C. Kopaska-Merkel (eds.) Pennsylvanian footprints in the Black Warrior Basin of Alabama. Alabama Paleontological Society Monograph no. 1, 147-152. Magwood, J.P.A. 1992. Ichnotaxonomy: a burrow by any other name...? In C.G. Maples and R.R. West (eds.) Trace fossils. Short courses in paleontology. No. 5. University of Tennessee, Knoxville, 15-33. McNamara, K.J. and Rudkin, D.M. 1984. Techniques of trilobite exuviation. Lethaia, 17: 153-173.

112 Meek, F.B. 1873. Preliminary palaeontology report, consisting of list and descriptions of fossils, with remarks on the age of the rocks in which they are found. Annual Report of the United States Geological Survey of Territories, 6: 429-518. Miller, S.A. 1889. North American geology and palaeontology for the use of amateurs, students, and scientists, Cincinnati, 718 pp. Morrissey, L.B. and Braddy, S.J. 2004. Terrestrial trace fossils from the Lower Old Red Sandstone, southwest Wales. Geological Journal, 39: 315-336. Owen, D.D. 1852. Report of a geological survey of Wisconsin, Iowa, and Minnesota, and incidentally, a portion of Nebraska Territory, Philadelphia, 638 pp. Pemberton, S.G., MacEachern, J.A. and Frey, R.W. 1992. Trace fossil facies models: Environmental and allostratigraphic significance. In R.G. Walker and N.P. James (eds.) Facies models: Response to sea level change. Geological Association of Canada, 47-72. Poire, D.G. and Del Valle, A. 1996. Trazas fosiles en barras submareales de la Formacion Balcarce (Cambrico/Ordovicico), cabo corrientes, mar del Plata, Argentina. Asociacion Paleontologica Argentina, Publicacion Especial 4: 89-102. Richter, R. 1932. Crustacea (Palaontologie). In R. Dittler, G. Joos, E. Korschelt, G. Linek, F. Oltmanns and K. Shaum (eds.) Handworterbuch der Naturwissenschaften. Gustav Fisher, Jena, 840-864. Romano, M. and Whyte, M. 1987. A limulid trace fossil from the Scarborough Formation (Jurassic) of Yorkshire; its occurrence, taxonomy and interpretation. Proceedings of the Yorkshire Geological Society, 46: 85-95. Romano, M. and Whyte, M. 1990. Selenichnites, a new name for the ichnogenus Selenichnus Romano & Whyte, 1987. Proceedings of the Yorkshire Geological Society, 48: 221. Rudkin, D.M., Young, G.A. and Nowlan, G.S. 2008. The oldest horseshoe crab: A new xiphosurid from Late Ordovician konservat-lagerstatten deposits, Manitoba, Canada. Palaeontology, 51: 1-9.

113 Saadi, M., Hilali, E.A. and Bensaid, M. 1974-1977. Carte Geologique du Maroc: Jbel Saghro-Dades (Haut Atlas central, sillon Sud-Atlasique et Anti-Atlas priental), 1/200 000. Editions du Service Geologique du Maroc. Notes et Memoires, No. 161. Royaume du Maroc. Ministere de I'Energie et des Mines. Direction de la Geologie, Rabat. Seilacher, A. 2008. Biomats, biofilms, and bioglue as preservational agents for arthropod trackways. Palaeogeography, Palaeoclimatology, Palaeoecology, 270: 252-257. Selley, R.C. 2000. Applied Sedimentology. Academic Press, San Diego, CA., 523 pp. Stormer, L. 1956. A Lower Cambrian merostome from Sweden. Arkivfor Zoologie, 9:507-514. Trewin, N.H. and McNamara, K.J. 1995. Arthropods invade the land: trace fossils and palaeoenvironments of the Tumblagooda Sandstone (?late Silurian) of Kalbarri, Western Australia. Transactions of the Royal Society of Edinburgh: Earth Sciences, 85: 177-210. Van Roy, P. 2006. An aglaspidid arthropod from the Upper Ordovician of Morocco with remarks on the affinities and limitations of Aglaspidida. Transactions of the Royal Society of Edinburgh: Earth Sciences, 96: 327- 350. Wang, G. 1993. Xiphosurid trace fossils from the Westbury Formation (Rhaetian) of southwestern Britain. Palaeontology, 36: 111-122. Weber, B. and Braddy, S.J. 2004. A marginal marine ichnofauna from the Blaiklock Glacier Group (?Lower Ordovician) of the Shackleton Range, Antarctica. Transactions of the Royal Society of Edinburgh: Earth Sciences, 94: 1-20. Westrop, S.R., Palmer, A.R. and Runkel, A. 2005. A new Sunwaptan (Late Cambrian) trilobite fauna from the Upper Mississippi Valley. Journal of Paleontology, 79: 72-88.

114 Figure 3-1. Schematic of area where Cambrian Moroccan traces are located. A. Overview of Morocco, with boxed region enlarged to highlight the geology. B. The geological overview of the region where Selenichnites tacfihtus n. isp. occurs, labeled as 'Cambrian locality' (Saadi et al., 1974-1977).

115 Figure 3-2. Stratigraphic section with occurrence of trilobites at the base and Selenichnites tacfihtus n. isp. at the top of the section.

116 ,. I *- Selenichnites langridgei bedding surface

100

90 - 1

—'-!• — — ' -1

_1

80 - t- i ~ "I

~ — . - . 1 j 70 -

r ,, , • ,,i — — — — ——- ~ ~ — -J - - --™™ -i 60 - M E ~J T _ _ —— - - — 1 E K S _ _ _, . . __ J

___- r r-i ^.-. H 40 - - — — — — — - - - — - -J

- —, i ,- — r ~M ,

1 Sandstone —T 30 - •i Bedded sandstone J , d

—f"-1 Ripple-bedded sandstone ' 20 - Sandy to Silty Shale

r 10 -

[_^ Z_z; d J -*— Occurrence of Trilobites n

117 ^ o t_; o

c

c ^ c U o c c^ ^

u ^ O c <^^ 1 meter ^ A

[->•

4; IV" /'?. *h*S J S lS B 1 meter . , ' l "$^-'' ^fe^-t»j5 ^^^*-->^

Figure 3-3. The Cambrian rock face with Selenichnites tacfihtus n. isp. A. Schematic of the traces upon the bedding surface, with arrows indicating direction of interference ripples. The thin mudstone bed overlying the ichnofossil bedding surface, and upper right corner is the next stratigraphically higher sandstone horizon. B. Image of the actual surface with boxed numbers corresponding to illustrated traces in subsequent figures (numbering corresponds to figure).

118 Figure 3-4. Selenichnites tacfihtus n. isp. all from the Azlag Formation, Tabanite Group, middle Cambrian, near Tansikht, south-central Morocco. A. Two traces with anterior directed to the top left corner, which is relatively consistent with the original orientation on the bedding surface. These traces are approximately 90° to the majority of the other traces, though aligned with one of the ripple trains (T21c). B. Three distinct traces, with two in the right portion of the image, and the upper trace is only the left half of the trace, with the third trace being the deeply incised anterior marking in the left of the image. The traces are not oriented according to original positioning on the bedding surface, but rotated 135° counter clockwise (T22c). C. Single trace rotated 55° counter clockwise from position on bedding surface (T24c). D. Single trace in approximately original position on bedding surface. Another occurrence that is 90° to the majority of traces, though aligned with a ripple train (T23c). E. Three traces, with two overlapping each on the bottom of the image, and the third trace is approximately 90° to the traces below. The image is rotated roughly 50° counter clockwise from position on the bedding surface (T25c). F. One prominent trace, and two less prominent with the image being rotated 150° clockwise from the bedding surface. * is the holotype (T20c). All scale bars are 1 centimeter in length.

119 120 Figure 3-5: Selenichnites tacfihtus n. isp. all from the Azlag Formation, Tabanite Group, middle Cambrian, near Tansikht, south-central Morocco. Oblique angle image of two 'trackways' and one single trace, white arrows indicate hypothetical direction of movement of the tracemakers. The single trace is to the far left of the image. The longest 'trackway' is just right of the single trace and the last trace of the six clear traces (eight traces in total to the 'trackway') are a series of possible appendage markings (bottom left corner of image). The second 'trackway' is on the right of the image, with two visible traces in the track, and three traces creating the total of the 'trackway'. Scale bar image is segmented into 1 centimeter increments.

121 CHAPTER 4: ORDOVICIAN OF MOROCCO

RUSOPHYCUS CARLEYI (JAMES, 1885), TRACE FOSSILS FROM THE LOWER ORDOVICIAN OF SOUTHERN MOROCCO, AND THE TRILOBITES THAT MADE THEM1

INTRODUCTION

It should not be a surprise that the best examples of trace fossils and the trilobites that may have made them are of species of Rusophycus. Rusophycus is usually regarded as a resting or hiding trace. The animal that made such a trace may well have spent enough time in the burrow that they had a reasonable chance of dying there, or they were trapped on or in the trace during a storm (by an obrution deposit) or by a mass wastage event (such as a mud flow or turbidite). Until now, the most persuasive example of this association has been the discovery by Osgood (1970) of three specimens of the trilobite Flexicalymene meeki (Foerste, 1910) on top of the trace Rusophycus pudicum Hall, 1852, from the Ordovician of the Cincinnatian Series. Many authors have suggested that Rusophycus and the similar trace Cruziana were made by trilobites. It is not until a trilobite, or any other organism, is found in close proximity to a trace fossil that the trace may be posited to have been formed by a specific organism (Fortey and Seilacher, 1997; Bertling et al., 2006). Trilobites have also been suggested as the makers of a variety of other traces, including Dimorphichnus, Monomorphichnus, and Thalassinoides. In many cases these suggestions are supported by limited amounts of evidence other than the fact that they are the sorts of traces that a benthic animal with numerous ventral appendages might have been expected to

1 A version of this chapter has been published. GIBB, S., CHATTERTON, B.D.E. & GINGRAS, M.K. 2010. Rusophycus carleyi (James, 1885), Trace fossils from the Lower Ordovician of southern Morocco and the trilobites that made them, Ichnos, 17,271-283. DOI: 10.1080/10420940.2010.535452, due to editorial error: GIBB, S., CHATTERTON, B.D.E. & GINGRAS, M.K. 2010. Erratum, Ichnos, 18,47-55. DOL10.1080/10420940.2011.559873. 122 make. In some cases trilobite fossils occur in the same beds as the ichnofossils, but in many cases they do not and may occur in adjacent rock units. Apart from the examples described by Osgood (see above), Cherns et al. (2006) demonstrated that some asaphids may have occupied burrows, such as Thalassinoides. Chatterton et al. (2003), and Chatterton and Fortey (2008) also wrote of trilobites inhabiting burrows as cryptic behavior. Fortey and Seilacher (1997) provided criteria for associating a trace such as Cruziana with its possible tracemaker. They then used these to suggest that Cruziana semiplicata Salter, 1853, was made by the trilobite Maladioidella cf. colcheni Shergold et al., 1983, from the upper Cambrian Andam Formation of Oman. Others (Birkenmajer and Bruton, 1971; Bergstrom, 1976; Fortey and Morris, 1982; Zylihska et al., 2006) have attempted to demonstrate this association between Cruziana and specific trilobites using some or all of the criteria listed by Fortey and Seilacher (1997). We first became aware of these Moroccan traces when we saw some for sale in Brahim Tahiti's museum shop in Erfoud, southeastern Morocco. Our interest was piqued by the presence of articulated trilobites and traces in the same small samples. We then tracked down the source of the specimens at a locality near Ouzina, close to the Algerian border in southern Morocco.

REGIONAL SETTING

Geographical Location

The traces and trilobites described herein were found in pits made by fossil miners close to the village of Ouzina in southeastern Morocco, to the southeast of the small town Taouz (Figs. 4-1.1, 4-1.2). The locality coordinates of one of these pits is N 30° 45' 14.8", W 04° 09' 04.3". All five shallow pits may be seen close to this point. Numerous traces and trilobites occur in the rocks excavated from these pits. These pits were only excavated to a depth of less than 2 metres (100-140 cm). While most of these specimens that we collected had

123 already been laid out on the ground adjacent to the pits, we assumed that they were collected from the base or near the base of the pits, since the miners would almost certainly have dug down to a productive level and no farther. We believe from the absence of obvious trilobites or traces in the sides of the trenches, above the bottoms, that they may only occur through a limited stratigraphic interval (possibly only one or a few thin beds). Therefore, we believe that the traces and trilobites were mainly found at a depth of 100-140 cm from the ground surface (no datum for a reference point is available, since the beds do not outcrop extensively at the surface, and the strata in this locality are flat lying and several kilometers from well-exposed Ordovician strata in cliffs to the east).

Geological Setting

The ichnofossils and trilobites are, lithologically, in a fine grained sandstone alternating with mudstone that has thin, horizontal laminations with small pyrite crystals throughout. The presence of pyrite presents an interesting lithological story, which is expanded upon below in the palaeoenvironmental reconstruction section of the paper. Based upon the lithological data and the trilobites present, one can deduce that the formation that the specimens occur in is the Upper Fezouata Formation (Destombes, 1962; 1976; Vidal, 1998a, b). This rock unit is Floian-Dapingian (Ordovician) in age. The Ouzina locality is located within the Tafilalt Basin (Belka et al., 1997).

PALAEOENVIRONMENTAL RECONSTRUCTION

Some features of the beds that contain the trilobites and Rusophycus traces lead us to believe that they may have been laid down in an environment where microbial mats formed periodically on the substrate surface. A range of Microbially Induced Sedimentary Structures (MISS) have been shown to indicate the presence of biomats and their stabilizing impact on sedimentary surfaces.

124 These biostabilization structures include wrinkle structures, mat chips, erosional pockets, kinneyia ripples, and diastasis cracks (see Noffke et al., 2001; Porada and Bouougri, 2007, for overviews). Of these, two features are observed on the bedding planes of the studied specimens: (1) poorly developed, low-relief wrinkle structures of variable orientation; and (2) abundant diastasis cracks that are most abundant descending from preservational planes (Fig. 4-2). Although wrinkle structures are more commonly linked to biostabilization, diastasis cracks in biomats — they are commonly observed together — have a less clear origin. In some cases, a link can be made with intertidal exposure and the subsequent desiccation of the biomat. However, many examples of subtidally formed cracks have been observed, the most common of which are syneresis cracks, which dominantly form as a result of volume reduction of clay associated with salinity changes or due to loading (Cegla et al., 1967; Plummer and Gostin, 1981). Increased cohesiveness of the sediment due to biofilm interactions cannot be discounted as a mechanism for promoting these development tears in the sediment. The increased tensile strength of the sediment would clearly encourage the occurrence, that is, due to various processes such as downslope sediment creep, wave shear, and preservability of the resulting diastasis cracks. The presence of biomats also may explain the excellent preservation of the studied fossils. Biofilms have been associated with high fidelity preservation in other Ediacaran and Lower Paleozoic strata (e.g., Seilacher, 2008; Buatois and Mangano, 2010). Indeed the close association of trilobite fossils and bedding planes dominated by diastasis cracks seems to support this hypothesis. Schieber (1998, 1999) outlined criteria for the recognition of microbial mats within a terrigenous clastic setting, particularly within the Mid-Proterozoic. These different textures/lithological features can be recognized during any time period and may have been relevant to the behavior of the trilobites and the preservation of their traces, and thus are considered herein. Schieber (1998, 1999) listed nine different attributes for the recognition of microbial mats in ancient strata, and two in particular may have played a key role in the preservation of the traces from the Ordovician of Morocco. One of these is

125 "lamina-specific distribution of certain early diagenetic minerals" (Schieber, 1999, p. 3). In the trilobite and trace bearing thin sandstone beds from Morocco, numerous small pyrite crystals are concentrated in certain laminae. As one would suspect, a setting rich in organics is essential for the formation of pyrite (by providing iron and sulfur and a reducing environment during decomposition). Microbial mats would enable the early diagenetic formation of pyrite. Another important feature in these Ordovician beds from Morocco is the presence of iron staining in a polygonal "cracked" pattern (like "mud cracks") on surfaces of some of the beds (Fig. 4-2). This would signify a "cohesive behavior of laminae" (Schieber, 1998, p. 105). The reducing conditions required for these sedimentary processes, of iron staining, to take place can be further expanded on when referring to Fortey's (1975) and further correlated to Fortey and Owens' (1978) Ordovician trilobite biofacies. Fortey (1975) described three different trilobite biofacies from the Arenig of Spitsbergen. The shallow water illaenid-cheirurid biofacies is of little relevance to the Ouzina locality, since the nominate trilobites do not occur or are rare and features of the sediment do not suggest a shallow or near shore environment. However, Vidal (1998a, b) pointed out that the raphiophorid/nileid biofacies does occur in Arenig (Floian- Dapingian) strata of southern Morocco, and the olenid biofacies is also of interest with regard to these strata (Fortey, 1975). Returning to Schieber's (1998, 1999) criteria for the formation of pyrite, the setting would have had to be rich in organics. Fortey (1975) stated, following Bulman (1964), that the Lower Ordovician Olenidsletta Member of Spitsbergen, with a number of asaphid trilobites occurring therein, is "characterized by an unusually high organic content" (p. 341). This statement was pertaining to an olenid biofacies, which is postulated to occupy the outer region of a continental shelf (Fortey, 1975). Vidal (1998a, b) placed the asaphids, more specifically Asaphellus fezouataensis, within the raphiophorid biofacies. Upon closer analysis of the summary that Fortey (1975) provided in his fig. 6, it can be observed that the asaphids are found in both the olenid and raphiophorid/nileid biofacies. Therefore, the asaphids occur in both zones: the distal raphiophorid community

126 and more proximal olenid community. Fortey (1975, p. 344), in listing the morphological character states exhibited by trilobites in the olenid biofacies, wrote: "isopygous or nearly isopygous trilobites of fairly low convexity, broad cephalic borders with genal spines, and large crescentic eyes." This is a close generalized morphological description of Asaphellus. Hence, it is thought that the community found at Ouzina is representative of a merging of a raphiophorid and olenid biofacies, originally living on a middle to distal offshore continental shelf (Fortey, 1975; Fortey and Owens, 1978; Vidal, 1998a, b). The lithology, of finely laminated mudstone with interbedded sandstone, can be classified into the "Archetypal" to "Distal Cruziana" ichnofacies, whereupon dominant deposit feeders and grazers/foraging organisms are typically found (Pemberton et al., 1992; Pemberton et al., 2001). We posit that the Rusophycus traces found at Ouzina were formed in a muddy setting with some influx of sand, with microbial mats, below the fairweather wave base, but above stonri wave base. Perhaps during storms trilobites would excavate a "burrow" = Rusophycus, to shelter from the drag produced by storm generated sea floor currents. During a storm, the trilobites would have been covered by sediment (Mangano et al., 2002). Some survived and were able to escape from the sediment, leaving a Rusophycus trace, while others were left entombed, allowing one to definitively assign an organism to the trace.

SMOKING GUN SCENARIO

When assigning an organism to a trace, particularly those organisms that were readily fossilized like a trilobite, it is vital to consider the body fossils that have been found in the same or adjacent rock units. It has been highlighted in a number of different papers that one active organism can create a number of different traces, and/or a number of different organisms can create similar ichnofossils (Frey, 1975; Osgood, 1975; Bromley and Fursich, 1980; Ekdale et al., 1984; Fortey and Seilacher, 1997; Pemberton and MacEachern, 1997). Fortey

127 and Seilacher (1997) outlined criteria to assign a trace fossil with to a body fossil: "(1) close association in the field; (2) concurrent stratigraphic range; (3) a minimal choice of candidates available that could have made the trace; (4) consistent size range between trace fossil population and that of the maker; and (5) consistent biogeographic ranges of both" (p. 105). We consider such evidence in our suggestion that Rusophycus carleyi (James, 1885) were produced by Asaphellus aff. fezouataensis Vidal, 1998b, thus not hiding behind any "smokescreen of 'ignorance of the producer'" (Donovan, 2010).

Hard Evidence

The hard evidence as to the organism that created the traces, Rusophycus carleyi, is plentiful. Fortey and Seilacher's (1997) first criterion is that there must be a close physical association between trace and the organism that made it. Numerous samples of the thin sandstones that occur in the locality near Ouzina contain both R. carleyi and articulated specimens of Asaphellus. Also, there are five specimens available where a specimen of the trilobite, Asaphellus aff. fezouataensis (Figs. 4-3.1 - 4-3.4, 4-4) appears to occur directly over a matching (size and orientation) specimen ofR. carleyi. The "distance" between the traces and the trilobites is no greater than 24.6 mm in the remaining specimens. Fortey and Seilacher's (1997) second and fifth criteria ("concurrent stratigraphic range" and "consistent biogeographic ranges of both") are more difficult to assess. Certainly, since these trilobites and traces occur together in these thin beds, their stratigraphic and geographic ranges overlap. R. carleyi is also known from other regions where asaphid trilobites occur, such as Draper's (1980) examples of similar age from central Australia, which occur close to specimens of the asaphid Lycophron Fortey and Shergold, 1984. We do not argue that all R. carleyi were made by a species of Asaphellus but do argue that some asaphid trilobites, including Asaphellus, did make burrows that are classified as R. carleyi. The third criterion of Fortey and Seilacher (1997) is "a minimal choice of candidates available that could have made the trace" and the fourth is a

128 "consistent size range between trace fossil population and that of the maker." Both are easy to consider for the Ouzina locality. Only one species of trilobite has been found at this locality—Asaphellus aff. fezouataensis. While others may have been present, they were not nearly as common. As for matching the size of the trilobite to the trace, this is done in a scatter plot of length versus width in Figure 4-5. In this diagram it can be seen that the sizes of the trilobites and the traces overlap consistently. This is, of course, particularly true for the five specimens where A. aff'. fezouataensis appear to overlie specimens of R. carleyi of corresponding size and orientation. Not all of the many trilobite specimens plotted in Figure 4-5 are complete, articulated trilobites. We have extrapolated the sizes of complete specimens from partial specimens and in some cases from single sclerites. In order to do so, we have carefully measured the constituent components of complete specimens of the species of Asaphellus from Morocco described by Vidal (1998b: A. fezouataensis, A. tataensis and ,4. aff'. jujuanus Harrington, 1937) and the new species illustrated and described by Fortey (2009: A. stubbsi). These were used to establish a number of different ratios between the entire animals and their constituent parts to be able to calculate the length and width of a complete exoskeleton trilobite from disarticulated and/or fragmented specimens. Asaphid trilobites are remarkably conservative in their proportions of length to width (for instance, they always have eight thoracic segments), and the sizes taken up by the pygidium, cranidium, and so on. The traces were of course measured directly, and did not require extrapolations. Further hard evidence of the trace and the tracemaker is from impressions of parts of the ventral and distal parts of the exoskeleton in and around the trace. These occur either as linear impressions of the margins of the trilobite encompassing the trace (margins of cephalon or pygidium or distal tips of thoracic segments) or impressions of the hypostome. What can be seen certainly matches what is known of the morphology of species of Asaphellus (Figs. 4-3.2 - 4-3.4). One specimen (UA 13656 — Fig. 4-4) exhibits a pygidial impression that matches the size and shape of a pygidium of Asaphellus. Several traces have thoracic tip impressions (Figs. 4^4, 4-7.3 - 4-7.4), though the exact number of segments

129 cannot be deduced from the traces since they do not seem to be clear and complete. They are consistent with, but not proof of, having been formed by a trilobite with eight thoracic segments. A hypostome impression was found in one specimen, UA13655 (Fig. 4- 6.1). This clearly matches in size and shape the Asaphellus hypostomes found at the same locality. The morphology of the hypostome, impressed into the trace, is similar in size and shape to Asaphellus aff. fezouataensis in the following traits: the anterior lobe of the middle body is marginally convex; a distinct pair of maculae is present; the lateral borders widen backward; and the posterior border is flat. Asaphids have eight thoracic segments. They presumably had eight thoracic and three post-oral endopodites (Whittington, 1997). Each pair of endopodites would have had a pair of coxa, for a total of 11 pairs of coxal impressions. Eleven pairs of coxal impressions can be clearly counted in some of the traces, as illustrated in Figures 4—6.2 and 4-6.4, supporting our suggestion that these traces were made by asaphids.

Soft Evidence

Evidence that is not definitively associated with a particular trilobite and/or other organism is considered herein as soft evidence. Soft evidence found in one trace is spinal extensions on the endopodites (Figs. 4-6.3, 4-7.1 - 4-7.4). Since it has been established that the Rusophycus carleyi, from Ouzina, were probably made by Asaphellus aff. fezouataensis, either one can then conclude that the endopodites of that trilobite had spines upon their podomeres to create these fine striations within the mesial opening, or just regard the ridges as fine ridges ignoring the morphology of the animal that created them. Thus far, only a limited number of trilobites have been identified with spinose endopodites. Two of them are Triarthrv.s eatoni (Hall, 1838) (see Cisne, 1981; Whittington and Almond, 1987) and Olenoides serratus (Rominger, 1887) from the Burgess Shale

130 (Whittington, 1975, 1980, 1997). Whittington (1980, p. 191) argued that the fact that O. serratus has spinose podomeres allowing "the entire series of limbs to seize and hold prey or decayed material." This hypothesis could be extended to A. fezouataensis, and it merely used its spines to shred algal/bacterial mats on the sea floor to feed on algae and/or microbial mats. Whittington (1980) was clear that he thought that O. serratus could have dug into the substrate, thus creating a trace such as Rusophycus. This, too, could be correlated with the behavior of A. fezouataensis. The criteria available at this one locality in the Floian-Dapingian of southwestern Morocco provide enough details to create the schematic of a trilobite/trace association that we have depicted in Figure 4-8.

SYSTEMATIC PALAEONTOLOGY

Ichnogenus RUSOPHYCUS Hall, 1852

Type species.—Fucoides biloba Vanuxem, 1842

RUSOPHYCUS CARLEYI (James, 1885)

Figs. 4-3.1 - 4-3.4, 4^1, 4-6, 4-7

Cruziana carleyi: James, 1885, p. 155, pi. 8, fig. 1.

Other material examined: UA 13655-13662 complete traces and traces not illustrated UA13666a-UA 13666m, from the Upper Fezouata Formation, Floian- Dapingian, Ordovician, Ouzina, southern Morocco. All housed at the University of Alberta and used for the trilobite/trace scatter plot.

Diagnosis: Elliptical shaped trace with two distinct convex hyporelief lateral

131 lobes from anterior to posterior. Mesial opening with subcircular to subrectangular impressions.

Description: Ellipsoidal trace. Encompassed by two convex hyporelief lateral lobes, which can be either conjoined anteriorly and/or posteriorly, typically lobes are closed posteriorly and open anteriorly. No obvious ridges on lobes. Defined mesial opening. Paired subcircular to subrectangular lobes flanking midsagittal plane and/or line, between lateral lobes. Striations/ridges may or may not be observed in mesial opening.

Discussion: The dimension ratios of the traces, Rusophycus carleyi, from Ouzina, are similar to Osgood's (1970) traces, therefore those of James (1885), and larger than those of Radwanski and Roniewicz (1963), also mentioned by Osgood (1970). The size of the "hunting burrow" (also R. carleyi) of Brandt et al. (1995) is 17.5 cm long and is on the outer dimensions of the Ouzina traces (Fig. 4-5). Their (1995, fig. 1) trace exhibits many of the morphological traits observed in the Moroccan traces, including cephalic impression, coxal impressions, thoracic tip impressions, and endopodal spine scratch markings. It is outstanding that the Ordovician is, thus far, the only time in which trilobites have been found, definitively within their own trace, with the five specimens illustrated/discussed herein and the three Flexicalymene meeki in Rusophycus pudicum of Osgood (1970). Seilacher (1970, 1991) classified Rusophycus carleyi within the '''Carleyi Group," which has some of the following characteristics: oval outline with coxal markings, paired lobes, genal spine impressions at the anterolateral end of the trace, and is an Ordovician trace group. All these characteristics are met by the traces from Morocco. As for the statement made by Seilacher (1970) that they were resting traces is obviously speculation, it is much less speculative than his (2007, p. 210) statement that "this leaves Cruziana carleyi as the only candidate for a real hunting scene. But as this is probably a molting burrow, it is unlikely that the trilobite was in the mood for hunting." Brandt et al. (1995, fig. 1) argued

132 the point that their traces ofR. carleyi were hunting burrows on a worm, which produced the Palaeophycus Hall, 1847, trace at the anterior end of the Rusophycus. From the previously proposed microbial mat environment in the middle to distal offshore that the trilobites inhabited, it is more likely that the trilobites were feeding and/or avoiding a storm. The event(s) could have entombed some trilobites in their traces, while others were able to escape the sediment, which had been deposited. A number of similar ichnofossils were eliminated as the ichnospecies found at the locality of Ouzina, based upon morphology, including Cruziana radialis Seilacher, 1991; Cruziana dilatata Seilacher, 1970; and Rusophycus moyensis Mangano, Buatois and Muniz Guinea, 2002. Cruziana radialis was described by Seilacher (1991) as a lebenspurren that displays a radial scratch pattern, and this scratch pattern, at the posterior portion of the trace, is medioanteriorly directed. Such is not the case with Rusophycus carleyi, for no radial scratch pattern is observed. The size of the trace is also outside the range of the traces found in Morocco. Seilacher (1991) stated that C. radialis found in the Toko Range of Australia reaches a width of 19 cm, while the widest trace from Morocco is only 9.1 cm, though size should not be a restriction on ichnospecies identification if all other morphological traits are similar. Another trace fossil described by Seilacher (1970) and also classified within the 'Carleyi Group' is Cruziana dilatata Seilacher, 1970. For a number of morphological reasons this ichnospecies is eliminated: C. dilatata was described as a flattened trace, while the Moroccan specimens have a reasonable degree of convexity; specimens lack longitudinal striations upon the lateral lobes; the lobes can be closed and/or gaping, dependent upon preservation; and no evidence exists of trifid or any type of terminal claw markings. The width of C. dilatata from Australia and Canada of 15 cm (Seilacher, 1970) is outside of the dimensions of the Moroccan specimens, though this is not a defining character state for an ichnospecies. Only one morphological character state that Seilacher (1970) presented fits with the specimens of Morocco and that is a median line within the

133 central region of the trace. This feature is present on a number of traces, which were not illustrated. Mangano, Buatois, and Muniz Guinea (2002) erected a new ichnospecies from Argentina, Rusophycus moyensis. This ichnospecies can be eliminated as a possible candidate for the Moroccan specimens from the following character states not exhibited by the Moroccan specimens but described for R. moyensis: endopodal lobes with defined directional markings, a defined V-shape to the posterior region of the mesial opening and pointed meeting of the right and left endopodal lobes, coxal impressions are subsquare/subrectangular (not circular), and coxal impressions span the mesial opening. R. moyensis does exhibit genal spine, thoracic tip, and pygidial impressions, though this would be more closely tied to preservation of the trace and not necessarily a morphological character trait (Mangano et al., 2002; Mangano and Buatois, 2003). It should be noted that all the traces compared and contrasted with R. carleyi above fall within Seilacher's (1970, 1991) 'Carleyi Group' of traces, with the presence of coxal impressions and an overall, generalized, coffee-bean shape.

Family Asaphidae Burmeister, 1843

Genus Asaphellus Callaway, 1877

Type species: Asaphellus (Isotelus?) homfrayi Salter 1866.

Diagnosis: See Fortey and Owens, 1978, p. 132.

Asaphellus aff'. fezouataensis Vidal, 1998b Figs. 4-3.1 -4-3.4,4-4,4-9.1-4-9.3

Other material examined: UA13657, 13661-13665 are illustrated (some

134 specimens are the same catalogued number, when the trilobites are within the trace, be they whole and/or incomplete trilobites) and UA1667a-UA1667q were not illustrated, from the Upper Fezouata Formation, Floian-Dapingian, Ordovician, Ouzina, southern Morocco. All are housed at the University of Alberta.

Diagnosis: Refer to Vidal, 1998b, pp. 43^16.

Description: Refer to Vidal, 1998b, pp. 46-49.

Discussion: Due to the poor preservation of the trilobite exoskeletons, from the Ouzina bed that contains the traces, we are unable to propose a new species or definitively assign our material to one of the Moroccan species from the same formation described by Vidal (1998b). Most of our specimens are crushed and/or incomplete. The characteristics available to us, details of facial sutures, the ratios of the specimens, and generalized exoskeleton morphology are most similar to Asaphellus fezouataensis. Platypeltoides magrebiensis Rabano, 1990, occurs in the Lower Ordovician, Fezouata Shales of Morocco. Because of its large size, it was considered a possible producer of this trace. We eliminated this species as a possible producer of R. carleyi because it only has seven thoracic segments (Moore, 1959). A trilobite with seven thoracic segments and three post-oral cephalic segments would be expected to show only ten paired coxal impressions. The presence of the trilobites with and/or within the traces also assists in eliminating many other potential candidates. With that, only Asaphellus species from the Tafilalt region are considered within this discussion (Vidal, 1998b; Fortey, 2009). Fortey (2009) described a new species of asaphid, Asaphellus stubbsi. Based purely on one morphological character trait, the angle of the genal spines, this species can be immediately excluded. The genal spines on A. stubbsi, as stated by Fortey (2009) "are unusual: from a broadly flattened triangular base they curve outward at an angle which may approach a right angle to the sagittal

135 line in some specimens before curving more or less strongly backwards, or even slightly recurved distally, sometimes with a sigmoidal flourish at the very tip" (p. 13). The elegant recurve and flourish of the genal spines are not exhibited by the specimens from Ouzina. Their genal spines are a continuation of the lateral border of the cephalon, running parallel to the thorax (sag.) to approximately the fifth to sixth thoracic segment. Other morphological traits that may be used to exclude A. stubbsi are the anteriorly curved distal thoracic tips and the gentle sigmoidal nature of the facial sutures posterior to the eye and the small size of the eye, in comparison to the dimensions of the cephalon. Asaphellus jujuanus Harrington, 1937, is distinctly smaller than the specimens from Ouzina, but differing environmental conditions could have an effect on the size. The main discriminating character trait of A. jujuanus that distinguishes it from the Ouzina specimens is the short length of the genal spine. Other defining traits of A. jujuanus cannot be observed in the weathered and taphonomically altered Ouzina specimens. Vidal (1998b) erected one other Asaphellus species, A. tataensis. The posterior facial sutures of that species are dissimilar to those of the specimens from Ouzina. A. tataensis has the posterior parts of the facial sutures diverging in a more sigmoidal pattern compared to the near perpendicular pattern of the Ouzina specimens. The palpebral lobes of A. tataensis also have a transverse foreshortened appearance when compared to those of the Ouzina specimens. Therefore, the closest species that we could associate with the taphonomically altered and/or incomplete specimens of Ouzina is Asaphellus fezouataensis. The following character traits support this conclusion: the posterior facial suture pattern, placement and size of the eyes, length of the genal spines, overall shape of pygidium and morphological landmarks on the pygidium, and the shape and orientation of the thoracic segments.

136 CONCLUSIONS Thus far, only the Ordovician has provided the palaeontological community with compelling evidence of the identities of the makers of two different ichnospecies of Rusophycus—trilobites. Fortey and Seilacher (1998) provided criteria to suggest an association between a trace and a producer. Every one of their criteria is met in our collections from the Floian-Dapingian of Ouzina.

137 REFERENCES

Belka, Z., Kaufmann, B., and Bultynck, P. 1997. Conodont-based quantitative biostratigraphy for the Eifelian of the eastern Anti-Atlas, Morocco. Geological Society of America Bulletin, 109: 643-651. Bergstrom, J. 1976. Lower Palaeozoic trace fossils from eastern Newfoundland. Canadian Journal of Earth Sciences, 13: 1613-163 3. Bertling, M., Braddy, S. J., Bromley, R. G., Demathieu, G. R., Genise, J., Mikulas, R., Nielsen, J. K., Nielsen, K. S. S., Rindsberg, A. K., Schlirf, M., and Uchman, A. 2006. Names for trace fossils: a uniform approach. Lethaia, 39: 265-286. Birkenmajer, K. and Bruton, D. L. 1971. Some trilobite resting and crawling traces. Lethaia, 4: 303-319. Brandt, D. S., Meyer, D. L., and Lask, P. B. 1995. Isotelus (Trilobita) "hunting burrow" from Upper Ordovician strata, Ohio. Journal of Paleontology, 69:1079-1083. Bromley, R. G. and Ftirsich, F. T. 1980. Comments on the proposed amendments to the International Code of Zoological Nomenclature regarding ichnotaxa. Bulletin of Zoological Nomenclature, 37: 6-10. Buatois, L. A. and Mangano, M. G. 2010. The trace-fossil record of organism- matground interactions in space and time. SEPM Research Conference on Microbial Mats, Denver, CO. Bulman, O. M. B. 1964. Lower Palaeozoic plankton. Quarterly Journal of the Geological Society of London, 120: 455^476. Burmeister, H. 1843. Die Organisation der Trilobiten, aus ihren lebenden Verwandten entwickelt; nebst einer systematishcen Ubersicht aller zeither beschriebenen Arten. Reimer, Berlin, 147 pp. Callaway, C. 1877. On a new area of Upper Cambrian rocks in South Shropshire, with a description of a new fauna. Quarterly Journal of the Geological Society of London, 33: 652-672. Cegla, J., Dzulynski, S., and Kwiatkowski, S. 1967. Fractures resulting from

138 liquid infiltration into dry powdered materials. Bulletin de I Academie Polonaise des Sciences. Serie des Sciences Geologiques et Geographiques, 15: 83-88. Chatterton, B. D. E., Collins, D. H., and Ludvigsen, R. 2003. Cryptic behavior in trilobites: Cambrian and Silurian examples from Canada, and other related occurrences. Special Papers in Palaeontology, 70: 157-173. Chatterton, B. D. E. and Fortey, R. A. 2008. Linear clusters of articulated trilobites from Lower Ordovician (Arenig) strata at Bini Tinzoulin, north of Zagora, southern Morocco. In Rabano, I., Gozalo, R., and Garcia- Bellido, D. (eds.). Advances in Trilobite Research. Cuadernos del Meseo Geominero, 9. Instituto Geologico y Minero de Espafia, Madrid, pp. 73- 78. Chatterton, B. D. E., Fortey, R. A., Brett, K., Gibb, S., and McKellar, R. C. 2006. Trilobites from the upper Lower to Middle Devonian Timrhanrhart Formation, Jbel Gara el Zguilma, southern Morocco. Palaeontographica Canadiana, 25: 1—177. Chatterton, B. D. E., Johanson, Z., and Sutherland, G. 1994. Form of the trilobite digestive system: alimentary structures in Pterocephalia. Journal of Paleontology, 68: 294-305. Cherns, L., Wheeley, J. R., and Karis, L. 2006. Tunneling trilobites: habitual infaunalism in an Ordovician carbonate seafloor. Geology, 34:657-660. Cisne, J. L. 1981. Triarthrus eatoni (Trilobita): anatomy of its exoskeletal, skeletomuscular, and digestive systems. Palaeontographica Americana, 9:95-142. Destombes, J. 1962. Stratigraphie et paleogeographie de I'Ordovicien de 1'Anti- Atlas (Maroc: un essai de synthese). Bulletin de la Societe Geologique de France, 4: 453-460. Destombes, J. 1976. The Ordovician of the Moroccan anti-atlas. In Basset, M. G. (ed.). The Ordovician System: Proceedings of a Palaeontological Association Symposium. University of Wales Press and National Museum of Wales, Cardiff, pp. 411^113.

139 Donovan, S. K. 2010. Cruziana and Rusophycus: trace fossils produced by trilobites. . . in some cases? Lethaia, 10.1111/j.l502-3931.2009.00208.x. Draper, J. J. 1980. Rusophycus (Early Ordovician ichnofossil) from the Mithaka Formation, Georgina Basin. BMR Journal of Australian Geology & Geophysics, 5: 57-61. Ekdale, A. A., Bromley, R. G., and Pemberton, S. G. 1984. Ichnology: Trace Fossils in Sedimentology and Stratigraphy. Society of Economic Paleontologists and Mineralogists, Tulsa, OK, 317 pp. Fetah, S. E. M., Bensaid, M., and Dahmani, M. 1986. Carte Geologique du Maroc, Tafilalt- Taouz, Editions du service geologique du Maroc, Notes et Memories, No. 244. Royaume du Maroc, Ministere de I'Energie et des Mines, Direction de la Geologique du Maroc, Rabat. Foerste, A. F. 1910. Preliminary notes on Cincinnatian and Lexington fossils of Ohio, Indiana, Kentucky, and Tennessee. Bulletin of the Scientific Laboratories of Denison University, 16: 17-100. Fortey, R. A. 1975. Early Ordovician trilobite communities. Fossils and Strata, 4: 339-360. Fortey, R. A. 2009. A new giant asaphid trilobite from the Lower Ordovician of Morocco. Memoirs of the Association of Australasian Palaeontologists, 37:9-16. Fortey, R. A. and Morris, S. F. 1982. The Ordovician trilobite Neseuretus from Saudi Arabia, and the palaeogeography of the Neseuretus fauna related to Gondwanaland in the earlier Ordovician. Bulletin of the British Natural Museum (Natural History) Geology, 36: 63-75. Fortey, R. A. and Owens, R. M. 1978. Early Ordovician (Arenig) stratigraphy and faunas of the Carmarthen district, southwest Wales. Bulletin of the British Museum (Natural History), 30: 225-294. Fortey, R.A. and Owens, R.M. 1987. The Arenig series in South Wales. Bulletin of the British Museum (Natural History), 41: 69-307. Fortey, R. A. and Seilacher, A. 1997. The trace fossil Cruziana semiplicata and the trilobite that made it. Lethaia, 30: 105-112.

140 Fortey, R. A. and Shergold, J. H. 1984. Early Ordovician trilobites Nora Formation, central Australia. Palaeontology, 27: 315-366. Frey, R. W. 1975. The realm of ichnology, its strengths and limitations. In Frey, R. W. (ed.). The study of trace fossils: A synthesis of principles, problems, and procedures in ichnology. Springer-Verlag, New York, pp. 13-38. Hall, J. 1838. Descriptions of two species of trilobites belonging to the genus Paradoxides. American Journal of Science, 13: 139-142. Hall, J. 1847. Palaeontology of New York. State of New York, Albany, 338 pp. Hall, J. 1852. Palaeontology of New York. State of New York, Albany, 362 pp. Harrington, H. J. 1937. Some Ordovician fossils from Northern Argentina. Geological Magazine, 74: 97-124. James, J. F. 1885. The fucoids of the Cincinnati Group, Pt. 2. Journal of the Cincinnati Society of Natural History, 1: 151-166. Mangano, M. G. and Buatois, L. A. 2003. Trace fossils. In Benedetto, J. L. (ed.). Ordovician fossils of Argentina. Secretaria de Ciencia y Techologia & Universidad Nacional de Cordoba, Cordoba, Argentina, pp. 507-553. Mangano, M. G., Buatois, L. A., and Muniz Guinea, F. 2002. Rusophycus moyensis n. isp. en la transition Cambrica-Tremadociana del norpeste Argentino: Implicancias paleoambientales y bioestratigraficas. Revista Brasileira de Paleontologia, 4: 35-44. McKellar, R. C. and Chatterton, B. D. E. 2009. Early and Middle Devonian Phacopidae (Trilobita) of southern Morocco. Palaeontographica Canadiana, 28: 1—110. Moore, R. C. 1959. Treatise on Invertebrate Paleontology. Part O. Arthropoda 1. Geological Society of America and University of Kansas, 560 pp. Noffke, N., Gerdes, G., Klenke, T., and Krumbein, W. E. 2001. Microbially induced sedimentary structures; a new category within the classification of primary sedimentary structures. Journal of Sedimentary Research,

11 :o^y—ODO. Osgood, R. G. J. 1970. Trace fossils of the Cincinnati area. Palaeontographica

141 Americana, 6: 281-^439. Osgood, R. G. J. 1975. The paleontological significance of trace fossils. In Frey, R. W. (ed.). The study of trace fossils: A synthesis of principles, problems, and procedures in ichnology. Springer-Verlag, New York, pp 87-108. Pemberton, S. G. and MacEachem, J. A. 1997. The ichnological significance of storm deposits: the use of trace fossils in event stratigraphy. In Brett, C. E. and Baird, G. C. (eds.). Paleontological events: Stratigraphic, ecological, and evolutionary implications. Columbia University Press, New York, pp 73-109. 'Yrriberton, S. G., MacEachem, J. A., and Frey, R. W. 1992. Trace fossil facies models: environmental and allostratigraphic significance. In Walker, R. G. and James, N. P. (eds.). Facies models: Response to sea level change. Geological Association of Canada, pp 47-72. Pemberton, S. G., Spila, M., Pulham, A. J., Saunders, T. D. A., MacEachem, J. A., Robbins, D., and Sinclair, I. K. 2001. Ichnology & sedimentology of shallow to marginal marine systems: Ben Nevis & Avalon Reservoirs, Jeanne d'Arc Basin. Short Course Notes. Vol. 15. Geological Association of Canada, St. John's, 343 pp. Plummer, P. S. and Gostin, V. A. 1981. Shrinkage cracks; desiccation or synaeresis? Journal of Sedimentary Petrology, 51:1147-1156. foMda, H. and Bouougri, E. H. 2007. Wrinkle structures—a critical review. Earth-Science Reviews, 81: 199-215. Pabasio, P. I. 1990. Trilobites del Museo GeoMinero. I. Platypeltoides magrebiensis n. sp. (Asaphina, Nileidae), del Ordovicico inferior del Anti- Atlas central (Marruecos). Boletin Geologico y Minero, 101: 21-27. Radwanski, A. and Roniewicz, P. 1963. Upper Cambrian trilobite ichnocoenosis from Wielka Wisniowka (Holy Cross Mountains, Poland). Acta Palaeontologica Polonica, 8: 259-279, 210 pi. Rominger, C. 1887. Description of primordial fossils from Mount Stephen, North West Territory of Canada. Academy of Natural Sciences of

142 Philadelphia, Proceedings, 12-19. Salter, J. W. 1853. On the lowest fossihferous beds of North Wales. Report of the British Association for the Advancement of Science (Belfast) for 1852. Geology and Physical Geography Section, 56—58. Salter, J. W. 1866. A monograph of the British trilobites from the Cambrian, Silurian, and Devonian formations. Monographs of the Palaeontological Society, 129-176. Schieber, J. 1998. Possible indicators of microbial mat deposits in shales and sandstones: Examples from the Mid-Proterozoic Belt Supergroup, Montana, USA. Sedimentary Geology, 120: 105-124. Schieber, J. 1999. Microbial mats in terrigenous elastics: the challenge of identification in the rock record. Palaios, 14: 3-12. Seilacher, A. 1970. Cruziana stratigraphy of "non-fossiliferous" Palaeozoic sandstones. In Crimes, T. P. and Harper, J. C. (eds.). Trace fossils. Seel House Press, Liverpool, pp. 447^176. Seilacher, A. 1991. An updated Cruziana stratigraphy of Gondwanan Palaeozoic Sandstones. In Salem, M. J., Hammuda, O. S., and Eliagoubi, B. A. (eds.). The geology of Libya. Elsevier, Amsterdam, pp. 1565-1580. Seilacher, A. 2007. Trace fossil analysis. Springer-Verlag, Berlin, 226 pp. Seilacher, A. 2008. Biomats, biofilms, and bioglue as preservational agents for arthropod trackways. Palaeogeography, Palaeoclimatology, Palaeoecology, 270: 252-257. Shergold, J. H., Linan, E., and Palacios, T. 1983. Late Cambrian trilobites from the Najerilla Formation, northeastern Spain. Palaeontology, 26:71- 92. Vanuxem, L. 1842. Geology of New York, pt. Ill, Comprising the survey of the 2d geological district. W. & A. White and J. Visscher, Albany, NY, 3-6 pp. Vidal, M. 1998a. Le modele des biofacies a Trilobites: un test dans I'Ordovicien inferieur de 1'Anti-Atlas, Maroc. Sciences de la terre et des planetes, 327:327-333.

143 Vidal, M. 1998b. Trilobites (Asaphidae et Raphiophoridae) de I'Ordovicien inferieur de 1'Anti-Atlas, Maroc. Palaeontographica. Abteilung A: Palaeozoologie-Stratigraphie, 251: 39-77. Whittington, H. B. 1975. Trilobites with appendages from the Middle Cambrian, Burgess Shale, British Columbia. Fossils and Strata, 4:97-136. Whittington, H. B. 1980. Exoskeleton, moult stage, appendage morphology, and habits of the Middle Cambrian trilobite Olenoides serratus. Palaeontology, 23: 171-204. Whittington, H. B. 1997. The Trilobite body. In Moore, R. C. and Kaesler, R. L. (eds.). Treatise on invertebrate paleontology, Part O, Arthropoda 1, Trilobita (rev.). The Geological Society of America, Inc. and The University of Kansas, pp. 87-135. Whittington, H. B. and Almond, J. E. 1987. Appendages and habits of the Upper Ordovician trilobite Triarthrus eatoni. Philosophical transactions of the Royal Society of London (series B), 317: 1^46. Zylihska, A., Szczepanik, Z., and Salwa, S. 2006. Cambrian of the Holy Cross Mountains, Poland; biostratigraphy of the Wisniowka Hill succession. Acta Palaeontologica Polonica, 56: 443—461.

144 FIG. 4-1. Locality maps. 1, Map of Morocco and sunounding area (modified from Chatterton et al., 2006; McKellar and Chatterton, 2009). 2, Ordovician (shaded regions) locality map for trilobite/trace locality (marked with yj within the Tafilalt basin (modified from Fetah et al., 1986).

145 FIG 4-2 Microbial mat "cracking" and preferential iron stammg along the crack lines from the Upper Fezouata Formation, Ouzina, southern Morocco 1, Asaphellus aff fezouataensis in bottom right comer (UA13664) 2, Asaphellus aff fezouataensis in middle (UA13667n) Scale bar is 1 cm

146 FIG. 4-3. The proximity of Asaphellus aff'. fezouataensis to Rusophycus carleyi. 1, A. fezouataensis with white arrow directed at a lobe of R. carleyi under the right pleural region of the trilobite (UA13664). The trilobite would have become trapped within the sediment. The left lobe is not visible because perhaps the trilobite had dug alongside another R. carleyi and only the right lobe of the most recent behavior is recorded. 2, R. carleyi, convex hyporelief, with white arrow pointing to the right lateral section of the cephalon of a trilobite (UA 13657). 3, R. carleyi, convex hyporelief, with white arrow directed towards the left genal spine of a trilobite (UA 13662). 4, R. carleyi, convex hyporelief, with the white arrow at the right ventral lateral section of the cephalon and the black arrow pointing at the transected portion of the hypostome of the trilobite, within the trace (UA13661). Specimens UA13662 & UA13661 were submerged in ethyl alcohol to enhance the exoskeleton within the trace, while UA 13664 & UA13657 were coated in ammonium chloride. Specimens are from the Upper Fezouata Formation, Ouzina, southern Morocco. Scale bar is 1 cm.

147 148 FIG. 4-4. Rusophycus carleyi (UA13658), convex hyporelief with left lateral border of pygidium (white arrow), and thoracic segment impressions (white arrows outlined by black), from the Upper Fezouata Formation, Ouzina, southern Morocco. Scale bar is 1 cm.

149 • * 90 • • • • 80

70 • • 60 • • E •

50 4-> 3 40

30 -

20

10

80 100 120 Length (mm)

FIG. 4-5. Scatter plot (length vs. width) of Rusophycus carleyi and the trilobites from the Ouzina locality. The length and width of the trilobite, when only fragments were available were inferred by extrapolating proportions from parts, with ratios taken from four Asaphellus species from the Ordovician of Morocco. The black diamonds are the trilobites and the grey squares are Rusophycus carleyi. All specimens are from the Upper Fezouata Formation, Ouzina, southern Morocco.

150 FIG. 4-6. Rusophycus carleyi (UA13655), convex hyporelief, from the Upper Fezouata Formation, Ouzina, southern Morocco. 1, Lateral oblique close-up of hypostome impression (black arrow). 2, Oblique lateral view of trace. 3, Lower posterior view of endopodal spine scratches within the medial opening. 4, Close-up of coxal impressions, with the left being the anterior end of the trace. Scale bar is 1 cm.

151 FIG. 4-7. Rusophycus carleyi. 1, UA13656 displaying endopodal spine scratches within the posterior lateral medial opening. 2, UA13660 displaying endopodal spine scratches within the lateral medial opening. 3, 4, UA13659. 3, Entire view of convex hyporelief of trace. 4, Close-up of thoracic segment impressions on trace. All specimens are from the Upper Fezouata Formation, Ouzina, southern Morocco. Scale bar is 1 cm.

152 fc»jr*«^ &». ^*^»s«' •: '•<* £$&

ivK\ is-'f-i^ &~

: ; •d ?y? J^r-V ^ ll'f

153 FIG 4-8. Schematic of the tracemaker (trilobite), Asaphellus aff. fezouataensis, removing itself from the substrate, resulting in the trace Rusophycus carleyi. From the striations/ridges within the mesial opening, it is assumed that A aff'. fezouataensis would have endopodomeral spines. Sketch created by Darrin Molmaro.

154 FIG. 4-9. Asaphellus aff. fezouataensis, all dorsal views. 1,UA13664, cephalon and majority of thorax (submerged in ethyl alcohol to enhance trilobite morphology). 2, UA13665, cephalon, minority of thorax and taphonomically altered hypostome (coated with ammonium chloride). Important note, with regard to this specimen is the pits observed on the posterior end of the cephalon, associated with the alimentary structures (diverticulae). The pits are similar in outline to trilobite alimentary diverticulae illustrated by Chatterton et al. (1994, Figs. 1.1-1.3, 2.1, 2.4 and 3.4). 3, UA13663, pygidium (coated with ammonium chloride). All specimens are from the Upper Fezouata Formation, Ouzina, southern Morocco. Scale bar is 1 cm.

155 CHAPTER 5: ORDOVICIAN OF AUSTRALIA

ARTHROPOD ICHNOFOSSILS FROM THE ORDOVICIAN STAIRWAY SANDSTONE OF CENTRAL AUSTRALIA1

INTRODUCTION

Ichnofossils typically ascribed to trilobites, such as Cruziana and Rusophycus, among others, are described from the upper unit of the Stairway Sandstone (Wells et al. 1970, Ritchie & Gilbert-Tomlinson 1977), Amadeus Basin of central Australia. This unit yields abundant biogenic sedimentary structures with rare fragmented and/or disarticulated trilobite sclerites and some bivalves and fishes (Ritchie & Gilbert-Tomlinson 1977). Though other traces do occur within this unit, only the Palaeozoic arthropod traces are addressed herein, due to this paper being part of a larger project on arthropod traces. The age of the upper Stairway Sandstone was initially considered to be of 'Upper Llanvimian to Llandeilian' age by Gilbert-Tomlinson (in Cook 1970, p. 71), that is, equivalent to the upper Middle Ordovician or in the upper portion of the Darriwilian Stage based on the revised global stratigraphic framework (Webby et al. 2004). However, Ritchie & Gilbert-Tomlinson (1977, p. 356) subsequently suggested a slightly older 'uppermost Arenigian-lower Llanvimian' age for the Stairway Sandstone, that is, earlier in the Middle Ordovician (lower- middle Darriwilian). Shergold (1986, p. 11, fig. 1) later recognised that the age within the Stairway Sandstone depended on what part of the tripartite subdivision had been sampled for fossils, and was able to show that the upper part belonged wholly within the 'Llanvimian', that is equivalent to the middle part of the Darriwilian Stage (Webby et al. 2004). But even that age may now need to be revised slightly upwards, for Zhang & Barnes (2003) recognised conodont zones

A version of this chapter has been published. GIBB, S., CHATTERTON, B.D.E. & PEMBERTON, S.G. 2009. Arthropod ichnofossils from the Ordovician Stairway Sandstone of central Australia. Memoirs of the Association of Australasian Palaeontologists, 37, 695-716. ISSN 0810-8889. 156 from the apparently conformably overlying Stokes Siltstone, and the lower zone in the sequence is likely to have an early Late Ordovician (early Caradocian) age. That is more or less equivalent to the upper portion of the recently approved Sandbian Stage (Bergstrom et al. 2006). Ichnofossil 'biological' names should be "based solely upon the morphological characteristics of the structure" (Kelly 1990, p. 424). The "ichnotaxa should be treated as non-biological form names only and their association with named organisms should be a matter of careful discussion, especially when there is no body fossil present. Even if there is a body fossil present, it may not be that of the original constructor" (Kelly 1990, p. 425). This methodology is followed in this paper, for as numerous authors have commented, a behaviour can be identical even when the organisms are different (Bromley & Ftirsich 1980, Ekdale et al. 1984, Kelly 1990). The categories of behaviour identified from the Stairway Sandstone are digging, scratching and walking, that is repichnia, or crawling and fodichnia or feeding, and/or possibly cubichnia or resting traces within or upon the substrate. Traces in the upper Stairway Sandstone are almost exclusively preserved in exogenic convex hyporelief (Seilacher 1964, Osgood 1970).

GEOLOGICAL BACKGROUND Geological focus is on sedimentary rocks of approximately 9144 metres (stratigraphic thickness), preserved in an area of 155000 square kilometres within an 800 kilometre long region referred to as the Amadeus Basin (Fig. 5-1). The basin is situated within central Australia (Wells 1970, p. 1; Cook 1970, p. 71; Shergold 1986, p. 10). The Stairway Sandstone, is the defined focus of the present study, within the Amadeus Basin and is part of the Larapinta Group (Chewings 1935, Prichard & Quinlan 1962, Wells et al. 1965, Wells et al. 1970) that was originally termed by Chewings (1935) in a table as the 'Stairway Quartzite'. Later, Prichard & Quinlan (1962, p. 21) referred to this lithological unit as the 'Stairway Greywacke'. Finally, Wells et al. (1965, p. 26) amended the formation name to the 'Stairway Sandstone'.

157 The regions of the Amadeus Basin that we focus upon are the Charlotte Range and Mount Watt (Fig. 5-1). We collected from eight localities on the Charlotte Range and one locality at Mount Watt (Table 5-1, coordinates with associated illustrated and unillustrated ichnofossils). We also examined and photographed a number of specimens housed in the Australian Museum collections that were collected on an earlier field excursion (in 1987) to the Charlotte Range and Mount Watt by Alex Ritchie, Robert Jones and Brian Chatterton. As the formation name implies, the clasts of the trace-bearing beds are sand. The sand grains are medium-sized, subrounded to rounded. The sandstones are an orange-buff colour on fresh surfaces and a darker orange-brown on weathered surfaces at the Charlotte Range. Thinner, interbedded mudstones are recessive and not usually visible in surface exposures. The Mount Watt specimens are similar in colour on weathered surfaces, but a slightly lighter buff colour on fresh surfaces, and the sands are of a finer grain size. The variation in colouration and grain size is minor between the Charlotte Range and Mount Watt. Ritchie & Gilbert-Tomlinson (1977, p. 354) reported that in the Charlotte Range, the Stairway Sandstone disconformably overlies the Jay Creek Limestone of Middle Cambrian age, while the Stairway Sandstone at Mount Watt disconformably caps the Winnall Beds of the late Proterozoic (Ritchie & Gilbert- Tomlinson 1977, p. 354). Based purely on the lithologies and the ichnotaxa found at these localities, the palaeoenvironment was reconstructed. The strata are interbedded sandstone and mudstone. A marine setting that fits this lithological signature is one that is distal lower shoreface and/or proximal upper offshore, or 'transition' between the lower shoreface and upper offshore (Pemberton et al. 2001, fig. 69, p. 89, p. 95) within the region of the fairweather wave base (Walker & Plint 1992, fig. 1, p. 219). Therefore, this would suggest a water depth of approximately 15 metres (Walker & Plint 1992, fig. 1, p. 219), in an epeiric sea (Cook 1970, p. 78). To substantiate this palaeoenvironmental setting, the ichnotaxa dominance and diversity confirms that the setting is that of an archetypal Cruziana ichnofacies with dominant deposit feeders and subordinate grazers and foragers (Pemberton et al. 2001, fig.

158 69, p. 89, p. 95, fig. 74, p. 96). The arthropod ichnotaxa collected from the Stairway Sandstone localities are approximately 72% Cruziana, 20% Monomorphichnus, 5% Diplichnites and 3% Rusophycus (see Appendix). These traces are indicative of the Cruziana ichnofacies. With regard to the preservation of the traces, Seilacher (1994, p.752) noted that one must have a "sharp interface between a sand veneer and an underlying mud layer. The erosional bases of thin stormsands provided such a tabula rasa". The traces of the upper Stairway Sandstone are near flawless in their preservation of behavioural activities of organisms digging, scratching and walking upon and within a sand layer that overlies a finer muddy substrate. It was previously thought the organisms were attempting to find shelter, though Seilacher (2007, p. 34) later stated the organisms "dug primarily for feeding, not to hide". Thus, after deposition of a mantling sand sheet upon a 'muddy' substrate, the trace may indicate that an organism was attempting to gain access to the possible food/microbial nutrients within the mud. Sand was later deposited to account for the thicker quartzitic sandstone layers in the Stairway Sandstone, though the ease with which the traces were found and/or extracted implies that a recessive layer normally existed under the quartz-rich sandstone bed of all preserved traces. Therefore, the organisms were moving the sandy layer to get to the mud layer in an attempt to find food. Since they disturbed a mud layer with an overlying sand layer, the sand immediately infilled the excavations within the mud thus preserving the trace (Seilacher 1982, p. 346; Goldring 1985). Fine mudstone/shale inclusions are sometimes found clinging to the surfaces of trace- bearing sandstone beds. Also some sandstone beds contain shale rip-up clasts, attesting to the original presence of mud layers that were eroded during storm events. Ritchie & Gilbert-Tomlinson (1977) and Ritchie (1985) described a variety of organisms from the uppermost Stairway Sandstone including: the fish Arandaspis prionotolepis and bivalves, along with the aforementioned Cruziana, which they identified as Cruziana cf. furcifera from the Mount Watt locality.

159 SYSTEMATIC PALAEONTOLOGY

The specimens we collected were all assigned catalogue numbers (from F133006-133027, F133861-F133958, F135853-F135856 and F58984, and are listed in the Appendix), and they are housed in the Australian Museum, Sydney. The ichnogenera and ichnospecies are based purely upon their morphological features. The ichnospecies is based upon the definition of an ichnotaxon: "A taxon based on the fossilised work of an organism including fossilised trails, tracks or burrows (trace fossils) made by an animal" (Ride et al. 1999, p.l 18). We however prefer the simpler definition for an ichnotaxon of Bertling et al. (2006, p. 238) which states: "a taxon based on a trace fossil, including fossilised trails, tracks or burrows". No mention of the organism that may have produced the lebensspuren (Trewin 1994, p. 812) is made other than in the discussion, in accordance with Osgood's suggestion "to base the names strictly on morphology" (Osgood 1970, p. 303). Specific morphological terminology follows Trewin (1994).

Cruziana d'Orbigny, 1842

Type ichnospecies. Cruziana rugosa d'Orbigny, 1842. Subsequent designation of ichnospecies by Miller (1889).

Diagnosis. An elongate, bilobate furrow that can vary in depth with respect to the bedding plane, may be straight and/or gently curved (not tightly or sharply) within or upon the bedding surface, and may be composed of repetitive sets (or series) of imprints (ridges) along its length (though in most cases these ridges cannot be differentiated into distinctive set patterns); ridges in central part of furrow usually aligned in a herringbone shape and more or less continuous along length of furrow; also, sometimes, outside herringbone-aligned ridges, the trace may exhibit pair of narrow, comparatively smooth outer zones with or without fine brush-like impressions, and additionally there may be presence of lateral ridges.

160 Remarks. Hantzschel (1975, p. W55) emphasised that Rusophycus is not at all equivalent to Cruziana. Cruziana consists of "elongate bandlike furrows covered by herringbone-shaped ridges" (Hantzschel 1975, p. W55) and not "short bilobate bucklelike forms, resembling [the] shape of coffee beans" (Hantzschel 1975, p. W101) as observed with Rusophycus. This is contrary to Seilacher's (1990, p. 651) statement that: "They are united under the ichnogeneric name Cruziana d'Orbigny, whether made in a stationary (coffee bean-shaped 'rusophyciform' expression) or in a bulldozing manner (band-shaped 'cruzianaeform' expression)." Consequently, the two forms are clearly distinguishable from their distinctive morphological character traits so the ichnogenera Cruziana and Rusophycus should not be synonymised but recognised as distinct and separate. This follows Fillion & Pickerill (1990, p. 24), who stated: "Although Seilacher (1970, p. 455) united both long furrows and short excavations (^Rusophycus) under Cruziana because similar scratch marks made it possible to 'attribute burrows of very different outline to the same animals', most subsequent authors considered, as we do, their morphologies to differ significantly and preferred to retain the two distinctive ichnogenera because knowledge of the tracemaker is not taxonomically significant in ichnology."

Cruziana barriosi Baldwin, 1977 (Fig. 5-2A-B)

1977 Cruziana barriosi; Baldwin, p. 17-19, pi. 1C. 1990 Cruziana barriosi; Fillion & Pickerill, p. 25, pi. 2, fig. 6. Material and locality. F133880 (CR2) and F133884 (CR7) from the Charlotte Range.

Diagnosis. See Baldwin (1977, p. 17).

Description. Refer to Baldwin (1977, p. 17).

Remarks. Due to the ridges running parallel to the midline of the trace, with little

161 divergence, many other Cruziana ichnospecies are easily eliminated from consideration in the specimens assigned to this taxon. The width of the Stairway Sandstone specimens are narrower (maximum width of 13.1 to 34.5 mm) than that of specimens collected by Baldwin (1977, p. 17), and Fillion & Pickerill (1990, p. 24). The morphological characteristics of the trace are, however, similar: little divergence from the midline, no lateral ridges and a possibility of a median groove are observed (though due to weathering it is difficult to ascertain if the adjacent trace is the left exopodite trace in F133884). The number of ridges is greater than those stated by Fillion & Pickerill (1990, p. 24) of up to sets of six, while Baldwin (1977, p. 24) stated that they occur "with at least 5 ridges per set". The specimens from Australia depict, possibly up to 11 (F133880: weathering has eliminated the definition of the ridges) and 12 ridges (F133884). Due to the morphological similarities of being 'parallel' to the midline striations, the striation count has little significance to the trace itself. Rather than create a number of similar species, with characteristics based on exact counts of striations, we include these Australian specimens in C. barriosi. Seilacher (1962, fig. 2, 1990, p. 652, fig. 32.2, 2007, p. 37, pi. 12) illustrated endopodal claw impressions and figured a possible Illaenus endopod with 12 'claws' on the limb, therefore presenting a possible trace maker of the specimen (F133884) found in Australia. Seilacher (2007, p. 37) suggested that Cruziana rugosa may have been produced by a species of Illaenus. Baldwin (1977, p. 17) stated for his C. barriosi that: "At a number of points the appendage impressions are interrupted and a surface morphology similar to that of C. rugosa is produced." This pattern is not observed by us, though it does correlate well with Seilacher's (2007, p. 37) findings of Illaenus and C. rugosa. Thus possibly it demonstrates the adoption of two different behaviours, by Illaenus, represented by the ichnospecies of C. rugosa and C. barriosi, as Baldwin (1977, p. 17) proposed. The feeding traces of primitive Ordovician fishes have not been described, although it is often assumed that some, if not most, primitive fishes were bottom feeders (Long, 1995). It is conceivable that Arandaspis or one of the other primitive fishes that occurred in the Stairway Sandstone could have produced a

162 trace that includes a series of parallel scratches. The mouth parts of these animals are not well known (Ritchie and Gilbert-Tomlinson, 1977). Herein, we only suggest that these animals could possibly have made such traces, but believe that it is more probable that they were made by an arthropod (fish specimens are certainly as common as, or more common than trilobites in this unit).

Cruziana furcifera d'Orbigny, 1842 (Fig. 5-2C-G)

1842 Cruziana furcifera; d'Orbigny, p. 21, pi. 1, figs 2-3. 1970 Cruziana furcifera; Seilacher, p. 464. 1977 Cruziana cf. furcifera; Ritchie & Gilbert-Tomlinson, p. 354. 1990 Cruziana furcifera; Fillion & Pickerill, p. 25, pi. 2, fig. 11 (see also for further synonymy). 2007 Cruziana furcifera; Egenhoff et al., p. 291-292, figs 3d-e, 4a-b. ?2007 Cruziana furcifera; Seilacher p. 194.

Material and locality. F133938 (CR7) and F133871 (CRR) from the Charlotte Range and Fl35852 (MW) from Mount Watt.

Diagnosis. The most comprehensive diagnosis was provided by Fillion & Pickerill (1990, p. 25), and they stated: "scratch marks are regular, some criss­ crossing in a faint rhombic pattern, acute angled, and typically associated in sets; they may swing towards parallelism with the median line in a median posterior direction."

Description. Refer to Fillion & Pickerill (1990, p. 25).

Remarks. The key feature is the parallelism of the scratch mark sets along the median line at the posterior of the trace. The Stairway Sandstone specimens (F133938, F133871 & F135852) lack lateral ridges, therefore discriminating them from the ichnospecies Cruziana goldfussi. The associated bundle sets are not well defined in the Stairway Sandstone traces, therefore they can not be assigned to C.

163 rugosa. Ritchie & Gilbert-Tomlinson (1977, p. 354) also identified Cruziana cf. furcifera from the Mount Watt locality (fig. 5 A, B). The images of Ritchie & Gilbert-Tomlinson (1977) indicate that the traces are C. furcifera. We also found and identified this ichnospecies at the same locality and bedding plane as a fish plate of Arandaspis prionotolepis Ritchie & Gilbert-Tomlinson 1977 (p. 367). Seilacher (1970, P. 462-464) classified Cruziana furcifera within the 'Rugosa Group' and he commented that the group was cosmopolitan and occurred in abundance. Cruziana furcifera also occurred in abundance, in the upper Stairway Sandstone of the Middle Ordovician, but members of the 'Rugosa Group' mainly occur in Early Ordovician of other Gondwana regions (Seilacher 2007, p. 191, pi. 66).

Cruziana goldfussi (Rouault, 1850) (Fig. 5-3A-F)

1970 Cruziana rugosa Seilacher p. 450, fig. 2a. 1990 Cruziana goldfussi Fillion & Pickerill p. 26, pi. 3, fig. 1 (see for further synonymy). 2007 Cruziana goldfussi Egenhoff et al., p. 292-294, figs 4c-e, 5a-b . 2007 Cruziana goldfussi Seilacher p. 194, pi. 68.

Material and locality. F133887 (CRR), F133936 and F133937 (CR7) from the Charlotte Range.

Diagnosis. After Fillion & Pickerill (1990, p. 26).

Description. The dimension ranges of the specimens are: maximum trace length = 113.6-180.5 mm; maximum trace width = 45.6-59.6 mm; maximum ridge width =1.8 mm; and distance between ridges = 0.5-1.3 mm.

Remarks. Crimes & Marcos (1976, p. 352-353) and Fillion & Pickerill (1990, p. 26) presented the most comprehensive descriptions and Seilacher's (2007, pi. 11)

164 sketch based on a specimen from the Lower Ordovician of France provided the depth and morphology this ichnospecies should display. Two slightly different 'forms' of Cruziana goldfussi have been recognised among the Stairway Sandstone samples. F133937 (Figs. 5-3A-D) is similar to the schematic representation figured by Seilacher (1970, p. 450, fig. 2a) as C. rugosa, though this was later amended to C. goldfussi. This Stairway form is a relatively deep trace, dug for some behavioural need, with a maximum recorded depth of 61.6 mm. The organism appeared to have dug into the substrate at a slight angle and upon reaching the 'bottom', extended its legs for maximum digging. The oddity of this trace (Figs. 5-3 A-D) is that the upper portion of the trace is slightly narrower than the lower part of the trace (measured from mid-length of the trace the 'bottom' has maximum width of 54.4 mm, and the 'top', a maximum width of 39.8 mm). One possible explanation is that the organism implemented maximum leg extension at the base of the trace and in retreat from the burrow, withdrew its legs and possibly flexed the body towards the midline and then pushed up with its legs. The trace has scratch marks laterally that were probably produced as the organism retracted itself from beneath the substrate. There are also lateral ridge markings, and posteriorly the trace has scratch markings that are more or less parallel to the median line/furrow. There are also 'cephalon' markings at the anteriormost end of the 'burrow'. The 'burrow' is angled into the substrate at approximately 30-35° to the horizontal (bedding plane). The other form (F133936: Fig. 5-3F) of the trace of Cruziana goldfussi is that of the 'standard' horizontal trace, whereas F133887 (Fig. 5-3E) seems to be an intermediate between both F133937 and F133936, with a slight angle (approximately 20°) to the trace at the anterior end. Both specimens (F133887: Fig. 5-3E & F133936: Fig. 5-3F) have a defined median line with little to no cross-over of scratch marks, with the scratches appearing as sets and approaching parallel at the posterior end of the trace (Seilacher 1970, p. 464, 1991, p. 1572). Seilacher's (1970, p. 462, 1991, p. 1572) C. goldfussi was classified by him in the 'Rugosa Group' and the other ichnospecies within this group are all from the Ordovician.

165 Due to the lateral ridges on all of the specimens, and other morphological characteristics already mentioned, they are assigned to Cruziana goldfussi. The specimens do not demonstrate outer lobe markings, therefore eliminating them from C. semiplicata, and they have lateral ridge markings, thus are not C. furcifera.

Cruziana omanica Seilacher, 1970 (Figs. 5-4A-F)

1970 Cruziana omanica; Seilacher, p. 466, fig. 9a, b. 1983 Cruziana warrisi; Webby, p. 63-65, fig. 2B. 1983 Rusophycus latus; Webby, p. 69-72, figs. 2A, 4A, 4D, 5A, 6. 1991 Cruziana cf. omanica; Seilacher, p. 1570, pi. 1, fig. a. 2004 Monomorphichnuspodolicus; Uchman et al., p. 75, fig. 5. 2007 Cruziana lata (Webby, 1983); Seilacher, p. 192. 2008 Cruziana lata (Webby, 1983); Seilacher, p. 36.

Material and locality. F133007, F133886, F133877, F133883, F133893 and F133895 (CR7) from the Charlotte Range.

Diagnosis. Shallow bilobate structure with equally shallow median furrow. Median furrow has scratch marks diverging into furrow, occasionally crossing to opposite side. Scratch marks that are long, blunt and tricuspidate, with median scratch that is stronger and more raised than lateral ones. Right and left markings are almost 180° to each other. Can be straight, curved, and/or partial ridge bundles.

Description. Large bilobate traces, that range mainly between about 7-12 cm in width and up to over 20 cm in length. Form extensive irregular to straight burrows preserved in convex hyporelief. Lateral lobes are fairly flat, with most of relief near middle and sides of traces. Lateral lobes are clearly separated by irregular median furrow. Considerable irregularity to median furrow is caused by

166 crossing of subparallel trifid scratch marks from both sides. Coarse scratch marks are markedly transverse to general axis of trace, only curving slightly backward near middle and sides of trace. Scratches show minor degree of bundling of 3-4 scratches in some traces. Surfaces with this ichnospecies are often intensely bioturbated, and adjacent traces interfere with another so sides are often quite irregular (Figure 5-4F).

Remarks. Cruziana cf. omanica was figured by Seilacher (1991) from the Pacoota Sandstone near Alice Springs. He considered the trace to have an Upper Cambrian age. This Pacoota trace is identical to the traces found in the Stairway Sandstone (Figure 5—4). Seilacher (2008, p. 36) stated "a form that is closely related or even identical is found in the lowennost Ordovician of Oman" even though he had previously stated that C. omanica is a Cambrian, more specifically an Upper Cambrian (1991, p. 1569, fig. 3) trace. This brings into question the use of this species as a biostratigraphic marker in the ichnostratigraphic concept of Seilacher for Gondwana, given that the Pacoota Sandstone specimen was first thought to be Upper Cambrian (Seilacher 1990, p. 660, fig. 32.5, 1991, fig. 3, 2007, p. 192, pi. 67), though later Seilacher stated it to be lowermost Ordovician (2008, p.36), and Lindsay & Korsch (1991, p.21) have also stated that the Ordovician sedimentary rocks of the Amadeus Basin include both the Pacoota and Stairway sandstones. Cruziana omanica may have a range from the Upper Cambrian (if this interpretation of age is the correct, or Lower Ordovician, if incorrect) and the younger age must range upwards into the Middle Ordovician. The range of Cruziana omanica may even be extended upwards if the synonymy of Monomorphichnus podolicus Uchman et al. (2004, p. 76) from the Lower Devonian is accepted. Monomorphichnus podolicus appears identical to C. omanica in most respects, but it is narrower than Seilacher's types (1970, p. 466), and size may not be a good reason for rejecting the synonymy. The specimens from the Stairway Sandstone are identified as Cruziana omanica based on the diagnosis provided by Seilacher (1970, p. 466) and his illustration (p. 465, fig. 9a, b) of "long, blunt tricuspidate endopodal markings, in

167 which the median scratch is stronger than the lateral ones". The width of the trace also corresponds with Seilacher's (1970, p. 466) definition of the trace of up to 8— 9 cm, though the range should be slightly expanded to be that of 7-12 centimetres to account for both the smaller and larger sized traces found in Australia. In terms of Seilacher's (1970, p. 465) taxonomic classification of groups within Cruziana, the Australian trace of C. omanica belongs to the 'Petraea Group' given the number of claw markings (between three and five). Cruziana kufraensis Seilacher et al. (2002) from the Early Silurian Tanezzuft Formation of SE Libya is similar in having near 180° between right and left lobe markings, but the authors of that ichnospecies indicated that these markings were made by legs that "had probably no more than two claws" (p. 262). Consequently C omanica, is defined as having trifid scratch marks, hence the scratches based on claw morphology discriminates between these two ichnospecies. Webby (1983, p. 63-65) erected the ichnospecies Cruziana warrisi. Since ichnofossils are based purely on morphology, we believe that C. warrisi and C. omanica are synonyms. In several works, Seilacher (1970, p. 466, 1991, p. 1571— 1572, 2007, p. 183, pi. 67, 2008, p.36) referred to C. omanica occurring in both Oman and Australia. It is not possible to differentiate a trace fossil into a distinct ichnospecies based upon geographical position. Scotese's (2001, p. 13) Middle Ordovician reconstruction of Gondwana has it encompassed by a large continental shelf, thus allowing for intracommunication along this eastern Paleo-Tethys Ocean shelf. A possible trilobite group, of the appropriate size range, found in the Stairway Sandstone that could make traces of this size is the Asaphida. Asaphids had planktonic larvae; the asaphoid protaspis (Fortey & Chatterton 1988, p. 178). The pelagic asaphoid protaspid could have dispersed these animals widely, including along the shallow shelf, thus allowing Oman and Australia to have similar ichnofossils; in this case, C. omanica. The adults were also large and active enough that, given time, they could also have dispersed widely. Ritchie & Gilbert-Tomlinson (1977, p. 353) found an endemic asaphid taxon in the uppermost Stairway Sandstone: 'Asaphus' thorntoni Etheridge, 1892 (=Basilicus7 thorntoni [Etheridge, 1892]).

168 The geographical argument can also be used for Monomorpichnus podolicus that was erected by Uchman et al. (2004, p. 75-77). Though the trace is found in Devonian strata in Ukraine, the morphological character states synonymise it with Cruziana omanica. Uchman et al. (2004, p. 75-77, fig. 5) recognised the trace as intermediate between Monomorphichnus and Cruziana, though from their figure and description, the ichnospecies is definitely Cruziana, more specifically, C omanica. The characteristics that synonymise the ichnospecies: parallel to subparallel striations, bundle of 4-6 striations (placing it within Seilacher's (1970) 'Petraea Group', though Seilacher had stated "3 to about 5", it is the 'about' that allows interpretation to account for 6 striations/bundle), "striae.. .whose axis is straight to slightly curved and perpendicular to the bundle" (Uchman et al. 2004, p. 77), and bilobate ridges. The trace also conforms to the adjusted size range mentioned above. Some specimens of Rusophycus latus Webby, 1983 are synonymised within Cruziana omanica for a number of morphological reasons. The diagnosis of the ichnospecies stated that the trace could consist of straight, sinuous, and/or partial bundles. A number of Stairway Sandstone specimens (Figure 5^1D-F), and Webby's (1983, fig. 2a) specimen, consist of bundles of ridges in a highly bioturbated sandstone. Due to the high degree of bioturbation, the 'typical' cruzianaeform morphology is overprinted a number of different times. When the bundle is a single representation, as seen by Webby (1983, fig. 6), it is still cruzianaeform though the organism dug deeper into the substrate at that point. Webby (1983, p. 71) stated that the ridges are bifid or trifid, therefore not being excluded from a taxon with a diagnosis of tricuspidate ridges. If the ridges were not so tightly overlapped, one would possibly observe that they are trifid. The size ratios cited by Webby (1983, p. 71), also fit within the ratios obtained from the Stairway Sandstone specimens. We consider that some specimens of R. latus represents short segments of C. omanica, some perhaps the result of slightly deeper digging while making a longer trail. The organism that created this particular trace may have been a trilobite, since trilobites are almost always interpreted as having made Cruziana and Rusophycus.

169 Few trilobites have been found in the Stairway Sandstone (only a few disarticulated sclerites have been found) due to the clastic nature of the unit and the fact that the upper unit was likely subjected to storm reworking and to comparatively high-energy regimes. The Horn Valley Siltstone is an Ordovician trilobite-bearing sedimentary unit that occurs below the Stairway Sandstone in some parts of the Amadeus Basin. The Horn Valley Siltstone outcrops neither at the Charlotte Range nor at Mount Watt. At the Charlotte Range, the Stairway Sandstone disconformably overlies the Jay Creek Limestone of Middle Cambrian age (Ritchie & Gilbert-Tomlinson 1977, p. 354). While at Mount Watt, the Stairway Sandstone overlies the Winnall Beds of late Proterozoic age (Ritchie & Gilbert-Tomlinson 1977, p. 354). The trilobite fauna of the Horn Valley Siltstone can still be considered for possible candidates for the traces in the Stairway Sandstone, at least at the family level. The most likely suspects are trilobite members of the Order Asaphida Salter, 1864, based on a few sclerites and fragments that have been found in close proximity to biogenic sedimentary structures within the Stairway Sandstone of the Charlotte Range (Figure 5-5) and that occur in abundance in other Ordovician clastic units within Australia. Another reason to consider asaphids as the trace makers is their size. Laurie (2006, p. 303-305, pi. 13) described and figured Lycophron howchini (Etheridge 1894), which has a thorax of comparable width to the large Cruziana omanica. Some of the asaphid sclerites in the upper Stairway Sandstone are large enough for the animals that secreted them to have made C. omanica (Figure 5-5).

Cruziana penicillata isp. nov. (Figs. 5-6A-H)

Material and locality. Holotype: F133868 (almost complete biogenic sedimentary structure, minus most posterior end); paratypes: F133018 (partial trace of posterior portion), and F133871 (only anterior portion of trace is visible, co-occurring with Cruziana furcifera, among other ichnofossils) (all occurring at CRR) from Charlotte Range; F135853, F135855 and F135856 all isolated

170 specimens from Mount Watt.

Etymology. Latin penicilla, brush, alluding to individual scratch mark sets that are similar to a brush mark.

Diagnosis. Ridges (scratch marks) occurring in distinct bundles that repeat and overlap previous bundlelike-sets, creating step-like appearance. Trace either angled -30° from horizontal bedding plane or horizontal.

Description. Ridges (or scratch marks) are clearly defined. Single, well-defined ridges (or scratch marks). Approximately five to seven ridges per bundle set. Ridges form bundles as overlapping sets that appear like a sequence of descending steps, when viewed as an entire specimen rather than as a convex hyporelief. Angle of trace relative to bedding plane varies from -30° inclination to horizontal. No lateral ridges. Faint cephalon markings may occur anteriorly. Ridges criss-cross median furrow. Median furrow is not clearly defined. Striations occur at approximately 45°-50° angle to median furrow (anterior to posterior). Dimensional ranges of specimens (even if incomplete or partially enclosed within the rock): maximum trace length = 14.5-52.0 mm; maximum trace width = 17.6-31.3 mm; maximum trace depth = 10.4-38.1 mm; maximum ridge width = 1.0—1.9 mm; maximum width between ridges = 0.3-1.0 mm and angle of trace from horizontal = 20°-35°.

Remarks. This species occurs in only one locality within the Charlotte Range, with some other traces in situ and from Mount Watt. The trace bears a slight similarity to C. rugosa, though the scratch marks of C. rugosa are more 'feather­ like' in appearance, having a posterior edge, and the angle of the scratch marks of C. rugosa is much greater relative to the median furrow (anterior to posterior). Also, C. rugosa differs in having a more clearly defined median furrow. The claw markings per leg are difficult to differentiate either due to the coarseness of the grain size of the sand relative to the size of the ridges, and/or dependent on some degree of post-production erosion. The holotype usually has

171 at least five claw markings per bundle, and this seems to be consistent, though in places a slight variation of up to seven striations per bundle is observed. Cruziana penicillata falls within Seilacher's (1970, p. 462-464, 1991, p. 1572) "Rugosa Group" due to the fact that it exhibits multiple and sharp scratches represented by up to 12 subequal claw-marks (Seilacher 1970, p. 462). It also occurs only in the form of a short cruzianaeform ribbon or deeper 'bath tub' furrow, rather than a stationary rusophyciform burrow (Seilacher 1991, p. 1572).

Diplichnites Dawson, 1873

Type ichnospecies. Diplichnites aenigma Dawson, 1873 by monotypy.

Remarks. Hantzschel (1975) provided a comprehensive account of the ichnogenus.

Diplichnites arboreus isp. nov. (Figs. 5-7A-C)

1983 Diplichnites sp. A; Webby p. 68, fig 3C.

Material and locality. Holotype: F58984 from the Charlotte Range (collected by A. Ritchie: recorded as R.d. CI); paratype: F133014 and F133022 from Mount Watt.

Etymology. Latin for tree: Trace resembling a stick drawing of a conifer, lacking a stem/trunk.

Diagnosis. Single, regular, almost straight scratch marks. Posterior to anterior tapering of markings. Paired rows of scratch marks to right and left form V-angle near 90°. No lateral ridge impression.

Description. Single scratch markings, though one or two scratches show forking

172 to produce bifid markings. Regularly spaced scratch marks, straight or with slight curvature. Posterior scratches are longer, whereas scratch marks along longitudinal median 'furrow' taper to anterior, where right and left scratch marks meet almost at a point. Longitudinal median furrow can be narrow and deep anteriorly, only slightly widening posteriorly (Holotype: F58984) or distinctly wider and shallow (F133014 and F133022) towards posterior end. Holotype dimensions: maximum width = 20.6 mm; maximum length = 28.9 mm; ridge width = 1.2-2.2 mm; and ridge spacing = 0.5-1.6 mm.

Remarks. Diplichnites arboreus isp. n. is not equivalent to Rusophycus crimesi Fillion & Pickerill, 1990 (p. 54) owing to the scratch marks not being strictly bifid. One or two scratch marks do give the appearance of bifid marking, though the majority of the scratches are either vertically or slightly inclined single markings. These taxa share some morphological traits, but are different enough to keep them as separate ichnospecies: the longitudinal median furrow in R. crimesi does not taper to the extent observed in D. arboreus and the scratch marks in that taxon are bifid. As Fillion & Pickerill (1990, p. 54) stated for their R. crimesi, it can be a rusophycid-like equivalent of a Cruziana ichnospecies, or an undertrace of a Cruziana or Rusophycus ichnospecies. Crimes (1970a, pi. 5d, 1970b, pi. 9d, e) illustrated two different forms of Diplichnites, but both were only identified at the generic level. Both of these examples, from the Cambrian of Wales, have some resemblance to D. arboreus from the Stairway Sandstone, but the width at the anterior of the trace is far too wide and never appears to come to a point. Crimes (1970a, b), illustrated Diplichnites with irregular spacing between the endopodal markings, and the angle between the right and left striations is greater than 90°.

Monomorphichnus Crimes, 1970b

Type ichnospecies. Monomorphichnus bilinearis Crimes, 1970b.

173 Remarks. See diagnosis provided by Crimes 1970b (p. 57) and discussion by Fillion & Pickerill (1990, p. 40-41).

Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya, 1977 (Figs. 5- 7D, 5-8A-C)

Material and locality. F133896 and F133931 (CR7) from the Charlotte Range and Fl33022 from Mount Watt.

Diagnosis. Refer to diagnosis by Crimes et al. (1977, p. 103), and Fillion & Pickerill (1990, p. 42).

Description. See Fillion & Pickerill (1990, p. 42).

Remarks. The description provided by Crimes et al. (1977, p. 103) presented the possibility of the trace being 'straight to slightly sigmoidal'. Traces collected from the Charlotte Range display a straight morphology. There is some variation among two traces of this species illustrated herein, though each was made by a single set of leg scratches. They do not occur as bifid scratches, as seen in Monomorphichnus bilinearis Crimes, 1970b, nor are they similar to M. multilineatus Alpert, 1976 (p. 234), which, as stated by Crimes et al. (1977, p. 103) exhibits central ridges that are "deeper than the outer ridges". The variation between the two specimens presented as Monomorphichnus lineatus is just a representation of the same morphological characteristics, presented by either different organisms or organisms of varying sizes. The largest of the two specimens (F133896: Fig. 5-8A) is very large: trace length = 68.5 mm, trace width = 48.0 mm, maximum ridge width = 3.1 mm, minimum ridge width =1.4 mm, maximum width between ridges = 11.0 mm, and minimum width between ridges = 6.1 mm. Crimes et al. (1977, p. 104) presented M. lineatus var. giganticus but based upon their description of the trace, it is composed of 10 ridges, and this is not the ridge count in any of the Stairway Sandstone specimens

174 found. Up to seven ridges occur within the Stairway ichnospecies sample of M. lineatus. The smaller of the two samples of Monomorphichnus lineatus (F133931, Fig. 5-8B) has the following sizes: trace length = 71.5 mm (though could be longer due to the trace being at the edge of the rock); trace width = 10.2 mm; maximum ridge width =1.5 mm; minimum ridge width = 0.6 mm; maximum width between ridges = 2.0 mm; and minimum width between ridges = 0.2 mm. As illustrated in Figure 5-8B and C, F133931 displays two types of Monomorphichnus, and the sketched Figure 5-8C demonstrates that these are two separate, partly superimposed ichnospecies: M. lineatus and M. sinus isp. nov. (see below).

Monomorphichnus multilineatus Alpert, 1976 (Fig. 5-8D)

Material and locality. F133943 (CR7) from the Charlotte Range.

Diagnosis. See Fillion & Pickerill (1990, p. 42).

Description. Refer to Alpert (1976, p. 234) and Fillion & Pickerill (1990, p. 42).

Remarks. The six parallel striations with the central ridges being slightly higher than the two lateral ridges fits the diagnosis of Alpert (1976, p. 234). The length of the trace is also in accordance with Alpert (1976), but the width is much greater in the Stairway specimen (F133943): width and length are almost equivalent, at approximately 16.5 mm. This specimen also resembles GSC 78158 from Fillion & Pickerill (1990, pi. 10, fig. 4), though they had one more striation in their specimen. It differs from Monomorphichnus lineatus in the following morphological character states: displays higher central ridges to the lateral ridges, whereas M. lineatus has an almost even depth to the ridges, and the lack of a continuous line, more like a digging than a drag-type structure.

175 Monomorphichnus sinus isp. nov. (Figs. 5-8B, C, E)

Material and locality. Holotype: F133931 (top right of Figs. 5-8B-C) from the Charlotte Range (CR7); paratypes F133931 (top left of Figs. 5-8B-C) from the Charlotte Range (CR7), and F133919 from the Charlotte Range (CR4).

Etymology. Latin sinus; curve.

Diagnosis. Six or seven parallel ridges, with slight to prominent curvature at midpoint.

Description. Bundles of six or seven curved ridges. Ridges angled at approximately 35°-45° from horizontal plane (when observed in convex hyporelief). Angled ridges: innermost ridge, of curve, is typically more shallow and ridges become higher/deeper and wider to fourth ridge and then taper to be more shallow. Curve can be slight to abruptly angled at midpoint. Blunt and/or pointed proximal/distal tips of ridges. Numerical dimensions of the traces are: trace length = 10.2-25.0 mm; trace width = 16.3-21.4 mm; maximum ridge width = 1.8-3.2 mm; minimum ridge width = 0.6-1.3 mm; maximum width between ridges = 1.5-1.7 mm; and minimum width between ridges = 0.6-0.7mm.

Remarks. Monomorphichnus sinus isp. nov. is distinct from other ichnospecies of Monomorphichnus due to the curvature of the ridges. Monomorphichnus sinus is definitely classified within the Monomorphichnus ichnogenus, due to the fact that it exhibits claw-like markings from a single side of the organism and no counterpart impressions of the opposite limb(s). Other ichnospecies appear to have a linear 'straightness' or are only slightly sinuous, which is typically defined as a slight flexion or extension as the appendage was dragging along the substrate, for example: M. bilinearis and M. lineatus. None appear to have the prominent curve seen in this ichnospecies. It has one character state that could link it closely with M. multilineatus for the innermost (lateral) ridge of the curve is less defined, as seen in M. multilineatus, but the 'other' lateral ridges are well defined, unlike

176 M. multilineatus.

Monomorphichnus spp. (Fig. 5-61)

Material and locality. Fl 33890 (CR7) from the Charlotte Range.

Remarks. The abundance of scratch marks covering this one rock creates difficulties in differentiating individual ichnospecies, other than Monomorphichnus sinus. The number of curved ridges and reworking creates issues as to which ridge may belong to the other ridge. There is a very high probability that M. lineatus and M. multilineatus are represented here, though due to the extreme reworking, they cannot be further differentiated herein.

Rusophycus Hall, 1852

Type ichnospecies. Rusophycus bilobatus Hall, 1852.

Diagnosis. Refer to Fillion &Pickerill (1990, p. 52).

Remarks. As mentioned previously, Seilacher (1970, p. 454, 1990, p.651) synonymised Cruziana and Rusophycus, thus creating problems within the classification of traces with cmzianae- and rusphyciform expressions. We follow Fillion & Pickerill (1990, p. 24) and Bertling et al. (2006, p. 281) in recognising these ichnotaxa as being distinct and useful in the classification of traces.

Rusophycus unilobus (Seilacher, 1970) (Fig. 5-8F)

1970 Cruziana uniloba; Seilacher, p. 473, fig. 7.26.

Material and locality. Fl 33920, one complete specimen (CR4) from the

177 Charlotte Range.

Diagnosis. Refer to Seilacher (1970, p. 473).

Description. Evidence of bilobate structure that lacks median furrow. Extensive scratch marks eliminate any possible median furrow, criss-crossing and creating interfingered braided structure. Angle of left and right scratch marks approximately 140°, being slightly offset. Length is longer than width of trace.

Remarks. Seilacher (1970, p. 473) originally named this trace Cruziana uniloba, though in the diagnosis he recognised it as a resting trace. Due to its form and since it is an accepted hypothesis that Rusophycus represents 'resting traces', this ichnospecies is assigned to Rusophycus. Seilacher (1970, p. 473, fig. 7.26) illustrated and diagnosed R. unilobus as lacking a median furrow due to the right and left striations interfingering across the midline. The width of the Stairway trace (F133920) is 54.9 mm, which is a little more than twice the width quoted by Seilacher (1970, p. 473) of 20—25 mm, but his trace was from the Lower Devonian, while ours is Middle Ordovician in age. This size difference between Ordovician and Devonian specimens of this taxon calls into consideration the Cruziana biostratigraphy that Seilacher has promoted in several works (Seilacher 1970, 1990, 1991, 1994, 2007 and Seilacher et al, 2002). It demonstrates that different organisms of different ages can produce morphologically similar traces, even though they may vary in size. Seilacher (1970, p. 471) also referred to R. unilobus as a member of the 'Pudica Group', he observed that "this group needs to be revised before it can possibly be used stratigraphically." Rusophycus unilobus is very distinctive, with no other ichnospecies previously identified in the literature that even closely resembles it. This particular specimen has a maximum length = 96.1 mm; maximum width = 54.9 mm; ridge width range = 1.3—4.7 mm and a maximum depth = 30.4 mm. The legs appear to be trifid, with the median claw, in some cases, being slightly higher than the two lateral claws, though in most cases they appear to be equal in size and shape. There are no lateral ridge marks, and the organism appears to

178 have reworked the substrate in a vertically downward direction.

CONCLUSIONS The uppermost part of the Stairway Sandstone from the Middle Ordovician (Darriwilian) of central Australia contains exquisite 'arthropod' traces. On some bedding surfaces, the bioturbation is so intense that individual ichnospecies cannot be identified. Eleven ichnospecies are identified. Arthropod traces are not the only biogenic sedimentary structures found in this unit, for a number of other traces such as Chondrites, Planolites and other nondescript ichnofossils, probably produced by worm-like animals, were found and but are not described herein. From the ichnofossils, the lithology occurring at these localities, and the palaeotectonic setting, we assign the upper Stairway Sandstone to the Cruziana ichnofacies, formed under an epeiric sea on a continental shelf. Identification of the organisms that created the traces described is still open for debate. Rare large asaphid and other trilobite exoskeletal sclerites occur in the same strata, as do primitive fishes. The trilobite sclerites are of a suitable size for the animals that secreted them to have made some of the larger traces. However, sclerites are so rare, the traces are so common, and our knowledge of the sclerotised and unsclerotised members of the Arthropoda that may have co­ existed with them in the Australian Middle Ordovician seas so incomplete, that we are unwilling to argue that trilobites must have been the only, or even primary, excavators of these traces. We even consider whether primitive bottom feeding fishes could have made one of the traces that we describe. One of the major problems faced by ichnologists who study Palaeozoic 'arthropod' traces, like those described in this work, is that many of the units that contain abundant traces contain few or no body fossils, and the units that contain abundant trilobites or other arthropods often contain few if any traces. Examples where trace makers and traces occur together, such as those of Osgood (1970) and possibly Fortey & Seilacher (1997), are the exception rather than the mle.

179 REFERENCES

ALPERT, S.P., 1976. Trilobite and star-like trace fossils from the White-Inyo Mountains, California. Journal of Paleontology 50, 226-239.

BALDWIN, C.T., 1977. The stratigraphy and facies associations of trace fossils in some Cambrian and Ordovician rocks or north western Spain. 9-40 in Crimes, T.P. & Harper, J.C. (eds), Trace fossils 2. Geological Journal, Special issue 9.

BERGSTROM, S.G., FINNEY, S.C., XU, C, GOLDMAN, D. & LESLIE, S.A., 2006. Three new Ordovician global stage names. Lethaia 39, 287-288.

BERTLING, M., BRADDY, S.J., BROMLEY, R.G., DEMATHIEU, G.R., GENISE, J.,

MIKULAS, R., NIELSEN, J.K., NIELSEN, K.S.S., RINDSBERG, A.K., SCHLIRF,

M. & UCHMAN, A., 2006. Names for trace fossils: a uniform approach. Lethaia 39, 265-286.

BROMLEY, R.G. & FURSICH, F.T., 1980. Comments on the proposed amendments to the International Code of Zoological Nomenclature regarding ichnotaxa. Bulletin of Zoological Nomenclature 37, 6—10.

CHEWINGS, C, 1935. The Pertatataka series in central Australia, with notes on the Amadeus sunkland. Transactions and Proceedings of the Royal Society of South A ustralia 59, 141—163.

COOK, P. J., 1970. Cambro-Ordovician. 60-86 in Wells, A.T., Forman, D.J., Ranford, L.C. & Cook, P. J., Geology of the Amadeus Basin, central Australia. Bureau of Mineral Resources, Geology and Geophysics, Bulletin 100.

CRIMES, T.P., 1970a. The significance of trace fossils in sedimentology, stratigraphy, and palaeoecology with examples from Lower Palaeozoic strata. 101-126 in Crimes, T.P. & Harper, J.C. (eds), Trace fossils. Seel House Press, Liverpool.

CRIMES, T.P., 1970b. Trilobite tracks and other trace fossils from the Upper Cambrian of North Wales. Geological Journal 7, 47-68.

CRIMES, T.P., LEGG, I., MARCOS, A. & ARBOLEYA, M., 1977. ?Late Precambrian-

180 low Lower Cambrian trace fossils from Spain. 91-138 in Crimes, T.P. & Harper, J.C. (eds), Trace fossils 2. Geological Journal, Special issue 9.

CRIMES, T.P. & MARCOS, A., 1976. Trilobite traces and the age of the lowest part of the Ordovician reference section forNW Spain. Geological Magazine 113,349-356.

DAWSON, J.W., 1873. Impressions and footprints of aquatic animals and imitative markings on Carboniferous rocks. American Journal of Science and Arts 105, 16-24. D'ORBIGNY, A., 1842. Voyage dans VAmerique meridionale (le Bresil, la Republique argentine, le Patagonia, Le Republique de Chile, le Republique de Bolivie, Le Republique de Perou) execute pendant les anee 1826-1833, 3 (Paleontologie), Paris, 188 p.

EGENHOFF, S.O., WEBER, B., LEHNERT, O. & MALETZ, J., 2007. Biostratigraphic precision of the Cruziana rugosa group; a study from the Ordovician succession of southern and central Bolivia, Geological Magazine 144, 289-303.

EKDALE, A.A., BROMLEY, R.G. & PEMBERTON, S.G., 1984. Ichnology. The use of trace fossils in sedimentology and stratigraphy. Society of Economic Paleontologists and Mineralogists Publication, Tulsa, 317 p.

ETHERIDGE, R., 1892. On a species of Asaphus from the Lower Silurian rocks of central Australia. South Australian Parliamentary Paper 23, 8-9.

ETHERIDGE, R. 1894. Further additions to the Lower Silurian fauna of central Australia. South Australian Parliamentary Paper 25, 23-26.

FILLION, D. & PICKERILL, R.K., 1990. Ichnology of the Upper Cambrian? to Lower Ordovician Bell Island and Wabana groups of eastern Newfoundland, Canada. Palaeontolographica Canadiana 7, 1-119.

FORTEY, R.A. & CHATTERTON, B.D.E., 1988. Classification of the trilobite Suborder Asaphina. Palaeontology 31, 165-222.

FORTEY, R.A. & SEILACHER, A., 1997. The trace fossil Cruziana semiplicata and the trilobite that made it. Lethaia 30, 105-112.

GOLDRING, R., 1985. The formation of the trace fossil Cruziana. Geological

181 Magazine 122, 65-72. HALL, J., 1852. Paleontology of New York. II. Organic remains of the lower middle division of the New York system. Geological Survey of New York, Albany, 362 p.

HANTZSCHEL, W., 1975. Trace Fossils and Problematica. 1-269 in Teichert, C. (ed.), Treatise on invertebrate paleontology, Part W, Miscellanea, Supplement 1. Geological Society of America, Boulder and University of Kansas, Lawrence.

KELLY, S.R.A., 1990. Trace Fossils. 423^125 in Briggs, D.E.G. & Crowther, P.R. (eds), Paleobiology: A synthesis. Blackwell Science, Oxford.

LAURIE, J.R., 2006. Ordovician trilobites from the Horn Valley Siltstone and basal Stairway Sandstone, Amadeus Basin, Northern Territory. Memoirs of the Association of Australasian Palaeontologists 32, 287-345.

LINDSAY, J.F. & KORSCH, R.J., 1991. The evolution of the Amadeus Basin, central Australia. Bureau of Mineral Resources, Geology and Geophysics, Bulletin 236, 7-32.

LONG, J., 1995. The rise of fishes. 500 million years of evolution. John Hopkins University Press, Baltimore, 220 p. MILLER, S.A., 1889. North American geology and paleontology for the use of amateurs, students and scientists. Western Methodist Book Concern, Cincinnati, 664 p. OSGOOD, R.G. J., 1970. Trace fossils of the Cincinnati area. Palaeontolographica Americana 6, 281^439.

PEMBERTON, S.G., SPILA, M., PULHAM, A.J., SAUNDERS, T.D.A., MACEACHERN, J. A., ROBBINS, D. & SINCLAIR, I.K., 2001. Short Course, Volume 15: Ichnology & sedimentology of shallow marine marginal systems: Ben Nevis & Avalon reservoirs, Jeanne D-Arc Basin. Geological Association of Canada, St John's, 343 p.

PRICHARD, C.E. & QUINLAN, T., 1962. The geology of the southern half of the Hermannsburg 1:250,000 Sheet. Bureau of Mineral Resources, Geology and Geophysics Report 61, 1-39.

182 RIDE, W.L., COGGER, H.G., DUPUIS, C, KRAUS, O., MINELLI, A., THOMPSON, F.C. & TUBBS, P.K., 1999. International Code of Zoological Nomenclature. International Trust of Zoological Nomenclature, London, 306 p.

RITCHIE, A., 1985. Arandaspisprionotolepis: The Southern Four-eyed fish. 95- 101 in Rich, P.V., van Tets, G.F. & Knight, F. (eds), Kadimakara: Extinct vertebrates of Australia. Princeton University Press, Princeton.

RITCHIE, A. & GILBERT-TOMLINSON, J., 1977. First Ordovician vertebrates from the Southern Hemisphere. Alcheringa 1, 351-368.

ROUAULT, M., 1850. Note preliminaire sur une nouvelle formation decouverte dans le terrain silurien inferieur de la Bretagne. Bulletin de la Societe Geologique de France [2] 7, 724-744.

SALTER, J.W., 1864-1883. A monograph of the British trilobites. Palaeontographical Society Monographs, 1-244. SCOTESE, C.R., 2001. Paleogeographie Atlas. PALEOMAP Progress Report 90- 0497, 45 p. SEILACHER, A., 1962. Form und Funktion des Trilobiten-Daktylus. Sonerausgabe zur Paldontologischen Zeitschrift I- VIII, 218—227.

SEILACHER, A., 1964. Sedimentological classification and nomenclature of trace fossils. Sedimentology 3, 253—256.

SEILACHER, A., 1970. Cruziana stratigraphy of "non-fossiliferous" Palaeozoic sandstones. 447-476 in Crimes, T.P. & Harper, J.C. (eds), Trace fossils. Geological Journal. Special Issue No. 3.

SEILACHER, A., 1982. Distinctive features of sandy tempestites. 333-349 in Einsele, G. & Seilacher, A. (eds), Cyclic and event stratification. Springer- Verlag, New York.

SEILACHER, A., 1990. Paleozoic trace fossils. 649-670 in Said, R. (ed.), Geology of Egypt. Balkema, Rotterdam.

SEILACHER, A., 1991. An updated Cruziana stratigraphy of Gondwanan Palaeozoic Sandstones. 1565-1580 in Salem, M.J., Hammuda, O.S. & Eliagoubi, B.A. (eds), The geology of Libya. Elsevier, Amsterdam.

183 SEILACHER, A., 1994. How valid is Cruziana stratigraphy? Geologische Rundschau 83, 752-758.

SEILACHER, A., 2007. Trace fossil analysis. Springer Verlag, Berlin, 226 p. SEILACHER, A., 2008. Fossil art. An exhibition of the Geologisches Institut. CBM-Publishing, Laasby, 98 p.

SEILACHER, A., LUNING, S., MARTIN, M.A., KLITZSCH, E., KHOJA, A. & CRAIG, J., 2002. Ichnostratigraphic correlation of Lower Palaeozoic elastics in the Kufra Basin (SE Libya). Lethaia 35, 257-262.

SHERGOLD, J.H., 1986. Review of the Cambrian and Ordovician palaeontology of the Amadeus Basin, central Australia. Bureau of Mineral Resources, Geology and Geophysics Report 276, 1—21.

TREWIN, N.H., .1994. A draft system for the identification and description of arthropod trackways. Palaeontology 37, 811-823.

UCHMAN, A., DRYGANT, D., PASZKOWSKI, M., POREBSKI, S.J. & TURNAU, E., 2004. Early Devonian trace fossils in marine to non-marine redbeds in Podolia, Ukraine: palaeoenvironmental implications. Palaeogeography, Palaeoclimatology, Palaeoecology 214, 67-83.

WALKER, R.G. & PLINT, A.G., 1992. Wave and storm-dominated shallow marine systems. 219-238 in Walker, R. G. & James, N. P., (eds). Facies models. Response to sea level change. Geological Association of Canada, St John's.

WEBBY, B.D., 1983. Lower Ordovician arthropod trace fossils from Western New South Wales. Proceedings of the Linnaean Society of New South Wales 107, 59-74.

WEBBY, B.D., COOPER, R.A., BERGSTROM, S.M. & PARIS, F., 2004. Stratigraphic framework and time slices. 41-51 in Webby, B.D., Paris, F., Droser, M.L. & Percival, I.G. (eds), The great Ordovician biodiversification event. Columbia University Press, New York. WELLS, A.T., 1970. Summary. 1^1 in Wells, A.T., Forman, D.J., Ranford, L.C. & Cook, P.J. (eds), Geology of the Amadeus Basin, central Australia. Bureau of Mineral Resources, Geology and Geophysics Bulletin 100.

184 WELLS, A.T., FORMAN, D.J. & RANFORD, L.C, 1965. The geology of the North- Western part of the Amadeus Basin, Northern Territory. Bureau of Mineral Resources, Geology and Geophysics Report 85, 1-45.

ZHANG, J. & BARNES, C.R., 2003. Early Ordovician conodonts from the Stokes Siltstone, Amadeus Basin, central Australia. Courier Forschungsinstitut Senckenberg 245, 1-37.

185 APPENDIX: SPECIMEN DETAILS

Number Location Identification F133018 CRR Paratype: Cruziana penicillata isp. nov. F133865 CRR Cruziana goldfussi (Rouault 1850) F133866 CRR Paratype: Cruziana penicillata isp. nov. F133868 CRR Holotype: Cruziana penicillata isp. nov. F133869 CRR Cruziana goldfussi (Rouault 1850) F133870 CRR Cruziana furcifera d'Orbigny 1842 F133871 CRR Cruziana furcifera d'Orbigny 1842 & Paratype: Cruziana penicillata isp. nov. F133887 CRR Cruziana goldfussi (Rouault 1850) F133008 CR1 Cruziana indet. (erosion) F133878 CR1 Cruziana barriosi Baldwin 1977 F133894 CR1 highly bioturbated F133881 CR1F Cruziana omanica Seilacher 1970 F133864 CR2 Rusophycus lunilobus (Seilacher 1970 F133867 CR2 Cruziana indet. (fragment) F133872 CR2 highly bioturbated F133873 CR2 Cruziana indet. (erosion) F133874 CR2 Cruziana furcifera d'Orbigny 1842 F133875 CR2 Cruziana indet. (erosion) F133876 CR2 Cruziana penicillata isp. nov. F133880 CR2 Cruziana barriosi Baldwin 1977 F133882 CR2 Cruziana barriosi? Baldwin 1977 or Cruziana furcifera? D'Orbigny 1842 F133947 CR3 Cruziana furcifera d'Orbigny 1842 F133948 CR3 possible fragment of thoracic segment F133949 CR3 thoracic fragment, molluscs F133950 CR3 thoracic segment, molluscs F133951 CR3 Cruziana omanica Seilacher 1970 F133953& F133952 CR3 partial Cruziana omanica Seilacher 1970 F133954 CR3 ammonoid F133955 CR3 molluscan fragments F133956 CR3 impression of pygidium F133957 CR3 pygidium F133010 CR4 Cruziana omanica Seilacher 1970 F133011 CR4 Cruziana omanica Seilacher 1970 F133913 CR4 highly bioturbated Monomorphichnus spp. F133914 CR4 Cruziana omanica Seilacher 1970 F133915 CR4 Cruziana omanica Seilacher 1970 F133916 CR4 Cruziana omanica Seilacher 1970 F133917 CR4 Cruziana omanica Seilacher 1970 F133918 CR4 half of Cruziana omanica (one lobe) Seilacher 1970 F133919 CR4 Paratype: Monomorphichnus sinus isp. nov.

186 F133920 CR4 Rusophycus unilobus (Seilacher 1970) F133921 CR4 highly bioturbated Monomorphichnus spp. F133922 CR4 Cruziana omanica Seilacher 1970 F133923 CR4 Cruziana omanica Seilacher 1970 F133924 CR4 Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya 1977 F133958 CR4 Cruziana omanica Seilacher 1970 F133903 CR5 Cruziana omanica Seilacher 1970 F133904 CR5 highly weathered Cruziana omanica Seilacher 1970 F133905 CR5 Cruziana omanica Seilacher 1970 F133907 CR5 Cruziana omanica Seilacher 1970 F133909 CR5 Cruziana omanica Seilacher 1970 F133910 CR5 Cruziana furcifera d'Orbigny 1842 F133911 CR5 Cruziana indet. F133023 CR6 highly bioturbated Monomorphichnus spp. F133024 CR6 highly bioturbated Monomorphichnus spp. F133025 CR6 Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya 1977 F133026 CR6 Diplichnites arboreus isp. nov. & Monomorphichnus spp. F133027 CR6 highly bioturbated Monomorphichnus spp. F133861 CR6 highly bioturbated Monomorphichnus spp. F133862& F133863 CR6 Diplichnites arboreus? & Monomorphichnus spp. F133902 CR6 Cruziana furcifera? d'Orbigny 1842 F133906 CR6 Cruziana omanica Seilacher 1970 F133908 CR6 Cruziana furcifera d'Orbigny 1842 F133912 CR6 Cruziana omanica Seilacher 1970 F133006 CR7 Cruziana omanica Seilacher 1970 F133007 CR7 Cruziana omanica Seilacher 1970 F133009 CR7 Cruziana omanica Seilacher 1970 F133877 CR7 Cruziana omanica Seilacher 1970 F133879 CR7 Cruziana omanica Seilacher 1970 F133883 CR7 Cruziana omanica Seilacher 1970 F133884 CR7 Cruziana barriosi Baldwin 1977 F133885 CR7 Monomorphichnus sinus isp. nov. & Monomorphichnus indet. F133886 CR7 Cruziana omanica Seilacher 1970 F133888 CR7 Cruziana omanica Seilacher 1970 F133889 CR7 Cruziana omanica Seilacher 1970 F133890 CR7 Monomorphichnus sinus isp. nov. & Monomorphichnus indet. F133891 CR7 Cruziana omanica Seilacher 1970 F133892 CR7 Cruziana omanica Seilacher 1970 F133893 CR7 Cruziana omanica Seilacher 1970 F133895 CR7 Cruziana omanica Seilacher with pygidial

187 impression F133896 CR7 Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya 1977 F133897 CR7 Cruziana omanica Seilacher 1970 F133898 CR7 Cruziana omanica Seilacher 1970 F133899 CR7 Monomorphichnus sinus isp. nov. & Monomorphichnus indet. F133900 CR7 Cruziana omanica Seilacher 1970 F133901 CR7 Cruziana omanica Seilacher 1970 F133925 CR7 Monomorphichnus lineatus? Crimes, Legg, Marcos & Arboleya 1977 F133926 CR7 Monomorphichnus sinus isp. nov. F133927 CR7 Cruziana omanica Seilacher 1970 F133929 CR7 Cruziana omanica Seilacher 1970 F133930 CR7 smooth Rusophycus indet. F133931 CR7 Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya 1977 & Holotype and Paratype: Monomorphichnus sinus isp. nov. F133932 CR7 Cruziana omanica Seilacher 1970 (one lobe) F133933 CR7 Cruziana omanica Seilacher 1970 F133934 CR7 Trilobite impression??? F133935 CR7 Cruziana omanica Seilacher 1970 F133936 CR7 Cruziana goldfussi (Rouault, 1850) F133937 CR7 Cruziana goldfussi (Rouault, 1850) F133938 CR7 Cruziana furcifera d'Orbigny 1842 F133939 CR7 Cruziana omanica Seilacher 1970 F133940& F133928 CR7 Monomorphichnus sinus isp. nov. & Monomorphichnus indet. F133941 CR7 Monomorphichnus sinus isp. nov. F133942 CR7 Cruziana furcifera d'Orbigny 1842 F133943 CR7 Monomorphichnus multilineatus Alpert 1976 F133944 CR7 smooth Rusophycus indet. F133945 CR7 Cruziana indet. F133946 CR7 Cruziana omanica Seilacher 1970 F133012 MW Cruziana penicillata isp. nov. & Cruziana indet. & Arandaspis Ritchie & Gilbert-Tomlinson 1977 impression F133013 MW Cruziana furcifera d'Orbigny 1842 F133014 MW Paratype: Diplichnites arboreus isp. nov. & Chondrites or Planolites F133015 MW Cruziana penicillata isp. nov. F133016 MW Cruziana furcifera d'Orbigny 1842 F133017 MW Bivalves from upper unit F133019 MW Cruziana penicillata? isp. nov. F133020 MW Cruziana penicillata isp. nov. & Diplichnites

188 arboreus? isp. nov. F133021 MW Cruziana penicillata isp. nov. & Cruziana furcifera? D'Orbigny 1842 F133022 MW Paratype: Diplichnites arboreus isp. nov. & Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya 1977 F135852 MW Cruziana penicillata isp. nov. F135853 MW Paratype: Cruziana penicillata isp. nov. F135854 MW Cruziana penicillata? isp. nov. F135855 MW Paratype: Cruziana penicillata isp. nov. F135856 MW Paratype: Cruziana penicillata isp. nov. F58984 R.d. CI Holotype: Diplichnites arboreus isp. nov.

189 Alice Springs

10 0 50 \ Kilometres

Figure 5-1 A Generalised map of mainland Australia B Locality map indicating positions of Charlotte Range and Mount Watt

190 LOCALITY GPS COORDINATES ICHNOTAXON/TAXA Cruziana furcifera, C. golfussi, C. Charlotte Range = CRR 24° 47' 27.2' S 133° 48' 57.6" E penicillata Cruziana barriosi, C. indet , highly Charlotte Range = CR1 24° 44' 04 7" S 133° 56' 43.7" E bioturbated specimen Charlotte Ranqe = CR1F 24° 43'51.1" S 133° 56'31 4" E Cruziana omanica Cruziana barriosi, C. furcifera, C. Charlotte Range = CR2 24° 43' 25.1' S 133° 59' 47.4" E penicillata, C. indet., Rusophycus ^unilobus Charlotte Ranqe = CR3 24° 44' 27.5" S 133° 55' 37 0" E Cruziana furcifera, C. omanica Cruziana omanica, Monomorphichnus lineatus, highly bioturbated Charlotte Range = CR4 24° 47'27 9" S 133° 49'05.1" E Monomorpichnus spp., Rusophycus unilobus Cruziana furcifera, C. omanica, C. Charlotte Range = CR5 24° 47' 35.5" S 133° 49' 07 6" E indet Cruziana furcifera, C. omanica, Diplichnites arboreus, Charlotte Range = CR6 24° 47' 34.0" S 133° 49' 07 4" E Monomorpichnus /meatus, highly bioturbated Monomorpichnus spp Cruziana barriosi, C. furcifera, C. goldfussi, C. omanica, C indet., Charlotte Range = CR7 24° 47' 20 1" S 133° 49' 28.6" E Monomorphichnus lineatus, M. multilineatus, M. sinus, M spp. Cruziana furcifera, C penicillata, C Mount Watt = MW 25° 19' 42.6" S 133° 53' 39 8" E indet., Diplichnites arboreus, Monomorphichnus lineatus R.d. CI = A Ritchie Collection 24° 50' 52" S 133° 41' 29" E Diplichnites arboreus

Table 5—1. GPS coordinates for the localities collected, with an ichnospecies list of traces collected from these localities, including some traces not mentioned within the paper.

191 Fig. 5-2. Specimens are all convex hyporelief. A-B, Cruziana barriosi Baldwin, 1977 from the Charlotte Range. A, F133884 from CR2. B, F133880 from CR7. C-G, Cruziana furcifera d'Orbigny, 1842. C-D, F133 871 from CRR from the Charlotte Range. C, anteroventral oblique view. D, ventral view. E-F, F133938 from CR7 from the Charlotte Range. G, F135852 from Mount Watt. Scale bar = 1 cm.

192 193 Fig. 5-3. Cruziana goldfussi (Rouault, 1850). Specimens are all convex hyporelief. A-D, F133937 from CR7. A, lateral view, arrow depicts scratch marks up the trace. B, oblique lateral view and arrow pointing to lateral ridge marking. C, opposite oblique lateral view of B. D, view of ventral surface of hyporelief. E, ventral surface of F133887 from CRR. F, ventrolateral view of F133936 from CR7. All from the Charlotte Range. Scale bar = 1 cm.

194 195 Fig. 5-4. Cruziana omanica Seilacher, 1970. Specimens are all convex hyporelief. A, F133893 (CR7). B, Fl33886 (CR7). C, F133895 with fragment of possible external mould of trilobite pygidium (CR7). D, F133007(CR7). E, portion of sample of F133877(CR7). F, F133883 (CR7). All from the Charlotte Range. Scale bar = 1 cm.

196 197 Fig. 5-5. Disarticulated and fragmented trilobite sclerites. A, impression of thoracic segment (F133950) from the Charlotte Range (CR3). B, external mould of axial region and portion of pleural region of thoracic segment, pleural furrow, anterior and posterior band have similar morphology to Lycophron howchini (Etheridge, 1894), indicated by the white arrow; black on white arrow is pointing at the typical gastropod found in the region (F133949 - CR3). C, external mould of left pleural region of fragmented pygidium, similar morphology to asaphid pygidia (F133895 - CR7). Scale bar = 1 cm.

198 Fig. 5-6. A-H, Cruziana penicillata isp. nov. All convex hyporelief. A-B, paratype (F133018) from the Charlotte Range. A, lateral view. B, ventral view of convex hyporelief. C, paratype: anterior portion of trace (F133871) from the Charlotte Range. D, paratype: F135855 from Mount Watt. E, holotype: F133868 from the Charlotte Range. F, paratype: F135853 from Mount Watt. G, paratype: F135856 from Mount Watt. H, paratype: Fl33021 from Mount Watt. I, Monomorphichnus spp. F133890 (CR7) from the Charlotte Range. Scale bar = 1 cm.

199 200 Fig. 5-7. A-C, Diplichnites arboreus isp. nov. Specimens are all convex hyporelief. A, holotype indicated by white arrow (F58984) from the Charlotte Range (R.d. CI) (coin for scale is 23.6 mm diameter). B, paratype underprint, highlighted by white arrows (F133014) from Mount Watt. C, paratype (F133022) from Mount Watt. D, Monomorphichnus lineatus Crimes, Legg, Marcos & Arboyleya, 1977 (F133022) from Mount Watt. Scale bar =1 cm.

201 8. All convex hyporelief. A, Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya, 1977 (F133896) from CR7 of the Charlotte Range. B, Monomorphichnus sinus isp. nov. (white, and black-on-white arrows) and Monomorphichnus lineatus Crimes, Legg, Marcos & Arboleya, 1977 (F133931) from CR7 in the Charlotte Range. C, Same as B: Monomorphichnus lineatus depicted as white straight lines and Monomorphichnus sinus traced by curved grey lines. D, Monomorphichnus multilineatus Alpert, 1976 (Fl33943) from CR7 in the Charlotte Range. E, Monomorphichnus sinus isp. nov. (F133919) from CR4 in the Charlotte Range. F, Rusophycus unilobus (Seilacher, 1970) (F133920) from CR4 in the Charlotte Range. Scale bar = 1 cm.

202 203 CHAPTER 6: DEVONIAN OF WALES

DEVONIAN SELENICHNITES (ICHNOFOSSIL) FROM PEMBROKESHIRE (WALES, UK)1

INTRODUCTION The coastal cliffs of the United Kingdom provide exquisite outcrops for geological and palaeontological research. One such location, the cliffs at Freshwater West in southwest Pembrokeshire, Wales, contains six ichnofossils that we assign to Selenichnites rossendalensis (Hardy, 1970). The rock containing these traces that we consider herein was extracted from the Heterolithic Association of the Skrinkle Sandstones of Upper Devonian age, part of the Upper Old Red Sandstone (ORS) (Dixon, 1933; 2000b; Marshall, 2000a). The six traces of S. rossendalensis were collected in one small region of a single bedding plane. Some of these traces are associated with crescentic scours created by horseshoe vortices caused by flow over/around bluff obstacles, once referred to as Blastophycus diadematus (Miller and Dyer, 1878). From the orientation of the long axes of the traces, the morphology of these scours and tool marks on the same bedding plane, the direction of a palaeocurrent is deduced and palaeoenvironmental settings are inferred. The morphological characteristics and sizes of the traces, coupled with the period of deposition, allow one to infer that xiphosurans are plausible trace makers of these ichnofossils.

1 A version of this chapter will be submitted for review. GIBB, S., SIVETER, D.J., CHATTERTON, B.D.E. and SHAW, J.. Devonian Selenichnites (ichnofossil) from Pembrokeshire (Wales, UK). Geological Journal. 204 GEOLOGICAL SETTING AND STRATIGRAPHY The rock that contains the traces described herein was found in the inclined strata of the Skrinkle Sandstones of the Upper Devonian, within the cliff at Freshwater West, southwest Pembrokeshire, Wales (SR 8875 9875 or approximately 51° 38' 53" N and 05° 03' 19" W) (See Figure 6-1.1 - 6-1.3). The Skrinkle Sandstones unconformably overlie the Ridgeway Conglomerate (Lower-Middle Devonian) and grade into the Lower Limestone Shales of the Lower Carboniferous (Dixon, 1933; Marshall, 2000a, b) (See Figure 6-1.4). The specimen is from the Heterolithic Association, Stackpole Sandstone Member, Gupton Formation of the Skrinkle Sandstones, approximately 12 metres from the base. Marshall (2000a), referred to a 'Heterolithic Association' but Marshall (2000b) called the same beds the 'Siltstone Association'. He considered that these beds represent a lacustrine environment (Marshall, 2000a, b). The geology of western Wales was depicted by Marshall (2000a, fig. 2, 2000b, fig. 1) and figured herein (See Figure 6-1.1). Williams et al. (1982, fig. 3) provided a detailed map of the Freshwater West locality, showing the Gupton Formation flanked to the north by the Ridgeway Conglomerate Formation and to the south by the West Angle Formation, as is also illustrated by Marshall (2000a, figure 3; 2000b, figure 2). Ziegler (1990) posited that the rocks of the Pembrokeshire region were situated at approximately 5-8°S during the Upper Devonian (Marshall, 2000a); Scotese (2001) supported this approximate palaeogeographic position.

SYSTEMATIC PALAEOICHNOLOGY

The methodology and terminology used to describe and name the traces has no bearing on the type of organism that created the trace, but is based purely on the morphology of the trace itself (Bromley, 1990; Magwood, 1992; Jensen, 2003; Bertling et al, 2006).

205 The material described herein is housed in the Oxford University Museum of Natural History, catalogued as OUMNH D.1819. The images were produced by coating a replica of the bedding surface with a sublimate of ammonium chloride, using low angle lighting and taking multiple photographs at various focal depths. These were then stitched together using the programs 'Helicon Focus' and 'Adobe PhotoShop CS'.

Ichnogenus Selenichnites (Romano and Whyte, 1987)

Type species. Selenichnites hundalensis (Romano and Whyte, 1987).

Diagnosis. Refer to Romano and Whyte (1987, p. 89).

Ichnospecies. S. rossendalensis (Hardy, 1970) (Carboniferous of the U.K.); S. cordoformis (Fischer, 1978) (Ordovician of Colorado); S. bradfordensis (Chisholm, 1985) (Carboniferous of the U.K.); S. hundalensis (Romano and Whyte, 1987) (Jurassic of the U.K.); S. langridgei Trewin and McNamara, 1995 (Silurian-Devonian of Western Australia); S. scagliai (Poire and Del Valle, 1996) (Cambrian/Ordovician of Argentina); S. antarcticus Weber and Braddy, 2004 (?Lower Ordovician of Antarctica); S. isp. Wang, 1993 (Upper Triassic of the U.K.); S. isp. Draganits et al., 2001 (Lower Devonian of Northern India); 5*. isp. Morrissey and Braddy, 2004 (Early Devonian of southwest Wales); and S. isp. Lucas and Lerner, 2005 (Lower Pennsylvanian of Alabama).

Selenichnites rossendalensis (Hardy, 1970) (Figures 6-2.1 - 6-2.4)

Material. The six ichnofossils are situated upon one rock slab that has been reassembled upon removal from the strata. They are preserved on the lower bedding surface of the bed (OUMNH D.1819).

206 Diagnosis. See Hardy (1970, p. 189).

Description. See Hardy (1970, p. 189), though ichnofossils are isolated traces and not trails of "irregularly meandering series of lunate casts" (Hardy, 1970, p. 189).

Remarks. The dimensions for six traces are within the size range provided by Hardy (1970) (See Figure 6-3). All six traces are preserved in convex hyporelief, occurring at various depths within the substrate. They may not have all been formed at exactly the same time but the similar orientations of the traces suggest that they were formed in a similar current regime. The traces are in various conditions of preservation due to the depth to which the organisms were able to burrow, erosional activities upon evacuation from the burrow and conditions while the organism was in the burrow. The specimen labelled V (Figure 6-2.1 - 6—2.2) is the best preserved trace of Selenichnites rossendalensis, with evident lateral ridges demonstrating a convex shape of the cephalon of the animal that made it, but showing limited detail of the remainder of the organism. Specimen 'f (Figure 6-2.1 and 6-2.4) also demonstrates that the cephalon was convex at the anterior, and shows that the organism had a telson, based on a drag mark present. Specimen 'c' (Figure 6-2.1 and 6-2.3) displays the same cephalon shape, together with posterior push marks created by appendages or legs digging into the substrate. Specimen 'd' (Figure 6-2.1) also has a semi-lunate anterior cephalon shape, though the preservation and morphological features are not as clear as in the aforementioned traces. Specimens 'b' and V (Figure 6-2.1) are poorer representations of this trace, but are identified mainly by their similarity to and close association with the other four traces. Unfortunately, the traces do not depict a series track. Therefore, it is not possible to determine whether one organism or more than one (whether at the same time or at different times) created the traces.

Discussion. Some ichnospecies have been reassigned to Selenichnites: S. bradfordensis (Chisholm, 1985) and S. cordiformis (Fischer, 1978). The

207 assignment of 5*. bradfordensis is controversial. This species was originally assigned to Auchlichnites?. Romano and Whyte (1987) stated that this species should not be assigned to Selenichnites due to the fact that they thought the ichnofossil was "a bedding plane representation through an oblique burrow" (p. 91). However, Draganits et al. (2001) then placed A? bradfordensis within Selenichnites, noting that only the traces that were found upon the bedding surface are S. bradfordensis. Chisholm (1985) illustrated a number of shallow surface bedding plane traces (pi. 75.3, 75.5 and 75.10), which we also refer to S. bradfordensis. Chisholm (1985) made reference to Selenichnites rossendalensis (then referred to as Limulicubichnus (Kouphichnium) rossendalensis by Eagar et al. (1985) and previously by Hardy (1970) as Kouphichnium rossendalensis), though he insisted that they were "extensive burrow systems" (Hardy, 1970, p. 626) into the substrate on the bedding surface, yet the specimens show no indication that they are anything more than a surface trace. The morphological details of the surface traces are relatively nondescript, aside from being isolated traces, with an overall oval shape, possible lateral ridges and posterior appendage impressions. Selenichnites hundalensis (Romano and Whyte, 1987) is an ichnospecies that displays morphological characteristics similar to those of the traces found at Freshwater West. The feature that most unites the two is the oval outline of the trace. The differences are that the traces do not consist "of three main parts, A, B and C" illustrated in figure 3.1 of Romano and Whyte (1987). Romano and Whyte (1987) highlighted the triangular shape of part A in their trace, which cannot be identified in any of the traces that we illustrate. One might recognize two paired crescentic lobes in specimens a, d, and/(Figure 6-2.1 - 6-2.2, 6-2.4), but they are not of the same morphology as the lobes described by Romano and Whyte (1987) for S. hundalensis. Romano and Whyte (1987) pointed out another difference between S. hundalensis and S. rossendalensis: the former, from the Jurassic of Yorkshire, is larger than specimens of the latter from the Carboniferous of Lancashire, northern England (Hardy, 1970). Size should not be a discriminating factor for ichnospecies and/or ichnogenera, but it should be used

208 as a purely a morphological observation of characters. Different species and/or the same species at different growth stages could create the same ichnospecies. One other distinguishing morphological characteristic of S. rossendalensis, figured herein (Figure 6-2.1 - 6-2.3), is the presence of distinct lateral ridges. Although these are not clear in all of the traces (in particular b and e), due to lack of preservation or the trace being a possible undertrack. The lateral ridges present in some of the traces are important as they help identify these traces. Such ridges are absent in S. hundalensis. The similarity of all the traces on the sample that we have, and the taphonomic variation between the traces strongly suggests they should all be assigned to the same species. Trewin and McNamara (1995) erected Selenichnites langridgei from the Tumblagooda Sandstone of the ?Late Silurian of Western Australia. These, like the Welsh traces, have an anterior crescentic continuous margin with the greatest depth at the anterior margin, but the Welsh traces lack the distinct median trefoil that helps to define S. langridgei. The ichnospecies Selenichnites scagliai (Poire and Del Valle, 1996) displays extraordinarily simple morphology: lacking defined lateral ridges and or any ventral morphology from the creator. Selenichnites cordiformis (Fischer, 1978) is a relatively nondescript lunate trace, with a continuous anterior margin. Fischer (1978) stated that the trace is heart-shaped, and this along with the limited morphological features of the ichnospecies, eliminates it from being conspecific with the specimens from Freshwater West. Selenichnites antarcticus Weber and Braddy, 2004 is a trace from Antarctica of uncertain geological age, though it is thought to be Early Ordovician. Selenichnites antarcticus has a continuous anterior margin, which is deeper on the anterior than the posterolateral edges of the horseshoe-shaped trace. A number of traces have been assigned as Selenichnites, though not into a specific ichnospecies, due to lack of distinguishing morphological characteristics. Morrissey and Braddy (2004, p. 330 and fig. 6.g) figured a trace that they stated probably depicted "the outline of a small arthropod resting shallowly in the

209 sediment or using its carapace to forage for food" (Morrissey and Braddy, 2004, p. 330). Once again, their trace from the Lower Old Red Sandstone of southwest Wales is horseshoe-shaped with a deeper anterior margin to the lateral flanks, though no ichnospecies was assigned. Martin and Rindsberg (2007, fig. 29.8) documented this behaviour in modern juvenile horseshoe crabs, which is further discussed in the section of 'possible trace-makers'. Wang (1993), Draganits et al. (2001), and Lucas and Lerner (2005) all described specimens of Selenichnites, which were not assigned to an ichnospecies. As with Morrissey and Braddy's (2004) specimens, they are simple traces lacking complex morphological detail, all showing a continuous anterior margin that created a lunate ichnofossil.

POSSIBLE TRACE-MAKERS In order to produce Selenichnites rossendalensis, the trace maker had to possess the following morphological features : a) Rounded anterior "cephalon". b) Lateral spines. c) A median spinose posterior extension, similar to, if not, a telson. d) At least one pair of "legs" or appendages capable of moving sediment and perhaps pushing the animal forward. e) Muscular apparatus to drive the anterior portion of the body into the substrate so as to not be carried off by the current. It has been suggested that the traces were formed in a lacustrine setting (Marshall, 2000a, b), therefore the list of possible Devonian trace makers is significantly reduced. Selenichnites rossendalensis has a history of being associated with xiphosurans. Hardy (1970) described Selenichnites rossendalensis as a non-marine ichnospecies and proposed two genera of horseshoe crabs (xiphosura) that could have created this trace —Bellinurus and Euproops (though he preferred Bellinurus since they had been found with freshwater bivalves). Eagar et al. (1985) also stated that S. rossendalensis could have been produced by Bellinurus or Euproops in a shoreline environment. We

210 do not believe the palaeoenvironment to be a shallow water setting due to the lithology and sedimentary structures, but we agree with the lacustrine interpretation. Other organisms from the Devonian are considered and ruled out as possible trace-makers. The first organism is the trilobite: trilobites have never been found in a non-marine environment. Another argument for discounting trilobites is that most, if not all, researchers would attribute a bilobed 'resting' trace such as Rusophycus to trilobites. Also, very few trilobites and none from the Upper Devonian, have long telson-like spine. Branchiopods exist in a non-marine setting, however, their traces are usually, if not always, bilobed (Gand et al, 2008) and they are seldom large enough to have created the traces from Wales. Eurypterids inhabited freshwater environments, though the lack of 'paddle' markings (sixth prosomal appendages) and the sheer size of even juvenile eurypterids would remove them as possible trace-makers. A number of other behavioural criteria have been posited for xiphosurans during the Palaeozoic. Waterston (1985) proposed that xiphosurans abandoned enrollment as a defensive behaviour in favour of burial/burrowing. Eldredge (1970) had already outlined the four stages of burrowing in Limulus polyphemus: in the first stage the prosoma is flexed and driven into the substrate, and any sediment that slips off the prosoma and blocks the channel between the prosoma and opisthosoma is cleared by the sixth walking leg; the second stage involves no movement of the cephalon and the gills are aerated through the open channel; the third stage is further burial and clearing of the channel; and in the final stage the horseshoe crab will remain almost totally inactive for up to 12 hours. Waterston (1985) discussed the need for efficient respiration while the organism is buried, hence the clearing of the channel by the specialized sixth walking leg as observed by Eldredge (1970). Fisher (1981) and Siveter and Selden (1987) have also noted the need for a fresh, clean water supply to aerate the gills while buried within the substrate. On closer examination of xiphosuran genera and their morphology, we can attempt to arrive at a possible creator or at least resolve that the trace maker was

211 in all likelihood a xiphosuran. The oldest known xiphosuran body fossil, Lunataspis aurora Rudkin, Young and Nowlan, 2008, is from Late Ordovician shallow marine deposits of Manitoba, Canada (Rudkin et al., 2008). Therefore it is clear that xiphosurans inhabited a shallow marine setting at that time. Another common hypothesis is an ichnofossil may be the first indication of an organism inhabiting an environment, well before a body fossil is identified, therefore suggesting a lower ghost range for a genus and/or species. This hypothesis was used by Alpert (1977, p. 3) with regard to noting the appearance of trilobites based upon noted trilobite-like traces to define the Precambrian-Cambrian boundary: "Rather than placing the basal Cambrian boundary at or just below the lowest trilobites, it is suggested that the boundary be lowered to the lowest trilobite trace fossils. Trilobite trace fossils commonly occur below the lowest trilobites in the Precambrian-Cambrian successions, worldwide, and should be considered to be Cambrian in age, as they indicate the existence of trilobites, or the arthropod grade of organisation". Crimes et al. (1977, p. 91) also noted that "most of the trace fossils come from below the lowest body fossils". Therefore, the premise of ichnofossil before body fossil is one that will be used to demonstrate that there are a number of possible freshwater xiphosuran candidates during the Devonian, with two being particularly plausible: Bellinurus and Euproops. Euproops morani Eller, 1938a was found in a marine setting, not lacustrine as represented by the Freshwater West locality, and it is from Upper Devonian strata of Pennsylvania, the 'Salamanca sandstone' (Eller, 1938a). In 1938, they were the oldest recorded "Euproops-like limuloids", but only the opisthosoma was recovered (Eller, 1938a). Eller (1938b) also reexamined Williams' (1885) Protolimulus eriensis from Upper Devonian strata of the Chemung Group of Pennsylvania. Aside from the noted poor preservation, and the fact that it is a marine organism, it shares the characteristic of a prosoma that has prolonged lateral spines and the presence of a telson. The overall shape of P. eriensis is similar to those of the Welsh traces, but the palaeoenvironment is not equivalent.

212 The first record of freshwater xiphosurans is from the Lower Carboniferous of the maritime provinces of Canada (Copeland, 1957). Copeland (1957) described Palaeocaris novascoticus and Euproops thompsoni (E. thompsoni could possibly be from a marine environment, though the Windsor group specimens are from an unknown environment). Palaeocaris Meek and Worthen, 1865, can be eliminated from the possible trace-makers, as its morphology does not correspond with the impressions of Selenichnites rossendalensis. The morphology of Palaeocaris is that of an elongate body, the prosoma does have a 'horseshoe' shape, though its ratio is approximately 1.5 times as long as it is wide, and it lacks any recognizable lateral spines, thus this animal would have been incapable of creating S. rossendalensis. Stonner (1952, fig. 1. h) and Selden and Siveter (1987a, fig 1. e) illustrated Bellinuroopsis Chernyshev, 1933, (following the spelling of Selden and Siveter (1987b)) from a Devonian marine environment of Russia. Once again this taxon was not collected from a lacustrine setting, and its morphology is not similar enough to have created the traces from Freshwater West. The anterior shape of the prosoma is similar in shape, as not being evenly rounded, but it is inferred to have had only small lateral spines or posteriorly pointed prosomal angles, not the required elongate spines suggested by the lateral edges of the trace Selenichnites rossendalensis. Rolfeia Waterston, 1985, was inferred by Selden and Siveter (1987a, fig. 2) to have a possible origin in the Devonian. A Devonian representative has yet to be found. Rolfeia, from the Lower Carboniferous, has a rounded prosoma, but lacks elongate lateral spines (Waterston, 1985). Other xiphosuran genera known from the Devonian, though lacking requisite elongate lateral prosomal spines include: Legrandella Eldredge, 1974, Weinbergina Richter and Richter, 1929, and Kasibelinurus Pickett, 1993. Legrandella is a particularly unlikely candidate as it has an anterior projection from the prosoma. A Lower Devonian genus, Willwerathia Stormer, 1969, is another possible candidate, though the morphology of the Devonian W. laticeps (Stormer, 1936)

213 does not correlate with the morphology of our traces. Morphological traits that W. laticeps possesses that do not match Selenichnites rossendalensis from the Freshwater West locality are: an anteriorly indented prosoma, evenly rounded prosoma, and short prosomal spines (Anderson et al, 1998). Bellinurus Koenig, 1820, and Euproops Meek, 1867 have been cited by both Hardy (1970) and Eager et al. (1985) as a possible trace-maker for Selenichnites rossendalensis. Both genera are roughly the same size as the traces, though they are found in younger Carboniferous strata (Copeland, 1957; Hardy, 1970; Ambrose and Romano, 1972; Fisher, 1977, 1979; Eagar et al, 1985; Selden and Siveter, 1987a; Anderson, 1994; Anderson and Selden, 1997). It should be noted that Bellinurus has been documented from the Upper Devonian by Selden and Siveter (1987a, fig. 1. f). Although slight morphological character states differentiate the two genera, it will probably never be determined which of the two, Bellinurus or Euproops, or some yet to be discovered xiphosuran made the traces. Euproops is sketched (See Figure 6-2.5) with the prosoma driven into the substrate, possibly after stage one in the burrowing process of Eldredge (1970). The most obvious feature of a trace maker that can be preserved in a convex hyporelief ichnofossil is the shape of the prosoma. The traces suggest the organism had an elongate anterior margin and not an exact symmetrical crescentric shape, suggesting that the trace-maker was shaped more like Bellinurus than Euproops. There are no other morphological identifying characters from the ichnofossils that allows one to differentiate between these two genera. Both genera have lateral spines, a telson, at least one pair of legs and the muscular apparatus for burrowing into the substrate. Thus, until a body fossil is found within the strata at Freshwater West of the appropriate size, morphological shape and the capacity to burrow, the trace-maker is hypothetical.

214 ASSOCIATED ICHNOFAUNA It is important to note that there are no other associated ichnofossils nor body fossils found with the six traces. This low diversity of traces may be cited as negative evidence in support of a non-marine environment.

PALAEOECOLOGY The outstanding preservation of the six biogenic sedimentary structures and the physical sedimentary structures in the form of crescentic scours generated by horseshoe vortices and tool marks, assist in establishing the alignment of the organisms that produced the trace fossils. Miller and Dyer (1878, p. 24, pi.1.1) assigned an ichnospecies to the morphology of the sedimentary scours, Blastophycus diadematus. Osgood (1970, p. 390), following the tenninology of Pettijohn and Potter (1964, pis. 9IB, 92A, 93A, 93B), stated that the markings are 'current crescent scars'. Osgood (1970, p. 390, pi. 67 fig. 2, pi. 81 fig. 1,3,5,10) demonstrated through his own flume experimentation that an enrolled trilobite produced two scour channels on the leeward side. Further to this research, Kirkel et al. (2005; 2008) illustrated the flow structure of horseshoe vortices from both modelling and observation. Figure 6-5.1 illustrates an instantaneous depiction of flow threads. The geometry of a vortex wrapping around the obstacle is quite clear as are the two vortices (legs) extending beyond the lee side of the obstacle. Figure 6-5.2 represents the time averaged distribution of shear velocity normalized with the free flow velocity (UT/U), a measure of bed shear stress which indicates areas of probable erosion, dark areas in the figure, with high values of normalized shear velocity. The dark areas are narrow around the upstream face of the obstacle and broaden into broader, more elongate zones to the lee of the obstacle. This time averaged depiction gives a clear picture of the erosional zone in areas of high normalized shear velocity. Siveter and Selden (1987), used flume tank research to establish that unidirectional currents create crescentic scour marks on the anterolateral side of the prosoma of a species of horseshoe crab, Xaniopyramis linseyi. They also

215 illustrated (Siveter and Selden, 1987, fig. 8) evidence of horseshoe vortices on the contralateroposterior of their specimen. Weber and Braddy (2004) were also able to demonstrate that Selenichnites antarcticus were oriented with their anterior facing directly into the current. Current lineations associated with the traces, mark prominent current events. The difference in the lineations associated with the traces illustrated by Weber and Braddy (2004, fig. 22a, e), and those described herein is that the traces they observed show anterior margins facing into the current, while the traces figured herein are either posteriorly oblique, or with complete posterior alignment (See Figure 6—4 and inset: a, d, e, and/have oblique alignment, while b and c are relatively close to a posterior palaeocurrent alignment). The orientation of the apex of Selenichnites isp. was illustrated and described by Draganits et al. (2001). Koupichichnium of Wright and Benton (1987) was shown to be travelling over ripple marks, therefore either with and/or against the tidal current. The six specimens found at Freshwater West demonstrate that through the observation of the most pronounced biogenic sedimentary structure (Figure 6—1.1, 6-1.2, 6-5) one can infer the palaeocurrent direction and the proposed palaeoenvironmental setting of the traces. Assuming that the traces were made by horseshoe crabs, the description by Eldredge (1970) of the burrowing technique of Limulus polyphemus becomes relevant. A quick summary would be that the horseshoe crab drives the prosoma into the muddy substrate by downward flexion, drives the substrate along the leading edge of the prosoma and then flexes the prosoma back to horizontal. This position can be observed in the study of neoichnology of a juvenile of L. polyphemus by Martin and Rindsberg (2007, fig. 29.8) and in the images for the proposed ichnospecies Limulicubichnus serratus with the trace-producer L. polyphemus (Miller, 1982, text-fig. 3). Eldredge (1970) further discussed the subsequent actions of burial of L. polyphemus, but the key point with regard to the Devonian Welsh traces is his statement that the channel between the prosoma and opisthosoma is kept clear by the animal (see above), allowing an inhalant water current to aerate the gills. The alignment of the traces examined in this paper is such that the current would enter the left

216 channel for aeration of the gills while the remainder of the current forms vortices around the organism and erodes in a pattern that is lateroposterior to the obstacle = organism. One can hypothesize that this pattern was caused by a turbidity flow (Figure 6-2.1 and 6—4). This is supported by the presence of sole markings and the horseshoe vortices on the surface of the bed. Upon cessation of the turbidity flow, the upstream scour mark is eradicated as the organism removed itself from the 'burrow'. Any subsequent sand grains from the flow that were covering the organism would have infilled the depressions, providing exceptional preservation. Buatois and Mangano (2007) discussed this type of preservation as "...particularly favored in low-energy areas of lacustrine systems. For example, alternation of very fine sand and mud deposited from underflow or turbidity currents is conducive to preservation of tiny surface to very shallow traces" (p. 299). The mention of lacustrine environment by Buatois and Mangano (2007) raises the question of: 'what was the palaeoenvironmental setting at Freshwater West when these traces were formed?' This interval has been interpreted by a number of different authors as a lacustrine depositional setting (Williams et al, 1982; Marshall, 2000b, a; Barclay and Williams, 2005; Hillier and Williams, 2006). One might conclude that the traces belong in the Scoyenia ichnofacies given that Buatois and Mangano (1995, p. 154; 2007, p. 286), citing Seilacher (1963; 1967), stated that the ichnofacies are typically "nonmarine sands and shales, often red beds". The Scoyenia ichnofacies, however, forms in a very shallow setting, and usually exhibits desiccation cracks. Because no desiccation cracks were found and the sedimentary structures present suggest a deeper water environment, the more likely ichnofacies is Mermia. The physical characteristics of the Mermia ichnofacies (Buatois and Mangano, 1995, 1998, 2004, 2007) are such that it may be represented in the Heterolithic Association of the Gupton Formation in the Skrinkle Sandstones. Buatois and Mangano (1995, p. 155) listed the following conditions for the Mermia ichnofacies: "Noncohesive, fine-grained sediment at well- oxygenated, low energy, permanently subaqueous zones; sedimentation rate is normally low, but punctuated episodic deposition (e.g., turbidity currents, density

217 underflows) may occur. Physical sedimentary structures may include parallel lamination and stratification, ripple cross-lamination, normal grading, tool and flute marks, as well as soft-sediment deformation structures". They described a palaeoenvironment that is at "the distal zones of sand lobes formed within a sublacustrine fan complex dominated by turbidity and underflow density currents" (Buatois and Mangano, 1995, p. 152). These are features that have been discussed throughout this paper, and hence suggest a Mermia ichnofacies for the beds that contain the Welsh traces. Thus, relatively deep-water lacustrine beds probably provided the palaeoenvironmental setting for the production of Selenichnites rossendalensis.

CONCLUSIONS The biogenic sedimentary structures present at the Freshwater West locality in the 'Heterolithic Association' of the Gupton Formation of the Skrinkle Sandstones are Selenichnites rossendalensis. From the morphological characteristics observed and the age of the traces, the likely trace-maker was a xiphosuran. The assignment of this trace maker is based on the morphological features of the trace and neoichnology. Due to the lack of body fossils, a putative trace-maker could not be determined with confidence. The sedimentology and physical sedimentary structures place the bed within a deep-water lacustrine setting, possibly that of the Mermia ichnofacies.

218 REFERENCES

Alpert, S.P. 1977. Trace fossils and the basal Cambrian boundary. In: Trace fossils 2: Geological Journal, Special issue 9, Crimes, T.P., Harper, J.C. (eds.). 1-8. Ambrose, T., Romano, M. 1972. New Upper Carboniferous Chelicerata (Arthropoda) from Somerset, England. Palaeontology 15: 569-578. Anderson, L.I. 1994. Xiphosurans from the Westphalian D of the Radstock Basin, Somerset Coalfield, the South Wales Coalfield and Mazon Creek, Illinois. Proceedings of the Geologists' Association 105: 265-275. Anderson, L.L, Poschmann, M., Brauckmann, C. 1998. On the Emsian (Lower Devonian) arthropods of the Rhenish Slate Mountains: 2. The synziphosurine Willwerathia. Paldontologische Zeitschrift 72: 325-336. Anderson, L.L, Selden, P.A. 1997. Opistosomal fusion and phylogeny of Palaeozoic Xiphosura. Lethaia 30: 19-31. Barclay, W.J., Williams, B.P.J. 2005. Freshwater West, Pembrokeshire (SR 884 996 -- SR 887 988). In: The Old Red Sandstone of Great Britain, Barclay, W.J., Browne, M.A.E., Mcmillan, A.A., Pickett, E.A., Stone, P., Wilby, P.R. (eds.). Peterborough: Geological Conservation Review Series, No. 31, Joint Nature Conservation Committee; 291-301. Bertling, M., Braddy, S.J., Bromley, R.G., Demathieu, G.R., Genise, J., Mikulas, R., Nielsen, J.K., Nielsen, K.S.S., Rindsberg, A.K., Schlirf, M., Uchman, A. 2006. Names for trace fossils: a uniform approach. Lethaia 39: 265-286. Bromley, R.G. 1990. Trace fossils: Biology and taphonomy. London: Unwin Hyman. Buatois, L.A., Mangano, M.G. 1995. The paleoenvironmental and paleoecological significance of the lacustrine Mermia ichnofacies: An archetypical subaqueous nonmarine trace fossil assemblage. Ichnos 4: 151-161. Buatois, L.A., Mangano, M.G. 1998. Trace fossil analysis of lacustrine facies

219 and basins. Palaeogeography, Palaeoclimatology, Palaeoecology 140: 367-382. Buatois, L.A., Mangano, M.G. 2004. Animal-substrate interactions in freshwater environments: applications of ichnology in facies and sequence stratigraphic analysis of fluvio-lacustrine successions. In: The application of ichnology to palaeoenvironmental and stratigraphic analysis, Mcilroy, D. (ed.). Geological Society, London, Special Publications, 228; 311-333. Buatois, L.A., Mangano, M.G. 2007. Invertebrate ichnology of continental freshwater environments. In: Trace fossils: concepts, problems, prospects, Miller, W. (ed.). Amsterdam: Elsevier; 285-323. Chernyshev, B.I. 1933. Arthropoda c Ypana H flpyrux Mecr C.C.C.P. ITeHHmpcuibHozo HayHHO-HccnedoeacmenbCKOzo Feonozopa3 eedoHHOZo HHcmumymam naneoumoRozuH u Cmpamuapaqbun C6opuuKb 1: 15-24. Chisholm, J.I. 1985. Xiphosurid burrows from the Lower Coal Measures (Westphalian A) of West Yorkshire. Palaeontology 28: 619-628. Copeland, M.J. 1957. The Carboniferous genera Palaeocaris and Euproops in the Canadian Maritime provinces. Journal of Paleontology 31: 595-599. Crimes, T.P., Legg, 1., Marcos, A., Arboleya, M. 1977. ?Late Precambrian-low Lower Cambrian trace fossils from Spain. In: Trace fossils 2: Geological Journal, Special issue 9, Crimes, T.P., Harper, J.C. (eds.). 91-138. Dixon, E.E.L. 1933. Notes on the geological succession in South Pembrokeshire. Proceedings of the Geologists' Association 44: 402-409. Draganits, E., Braddy, S.J., Briggs, D.E.G. 2001. A Gondwanan coastal arthropod ichnofauna from the Muth Formation (Lower Devonian, Northern India): Paleoenvironment and tracemaker behavior. Palaios 16: 126-147. Eagar, R.M.C., Baines, J.G., Collinson, J.D., Hardy, P.G., Okolo, S.A., Pollard, J.E. 1985. Trace fossil assemblages and their occurrence in Silesian (Mid-Carboniferous) deltaic sediments of the central Pennine Basin, England. In: Biogenic structures: Their use in interpreting depositional environments, Curran, H.A. (ed.). Tulsa: Society of

220 Economic Paleontologists and Mineralogists, Special Publication No. 35; 99-149. Eldredge, N. 1970. Observations on burrowing behaviour in Limulus polyphemus (Chelicerata, Merostomata), with implications on the functional anatomy of trilobites. American Museum Novitates 2436: 1-17. Eldredge, N. 1974. Revision of the suborder Synziphosurina (Chelicerata, Merostomata), with remarks on merostome phylogeny. American Museum Novitates 2543: 1-41. Eller, E.R. 1938a. A new xiphosuran, Euproops morani, from the Upper Devonian of Pennsylvania. Annals of the Carnegie Museum 27: 151-153. Eller, E.R. 1938b. Note on the xiphosuran, Protolimulus eriensis Williams. Annals of the Carnegie Museum 27: 155-159. Fischer, W.A. 1978. The habitat of the early vertebrates: trace and body fossil evidence from the Harding Formation (Middle Ordovician), Colorado. The Mountain Geologist 15: 1-26. Fisher, D.C. 1977. Functional significance of spines in the Pennsylvanian horseshoe crab Euproops danae. Paleobiology 3: 175-195. Fisher, D.C. 1979. Evidence for subaerial activity of Euproops danae (Merostomata, Xiphosurida). In: Mazon Creek Fossils, Nitecki, M.H. (ed.). New York: Academic Press; 379-447. Fisher, D.C. 1981. The role of functional analysis in phylogenetic inference: examples from the history of the Xiphosura. American Zoologist 21: 47- 62. Gand, G., Garric, J., Schieber, J., Walter, H., Lapeyrie, J., Martin, C, Thiery, A. 2008. Notostraca trackways in Permian playa environments of the Lodeve basin (France). Journal of Iberian Geology 34: 73-108. Hardy, P.G. 1970. New xiphosurid trails from the Upper Carboniferous of Northern England. Palaeontology 13: 188-190. Hillier, R.D., Williams, B.P.J. 2006. The alluvial Old Red Sandstone: fluvial basins. In: The Geology of England and Wales, Brenchley, P. J., Rawson, P.F. (eds.). London: The Geological Society; 155-171.

221 Jensen, S. 2003. The Proterozoic and earliest Cambrian trace fossil record; patterns, problems and perspectives. Integrative and Comparative Biology 43:219-228. Kirkil, G., Constantinescu, G., Ettema, R. 2008. Coherent structures in the flow field around a circular cylinder with scour hole. Journal of Hydraulic Engineering 134: 572-587. Koenig, C. 1820. Icones Fossilium Sectites. Lucus, S.G., Lerner, A.J. 2005. Lower Pennsylvanian invertebrate ichnofossils from the Union Chapel Mine, Alabama: A preliminary assessment. In: Pennsylvanian footprints in the Black Warrior Basin of Alabama. Alabama Paleontological Society Monograph no. 1, Buta, R.J., Rindsberg, A.K., Kopaska-Merkel, D.C. (eds.). 147-152. Magwood, J.P.A. 1992. Ichnotaxonomy: a burrow by any other name...? In: Trace fossils. Short courses in paleontology. No. 5, Maples, C.G., West, R.R. (eds.). Knoxville: University of Tennessee; 15-33. Marshall, J.D. 2000a. Fault bounded basin fill: Fluvial response to tectonic controls in the Skrinkle Sandstones of SW Pembrokeshire, Wales. In: New perspectives on the Old Red Sandstone, Friend, P.F., Williams, B.P.J. (eds.). Geological Society, Special Publications; 401-416. Marshall, J.D. 2000b. Sedimentology of a Devonian fault-bounded braidplain and lacustrine fill in the lower part of the Skrinkle Sandstones, Dyfed, Wales. Sedimentology 47: 325-342. Martin, A.J., Rindsberg, A.K. 2007. Arthropod tracemakers of Nereites? Neoichnological observations of juvenile limulids and their paleoichnological applications. In: Trace fossils: concepts, problems, prospects, Miller, W. (ed.). Amsterdam: Elsevier; 478-491. Meek, F.B. 1867. Notes on a new genus of fossil Crustacea. Geological Magazine 4: 320-321. Meek, F.B., Worthen, A.H. 1865. Notice of some new types of organic remains from the Coal Measures of Illinois. Proceedings of the Academy of Natural Sciences of Philadelphia 17: 41-45.

222 Miller, M.F. 1982. Limulicubichnus: a new ichnogenus of limulid resting traces. Journal of Paleontology 56: 429-433. Miller, S.A., Dyer, C.B. 1878. Contributions to paleontology, no. 1. The Journal of the Cincinnati Society of Natural History 1: 24-39. Morrissey, L.B., Braddy, S.J. 2004. Terrestrial trace fossils from the Lower Old Red Sandstone, southwest Wales. Geological Journal'39: 315-336. Osgood, R.G.J. 1970. Trace fossils of the Cincinnati area. Palaeontographica Americana 6: 281-439. Pettijohn, F.J., Potter, P.E. 1964. Atlas and glossary of primary sedimentary structures. New York: Springer-Verlag. Pickett, J.W. 1993. A Late Devonian xiphosuran from near Parkes, New South Wales. Memoirs of the Association of Australian Palaeontologists 15: 279-287. Poire, D.G., Del Valle, A. 1996. Trazas fosiles en barras submareales de la Formation Balcarce (Cambrico/Ordovicico), cabo corrientes, mar del Plata, Argentina. Asociacion Paleontologica Argentina, Publicacion Especial 4 89-102. Richter, R., Richter, E. 1929. Weinbergina opitzi, n.g., n.sp., ein Schwerttrager (Merost., Xiphos.) aus dem Devon (Rheinland). Senckenbergiana 11: 193- 209. Romano, M., Whyte, M. 1987. A limulid trace fossil from the Scarborough Formation (Jurassic) of Yorkshire; its occurrence, taxonomy and interpretation. Proceedings of the Yorkshire Geological Society 46: 85-95. Romano, M., Whyte, M. 1990. Selenichnites, a new name for the ichnogenus Selenichnus Romano & Whyte, 1987. Proceedings of the Yorkshire Geological Society 48: 221. Rudkin, D.M., Young, G.A., Nowlan, G.S. 2008. The oldest horseshoe crab: A new xiphosurid from Late Ordovician konservat-lagerstatten deposits, Manitoba, Canada. Palaeontology 51: 1-9. Scotese, C.R. 2001. Paleogeographie Atlas, PALEOMAP Progress Report 90-

223 0497. Arlington, Texas: Department of Geology, University of Texas at Arlington. Seilacher, A. 1963. Lebensspuren und Salinitatsfazies. Fortschritte in der Geologie Rheinland und Westfalens 10: 81-94. Seilacher, A. 1967. Bathymetry of trace fossils. Marine Geology 5: 413-428. Selden, P.A., Siveter, D.J. 1987a. The origin of the limuloids. Lethaia 20: 383- 392. Selden, P.A., Siveter, D.J. 1987b. The status of Bellinuroopsis Chernyshev, 1933, and Neobelinuropsis Eller, 1938 (Xiphosura, Bellinuroidea). Journal of Paleontology 61: 1285. Siveter, D.J., Selden, P.A. 1987. A new, giant xiphosurid from the lower Namurian of Weardale, County Durham. Proceedings of the Yorkshire Geological Society 46: 153-168. Stormer, L. 1936. Eurypteriden aus dem Rheinischen Unterdevon. Abhandlungen der Preufiischen GeologischenLandesansta.lt, Neue Folge 175: 1-74. Stormer, L. 1952. Phylogeny and taxonomy of fossil horseshoe crabs. Journal of Paleontology 26: 630-639. Stormer, L. 1969. Eurypterids from the Lower Devonian of Willwerath, Eifel. Senckenbergiana Lethaea 50: 21-35. Trewin, N.H., Mcnamara, K.J. 1995. Arthropods invade the land: trace fossils and palaeoenvironments of the Tumblagooda Sandstone (?late Silurian) of Kalbarri, Western Australia. Transactions of the Royal Society of Edinburgh: Earth Sciences 85: 177-210. Wang, G. 1993. Xiphosurid trace fossils from the Westbury Formation (Rhaetian) of southwestern Britain. Palaeontology 36: 111-122. Waterston, CD. 1985. Chelicerata from the Dinantian of Foulden, Berwickshire, Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences 76: 25-33. Weber, B., Braddy, S.J. 2004. A marginal marine ichnofauna from the Blaiklock Glacier Group (?Lower Ordovician) of the Shackleton Range, Antarctica. Transactions of the Royal Society of Edinburgh: Earth Sciences 94: 1-20.

224 Williams, B.P.J., Allen, R.J.L., Marshall, J.D. 1982. Old Red Sandstone facies of the Pembroke Peninsula, south of the Ritec Fault. In: Geological excursions in Dyfed, south-west Wales, Bassett, M.G. (ed.). Cardiff: National Museum of Wales; 151-174. Williams, H.S. 1885. Notice of a new limuloid crustacean from the Devonian. American Journal of Science 30: 45-49. Wright, A.D., Benton, M.J. 1987. Trace fossils from Rhaetic shore-face deposits of Staffordshire. Palaeontology 30: 407-428. Ziegler, P.A. 1990. Geological Atlas of Western and Central Europe. The Hague: Shell Internationale Petroleum Maattschappij.

225 Figure 6-1. (1) Geological map of the region (modified from Marshall, 2000a, b); (2) 'Freshwater West' locality in relation to the west of Wales; (3) Section image of the 'Freshwater West' locality, with geologist of approximate height of 170 centimetres, with hand on location in which rock was removed with the five traces of Selenichnites rossendalensis. The division of the Stackpole Sandstone Member, within the Gupton Formation, into the older 'Heterolithic Association' and younger 'Quartzarenite Association' (Marshall, 2000b); (4) Stratigraphic columns of the Freshwater West locality (modified from Marshall, 2000a, b).

226 \ Quartzarenite Milford Haven sociation \ Assoclatlon

LOWER SANDSTONE MEMBER

Trough Cross-Bedding Bedded sandstone Limestone

. Conglomerate Ripple-bedded sandstone Mudstone/Siltstone

227 Figure 6-2. (1) Image of the complete slab replica with the 6 traces of Selenichnites rossendalensis; (2) Trace 'a' from corresponding image. Note scour marks (small arrows); (3) Trace 'c'; (4) Trace '/, with all scale bars - 1 centimetre; (5) Euproops depicted in stage one (Eldredge, 1970) of burrowing into substrate (stippling is substrate covering the anterior of the prosoma).

228 229 Specimen Max. Length Max. Width Prosoma Length Max. Depth A 20.9 16.5 17.1 3.4 B 20.6 13.8 n/a 1.9 C 17.9 18.5 13.2 2.4 D 12 12.6 10.6 2.5 E 12.9 15.1 n/a 2.1 F 21.4 17.1 12.9 2.5

Figure 6-3. Measurements of each ichnofossil, and letters correspond to Figure

230 Figure 6—4. Sketch of the rock with Selenichnites rossendalensis and physical sedimentary structures depicted as stippled. The arrows delineate the hypothesized palaeocurrent direction (i: is a horseshoe vortex with missing obstacle; iii, iv and v: tool marks). There is only a slight variance in current direction observed from the physical sedimentary structures. Inset: Rose diagram depicting the long axes orientation of the traces (shaded blocks) and the mean palaeocurrent direction (black arrow), orientation is based purely on the illustrated surface, an arbitrarily defined direction for the purposes of the description. Scale bar = 1 centimetre.

231 •^wi^^ww^sfy Figure 6-5. Horseshoe vortex system. (1) Instantaneous flow ribbons showing strong circulation around the obstacle and weaker, more random flow to the lee of the obstacle (modified from Kirkil et al (2005, fig. 2); (2) Time averaged shear velocity associated with a horseshoe vortex The shear velocity gives a realistic estimate of the zones of erosion (dark). Note the coincidence of the zones of erosion and the scours associated with Selenichnites rossendalensis (modified from Kirkil et al (2005, fig 7b).

233 CHAPTER 7: CONCLUSION

SUMMARY OF WORK

This dissertation addresses ichnological questions related to traces left by arthropods at a number of localities, located on four different continents. The strata containing these traces range in age from lower Cambrian to Upper Devonian. While most of this thesis is systematic in nature, describing, naming and illustrating ichnotaxa putatively made by members of the Phylum Arthropoda, other aspects deal with using the traces to determine the age of the rocks, deducing the palaeoenvironment and understanding the evolution of behaviour are also addressed in the five chapters (chapters 2 - 6). Ichnological studies are intricate. They involve studying many factors: the form of the traces, the sediments and physical sedimentary structures, the morphologies and behaviours of the organisms that may have made the traces that existed or may have existed in the time period when the traces were made, and the taphonomy of the traces and how taphonomic events may have affected preservation and the forms of the traces. Looking more closely at the environments, the Cambrian Gog Group, the Cambrian Azlag Formation and the Ordovician Stairway Sandstone are all in a shallow marine palaeosetting. The Ordovician of the Upper Fezouata Formation was a more middle to distal offshore marine setting, and the Devonian of the Stackpole Sandstone Member of the Gupton Formation was a lacustrine setting. Therefore, the lacustrine palaeoenvironment of the Devonian is the only non-marine setting observed. One aspect that has been considered in this work is the relationship of current flow direction on the sea floor to the form and orientation of various ichnotaxa. The Cambrian Azlag Formation in a shallow marine setting has interference ripples preserved in the sandstone with a thin mudstone overlying. The prominent direction of the traces, at this locality, follows one of the two wavetrains that caused the interference ripples. The Devonian locality in Wales also displayed both physical and biogenic sedimentary structures, providing a

234 prominent direction of the flow of water and distinct orientations of the traces in relation to that flow direction. 'Microbially induced sedimentary structures' (MISS) were observed a number of localities that contained the traces studied and this microbially-rich environment apparently assisted in the preservation of the traces. Kinneyia structures were observed in the lower Cambrian Gog Group, demonstrating that microbial mats were possibly present during this time. Also observed in the Gog Group material are mica-rich mudstones. The microbial biofilms and/or mats would have been sticky and the mica flakes would have collected on and in the microbial setting. The binding of sediment by bacteria assisted in the preservation of the traces and the bacteria probably also directly or indirectly provided food for the organisms that left the traces. The Cambrian in Morocco has been interpreted to be a shallow, tidally-influenced, marine setting with interference ripples. It was proposed that the sand would have had a microbial veneer, that helped to preserve the physical sedimentary structures. The organism(s) then produced Selenichnites traces by prodding into the sand, possibly through a thin mud deposited due to the settling fines after a storm, in search of food. The Ordovician of Morocco also presented microbial structures in the way of syneresis cracks from the loading of the mud in a more middle to distal offshore. These rocks also contain mica-rich and pyrite-rich laminae, which support the original presence of sticky, organic-rich laminae on the sediment surface during deposition. The microbial films/mats would have aided in the stabilization of the muds and the preservation of the traces. Some of the Rusophycus carleyi traces still preserve the articulated exoskeletons of the plausible trace-maker: a trilobite. It is proposed that many of the ichnofossils discussed, 25 ichnospecies from the Cambrian Gog Gog, one ichnospecies from the Cambrian of Morocco, one ichnospecies in the Ordovician of Morocco, 10 ichnospecies in the Ordovician of Australia, and one from the Devonian of Wales were produced in an instratal setting. Where this is not the case, it is suggested that bacterial films/mats on the surface of, or sometimes within, the sediment may have played

235 an important role in the preservation of the traces. The Cambrian of Morocco and the Devonian of Wales could possibly have been more of a surfical trace, with a lack and/or limited interface between disparate sediments at the sediment-water interface. The five core chapters of this thesis explore the use of ichnology in enhancing our understanding of evolution in the behaviours of organisms. The localities where ichnofossils were obtained for this thesis were chosen for a variety of reasons, including: quality of preservation of the ichnotaxa; opportunity (I was able to visit the site and collect the fossils); novelty (the arthropod traces had not been described previously or had been inadequately described); the possibility to deduce the nature of the trace maker(s); the opportunity to shed new light on our understanding on the history of life and evolution of behaviours through time. A complete summary of each chapter was provided in the introductory chapter and this chapter will concentrate on areas of future work that were stimulated by the research in this dissertation.

FUTURE WORK I have been fortunate enough to visit enough trace fossil-bearing localities throughout the Palaeozoic to understand how much more research on ichnology remains to be done. Our libraries contain many works on palaeontology, sedimentology, taphonomy, palaeoecology, stable isotopes, functional morphology/biomechanics and fluid dynamics. These all need to be brought together for robust studies in ichnology. Another area that will be very fruitful for future work is neoichnology. As noted by Cadee and Goldring (2007, p. 3): "The study of recent traces of invertebrates in the Wadden Sea played an important role in the development of ichnology. In 1928, Rudolf Richter (1881-1957) founded the first marine institute devoted entirely to Aktuogeologie and Aktuopalaontologie^. Neoichnology is obviously not a novel concept, and it is practiced by many ichnologists and palaeontologists at present, and has been in the past, though more must be done to gain a more thorough understanding of the

236 history of life. While not included as a chapter in this thesis, a number of experiments were carried out using the notostracan Triops, living above a variety of different substrates. Some of the data from these experiments were useful in understanding why some ichnospecies of Cruziana and Rusophycus are not constrained within the time ranges that have been suggested in some publications. I intend to complete this work and publish it in the future. Each of the chapters in the thesis provided me with ideas for future work.

Lower Cambrian Gog Group The abundance and diversity of ichnofossils within the Gog Group, occurring near the beginning of the 'Cambrian Explosion', is absolutely outstanding. Not only is there a high diversity of traces putatively made by arthropods, but also the complexity of the traces is equivalent to those found in younger rocks. Therefore this was truly an explosion of not just complex, hard- bodied organisms, but also behavioural adaptations on and in the substrate. A listing of the entire ichnofauna from the various units within the Gog Group would provide important additional information on the "Cambrian Explosion" in shallow near shore environments. On Babel Mountain, adjacent to Moraine Lake in Alberta there are numerous large slabs of Gog Group quartzites that have numerous extensive trackways of Cruziana and or Diplichnites. These often transform backwards and forwards into one another, depending upon how deeply the organisms were burrowing at the time. In some of the same beds, cross beds can be used to determine the direction of current flow when the original sands were being deposited. There seems to be distinctive orientations of these trackways, with trackways often running subparallel to one another. It will be very interesting to map out directions of the long axes of the trackways, in rose diagrams, and compare them with sea floor current directions deduced from cross-bedding within the same beds. Evidence for the existence of Microbially Induced Sedimentary Structures (MISS) are found in the Gog Group and the Upper Fezouata Formation in

237 Morocco. Further studies of these from the Gog Group and various other localities containing younger strata (such as the Ordovician Bell Island Group of Newfoundland) will be part of a more extensive study on the importance of microbial mats in causing and preserving various types of trace fossils.

Cambrian of Morocco The study of Middle Cambrian Selenichnites traces from Morocco involved an extensive literature search into the geology of Morocco, and the organisms that have been found there of Cambrian age. I was fortunate to come across a short paper, which stated that in one region of the Atlas Mountains trilobites have found in Rusophycus traces that they had made. Communication with the author, Gerd Geyer, provided me with locality data, and the reason why they have not been illustrated and described (the rocks are highly friable). Because I have had extensive training in the preparation and restoration of fossils, from a former technician at the University of Alberta (Allan Lindoe), I believe that I will be able to extract these fossils and their traces from the rocks and strengthen them so that they can be collected, studied and published on. This is potentially another rare example of trilobites in direct association with the traces they have made.

Ordovician of Morocco The chapter on the Ordovician of Morocco where I was able to show that examples of Rusophycus carleyi are directly associated with and were almost certainly made by asaphid trilobites allowed me to suggest some details of the morphology of the appendages of these trilobites that are otherwise unknown. Careful examination of detailed studies where Rusophycus or Cruziana ichnospecies that have been associated with some degree of confidence with particular trilobite taxa may provide useful morphological data on the soft parts of other trilobites whose appendages are otherwise unknown. Examination of traces of trilobites whose appendages are known would also prove to be highly informative in understanding how and why these traces are made.

238 Ordovician of Australia It was one of the Ordovician traces from Australia, Cruziana penicillata, that influenced my interest in and understanding of the activities of modern arthropods digging into the substrate at an angle greater than horizontal. The study of this trace and my observations of the Recent notostracan Triops burrowing, showed me that some vertical burrowing involves forward and/or backward movement. Thus, I assigned traces to Cruziana that involve some forward or backward travelling, with or without vertical movement. Where burrowing was only vertical, with perhaps minor side-to-side rocking, and the number of scratch marks is more directly related to the number of appendages borne by the animal that made the trace, I assigned the trace to Rusophycus. The concept of horizontal and vertical movement to distinguish Cruziana from Rusophycus, which was caused by purely vertical movement, with no horizontal component, introduces the need to scour the literature for the possible reassignment of a number of ichnospecies.

Devonian of Wales The Welsh traces from the Devonian were an entertaining introduction to fluid dynamics and the possible adaptations of organisms to exist in an environment which can present occasional hostile turbidity flows. This project reintroduced me to the concept of 'ghost lineages' that become more visible when they make trace fossils. The absence of body fossils of the correct size and shape for this time period in a freshwater setting required a hypothesis on the timing of the movement of the organisms that made the traces from a marine to freshwater setting. A search for other traces made by 'ghost lineages' will help to illuminate past migrations of organisms from one environment to another.

Like a geological striptease, veils of rock were stripped slowly away to ever more fundamental levels with the ancient chain.

239 Eventually it was stripped naked to its interior, and the show was over. - Fortey (2000)

240 REFERENCES

CADEE, G. C, AND R. GOLDRING. 2007. The Wadden Sea, Cradle of invertebrate ichnology, p. 3-13. In W. Miller (ed.), Trace Fossils: Concepts, problems, prospects. Elsevier, Amsterdam. FORTEY, R. A. 2000. Trilobite! An eyewitness to evolution. Alfred Knopf, New York, 284 p.

241