h. .47.“. LP: . x . L. fiéufifi .. . . udwé . mfiimmfi ... .w. Mu... «HQJNA é “a? . m...» #5..“ “Lawn-c... ‘ ‘Afi‘hg‘ : M‘ ‘fir ; $4 .1 .. .w......\. 2..” , ting... :5 .. z . a .3. . . . _ , 5J,.§\lu\..x‘. :10... {:51 s I ‘IL' iii-v .. fl... 2.: y I hf. _. 2s“: Jan. 1. v.5 IL A.) VI. 3 a «11"... :21... . .Juu; . ms LIBRARY 3 Michigan State 2.43 0 University This is to certify that the dissertation entitled A RECONSIDERATION OF THE ARTHROPHYCUS ICHNOGENUS: DEFINITION, BIOSTRATIGRAPHIC UTILITY, AND A PROPOSAL TO DEVELOP A NUMERICAL ICHNOTAXONOMY presented by Megan E. Seitz has been accepted towards fulfillment of the requirements for the PhD. degree in Geological Science \ I . K . J— _/' , z” A \ .’ l. I 1(4" 2 1’7 (‘ / 1' * I . f ’ 1 V "‘ ‘ ‘ ' ' 5“ ' I Major Professor's Signature {’1 , "I {I I} I '1’, / 5! :I ‘I I.“ /, Date MSU is an Aflinnative Action/Equal Opportunity Employer — -.—--o-----u---n—-—--n-n— PLACE IN RETURN BOX to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 5/08 K:lProj/Acc&Pres/CIRC/DateDue.indd A RECONSIDERATION OF THE ARTHROPHYCUS ICHNOGENUS: DEFINITION, BIOSTRATIGRAPHIC UTILITY, AND A PROPOSAL TO DEVELOP A NUMERICAL ICHNOTAXONOMY By Megan E. Seitz A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Geological Science 2010 ABSTRACT A RECONSIDERATION OF THE ARTHROPHYC US ICHNOGENUS: DEFINITION, BIOSTRATIGRAPHIC UTILITY AND A PROPOSAL TO DEVELOP A NUMERICAL ICHNOTAXONOMY By Megan E. Seitz Arthrophycus Hall 1852 is a long-studied ichnogenus reported from localities worldwide, including all seven continents and twenty-eight countries. It is most abundant in Ordovician and Silurian strata and is regarded by some ichnologists as having biostratigraphic utility, although other occurrences of the ichnogenus have been reported from the Proterozoic to the Miocene. Imprecise or overly-brief descriptions and ambiguous or unclear illustrations, drawings, and photographs of specimens regarded as Arthrophycus have resulted in a taxonomic “wastebasket” and have led to confusion over the biostratigraphic utility of the ichnogenus. A review of the primary Arthrophycus literature converged on the following diagnostic characters for the ichnogenus: annulations, simple form, branches, bundles, shape of cross-section, median groove, and internal structures. Characters such as annulations, branching, and median groove encompass variability not previously discussed in reviews of the ichnogenus. Refining the definition of these characters produced sixteen characters that could be coded for a numerical phenetic analysis of Arthrophycus ichnospecies. Cluster analyses revealed two main clusters: a cluster of six (A. alleghaniensis, A. brongniartii, A. Iinearis, A. Iateralis, A. minimus, A. parallelus) considered to be valid Arthrophycus and a second cluster of discredited ichnospecies, members of which probably belong in other ichnogenera. The cluster analyses were supported by PCO and cladistic analyses. Only Paleozoic occurrences of Arthrophycus are considered valid, including specimens fi‘om the Cambrian, Devonian, and Carboniferous, challenging previous conclusions that Arthrophycus is confined to the Ordovician and Silurian. This analysis also confirms an hypothesis that Arthmphycus originated in the southern continents in the Cambrian, persisted there in the Ordovician, and then expanded to the northern continents during the late Ordovician and Silurian. ACKNOWLEDGMENTS I would like to acknowledge the help and support of many fine people: My committee members: Dr. Danita S. Brandt, Dr. Robert L. Anstey, Dr. Brian A. Hampton, and Dr. Richard Snider My parents and sister and all of my wonderful family The fine people of ILLiad who helped find obscure papers and books Jane Wei, Gabe Yedid, and Dr. Yue Li, who translated the papers in Chinese Kathy Linta and family, who hosted me during my visit to Yale Dr. Susan Butts, Senior Collections Manager at the Yale Peabody Museum Bushra Hussaini, Senior Scientific Assistant at the American Museum of Natural History Dr. Ed Landing, State Paleontologist and Paleontology Curator, and Linda Hemick, Paleontology Collections Manager, at the New York State Museum Dr. Richard Laub, Curator of Geology at the Buffalo Museum of Science And especially to Tiger, T. rex, Hanna, Heidi, Tish, Pincushion, Nikki, Trouble, and Red Dog. I love you all. iv “L’etude de ces fossiles plus que problematiques est un travail tres ingrat. . .” (The study of these more than problematic fossils is a very thankless job...) ~ Fritsch, 1908 “A single slab in your cabinet, ornamented in relief with groups of this remarkable fossil, whose figured surface reminds us of the Gothic tracery of ancient sculpture, is of itself an object of admiration.” ~ Taylor, 1835 “We will be known forever by the tracks we leave.” ~ Native American Proverb TABLE OF CONTENTS LIST OF TABLES ................................................................................ ix LIST OF FIGURES ............................................................................... x CHAPTER I INTRODUCTION ................................................................................. 1 Rationale for Study ........................................................................ 1 Objectives .................................................................................... 2 History of Trace Fossils ..................................................................... 2 Ichnotaxonomy ............................................................................ 4 Background of Phenetics .................................................................. 7 Methods .................................................................................... 7 CHAPTER H HISTORY OF ARTHROPHYC US AND DEFINITION OF CHARACTERS ........... 14 History of Arthrophycus .................................................................. 14 Identity of Arthrophycus ............................................................... 17 “Official” Diagnosis of Arthrophycus ................................................. 19 Taxonomically Important Characters fiom Original Diagnoses .................. 20 Additional Characters ................................................................... 25 Gradation Between Arthrophycus and Other Ichnogenera ......................... 29 Suprageneric Classification of Arthrophycus ......................................... 3O Morphological Issues Raised in Previous Work .................................... 31 CHAPTER III DEFINITION OF ART HROPHY C US, PART I.: QUALITATIVE CRITIQUE OF PREVIOUS WORK .............................................................................. 58 A. alleghaniensis ......................................................................... 58 A. harlani and Harlania halIi ........................................................... 74 A. brongniartii ............................................................................ 75 A. [means ...................... 78 A. montalto ................................................................................ 80 A. siluricus ................................................................................. 81 A. elegans ................................................................................... 82 A. corrugatus ............................................................................. 82 A. flabelliformis .......................................................................... 83 A. krebsi ................................................................................... 84 A. minoricensis ........................................................................... 85 A. annulatus ............................................................................... 87 A. strictus .................................................................................. 87 A. dzulynskii ............................................................................... 89 A. tennis .................................................................................... 89 vi A. qiongzhusiensis ........................................................................ 91 A. tarimensis .............................................................................. 92 A. hunanensis ............................................................................ 93 A. unilateralis ............................................................................ 93 A. Iateralis ................................................................................ 93 A. simplex ................................................................................. 95 A. minimus ................................................................................ 96 A. parallelus .............................................................................. 97 A. isp ....................................................................................... 98 Taxonomic Conclusions ............................................................... 114 CHAPTER IV DEFINITION OF ARTHROPHYCUS, PART II.: NUMERICAL ICHNOLOGY.. .. 132 Previous Use in Ichnology ............................................................. 132 Basic Methods .......................................................................... 132 Character States ......................................................................... 134 Missing Data ............................................................................ 137 Cluster Analysis Results ............................................................... 138 Principal Coordinates Analysis (PCO) .............................................. 141 Cladistic Analysis ...................................................................... 142 Potential Errors ......................................................................... 145 Numerical Taxonomy Conclusions .................................................. 146 An Application of the Method ....................................................... 148 CHAPTER V SYSTEMATIC ICHNOTAXONOMY ...................................................... 160 Generic Synonymy ..................................................................... 160 Synonymy .............................................................................. 160 Type Ichnospecies ..................................................................... 166 Original Diagnoses ..................................................................... 166 Emended Diagnosis .................................................................... 167 Discussion ............................................................................... 167 Included Ichnospecies .................................................................. 168 Stratigraphic and Geographic Range ................................................ 168 CHAPTER VI ART HROPHYCUS IN TIME AND SPACE ................................................ 169 Biostratigraphy ......................................................................... 169 Biogeography ........................................................................... 175 Paleobiogeography ..................................................................... 1 77 Facies Control of Arthrophycus Distribution ....................................... 179 CONCLUSIONS ............................................................................... I96 APPENDICES .................................................................................. 199 Condensed Data for All Reports of Arthrophycus ................................. 199 vii Character States for Quantitative Analyses ......................................... 206 Coded Characters ....................................................................... 207 LITERATURE CITED ........................................................................ 208 viii 2.1 2.2 2.3 2.4 3.1 3.2 4.1 6.1 6.2 LIST OF TABLES Authors who have suggested some kind of worm-like animal as the trace-maker of Arthrophycus .......................................................................... 38 Additional suggestions for the Arthrophycus trace-maker .......................... 39 Reported sizes of A. alleghaniensis specimens ...................................... 40 Reported sizes of Arthrophycus specimens, all ichnospecies except A. alleghaniensis ............................................................................ 42 Published ichnospecies of Arthrophycus ............................................ 1 16 Guide to assessments regarding the status of each proposed ichnospecies of Arthrophycus ........................................................................... 1 1 7 Size range of Arthrophycus ichnospecies. Yellow indicates the smallest size, green is intermediate, and blue is the largest size. Blocks in pink indicate an overlap of ranges or, in the case of A. alleghaniensis, A. brongniartii, and A. Iinearis, a very large range. The maroon color indicates the “more typical” range of widths for that ichnotaxon and the X’s mark the approximate midpoints ................................................................................ 149 Occurrences of Arthrophycus through time by ichnospecies. Number of reports may not equal the total of the row because all periods are counted if a single report spans more than one period (e. g. Ordovician-Silurian). The “Other” column contains reports of uncertain age ........................................... 181 Coding for geologic periods and regions used in the bubble charts ............. 182 ix LIST OF FIGURES Images in this dissertation are presented in color. 1.1a 1.1b 1.2 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 Sketch of Arthrophycus allegham'ensis, fi'om Hall (1852). Silurian, Medina Sandstone, New York ................................................................... 12 Sketch of A. alleghaniensis, from Hall (1852). Silurian, Medina Sandstone, New York ................................................................................. 12 Data sheet used to characterize museum specimens of Arthrophycus ............ l3 Specimen of a “typical” Arthrophycus from the Silurian of New York, Medina Sandstone. Note regular annulations running perpendicular to the long axis. BMS E20774. Scale bar represents 1 cm ............................................. 43 Specimen of A. alleghaniensis showcasing the rather simple form and fmger- like shape of the traces. Silurian of New York, formation unknown. OSU8391. Scale bar represents 1 cm ............................................................... 43 Example of branching (arrows) in A. alleghaniensis. Silurian of New York, formation unknown. OSU8391. Scale bar represents 1 cm ....................... 44 Example of A. alleghaniensis with pseudo-branching, in which two otherwise independent traces cross one another. Silurian of New York, formation unknown. OSU8391. Scale bar represents 1 cm .................................... 44 Wide palmate bundling in a specimen of A. alleghaniensis. Silurian of New York, Medina Sandstone. YPM38353. Scale is in mm ............................ 45 Specimen with narrower bundles of Arthrophycus. Silurian of Martinsburg, PA, Clinch Sandstone. CMNH3819. Scale bar is 2 cm long ...................... 45 Specimens of Arthrophycus exiting various intergrading branching and bundling patterns. Silurian of Rochester, NY, Grimsby Sandstone. YPM207001. Scale bar represents 2 cm ............................................. 46 Specimens of Arthrophycus that disappear almost immediately from their bundles and descend into the matrix (arrow). Silurian of Rochester, NY, Grimsby Sandstone. YPM207001. Scale bar represents 2 cm ................... 46 Specimen of Arthrophycus with a clear median groove (arrow). Silurian of western New York, Medina Group. YPM7365 ..................................... 47 2.10 2.11 2.12 2.13a 2.13b 2.14 2.15 2.16 2.17 2.18 2.19 2.20 Arthrophycus with faint or indistinct median groove. This close-up of YPM7365 shows that even clear median grooves can appear indistinct depending on the magnification or lighting. Scale bar represents 1 cm .......... 47 Arthrophycus with spreiten (arrow) revealed on a polished cross-sectional surface. Silurian of New York, Medina Group. YPM150650. Scale is in mm ......................................................................................... 48 Specimens of Arthrophycus with fine striations (arrow) on top of the individual annulations. These annulations also display a pinched shape toward the middle, lending to the impression that a groove is present. Silurian of New York, Medina Sandstone. YPM150639. Scale is in mm ................................. 48 Specimen of Arthrophycus displaying vertical arch or bowing (arrow) as it passes over another specimen (or under, if the specimen is inverted as most are). Silurian of Rochester, NY, Grimsby Sandstone. YPM207001. Scale bar represents 1 cm ........................................................................... 49 Bundled specimen of Arthrophycus displaying vertical arch. Silurian of Para, Brazil, sandy shale. NYSM6176. Scale is in mm .................................. 49 Sample of Arthrophycus in which many primarily flat-lying traces cross over one another. Silurian of Grimsby, Ont., Grimsby Sandstone. BMS E3799. Scale bar represents 2 cm ............................................................... 50 Complicated arrangement of Arthrophycus in which traces go in many directions. Silurian of Medina, NY, Medina Sandstone. NYSM2. Scale bar represents 1 cm ........................................................................... 50 Flat-lying Arthrophycus traces with other traces perpendicular to the bedding. Silurian of Lockport, NY, Medina Group. YPM508647. Scale bar represents 1 cm ....................................................................................... 51 Twisting of Arthrophycus traces with respect to one another, in three dimensions. Silurian of Rochester, NY, Grimsby Sandstone. YPM207001. Scale bar represents 2 cm ............................................................... 51 Large loop pattern of Arthrophycus traces. Silurian of Rochester, NY, Grimsby Sandstone. YPM207001. Scale bar represents 2 cm .............................. 52 Arthrophycus (arrows) co-occurring with Rusophycus (circle). Silurian of Lewiston, NY, found loose. BMS E25610. Scale bar represents 2 cm ......... 52 A. parallelus co-occurring with Lockeia (arrow). Pennsylvanian of Michigan, Grand River Formation. University of Michigan Museum of Paleontology (UMMP) 73822. Traces are approximately 4 mm wide ........................... 53 xi 2.21a 2.21b 2.21c 2.22 2.23 2.24 2.25 3.1 3.2 3.3 3.4 3.5 3.6 3.7 Sketches of three different ichnospecies of Ordovician Daedalus, from Seilacher (2000). Scale bars are in centimeters ..................................... 53 Polished cross-section of Daedalus archimedes, showing the spreiten of the trace (arrows). Silurian of Medina, NY, Medina Sandstone. YPM35825. Scale is in mm ................................................................................... 54 Non-spiral specimen of D. archimedes, which appears more similar to Arthrophycus than the spiral version does. Silurian of Medina, NY, Medina Sandstone. YPM35822. Scale bar represents 1 cm ................................ 54 Specimen of Arthrophycus. Silurian of Lockport, NY, Medina Group. YPM35814. I suggest that morphotypes known only from a few individuals, such as this one, be termed “Arthrophycus-like.” Scale is in mm ................ 55 Example of a specimen referred to Arthrophycus (arrow), almost certainly in error, along with specimens of Gordia. Devonian of Ohio, Chagrin Shale. CMNH3705. Scale is in cm ............................................................ 55 Size of ichnospecies of Arthrophycus over time. Gridlines mark every 5 mm in width and start from 0 mm ............................................................. 56 Inset of crowded area of graph in Fig. 2.24. Grid lines mark every 5 mm and start from 0 mm .......................................................................... 57 Drawings of F ucoides alleghaniensis from Harlan’s original work (1831). Silurian of Pennsylvania, found in a tavern wall. No scale given ............... 118 The first known drawing of F. brongniartii, fi'om Taylor (1835). Silurian of Lewiston, PA. No scale given ....................................................... 118 Seilacher’s original drawings of A. linearis and A. unilateralis (1997). No locality information or scale given ................................................... 119 A. Iinearis in Seilacher (2000), along with a figure of wrinkles on that specimen. Silurian of Rochester, NY, Medina Sandstone. Scale is in cm for (a), m for (b) .......................................................................... 119 A. montalto from Lesley (1889). Cambrian of Pennsylvania, found in a sawmill wall. No scale given .................................................................. 120 A. elegans from Herzer (1901). Carboniferous of Marietta, OH. Thinner traces are approximately 1/ 16” wide, wider traces are approximately 3/16” wide. . . 120 A. corrugatus (as Radicites rugosus) from Fritsch (1908). Silurian of Czech Republic. No scale given ............................................................. 121 xii 3.8 3.9 3.10a 3.10b 3.11 3.12 3.13 3.14 3.15a 3.15b 3.16 3.17 3.18a 3.18b 3.19a A. flabelliformis from Hundt (1940). Ordovician of Wuenschendorf ad Elster, Germany. No scale given ............................................................ 121 A. Irrebsi (arrow with base line) from Hundt (1941). Ordovician of Wuenschendorf ad Elster, Germany. No scale given ............................. 122 A. minoricensis from Orr (1994). Lower Carboniferous of Menorca Island. Traces are approximately 1 cm wide ................................................ 123 A. minoricensis from Llompart and Wieczorek (1995). Carboniferous of Minorca Island, Culm siliciclastic sequence. No scale given ................... 123 A. annulatus from KsiaZkiewicz (1977). Lower Eocene of Poland, Ciezkowice Sandstone (8) and Turonian, Inocerarnian Beds (9). Traces are approximately 1 cm wide ............................................................................... 124 A. strictus from Ksiazkiewicz (1977). Cretaceous of Poland, Lgota Beds (11), Ropianka Beds (12). Traces are 1.5 to 2.0 mm wide .............................. 124 A. dzulynskii from Ksiazkiewicz (1977). Oligocene of Poland, Krosno Beds. Traces are approximately 1 cm wide ................................................ 125 A. tenuis (as Sabularia tenuis) from Ksiazkiewicz (1977). Oligocene of Poland, Krosno Beds. Traces are 0.5 to 1.0 mm wide ...................................... 125 A. qiongzhusiensis from Luo et al. (1994). Early Cambrian of China, Yu’anshan Member of the Chiungchussu Formation. It was not clear from the paper which structure is Arthrophycus; the most likely candidate is circled... 126 A. qiongzhusiensis, also from Luo et al. (1994). Early Cambrian of China, Yu’anshan Member of the Chiungchussu Formation. No scale given ......... 126 A. tarimensz‘s from Yang (1994). Lower Ordovician of the Tarim Basin, China, Ulikeztag Formation. No scale given ............................................... 127 A. hunanensis from Zhang et al. (1996). Silurian of China, Xiaoheba Formation. Traces are approximately 1 cm wide ................................. 127 A. lateralis from Seilacher (2000). Lower Silurian of Libya, Akakus Sandstone. Scale bar in comer represents 3 cm .................................... 128 Sketch of A. Iateralis fi'om Seilacher (2000) ....................................... 128 Loop of Arthrophycus showing asymmetry of the bundling toward the bottom of the photograph. Silurian of Rochester, NY, Grimsby Sandstone. YPM207001. Scale bar represents 2 cm ............................................ 129 xiii 3.19b Asymmetrical bundle in an Arthrophycus specimen, labeled as A. harlani but 3.20 3.21 3.22 3.23 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 possibly representing A. lateralis. Silurian of Tyrone, PA, Tuscarora Sandstone. PR1, unnumbered specimen. Scale bar is in cm ..................... 129 A. simplex (arrow) from Seilacher et a1. (2002), with specimen of Cruziana goldfirssi. Ordovician of the Kufra Basin, Libya, Hawaz Formation ........... 130 A. minimus from Mangano et al. (2005a). Upper Cambrian of Argentina, Alfarcito Member of the Santa Rosita Formation. Scale bars represent 1 cm ...................................................................................... 130 Unnumbered specimen referred herein to A. parallelus, from the New York State Museum. Devonian of New York, possibly Hamilton Group. Scale bar represents 1 cm ......................................................................... 131 Arthrophycus isp. Late Devonian to Early Mississippian, Rowlesburg, WV, Oswayo Member of the Price Formation. CMNH8340. Scale bar represents 1 cm. Specimen is the same as that figured in Bjerstedt (1987) ................... 131 Different types of annulations observed in ichnospecies of Arthrophycus. A) rings; B) concavo-convex; C) chevrons; D) biconvex ........................ 150 Cluster diagram of twenty OTUs of Arthrophycus ................................. 151 Cluster diagram of twenty OTUs of Arthrophycus, without the width character ................................................................................. 152 Two-way cluster diagram of characters with Arthrophycus OTUs. Non-zero numbers in the data matrix indicate which characters are most important for defining clusters of taxa. The vertical line divides the numbers in the matrix into those for the Arthrophycus cluster on the lefi and all others on the right. See key on next page for abbreviations ............................................. 153 PC0 graph, coordinates 1 and 2. The cluster of six is represented by the + symbols; all other OTUs are represented by the dots .............................. 155 PC0 graph, coordinates 1 and 3. The cluster of six is represented by the + symbols; all other OTUs are represented by the dots. The dot near the top of the graph represents A. minoricensis; this point was not separate fi'om the others in the previous graph .................................................................. 156 Cladogram (third of three most parsimonious trees) of twenty OTUs of Arthrophycus, using A. strictus (P) as the outgroup ............................... 157 Strict consensus cladogram using A. strictus (P) as the outgroup ............... 158 xiv 4.9 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10 6.11 Cladogram (first of nineteen most parsimonious trees) of twenty OTUs of Arthrophycus, using A. qiongzhusiensis as the outgroup ......................... 159 Histogram of reported occurrences of Arthrophycus through time .............. 182 Histogram of occurrences of all Arthrophycus reports through time, divided into the three taxonomic assessment categories ......................................... 183 Histogram of occurrences of Arthrophycus reports through time, confirmed reports only .............................................................................. 184 Histogram of occurrences of Arthrophycus reports in time, confirmed and unverifiable reports only ............................................................... 185 Histogram of occurrences of valid ichnospecies of Arthrophycus through time, including “Arthrophycus isp.,” broken into the three taxonomic assessment categories ................................................................................ 186 Histogram of occurrences of valid ichnospecies of Arthrophycus through time, 1!); including “Arthrophycus isp.,” broken into the three assessment groups. 187 Histogram of valid Arthrophycus-bearing formations throughout geologic time ....................................................................................... 187 Histogram of all Arthrophycus occurrences, by geographic region ............. 188 Histogram of all Arthrophycus occurrences, by geographic region, divided into the three taxonomic assessment categories .......................................... 188 Histogram of regions reported for all Arthrophycus occurrences, confirmed occurrences only ........................................................................ 189 Bubble chart of Arthrophycus reports over time and space and divided into categories of accepted and present. The vertical line divides the southern and northern regions, the lower horizontal line marks the middle of the Ordovician, and the upper horizontal line marks the middle of the Silurian. The central point of each circle marks the time period in which it belongs and the size of the circle indicates the number of reports at that point in time and space. See Table 6.2 for key to regional identification.....................................~.... 190 XV 6.12 6.13 6.14 6.15 6.16 Bubble chart of accepted Arthrophycus reports over time and space and by ichnospecies. The vertical line divides the southern and northern regions, the lower horizontal line marks the middle of the Ordovician, and the upper horizontal line marks the middle of the Silurian. The central point of each circle marks the time period in which it belongs and the size of the circle indicates the number of repOrts at that point in time and space. See Table 6.2 for key to regional identification ..................................................... 191 Bubble chart of accepted Arthrophycus over time and space, by ichnospecies and including “Arthrophycus isp.” reports. The vertical line divides the southern and northern regions, the lower horizontal line marks the middle of the Ordovician, and the upper horizontal line marks the middle of the Silurian. The central point of each circle marks the time period in which it belongs and the size of the circle indicates the nmnber of reports at that point in time and space. See Table 6.2 for key to regional identification .................................... 192 Paleogeographic map of the southern continental positions during the Late Ordovician. Ice sheet is shaded and stars represent approximate locations of Arthrophycus specimens. From Turner et al. (2005) .............................. 193 Paleogeographic map of all the continental positions during the Early Ordovician. Arrows indicate plate movements and stars represent approximate locations of Arthrophycus specimens. From Scotese (2002) .................... 194 Paleogeographic map of all the continental positions during the Early Silurian. Arrows indicate plate movements and stars represent approximate locations of Arthrophycus specimens. From Scotese (2002) ................................... 195 xvi Chapter I. Introduction and Background Rationale for Study Arthrophycus is a well-known ichnogenus (Figure 1.1) with specimens reported from localities around the world (Seilacher 2007a), and is often cited as having biostratigraphic utility. Imprecise descriptions and ambiguous or unclear illustrations, drawings, and photographs of specimens regarded as Arthrophycus have resulted in a “wastebasket” taxon (as used by Plotnick and Wagner 2006) and have led to confusion over the biostratigraphic utility of the ichnogenus. Hantzschel’s (1975) description of the ichnogenus in the Treatise of Invertebrate Paleontology has been the standard until a recent, sweeping reconsideration of “arthrophycid” burrows (Seilacher 2007a). Seilacher’s (2007a) interpretation goes far beyond a strict morphological taxobasis advocated by other ichnologists (e. g. Keighley and Pickerill 1996), and his “fingerprint” for the arthrophycids is not observable in many published descriptions of trace fossils assigned to Arthrophycus, further suggesting that a thorough review of the ichnogenus is in order. Seilacher (2007b) and other authors have suggested that Arthrophycus can be used in biostratigraphy. Relatively short-lived fossil species, known as index fossils, are often used to date formations in which they are found. Arthrophycus has been used in this manner (e.g. Silva 1951, Kumpulainen et al. 2006), but this technique is useful only if the ichnospecies is truly stratigraphically limited, or the dates assigned will be of little worth. Beginning at least as early as Conrad (1839), several authors have reported Arthrophycus as an index fossil for the Ordovician and Silurian (e. g. Moore 1933, Shimer and Shrock 1944), but others have questioned this use of the ichnogenus (Young 1955, Femandes and Borghi 1996). Stratigraphically, Arthrophycus may be best known fiom Ordovician and Silurian strata (e.g. Hantzschel 1975, Seilacher 2000), but other specimens have been reported from the Proterozoic (Mukherjee et al.1987) to the Cretaceous (e.g. Frey and Howard 1970, Banerjee 1982) and even to the Miocene (Uchman and Demircan 1999). However, reports of Arthrophycus from Devonian and Carboniferous strata have been questioned (Méngano et al. 2005a), and post-Paleozoic occurrences of Arthrophycus are highly suspect (Hantzschel 1975, Mangano et al. 2005a). Objectives The goals of this project are to (1) bring diagnostic coherence to the ichnogenus Arthrophycus, (2) clarify whether the ichnotaxon is a valid index fossil for the Ordovician/Silurian, and (3) evaluate the use of a numerical approach to ichnotaxonomy. History of Trace Fossils Essentially, the history of trace fossils begins with fucoids. Of 258 genera named before 1900 that are now known as trace fossils, 120 were called fucoids, interpreted as fossil seaweed or algae (Osgood 1975a). From 1822 to 1881, at least 130 species of the genus F ucoides were named (James 1893b). Some ichnogenera, such as Chondrites and Cruziana, did appear plant-like, and the drive to publish likely supported the large numbers of overly facile descriptions (Osgood 1975a). Dawson (1864) was among the first to recognize that not all of the supposed “fucoids” were plants, but that some were the work of trilobites. The papers of the following years focused mainly on the genera Rusophycus (as had Dawson) and Cruziana (Osgood 1975a). Nathorst (1881) was the first to find widespread, though not universal, acceptance for his view that these structures were fossil traces. Nathorst worked on modern depositional environments along the Swedish coastline and argued that a fiagile little alga was unlikely to make convex impressions in sandstone, but would most likely have been crushed flat. However, not everyone was convinced. In an encyclopedic work, Schimper and Schenk (1879-1890) classified trace fossils such as Arthrophycus with the “Algae incertae sedis,” a large group within the seaweeds and algae that spanned twenty-five pages. Progress toward understanding the true nature of trace fossils came in the 19203 with works by Richter (Osgood 1975a). Further understanding resulted from Abel’s (1935) large and standard reference book Vorzeitliche Lebensspuren, in which he discussed a large number of trace fossils, both vertebrate and invertebrate, and even included coprolites. German authors then held sway for a number of years, writing many influential trace fossil papers (Osgood 1975a). Perhaps chief among these was Seilacher, who developed an ethological (behavioral) scheme of classifying traces (1953) and recognized a number of ichnofacies grouping ichnofossils by depth (1964). Finally, in 1962, Hantzschel wrote the first version of the portion of the Treatise of Invertebrate Paleontology that dealt with trace fossils. The great advantage of this work was that it effectively organized many genera of trace fossils, reducing their synonyms and providing a useful encyclopedia with illustrations (Osgood 1975a). Currently, trace fossils are defined most basically as ancient structures produced in or on the substrate by animals (Frey 1973, Bromley 1990). The International Code of Zoological Nomenclature defines an ichnotaxon as “the fossilized work of an organism” and, as do most authorities, restricts trace fossils to “the result of the activity of the animal” and not animal parts or molds of the inactive body (ICZN 1999). Though a few details need to be hammered out, most workers agree with these basic definitions (e.g. Hantzschel 1975, Bromley 1990). Trace fossils remain important fossils to study. They add to our knowledge of both soft and hard-bodied animal diversity (Osgood 1975b) and can be used in paleoecological studies (e.g. Paczesna 1996). Other examples of problems that trace fossils can solve include oxygen levels, salt levels, current direction, and deposition rate (Seilacher 1964a, Bromley 1990). Trace fossils generally cannot be transported without destroying the trace (Seilacher 1964a), so the state of the traces is evidence of how much the substrate has changed (Bromley 1990). Ichnofossils are more reliable records of behavior than body fossils are (Osgood 1975b) and are common in strata that do not preserve body fossils (Bromley 1990). A notable recent example, and one that highlights the importance of trace fossils, is the report of new trackways of Devonian tetrapods, eighteen million years before the first known body fossils of tetrapods and ten million years before the first known elpistostegids such as T iktaalik (Niedzwiedzki et al. 2010). Ichnotaxonomy The goal of any classification scheme is to group similar things together while separating them from unlike firings (Simpson 1975). Trace fossils are classified the same way as biologic taxa, using a basic Linnaean binomial system of ichnogenus and ichnospecies (Bertling et al. 2006, Bertling 2007). Significant differences in morphology are taken to exhibit differences at the ichnogenus level and lesser differences differentiate ichnospecies (Bertling et al. 2006). However, in practice, differences between ichnogenera and ichnospecies depend largely on the individual researcher (Magwood 1992). There is no widely accepted suprageneric classification for trace fossils; some suggest that ichnofamilies be based solely on morphology and used only for convenience, with no requirement that every ichnogenus be placed in an ichnofamily (Rindsberg 1998, Bertling et a1. 2006). Recognizing and diagnosing trace fossil taxa has always been problematic, as different researchers may disagree on what constitutes a valid group. Many bases for classification have been proposed and/or used, including morphology, supposed behavior, substrate, size, preservation, supposed producer, type of fill, geologic age, and location (Magwood 1992, Pickerill 1994, Demathieu and Demathieu 2002, Bertling et al. 2006, and Bertling 2007). Most of these proposed ichnotaxobases are not useful for all types of trace fossils and are subject to the interpretation of the worker (Magwood 1992). Diagnoses of trace fossils should therefore “be restricted to observations” rather than subjective interpretations (Minter et al. 2007). Although complete consensus among ichnologists is still lacking, trace morphology has emerged as the exclusive acceptable ichnotaxobase (F firsich 1974, Keighley and Pickerill 1996), as the interaction of an animal with its environment is preserves (Fttrsich 1974). Substrate criteria are still used in special cases, mostly for borings in hard substrates or wood (Bertling et al. 2006). In the case of Arthrophycus, the trace is not a boring into hard substrate, so the substrate criterion is not applicable; morphology of the trace is the lone remaining criterion on which to base its taxonomy. Morphological criteria can be further broken down into subcategories, including overall shape (especially type of branching, if present), cross-section, orientation with respect to substrate, surface features, internal structure, and wall lining (Bertling et al. 2006, Bromley 1990, Magwood 1992, Rindsberg 1998, Pickerill 1994, and Bertling 2007). These features must be easy to preserve and identify, or they lose their effectiveness in ichnotaxonomy (Magwood 1992, Bertling et al. 2006, Minter et al. 2007); the ichnotaxobases should be features that show little variability (Demathieu and Demathieu 2002). Some workers (e. g. Magwood 1992) persisted in reading specific producer behavior into the trace morphology, but behavioral inferences are not always clear, objective, or supported by other workers (e. g. Bertling et al. 2006). However, behavior is still an important consideration in trace fossil studies, as behavior is considered to be a major component of what might be regarded as the “extended phenotype” (Dawkins 1982) of a trace-making organism, along with morphology (Bertling et al. 2006). Some characters may provide clues to the trace ethology, and would then be acceptable as ichnotaxobases (Keighley and Pickerill 1996). Although the specific trace-maker is seldom known, this is no different conceptually than studying a vertebrate animal using only teeth. Because so many proposed observations and criteria for ichnotaxobases are subject to researcher interpretation, it is desirable to develop an objective, numerical means of trace fossil classification. The advantages of a numerical method over a subjective method are that it should be reproducible and systematic, and can introduce some controlled subjectivity that is preferable to uncontrolled subjectivity. The morphological variation among trace fossils will necessitate a slightly different list of positively scored characters to be encoded for each group of trace fossils studied, but the same basic techniques can apply to all such numerical analyses in ichnotaxonomy. Background of Phenetics Numerical phenetics, previously known as numerical taxonomy, is the quantitative method proposed herein as a new technique for the study of ichnotaxonomy. The technique of phenetics uses large numbers of characters to classify items (Sokal 1966), and thus is suitable for quantitatively analyzing the trace fossil morphological characters. Cluster analysis or other measures of similarity, using phenetic characters, might be used to reveal similarities among the ichnospecies of Arthrophycus without assuming patterns of descent (Blackith and Reyment 1971). Cladistics, the type of phylogenetic analysis now popular for biological taxa, might be considered problematic in the study of trace fossils. This is because an evolutionary “history” or relationships cannot be assumed for trace fossils that bear a resemblance to each other because of the multiple ways that many organisms can make very similar traces and because trace fossils do not breed or produce “offspring” that can have a meaningful history. Cladistics uses only shared derived charactes, which introduces polarity to the character states and emphasizes the effect of the outgroup, but phenetics bases affinities on both primitive and derived states, without weighting either. Methods The main portion of my research involved gathering all reports of Arthrophycus in the literature to create a database of all known occurrences. To do this, I created an annotated bibliography and worked outward, searching for papers mentioned in each source. I also used numerous intemet searches and databases, including Georef, Google Scholar, ION (Index to Organism Names), the Zoological Record, and the Paleobiology Database. These references then formed the data for my synthesis of previous work done on Arthrophycus and for histograms of reports of the ichnogenus throughout geographic time and space. I also entered these records into Microsoft Excel databases to condense the bibliography into essential points and organize references efficiently by time, space, and ichnospecies. A subset of these references was in languages other than English, requiring translation for proper evaluation of the specimens described therein. Two of Harlan’s original descriptions and part of depert’s involved some Latin, which Dr. Robert Anstey of Michigan State University kindly translated. I was able to use “Google Translate” for languages that are based on the Roman alphabet, including seven papers in French, eight in German, fifteen in Portuguese (mostly Brazilian), six in Spanish, and one in Romanian. For all of these, I transcribed the papers into Microsoft Word by hand and then fed them into the translation program online. Foru' references in Chinese, including two descriptions of new species, were beyond the powers of Google Translate. Dr. Yue Li of the Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, kindly translated the relevant portions of these sources. All of these papers were potentially subject to error throughout the transcription and/or translation process. Although I made every effort to transcribe papers accurately, I had little familiarity with most of these languages and was unlikely to catch any typing mistakes. “Google Translate” and Drs. Anstey and Li may also have been fallible. However, I did find several mistakes when the translation program returned a nonsense answer, and the resulting translations made enough sense that I am confident that they retained the salient points. In addition, only four of the papers for which I used “Google Translate” described new species, for which detail and precise wording is most important; other papers that were translated were only reports of Arthrophycus occurrences. Of the four papers describing new species, three were in German, the language with which I am most familiar and would be most likely to catch any errors in transcription or translation. Besides the literature search, I traveled to seven museum collections to examine I as many specimens of Arthrophycus in person as possible. These visits were to the Cleveland Museum of Natural History (CMNH), the Ohio State University (OSU), the Geier Collections of the Cincinnati Museum (U CGM), the Yale Peabody Museum (YPM), the American Museum of Natural History (AMNH), the New York State Museum (N YSM), the Buffalo (NY) Museum of Science (BMS), and the Paleontological Research Institution (PR1). During each visit, I took digital photographs of Arthrophycus and noted qualitative characters on a data sheet designed for this purpose (Figure 1.2). I measured quantitative characters with a simple ruler at all the collections except for that of Yale Peabody, where I used digital calipers. I chose the individual candidates for the measurements of smallest and largest diameter mostly by eye, but made several measurements if I had any doubt as to which was the smallest or largest individual. While at these museums, I also examined specimens of genera supposedly related to Arthrophycus, including F ucoides, Daedalus, and the occasional unmarked mystery cabinet. I examined specimens of F ucoides because Arthrophycus was originally classified under that name; the search produced a number of specimens that were actually Arthrophycus, as well as a few specimens of true seaweed fucoids that served as usefirl comparisons and revealed insights into the ichnogenus history. Ichnospecies of Daedalus may be related to Arthrophycus by morphology or trace- maker and thus merited inspection for purposes of comparison. The mystery cabinets occasionally produced unlabelled specimens of Arthrophycus, including an unnamed and unnumbered specimen found at the New York State Museum and described herein. For most specimens, I had to rely on published photographs and drawings; many of these are grainy or otherwise indistinct. In an attempt to surmount the limitations of poor reproductions, I requested loans or better pictures from authors and museum staff, but rarely received a response. Efforts to request loans of specimens were not successful either; partly because trace fossils commonly are not given catalog numbers. As a result, I have had to rely on the published figures for many Arthrophycus specimens. Once I had collected my data and published reports of Arthrophycus, I began to condense the data and establish a list of diagnostic characteristics for the ichnogenus, which I used to evaluate each proposed occurrence and ichnospecies. I assessed each occurrence as belonging to one of three categories: conformable to Arthrophycus, unverifiable (usually noted only in a list without a detailed description or figures), and not conformable to Arthrophycus. If the published photographs were of poor quality or questionable as to their characters, but the written description included references to key morphological characteristics of Arthrophycus, I made decisions based primarily 10 on the written description. A new ichnospecies, however, must have a good quality photograph that clearly shows salient features of Arthrophycus to be considered as a valid member and new species of the ichnogenus. 11 Figure 1.1a: Sketch of Arthrophycus alleghaniensis, from Hall (1852). Silurian, Medina Sandstone, New York. Figure 1.1b: Sketch of A. alleghaniensis, from Hall (1852). Silurian, Medina Sandstone, New York. Arthrophycus Data Sheet Date: ID# Taxon Locality Age Stratigraphic unit and lithology Diameter max/min Approx. # of traces Length of “full” traces Branching (true, pseudo, not) Spreite or backfill Shape of cross-section Shape of overall structure Bilobate Chevrons Annulations (width if present) # planes Orientation with respect to others Other structures, trace or body fossils, features Sketch: Figure 1.2: Data sheet used to characterize museum specimens of Arthrophycus. l3 Chapter 11. History of Arthrophycus and definition of characters History of Arthrophycus The history of Arthrophycus begins with initial descriptions of specimens as fucoids by Eaton (1820), but he did not name the specimens formally. A formal name had to wait until 1831, when Harlan described the first ichnospecies of what is now Arthrophycus under the name F ucoides. He described F ucoides alleghaniensis n. sp. in that year and then followed it with a description of F. brongniartii n. sp. in December 1831 (published in 1832). As was common at the time, Harlan placed both new species into the group of fucoids, or “seaweeds.” However, the fucoid group was a common wastebasket designation at the time (Dawson 1964) and often included fossils that were not “seaweeds” at all (Shimer and Shrock 1944). Eaton (1832) referred to fossil specimens that have since been placed in Arthrophycus on the basis of the descriptions (the actual types are lost) (Rindsberg and Martin 2003). Eaton’s fossils, from the Medina Sandstone, were described as branched, long, curved, and occurring in layers. Eaton placed the specimens into Encrinus giganteus Eaton, 1832, later placed in synonymy with Arthrophycus (his specimens may have been either A. alleghaniensis or A. brongniartii) (Bassler 1915, Rindsberg and Martin 2003). The next account of Arthrophycus (as F ucoides) came fi'om Taylor in 1834, who described abundant specimens of F alleghaniensis in Pennsylvania, near the locality where Harlan had found his original specimens. Taylor followed this paper with another in 1835, in which he described finding numerous F. alleghaniensis specimens in situ, as well as specimens of the rarer F. brongniartii. 14 In a paper read in 1834 but published in 1836, Harlan briefly mentioned the two species of F ucoides that he had previously published and noted some particular points in which his species differed from specimens that other authors had referred to them. In this critique, he distinguished three important characters of F. brongniartii that those other specimens lacked: grooves, branching, and a rougher texture of the surrounding sandstone matrix. Harlan also observed that the specimens he excluded from F. alleghaniensis contained a filling of “convex layers of sandstone,” which puzzled him, because he thought that his F ucoides specimens were similar to aquatic cryptogarnae (algae). In a last note to this paper, Harlan recorded that Mantel] (1834) had given an unrelated fossil the name F ucoides brongniartii, but that the specimens were not of the same form as those he (Harlan) described. Conrad also mentioned Harlan’s F ucoides in several reports (1837, 1838, 1839). In the first (1837), he referred to both F. alleghaniensis and F. brongniartii as present in the Medina Sandstone. In 183 8, he used the name F. harlani for the first time in the literature (“F ucoides harlani nobis (F. brongniartii Harlan)”) without giving a reason for the change in specific name. In 1839, Conrad again used the name F. harlani to describe Harlan’s fossils, which he found in abundance throughout New York, Pennsylvania, and Virginia. However, he noted that F. harlani was restricted to the Medina Sandstone, thus recording the first opinion that Arthrophycus is limited biostratigraphically. One author described Arthrophycus as a crinoid. De Castelnau (1843) placed specimens into the crinoid genus Crinosoma, noting that the specimens were poorly preserved and different fi'om any known crinoid form, and thus deserving of a new 15 generic name. No other authors have used the name since 1843, but Arthrophycus was used many times after Hall coined it in 1852, so Crinosoma is a nomen oblitum and Arthrophycus is a nomen protectum (Rindsberg and Martin 2003). Hall (1852) redescribed Harlan’s fossils as Arthrophycus in his great work on the paleontology of New York and referred to A. harlani, making no distinction between Harlan’s F. alleghaniensis and F. brongniartii (James 1893a also considered the two ichnospecies to be the same form). In the same year, G6ppert (1852) described Harlan’s F. alleghaniensis as Harlania halli (and also placed F. brongniartii under Harlania halli). In a case in which two taxonomic names are published in the same year, the first reviewer (Miller 1877) decides upon the name to be preferred in the future; Hall’s Arthrophycus prevailed (Rindsberg and Martin 2003). It is also possible that the relevant section of Hall’s (1852) paper was actually composed in 1850, thus taking precedence over the later work of depert (1852) (James 1893a). However, in Europe, Hall’s work was mistakenly dated as 1853, leading researchers there to use the name Harlania as late as 1966 (Gubler et al. 1966), but the practice is no longer common. The name Harlania is widespread only in Jordan, where an Ordovician layer there has been informally named the “Harlania Shale” (Selley 1972) and thus the name Harlania remains in use for the fossils as well (e.g. Eschard et al. 2005; Turner, Makhlouf, and Armstrong 2005). James and other researchers ultimately supported A. alleghaniensis as the best name for the species, noting that F. alleghaniensis was a valid name, adequately illustrated and with correct format by Harlan (1831), so there was no need for Conrad (1838) and Hall (1852) to change from this earliest species designation (James 1893a, 16 Rindsberg and Martin 2003). Some confusion remains: museum specimens are cataloged variously as F. alleghaniensis, F. harlani, A. harlani, A. alleghaniensis, H. alleghaniensis, and so on. Though most authors adhere to one name, at least one used two apparently synonymous names (A. harlani and A. alleghaniensis) in the same paper (Moneda 1963). Identity of Arthrophycus Some authors, beginning with Seilacher (1955), have proposed a merger of Arthrophycus with the ichnogenus Phycodes. However, the two ichnogenera possess distinctive features and differences (e. g. size, prominence of annulations, relation to sediment as seen through internal structure) that merit separate ichnogeneric designations (Osgood 1970). Individual Arthrophycus specimens are not always observed in bundles, which is additional evidence against a synonymy with Phycodes, which is always bundled (Uchman 1998). The idea of merging the two ichnogenera has now largely disappeared (Hantzschel 1975, Pickerill et al. 1984). Even once the fossils now known as Arthrophycus had a name that most ichnologists could agree upon, their biological affinity still remained in question. Along with being considered blooms or stalks of fucoids or other plants, Arthrophycus has also been identified as traces of many different types of other organisms or even as inorganic forms (Hantzschel 1975). The plant idea had its heyday from 1831 to 1901 (e.g. Harlan 1831, Schimper and Schenk 1890, Herzer 1901) and was resurrected briefly in the mid-twentieth century (Becker and Donn 1952, Duimovich 1963), but the hypothesis has since received no support. 17 The suggestion that Arthrophycus could be inorganic is actually far simpler than it first appears. No one has ever proposed that the entire ichnogenus is inorganic; rather, a few isolated specimens have been dismissed as actually inorganic. This is limited to two reports: that of Schiller (1930) and Kulkarni and Borkar (2002). Schiller’s paper described some putative Arthrophycus specimens from Argentina that he determined were tectonic products rather than organic remains, but he allowed other specimens of Arthrophycus in other rocks to be true fossils. In the second case, Kulkarni and Borkar (2002) determined that reputed Arthrophycus from the Proterozoic of India (described by Mukherjee et al. 1987) were inorganic in nature. Debate still rages as to the producer of Arthrophycus; suggestions include worms of many types, trilobites and other arthropods, gastropods, echinoderms, and various unknown animals (see Tables 2.1 and 2.2). Imprints of non-moving animals have also been suggested, including worms (Borrello 1966), crinoids (de Castelnau 1843), and octocorals (Fauchald, pers. comm). Though the most popular suggestion is that some kind of worm-like animal made the traces known as Arthrophycus (including works up to 2009, e. g. Miller et al. 2009), many of the more recent publications favor an arthropod producer, whose appendages would have been best-suited to make the ridge-like annulations seen in most specimens of the ichnogenus. In a personal communication, Kristian Fauchald of the Smithsonian Institute, an expert on modern polychaete annelids, indicated that an annelid would be unlikely to make these annulations while moving (Brandt et al. 2008). An interesting new candidate for the trace-maker is T anazios dokeron, the Silurian crustacean-like creature fiom England 18 figured in Siveter et al. (2008). However, a full discussion of the trace-maker is outside of the scope of this research. “Official” Diagnosis of Arthrophycus Given its varied and confirsing history, the diagnostic characteristics required for Arthrophycus have become rather blurred over time. Therefore, a review of the original descriptions is in order. Harlan’s original Latin (1831) description of F. alleghaniensis reads as follows: “F ucoides alleghaniensis: F rande compressa, rugata; apice recurva, obtuse; ramis inequalibus, digitatis etfastigiatis, enervibus, nudatis.” This translates to “frond compressed, wrinkled, recurved bluntly at its apex; branches unequal, fingerlike structure in a separated bundle, weakly differentiated and stripped of any coverings.” He then went on to say: “They lie upon each other three or four layers deep. . .project in bold relief from the surface, with their distal extremities disposed in every direction; they appear to have been of different ages, and vary in size accordingly from two to five inches in length, the largest being eight tenths of an inch in thickness. In breadth they vary from one to five tenths of an inch. . . gently arched from the base towards the apex, and more or less recurved at top; in every instance the apex is curved downwards and sinks into the stone. The superior surface of both the stalk and branches is cylindrical, transversely wrinkled by irregular channels, and marked by a longitudinal depressed line. . .They have grown in such profirsion and are so crowded together that the commencements or bases of the stalks are for the most part concealed. . .The branches are compressed laterally as well as the stalk.” Hall’s (1852) original description of Arthrophycus is as follows: “Stems simple or branching, rounded or subangular, flexuous, ascending, transversely marked by ridges or articulations. The species of this genus yet known consists either of simple elongated stems of nearly equal dimensions throughout, or those which divide near the root into several branches and afterwards remain simple.” l9 Gfippert’s (1852) description of Harlania, translated from the Latin, is: “straplike simple turflike aggregate or dichotomous branch, in younger states longitudinally sulcate, branches of adults subcylindrical interrupted transversely by elevated ridges.” In the Treatise of Invertebrate Paleontology, Hantzschel (1975) described ' Arthrophycus as: “Bundles of armulated curved burrows, simple or branched, subquadrate in cross section, mostly I to 2 cm. in diameter, up to 60 cm. long, commonly bilobate with median longitudinal depression; surface showing strong, very regularly spaced transverse ridges; internal chevron-shaped filling.” When analyzing the four above descriptions, several important points emerge. All four authors mentioned five characteristics: transverse ridges or annulations, simple overall form, branches, bundles, and a somewhat compressed cross section. Therefore, to be considered a valid ichnospecies or specimen of Arthrophycus, the specimen should possess most or all of these five characters. Harlan (1831), depert (1852), and Hfintzschel (1975) mentioned a longitudinal depression, as did Hall (1852) in his separate description for A. harlani. In addition, Hamzschel (1975) mentioned an internal filling and Hall (1852) mentioned nearly equal dimensions throughout, both of which have been noticed by subsequent authors. Finally, three of the authors agreed on an approximate diameter of 1-2 cm for their specimens. These characters shall all be discussed more fully below. Taxonomically Important Characters from Original Diagnoses The following characters are shared among the original descriptions of Arthrophycus by Harlan (1831), Hall (1852), depert (1852), and Hiintzschel (1975). 20 Annulations Annulations are present in all the proposed ichnospecies of Arthrophycus examined herein, with only a few exceptions in which the character is faint or not mentioned (e.g. some specimens of A. tenuis); no proposed ichnospecies is without them completely. Harlan (1831) referred to transversely wrinkled channels, Hall (1852) to transverse ridges or annulations, and depert (1852) and Hantzschel (1975) mentioned transverse ridges. Possession of this character state is thus an important requirement to be Arthrophycus (Figure 2.1); specimens lacking annulations are very questionable (e.g. Pettijohn and Potter 1964) and may instead be Paleophycus, Phycodes, or other traces that are otherwise similar in form to Arthrophycus. Simple form Simple form or structure is another common theme in Arthrophycus descriptions (Figure 2.2). Hall (1852), G6an (1852), and Hantzschel (1975) used the term “simple,” while Harlan (1831) described his fossils as “fingerlike” and “stripped of any coverings.” These structures may be branched or form bundles, but ultimately the individual trace is simple in form. This shape is rarely perfectly straight but curves or arches, as both Harlan (1831) and Hantzschel (1975) wrote in their descriptions and Hall (1852) added in his description of A. harlani. Branches Branches are also a common feature: Harlan (1831), Hall (1852), depert (1852), and Hantzschel (1975) all included branching for Arthrophycus and most other authors do as well. However, some proposed ichnospecies of Arthrophycus (e.g. A. brongniartii) do not exhibit branching, and the lack of branching is often cited as one 21 of the reasons for a distinct new ichnospecies designation. The presence or absence of branching need not be a problem, for both Hall (1852) and Hantzschel (1975) observed that specimens may be simple or branched, likely alluding to this variation. If branching is present, it may manifest itself as “simple branching” with few (typically two to three) structures involved (Figure 2.3), “pseudo-branching” (unrelated branches cross over one another) (Figure 2.4), or “palmate branching” (many branches form a bundle or fan, as discussed below) (Figure 2.5). M Bundles or aggregates of traces feature in all four of the above original descriptions. Traces may appear to converge into one thick and relatively narrow bundle (Harlan’s (1831) “stalk”) (Figure 2.6), or into a fan-like structure that is largely two-dimensional (Figure 2.5). As with branching, bundling is a common and well- known character, but is not essential to all Arthrophycus (e. g. A. brongniartii). The particular type of bundling in A. Iateralis Seilacher 2000 is part of the justification for its designation as a separate ichnospecies (Seilacher 2000). The bundling prominent in A. alleghaniensis, the most familiar ichnospecies of the ichnogenus, likely led to the ichnospecies’s initial description as plant-like. The different types of bundling (and also branching) may intergrade, as an individual trace may emerge from its bundle to be predominantly straight and independent or may cross other traces (Figure 2.7). It is often impossible to trace a particular trace for very long before it descends into the matrix (Figure 2.8), further complicating the issue. 22 Cross-section Cross-section is the final common character from the four original descriptions of Arthrophycus. Although Harlan (1831) referred to his specimens as “cylindrical,” he also mentioned compression of the branches and stalk. Hall (1852) described his specimens as “rounded or subangular,” depert’s (1852) were subcylindrical, and Hantzschel (1975) used the term “subquadrate.” Of the twenty-three Arthrophycus ichnospecies, seven are characterized as subquadrate, eight are cylindrical or elliptical, and eight do not include a description of the cross-section. Median mve A median groove or longitudinal depression (Figure 2.9) was mentioned in the works of Harlan (1831), depert ( 1852), and Hantzschel (1975), while Hall (1852) included the feature in his description of A. harlani. The groove is a common feature in Arthrophycus specimens and is found in nearly half of the purported ichnospecies (nine of them show it, nine do not, and the remaining four are unclear in the published photograph or not mentioned in the written description). Some authors have mentioned the difficulty of preservation of such a delicate feature (MacNaughton and Pickerill 2003, Miller et al. 2009) and even specimens that are otherwise clearly Arthrophycus do not have a groove or have only an indistinct median groove (Figure 2.10). Regardless, the feature is both common in specimens and well-known. Internjgl structur_gs Internal structures are not mentioned often in published descriptions of Arthrophycus, as recognition of them requires cut and polished or broken specimens in which a cross-section is observable (Figure 2.11). Nevertheless, internal structure may 23 prove to be a diagnostic feature for Arthrophycus. Among the primary descriptions of Arthrophycus given above, only Hantzschel (1975) described the filling, as “chevron- shaped,” though Harlan noted internal structures in a later paper (1836). Sarle (1906) presented the internal structures of Arthrophycus to determine how the animal burrowed and then used those structures as evidence that Arthrophycus was a trace fossil rather than a plant or organism. Seilacher (2000) has used variation in internal structures as a basis for subspecies of Arthrophycus (A. linearis subsp. protrusiva and subsp. retrusiva). Constant diameter Constant diameter of individual traces featured in only Hall’s (1852) work among the four primary descriptions, but it is used often as an important characteristic of Arthrophycus. Except where branching and bundling make the diameters of individual burrows indistinct, Arthrophycus has a very consistent diameter along its length, as subsequent authors have observed (e. g. Dalloni 1934, Neto de Carvalho et al. 2003, Kumpulainen et al. 2006). __S_iz_e Size (width or diameter) of individual Arthrophycus traces is the final common character among three of the four authors above: the descriptions of Harlan (1831), Hall (1852), and Hantzschel (1975) agreed on an approximate diameter of 1-2 cm. This approximate diameter remains typical of the Arthrophycus ichnogenus (Tables 2.3 and 2.4), but reports for A. alleghaniensis alone range from a minimum of 0.25 cm (Lifian 1984) to a maximum of 6 cm (Burjack and Popp 1981). However, the smallest and largest measurements are not typical of the ichnospecies and some of them may 24 not even be validly assigned to Arthrophycus. A more typical range for A. alleghaniensis is 0.5 to 1.5 cm (Table 2.3); other ichnospecies tend to have rather small ranges of reported sizes, often with a difference of only a few millimeters between the minimum and maximum, because fewer authors have remarked upon them or reported new occurrences (Table 2.4). Preserved burrow length is not always reported, and this measurement is far less reliable than that of diameter because burrows of Arthrophycus do not have distinct starting or ending points, but weave under one another or disappear off the edge of the preserved rock slab (Sarle 1906). Because the full length of Arthrophycus traces is seldom preserved, authors who do report length measurements often include very large ranges or simply note the longest preserved burrow that they could follow without doubt as to its identity (e. g. Hantzschel 1975). Although size is not usually considered a valid ichnotaxobase because it may be more closely correlated with the size of the trace-maker than the actual trace morphology (F firsich 1974, Bertling et al. 2006, Bertling 2007), authors have used it nonetheless (e. g. Osgood 1970, to differentiate Arthrophycus from Phycodes; also Magwood 1992); size is also an easy character to measure and may help to reveal the trace-maker’s identity. At least one ichnotaxonomist allows that size may be acceptable for ichnospecific designations, but probably not for ichnogenera (Pickerill 1 994). Additional Characters Other characters appear in subsequently published descriptions of Arthrophycus but were not mentioned in the cited primary descriptions. These other characters 25 include wrinkles or fine ornamentation, orientation with respect to the substrate and to other individuals of Arthrophycus, and association with non-Arthrophycus traces. Some authors have questioned whether a single trace can be classified as Arthrophycus or if Arthrophycus must refer to an aggregate of many traces, preferably in bundles (Pickerill et al. 1991, Femandes et al. 2002). Furthermore, some workers have wondered whether Arthrophycus is always found in monoichnospecific assemblages or if it can be associated with other body or trace fossils (Méngano et al. (2005a)). Because none of these characters was mentioned in any of the four original descriptions, it remains in question whether these additional characters should be accepted as diagnostic or not. These additional characters are discussed more fully below. Wrinkles Wrinkles or fine ornamentation on the surface of the trace require exceptional preservation and photography. In most cases, one or both of those is lacking in published descriptions. However, fine wrinkles along the annulations have been seen in A. alleghaniensis (Figure 2.12) and A. brongniartii (Rindsberg and Martin 2003) (and as A. linearis, by Seilacher 2000). Oriept_ation with respect to substrag The orientation with respect to the substrate is a feature that may elucidate what the trace-maker was doing in or on the sediment. Arthrophycus traces commonly lie more or less parallel to bedding; some specimens show gentle arching in the vertical dimension (Figure 2.13). The traces may have several layers of primarily flat-lying traces atop one another (Figure 2.14). Others, however, twist in many directions and in 26 all three dimensions (Figure 2.15) and some traces may run perpendicular to the bedding and to surrounding Arthrophycus (Figure 2.16). Orient_ation relgtive to omega—es The orientation with respect to other traces is also important in Arthrophycus and may help to differentiate among ichnospecies. These include A. lateralis (Seilacher 2000), which has a particular pattern to its palmate form, and A. parallelus (Brandt et al. 2010), which is primarily parallel. Apart from branching or bundling, described Arthrophycus traces can have different orientations with respect to one another. Some are markedly parallel, others form a radial pattern, and most intersect each other in a way that looks similar to branching (termed pseudobranching above). In large slabs containing many Arthrophycus, there are even more patterns of ’ orientation, such as “twisting” (Figure 2.17) and “loops” (Figure 2.18). Of course, in instances of only one trace, this character does not apply. Number of traces The number of traces necessary to be considered Arthrophycus has been a subject of debate. As Pickerill et al. (1991) stated, isolated specimens of Arthrophycus are “atypical and not clearly understood.” At least thirteen examples of published isolated Arthrophycus specimens exist, including both those placed in Arthrophycus with no ichnospecies designation and others that have been designated as new ichnospecies. New ichnospecies herein called “singletons” include both A. corrugatus F ritsch (1908) and A. dzulynskii Ksiazkiewicz (1977), while those placed in “Arthrophycus isp.” include Akpan and Nyong (1987), Bhargava et al. (1984), Lin et al. (1986), Stanley and Feldman (1998), and Terrell (1972). Oddly, the specimen that 27 Pickerill et al. (1991) placed under “Arthrophycus isp.” was also an isolated specimen. It should be noted that the “isp.” designation indicates that the authors were not sure of species designation, and thus it may also be questionable whether many of the specimens are correctly assigned to Arthrophycus. Conversely, Pickerill et al. (1991) classified as singletons specimens from Bjerstedt (1987) and Durand (1985), which in fact have multiple individuals of Arthrophycus. Pickerill et al. (1991) also misinterpreted a specimen described by Legg (1985), which may be a single stack of several traces of Arthrophycus. Some traces reported as Arthrophycus have very few individuals, perhaps as few as two or three forming a single branch structure (e. g. Luo et al. 1994). This is particularly characteristic of the Cretaceous examples described in several publications by Howard and Frey in the 19608 and 19708 (e.g. Howard 1966, Frey and Howard 1970, Frey 1972). Associated fossils Associated trace or body fossils are not common in slabs with Arthrophycus specimens. Arthrophycus is usually included in the Cruziana ichnofacies, which is known for a predominance of cubichnia (resting traces) and arthropod tracks and burrows, without many grazing tracks, and is characteristic of shallow marine deposits above wave base (Seilacher 1964a). Unsurprisingly, Cruziana and the related Rusophycus (Figure 2.19) are the most common ichnofossils cited in association with Arthrophycus (e.g. Turner and Benton 1983, Seilacher et al. 2002, Poire' et al. 2003, Mangano et al. 2005a), as well as specimens of Daedalus (Miller et al. 2009). Arthrophycus has been noted in the same assemblage as Teichichnus (Paczesna 1996) 28 and specimens of Dictyodora (Orr 1994) and the lump-like Lockeia have also occurred with Arthrophycus (Brandt et al. 2010) (Figure 2.20). The only reported case of a body fossil occurring “from sedimentary sequences containing analyzed A. alleghaniensis” is the centipede-like fossil found by Baldwin and Strother (2004), who suggested centipede-like animals as the trace-maker of Arthrophycus in part as a consequence of that association. Gradation Between Arthrophycus and Other Ichnogenera Part of the joy in interpreting trace fossils is that one animal can make numerous traces and numerous animals can produce or use the same trace (Osgood 1975b, Bromley 1990, Magwood 1992, Minter et al. 2007). Many animals make similar burrows for similar reasons or with similar behaviors (Bromley 1990); the simple U-shape found in burrows such as Diplocraterion is an example (Osgood 1975b). In addition, traces made by one animal can also be preserved differently in different substrates (Osgood 1975a, Bromley 1990). Any of these problems may produce burrows that appear to be the same, or make unreal differences apparent (Osgood 1975b). Gradational traces are particularly well-known between the probable trilobite traces Cruziana and Rusophycus (Keighley and Pickerill 1996). Arthrophycus is hardly immune to these problems of gradation, as the plethora of potential trace-makers indicates. Similarity of body plans among such large groups as the annelids, priapulids, and nemerteans makes a positive identification of individual producer difficult and differentiation among similar makers with similar behaviors almost impossible (Osgood 1975b). Some evidence of intergradations between Arthrophycus and other ichnogenera exists in the Silurian of Alabama, where 29 specimens of Arthrophycus show a similarity to Nereites biserialis, Rusophycus-like structures, and possibly others (Rindsberg and Martin 2003). However, such intergradations are rare in Arthrophycus. Suprageneric Classification ofArthrophycus Although trace fossils need not be organized beyond the generic level (ICZN 1999), a few authors have attempted higher hierarchy levels. In one such scheme, Arthrophycus was grouped most closely with Rhabdoglyphus Vassoevich, 1951, and Scalarituba Weller, 1899, in the superichnofamily Unilobatoidea, with Cruziana placed in the same ichnoorder (Mikulas 1992). However, such schemes are uncommon and other authors have not chosen to follow them; descriptions of ichnogenera are instead listed in alphabetical order or grouped by ethology (repichnia, cubichnia, etc.). In a recent attempt at a suprageneric classification, Seilacher (2000 and 2007a) grouped Arthrophycus with Daedalus and Phycodes to create the family-rank arthrophycids, together with the teichichnids. He grouped these ichnofossils based on what he perceived as their similar “fingerprint” of regular annulations, backfill structures, and blind tube endings. ’ Other authors support Seilacher on this point, including Sarle (1906) and Brett (pers. comm), noting the similarity between Daedalus and Arthrophycus in particular. Although some of the spiral-fonn ichnospecies of Daedalus initially appear to be very different from the simple structures of Arthrophycus, others are quite similar, possessing annulations, a possible median groove, and occasionally a bundle or fan-like structure (Durand 1985) (Figure 2.21). 30 Morphological Issues Raised in Previous Work Mtons vs. multiple tra_c_e_s Although Arthrophycus is commonly described as a gregarious trace, occurring in the hundreds on larger specimens (e. g. Miller et al. 2009), other authors have attempted to include in the ichnogenus specimens that consist of a single trace (e. g. Ksiazldewicz 1977, Bhargava et al. 1984, Pickerill et al. 1991). In their descriptions, Harlan (1831), Hall (1852), and Hantzschel (1975) referred to groups of traces and their orientations to one another (branching and bundling), although the first two authors interpreted Arthrophycus as a plant or mass of plants rather than groups of traces. More recent authors have excluded reports of singletons fi‘om Arthrophycus (e. g. Mangano et al. 2005a). In some cases, a reported specimen is a singleton based on the sampled rock size — only one or a few traces can fit on the surface, or a trace may have broken from the rest of the matrix (Figure 2.22). However, this is not the case for most published specimens — the rest of the rock surface is devoid of traces (e. g. Akpan and Nyong 1987, Pickerill et al. 1991). Many questionable assignments to Arthrophycus are either singletons (e. g. Akpan and Nyong 1987) or single branches of one or two traces (e.g. Howard and Frey 1966). Some museum collections include single traces or structures that have been referred to Arthrophycus, likely in error as they lack diagnostic features of the ichnogenus (Figure 2.23). Published descriptions of putative Arthrophycus singletons encourage proliferation of the concept despite the departure from the type concept that these specimens represent. 31 If trace fossils are understood to be records of ancient behavior, then a propensity to be gregarious is a part of that behavior, whether it is true “social behavior” (many individuals making traces at close to the same time) or merely apparent (a few individuals making many traces in the same spatial area). Behavior can be a morphologic character (and thus a legitimate ichnotaxobase) because it is part of the “extended phenotype” (Dawkins 1982) of the maker of Arthrophycus. Therefore, gregarious behavior, as evidenced by multiple co-occurring traces, should become part of the ichnogenus diagnosis, and that traces that are too small 0r broken to properly display that behavior but containing all other characteristic traits be termed “Arthrophycus-like.” Median groove Much of the Arthrophycus research summarized above raised the question of the taxonomic importance of the median groove. A median groove was included in the initial descriptions of Harlan (1831), Hall (1852), depert (1852), and Hantzschel (1975), as well as in the diagnoses of many other authors, for A. alleghaniensis and most of the other ichnospecies and Arthrophycus isp. reports. However, a groove can be effaced, either by a lack of substrate cohesiveness or dewatering before lithification, or by post-lithification weathering. Some authors reported a median groove in their specimens, but wrote that it was faint or not present along the full length of the trace or traces (e.g. Metz 1998, F emandes et al. 2000, Kumpulainen et al. 2006). In addition, recognition of a “groove” can disappear upon closer inspection. This is evident in a number of the specimens of Arthrophycus from the classic Medina Sandstone examined during museum visits. The annulations in these specimens are not 32 really “ring” structures, but exhibit a “pinched” appearance toward the middle of the trace (Figure 2.12). The ridges still have a faint dip toward the middle as in the groove, but the shape of the annulations is the primary contributor to the appearance of a full groove. One of the figures of Harlania in Bender (1963) also shows this, demonstrating that the feature is not limited to the Silurian of New York, and it is likely that “pinching” of annulations creating the illusion of a median groove exists in other Arthrophycus specimens but has not yet been recognized; the poor quality of many published figures makes that even more difficult to determine. Because of problems with preservation, recognition, and reporting, some authors (e. g. Pickerill et al. 1991) suggested that a median groove need not be a diagnostic criterion for inclusion in Arthrophycus. In the review of Arthrophycus literature, presence of a median groove was considered support for inclusion in Arthrophycus, but, because of the preservation and recognition factors discussed above, the apparent absence of a median groove was not considered evidence that a specimen should be excluded fiom Arthrophycus. The presence or absence of a median groove should remain in the diagnosis for the ichnogenus because it is so common and should not be excluded just because it may be lost occasionally. As a final note, the median groove, where present, may influence the choice of the most likely trace-maker. A median groove or furrow is often interpreted as the work of a multi-limbed animal that cast sediment toward the midline of its body as it processed the substrate (in fodinichnia) or moved along the substrate (in repichnia). 33 Emhingand bunidles Although branching and bundling were mentioned explicitly by Harlan (1831), . Hall (1852), Goppert (1852), and Hantzschel (1975) in their diagnoses of Arthrophycus, a few authors have suggested that neither characteristic is diagnostic for Arthrophycus. This claim originated with Pickerill et al. (1991), from those who have cited that work (e. g. Femandes and Borghi 1996, Femandes et al. 2002), or from those who have placed singletons into Arthrophycus. However, the branched nature of Arthrophycus should remain part of the ichnogeneric diagnosis. Even the name Arthrophycus means “jointed plant,” referring to the perceived branching of the structures. Some branching may be pseudo-branching that occurred when individual traces crossed over one another, but branching as a character is present in anything that can be considered Arthrophycus. The type of branching, however, can differ among ichnospecies of Arthrophycus. Ichnospecies may have prominent or minor branching, but branches must be present in some form. Branching can take the form of prominent bundling and palmate shapes in A. alleghaniensis and A. lateralis or the cross-over pseudo-branches in the more independent A. brongniartii and A. minimus. Even A. parallelus, noted for its rather independent and predominately parallel traces, exhibits pseudo-branching in places. Some authors have noticed a difference in branching that they did not choose to use as evidence for a new species. Femandes and Borghi (1996) and Femandes et al. (2000) mentioned both a dichotomous form and one with “several new branches starting from a single point.” This may reflect the difference between the palmate 34 bundles of A. alleghaniensis and A. Iateralis and the simpler branching or pseudo- branching of A. brongniartii (all of these have been found by these authors in Brazil), or it may indicate something different. Size of traces over time As stated above, reports of width in Arthrophycus can vary tremendously, from 2.5 to 60 mm in one ichnospecies alone, and 0.5 mm to 60 mm in the entire ichnogenus (Tables 2.3 and 2.4). Small size may be primitive (Méngano et al. 2005a), or reflect environmental stress (MacEachem et al. 2005) or nutrient availability (Femandes 2001), or perhaps the age or growth stage of the trace-maker. A number of the smaller proposed ichnospecies of Arthrophycus are known from the Cambrian. These include A. qiongzhusiensis, A. strictus (specimens from Paczesna 1996 only), and A. minimus. Mangano et al. (2005a) suggested that small size is therefore a primitive characteristic of Arthrophycus. However, size may also reflect environmental stress (MacEachem et al. 2005). A number of the authors working in Brazil noticed a smaller size in Arthrophycus in the latter portion of the time range there (Early Devonian) (F emandes et al. 2000) and have posited that the small size is a result of nutrient availability (Femandes 2001). Mangano et al.’s (2005a) assertion that small size is characteristic of geologically older Arthrophycus is further countered by small ichnospecies from younger strata. Those ichnospecies in the Cretaceous and Cenozoic (A. strictus and A. tenuis) may not be valid Arthrophycus as discussed above, but A. parallelus, from the Carboniferous, has a size range within that of A. minimus. A. elegans, while not a valid ichnospecies but possibly validly included in Arthrophycus, also exhibits small size 35 and is from the Carboniferous. Small size is therefore not necessarily primitive in Arthrophycus. To test for trends, or lack thereof, of the size of Arthrophycus over time, I produced a graph of width over time (Figure 2.24). To do this, I recorded a list of all reported widths of Arthrophycus, excluding those that were designated only as “Arthrophycus isp.” and recording all reports for whatever ichnospecies name the authors had originally used (c. g. both A. brongniartii and A. linearis are in the graph separately) (see Tables 2.3 and 2.4). I also included an unidentified specimen fiom the New York State Museum in Albany, NY, designated as “Albany.” (See Chapter IV for further discussion of this specimen.) Most of the width measurements were reported as ranges (e.g. 3.5 to 4.5 m); I used the midpoint for each one. The dates of the occurrences are reported in the literature only as periods (Cambrian, Ordovician, etc.) rather than as absolute dates. To portray the time distances between periods as accurately as possible, especially in the case of the distances between the Paleozoic and the Cenozoic occurrences, the midpoint of the reported time period was used, using the International Stratigraphic Chart (Ogg 2008) for the dates. For example, the Cambrian period is calculated as [(542.0 - 488.3) / 2] + 488.3 to get a midpoint age of 515.15 million years. For reported ages that encompassed two periods, the boundary between those ages was used (e. g. “Ordovician to Silurian” is 443.7 million years). These dates are imperfect, but they are preferable to a scheme without numbers at all, which would require a non- numerical coding along the time axis. 36 The resulting graph showed some interesting trends (Figure 2.24). A. alleghaniensis was notable for its large range over both time and width range, but one must also note that the ichnospecies also had the largest number of measurements, by far (n = 27, compared to the next largest n = 4 for both A. linearis and A. tenuis). As noted earlier, there were small individuals throughout the range of Arthrophycus and no trend in size over time, though the larger individuals were concentrated in the middle of the time range (Silurian) (see Figure 2.25 for a close-up of this crowded region). A few of the ichnospecies, such as A. linearis and A. brongniartii, had a larger range in size, but most ichnospecies had smaller ranges or few reports. Annulations Although most authors noted annulations in their photographs or drawings and mentioned it in their descriptions, defining what is meant by the term “annulations” was not self-evident. In the four primary descriptions of Arthrophycus, Harlan (1831), Hall (1852), depert (1852), and Hantzschel (1975) all observed annulations in their specimens. However, a definition of the term was rarely, if ever, given. Annulations are raised bands separated by thin grooves, perpendicular to the long axis of the trace, and are usually regularly spaced. With closer inspection, it was evident that not all annulations in ichnospecies placed in Arthrophycus have the same shape: some are concavo-convex like cupped shells, some have a zippered chevron appearance that has more of a v-shape than do the concavo-convex annulations, and some are biconvex like durnbbells. This biconvex shape is the form common in A. alleghaniensis and A. brongniartii and that appears to be pinched in the middle. 37 Table 2.1: Authors who have suggested some kind of worm-like animal as the trace- maker of Arthrophycus. Reference Specific Suggestions Abel 1935 Lanice-like Borghi et al. 1996 polychaete Downey 1980 segmented annelid, e. g. Nephtyidea Femandes et al. 1995 sedentary polychaete Femandes et al. 2002 worm-like Hantzschel 1975 Konate et al. 2003 Ksiazkiewicz 1977 polychaete Kumpulainen et a1 . 2006 Lin et al. 1986 Mangano et al. 2005 coelomate Miller et al. 2009 endobenthic annelid or annelid-like Moore 1933 Neto de Carvalho et al. 2003 coelomate Pickerill et al. 1991 sedentary polychaete Sarle 1905 Sarle 1906 polychaete Schuchert 1916 lobworm-like Seilacher and Alidou 1988 Seilacher et al. 2003 Seilacher 2007a 38 Table 2.2. Additional suggestions for the Arthrophycus trace-maker. Reference Organism Specific Notes Konate et al. 2003 Trilobite Kumpulainen et al. 2006 Trilobite Neto de Carvalho et al. 2003 Trilobite Rindsberg and Martin 2003 Trilobite trinucleine trilobite Baldwin and Strother 2004 Arthropod centipede-like Borghi et al. 1996 Arthropod Brandt et al. 2008 Arthropod non-trilobite, long-bodied arthropod F emandes et al. 1995 Arthropod Hantzschel 1975 Arthropod Kumpulainen et al. 2006 Arthropod xiphosuran Mangano et al. 2005 Arthropod Neto de Carvalho et al. 2003 Arthropod (for A. brongniartii) Pickerill et al. 1991 Arthropod Ksiazkiewiecz 1977 Gastropod (for A. dzulynskii) KsiaZkiewicz 1977 Echinoderm (for A. annulatus) Sarle 1906 Echinoderm ophiuroid Brandt et al. 2008 Unknown poorly-fossilized Pemberton and Risk 1982 Unknown soft-bodied “experiment” Schimper 1869 Unknown no modern analogue 39 m2: c Eat oofiEemo no: 56> moor no 9 ow ow 9 or :oEom SEE. moor mw 2 m 595 .oEEw 39 m2: : Eat voumEemo ow 2 e7 Locom=om mwoew m2: e E9: boomEemo or: _cm_omc_o Locom=om meow or 9 m 3:038 eoew . or 9 m Ntmzw x505 ewe? me: e E0: uoLmEemm or Be: m when. meow w 2 v._. em 2 mm 3 9 o ocmEom =5on 52 On eeor oumsl oo 9 ow “85:30:; go: 02 moor : E E0: uBmEemo mw ww 9 m moucmEon. o:_m>._mo me 202 meow m9: c Em. o7 eno>> Lo==2 eoew w 2 F oer 9 m: o 8 v Hos. moor v 2 mw cmcj veer w 2 F or o“ m 5&5: 559: E3. oeow 3:05 E cozm wwv 9 mom mow... coco: 52. eoo 9 m: ew 9 e7. _ocomuch one? moses c. :96 Ewe 9. n.0,. __mI weer m9: e E0; umfiEemo w 2 or E com moucmEo... weew wfiw 9 one we? em: 9 _.w.m Emuom moecmEon. eoew em 9 ow 9550 moor m2: c Eat ecumEemo or 2 or 8260 ewe? ow 9 m E260 32 ee 9 or on xomEm 52. oer 9 on we 2 m o=otom 89 m_. 96.22 Emcom moor was to oewr 526.3 Rev nouoz 523 .::< ~EEM £95.. .3255" w 353‘ w ..o£:< 3mm. :55 523 .maofimoomm fleeceeeemfie .Vmo mom? entomom “Ow 033. 40 xx xx xx seam 08:82 8825 F 8 2 8 N or one 03 32 28288 :2 323 3. 22.. as 685 88 43% xx xx xx 4% 82 2282 2 to 32 28885:. xx xx xx c.2885) 5 So: 82 geoléc- of. 55258 82 228828 8.0 2 3 08 2 a 3. 2 3 2225 25 :22 88 assess and 23 08 2 a we 2 3 «85.8 Ileana.» 88 3.252 xx xx xx 5528 52 2222.8 xx xx xx gm 88 ”eases 2: Eco: 33:8 8 28 or a 82.28”. 88 ”2922 xx xx xx 5%: :8 .58: o? 2 8 8 2m mg 88 82 3 xx xx xx 6%: 82 accesses: 855 5 5% 3o 3. 2 3 .85: 52 gimme” F 8 :22 lflgx E: leaga N 82 S 29 8.3:: 82 £3an xx xx xx 52E 8? alsaeraao .m 2: 220285528 82 8. ans so 558:8 88 88% 2.85. "6. E2; 8: a: as 5528 88 885‘ 2: 25232328 87 2 3 a- 558:8 88 88% xx xx xx 5828 32 88:: 2: cht uQmEewo ww Ber moocmEou 9:950 on 062 meow @Molér xx xx xx 8%: 82 8% is 23 £222 lug 88 Illegal 0.8 220 Sam 856: a 88 22E 8 2 ca 2: 2. use: ax :2 ”2288 use: 5.23:5 433.3% «353 22.3 33 8. .flaemwacfiozo .V 585 momoommoenom =m £586on SSERESKVE mofim eotomom ”1w 29a. 41 mm 9 : mé 2 e.w cmeroo cmEcoD eoew 2:22 go or 2 m we 2 we cmEcoD 89 £26.. ow 2 e_. e.w 9 me NoEoiwmfix :2. 23:8 .m. on 2 m w 2 v 562m somoeecmxoz veer 23:3 .m. we v 2 m :9 woow MacoEtS mw 2 we e 2 m mcmonomn. moor «$8.5m me oo 2 on e.w 2 m... Noioiumfix R2. macaw 2: e E0: eofiEewo em... or .6 Lo Lozom=ow woow 3&an on: e E0: uQmEemo oe_.r er 2 m... 0:950 ounce. eoew xoamEEm owzczcoo TN om... 42 Figure 2.1: Specimen of a “typical” Arthrophycus from the Silurian of New York, Medina Sandstone. Note regular annulations running perpendicular to the long axis. BMS E20774. Scale bar represents 1 cm. Figure 2.2: Specimen of A. alleghaniensis showcasing the rather simple form and finger-like shape of the traces. Silurian of New York, formation unknown. OSU8391. Scale bar represents 1 cm. 43 Figure 2.3: Example of branching (arrows) in A. alleghaniensis. Silurian of New York, formation unknown. OSU8391. Scale bar represents 1 cm. Figure 2.4: Example of A. alleghaniensis with pseudo-branching, in which two otherwise independent traces cross one another. Silurian of New York, formation unknown. OSU8391. Scale bar represents 1 cm. Figure 2.5: Wide palmate bundling in a specimen of A. alleghaniensis. Silurian of New York, Medina Sandstone. YPM3 8353. Scale is in mm. Figure 2.6: Specimen with narrower bundles of Arthrophycus. Silurian of Martinsburg, PA, Clinch Sandstone. CMNH3819. Scale bar is 2 cm long. 45 Figure 2.7: Specimens of Arthrophycus exiting various intergrading branching and bundling patterns. Silurian of Rochester, NY, Grimsby Sandstone. YPM207001. Scale bar represents 2 cm. Figure 2.8: Specimens of Arthrophycus that disappear almost immediately from their bundles and descend into the matrix (arrow). Silurian of Rochester, NY, Grimsby Sandstone. YPM207001. Scale bar represents 2 cm. 46 Figure 2.9: Specimen of Arthrophycus with a clear median groove (arrow). Silurian of western New York, Medina Group. YPM7365. Black portion of scale is 5 cm long. Figure 2.10: Arthrophycus with faint or indistinct median groove. This close-up of YPM7365 shows that even clear median grooves can appear indistinct depending on the magnification or lighting. Scale bar represents 1 cm. 47 Figure 2.11: Arthrophycus with spreiten (arrow) revealed on a polished cross-sectional surface. Silurian of New York, Medina Group. YPM150650. Scale is in mm. Figure 2.12: Specimens of Arthrophycus with fine striations (arrow) on top of the individual annulations. These annulations also display a pinched shape toward the middle, lending to the impression that a groove is present. Silurian of New York, Medina Sandstone. YPM150639. Scale is in mm. 48 Figure 2.13a: Specimen of Arthrophycus displaying vertical arch or bowing (arrow) as it passes over another specimen (or under, if the specimen is inverted as most are). Silurian of Rochester, NY, Grimsby Sandstone. YPM207001. Scale bar represents 1 cm. Figure 2.13b: Bundled specimen of Arthrophycus displaying vertical arch. Silurian of Para, Brazil, sandy shale. NYSM6176. Scale is in mm. 49 Figure 2.14: Sample of Arthrophycus in which many primarily flat-lying traces cross over one another. Silurian of Grimsby, Ont., Grimsby Sandstone. BMS E3799. Scale bar represents 2 cm. Figure 2.15: Complicated arrangement of Arthrophycus in which traces go in many directions. Silurian of Medina, NY, Medina Sandstone. NYSM2. Scale bar represents 1 cm. 50 Figure 2.16: F lat-lying Arthrophycus traces with other traces perpendicular to the bedding. Silurian of Lockport, NY, Medina Group. YPM508647. Scale bar represents 1 cm. Figure 2.17: Twisting of Arthrophycus traces with respect to one another, in three dimensions. Silurian of Rochester, NY, Grimsby Sandstone. YPM207001. Scale bar represents 2 cm. 51 Figure 2.18: Large loop pattern of Arthrophycus traces. Silurian of Rochester, NY, Grimsby Sandstone. YPM207001. Scale bar represents 2 cm. Figure 2.19: Arthrophycus (arrows) co-occurring with Rusophycus (circle). Silurian of Lewiston, NY, found loose. BMS E25610. Scale bar represents 2 cm. 52 Figure 2.20: A. parallelus co-occurring with Lockeia (arrow). Pennsylvanian of Michigan, Grand River F orrnation. University of Michigan Museum of Paleontology (UMMP) 73 822. Traces are approximately 4 mm wide. Spiral Daedalus; L.0rdovician ‘9,.v---u.,_‘ protrusive I *,-—q-.. /.’i ' D. desglandi Figure 2.213: Sketches of three different ichnospecies of Ordovician Daedalus, from Seilacher (2000). Scale bars are in centimeters. 53 Figure 2.21b: Polished cross-section of Daedalus archimedes, showing the spreiten of the trace (arrows). Silurian of Medina, NY, Medina Sandstone. YPM35825. Scale is in mm. Figure 2.21 c: Non-spiral specimen of D. archimedes, which appears more similar to Arthrophycus than the spiral version does. Silurian of Medina, NY, Medina Sandstone. YPM35822. Scale bar represents 1 cm. 54 Figure 2.22: Specimen of Arthrophycus. Silurian of Lockport, NY, Medina Group. YPM35814. I suggest that morphotypes known only from a few individuals, such as this one, be termed “Arthrophycus-like.” Scale is in mm. Figure 2.23: Example of a specimen referred to Arthrophycus (arrow), almost certainly in error, along with specimens of Gordia. Devonian of Ohio, Chagrin Shale. CMNH3705. Scale is in cm. 55 SE: .55 Someuoafiéxmo mowooomoS—om .«o onm “vww 93me SE: 2:: eew- eon- 8? com. 80. .85 e 58m gm e3 fie? E SE n b9»... 288 3585 ll Pam?“ «Seek . o 8 t manages... ceases L xo_aE_w 22232555 x 3.2.29". 22855.2 SEES 2.223 I 22955: , . mcwoom + __22289 a 3.8230 x 282.: x .3855 maisccf ., (“ll“) I-DPIM 25... .03 due. unexcnoéwi .o 35 56 .58 o 88m Sam one 88 m bog v.88 3:: 2.5 .vww .wE e. nafiw mo 8.8 .8638“. me Home. ”mww 05%.“. ASE. 25... com. albeit». 23:3 . $58.5... 2.3.2.5 e g 30.25 . 22238920 x.. 3.2.93 . 228.555. fl 35.5.... W 2223 J. 229.23... _... _ mcmeom. L 52280 e . mementoo X” 2.8:... x m ..ta_ceco._m A 3.225. N, , f m.mco.cm..eo__<0m oov. oom- 57 Chapter 111. Definition of Arthrophycus, Part 1.: qualitative critique of previous reports of Arthrophycus There have been twenty-three ichnospecific names included under Arthrophycus, but many of these assignments have been questioned (e.g. Uchman 1998, Rindsberg and Martin 2003, Mangano et al. 2005a). This section details the history of each ichnospecies name and evaluates its characters to determine whether it should be included within Arthrophycus. Table 3.1 provides a guide to the first publication of each proposed ichnospecies, Appendix I sums up all the reports, and Appendix II is a guide to all the character states possessed by each ichnospecies. A. alleghaniensis A. alleghaniensis (Harlan 1831) Hall 1852 was the first species published and therefore established the type species of the ichnogenus by original monotypy. Harlan’s original specimens, taken from a decorative stone in front of a Pennsylvanian tavern, were lost (Rindsberg and Martin 2003), leaving only his drawings (Figure 3.1). However, ample numbers of other A. alleghaniensis specimens remain, as the ichnospecies is common in outcrops of the Silurian MedinaSandstone. The Peabody Museum at Yale University alone has 58 specimens, for example. This is also the ichnospecies that is most ofien identified elsewhere, and its morphological characteristics are those defined by Harlan (1831) and Hall (1852), and discussed more fully in Chapter II. A. alleghaniensis has appeared in many works since its original descriptions by Harlan (1831) and Hall (1852). The following are critiques of reports of A. 58 alleghaniensis and its synonyms in the literature. These are listed alphabetically by author and chronologically in cases of multiple papers by the same author. Acefiolaza and Acefiolaza (2003) reported A. alleghaniensis from the Cambrian-Ordovician Balcarce Formation of Argentina. The authors did not include a detailed description, but the figure of Arthrophycus cf. alleghaniensis conforms to the original concept, as observable features include a mass aggregate of finger-like shapes with annulations and possible median grooves. Their figured Arthrophycus isp. is probably also A. alleghaniensis, for the specimen has those same features. Baldwin (1977) described A. alleghaniensis fi'om the Cambrian-Ordovician Barrios Formation of Spain. He noted the regular annulations, general curving shape with some bundles, and retrusive spreiten in cross-section. He also reported a large number of individuals of varying sizes (5 to 25 mm wide). Baldwin’s figure shows a mass of these structures, which conform to A. alleghaniensis on the basis of the annulations, shape, bundles, and retrusive spreiten, though Méngano et al. (2005a) suggested that, for their lack of bundles, they might be more properly assigned to A. brongniartii. It is also possible that the specimens represent an intergradational form. Bender (1963) reported “Harlania alleghaniensis” from the Silurian Sabellarifizx Sandstone of Jordan. He included no written description, but did discuss where and when the ichnogenus had been reported in previous work. His two figures show traces with annulations, median groove, and the general shape found in Arthrophycus; one even shows the “pinching in” of the annulations discussed below. The general shape of the specimens figured may indicate A. brongniartii rather than A. 59 alleghaniensis, but these specimens can be regarded as Arthrophycus on the basis of the presence and shape of annulations, median groove, and general shape of the trace. Bender (1968) also reported Harlania alleghaniensis from the Silurian “worm- burrows sandstone” of Jordan and included two figures. One of these shows A rthrophycus from the middle Ordovician of Jordan; the other shows a cross-section from the same location. The first figure shows an A. brongniartii-like shape and orientation, with annulations and a median groove. The cross-section shows spreiten, but Bender did not discuss them. These specimens conform to the definition of Arthrophycus. Borghi et al. (1996) reported A. alleghaniensis from the Silurian Pumas Group of Brazil. The authors thought of Arthrophycus as monospecific and so referred their specimens to A. alleghaniensis, but the relatively isolated nature of the individuals described is probably better suited to A. brongniartii. Without a more detailed description and figure, this report remains doubtful to some degree. Borrello (1966) described A. alleghaniensis fi'om the Ordovician-Silurian (‘?) La Tinta Formation of Argentina. The dating of the La Tinta Formation has been questioned (e. g. Mangano 2005a), and Borrello himself was not at all precise in his paper, indicating only that the formation was early Paleozoic. I have used an approximate Ordovician-Silurian date for Bollerro’s specimens based, on the diagrams presented in Zirnmerrnann and Spalletti (2009) and on support from Burj ack and Popp (1981). From his specimens, Borrello (1966) identified transverse annulations, the median groove, a variety of cross-sectional shapes, and a variety of branching patterns that included palmate fans. Most of his measured sizes were on the small side, with a 60 range of 3-12 mm in width. His impressive eleven illustrated plates of Arthrophycus specimens show both bundled and independent individuals, sometimes even grading into one another. He speculated about the trace-maker, thinking that a particularly curvy vermiform specimen resembled “true type annelids” more than it did trace fossils. Regardless, the specimens clearly belong in Arthrophycus based on their transverse annulations, median groove, variety of cross-sectional shapes including subquadrate, and variety of branching patterns including palmate fans. Burjack and Popp (1981) described large A. alleghaniensis specimens from the Silurian Vila Maria Formation of Brazil. Their specimens were sinuous, with some branching, and had regular annulations, a sub-circular cross-section, and sometimes a median groove. The diameters of the specimens ranged from 1.5 to 6 cm, which is unusually large for Arthrophycus (but see Nogueira et al. 1999). The photographs are somewhat blurry and include only a blurry coin as an indication of scale, so the claims of exceptional size cannot be verified. The lack of palmate structures or bundling may indicate an assignment to A. brongniartii rather than A. alleghaniensis, but the specimens can be placed in Arthrophycus based on their regular annulations, cross- section, branching, median groove, and general shape, regardless of which ichnospecies designation might be more appropriate. Conrad (1837) noted F. alleghaniensis from the Silurian of New York, probably of the Medina Group. He did not include a figure or description, so this record is not fully confirmed. In 1838, Conrad referred to the ichnospecies as F. harlani. In 1839, Conrad again referred to E harlani, expanding his report to 61 Pennsylvania and Virginia. Both of these records are also not completely confirmed, lacking good descriptions or figures. Dalloni (1934) observed Harlania in the Silurian sandstones of Chad. He observed the annulations, median groove, constant diameter of individuals, and rectangular cross-section; his figured specimens also show bundling. These specimens conform to the definition of Arthrophycus based on both the description and figured specimens. Downey (1980) described specimens of A. alleghaniensis from the Lower Silurian Tuscarora Sandstone of Pennsylvania. His specimens exhibit the annulations, median groove, general shape, bundles and branches common to Arthrophycus. He noticed an iron “stain” on the edges of the traces that he suggested could represent the remnants of ancient hemoglobin secreted by the trace-maker, which he thought might be a worm similar to Agaophamus circinata, a segmented marine annelid. Downey further suggested that asexual reproduction in the annelids could explain the branched and bundled patterns of Arthrophycus. Regardless of that speculation, these fossils conform to the definition of A. alleghaniensis based on the annulations, median groove, general shape, bundles, and branches. Eschard et al. (2005) reported Harlania from the Lower Ordovician “Vire du Mouflon” of Algeria. Their report consisted of one sentence regarding the ichnospecies, observing it as part of the Cruziana ichnofacies present in the sandstones. This occurrence is therefore unverifiable. Fenton and F enton (1958) mentioned A. alleghaniensis from the Silurian Medina Sandstone of eastern North America. Their work was an elementary catalog 62 rather than a technical publication, but the figured specimens are confirmable as A. alleghaniensis for their simple shape, regular annulations, and bundled forms. Femandes (1999) reported A. alleghaniensis from the Silurian-Devonian of Brazil. This work is really just a compilation of work that had been published earlier on four Brazilian formations, and Arthrophycus was reported from two (earlier reports from a third formation rejected) with no figures or specific description. Femandes (2001) reported A. alleghaniensis again in the Silurian-Devonian of Brazil. Like the previous work, the author listed the ichnogenus again in several formations of Silurian-Devonian age in Brazil, but he also saw some differences between the Brazilian Silurian and Devonian forms and suggested that differences in behavior might reflect nutrient availability. Because Femandes included no description or figures, this report is regard as providing no additional confirmation beyond what had been published earlier. F emandes et al. (1995) mentioned A. alleghaniensis in the Silurian Vila Maria Formation of Brazil, along with Paleophycus. In this report, the first of many by the same authors, they used the presence of Arthrophycus to understand the paleoenvironment of the Vila Maria Formation, which they interpreted as shallow marine from the presence of the Cruziana ichnofacies. As the authors included no figures, this report is not verifiable. Femandes and Borghi (1996) noted A. alleghaniensis in the Ordovician- Devonian (1’) of Brazil. The authors included no figures, but rather discuss the history and distribution of Arthrophycus, with particular attention to Brazil. Their stratigraphic resolution is somewhat confused, because they mentioned Arthrophycus in the 63 Ordovician and possibly early Devonian, but then in their conclusion the authors stated that the ichnospecies is restricted to the Silurian. However, with regard to assessing their taxonomy, these authors seem to have considerable experience in dealing with Arthrophycus, so their report here depends upon the strength of their other publications. F emandes et al. (2000) described A. alleghaniensis from the Silurian-Devonian Furnas Formation of Brazil. The figured specimen shows annulations, straight form, and bundling, while the description mentions a sub-circular cross-section and sporadic faint median groove. The authors noted some differences in branching between various formations; this may indicate the presence of both A. alleghaniensis and A. brongniartii. This material conforms to A. alleghaniensis based primarily on the illustration, but also on the written description. Femandes et al. (2002) described several ichnospecies of Arthrophycus, including A. alleghaniensis, in the Trombetas Group and the Nhamunda, Vila Maria, and Furnas Formations, of Ordovician-Devonian (?) age in Brazil. The figures of A. alleghaniensis show bundled structures twisting about in three dimensions, with annulations and typical shape and size expected for Arthrophycus. These specimens conform to A. alleghaniensis on the basis of annulations, shape, and bundling. Fritel (1925) remarked upon Harlania from the Silurian of Chad. He used the ichnogenus to assist him in dating the formations, noting that the “Ennedi formations” should correlate to the Medina Sandstone. He included no figures and described the fossils only as “slightly flattened cylinders” that “intersect in every direction.” Crimes (1981) seemed to support this report, but without figures or a detailed description, its assignment cannot be confirmed. In his review of the strata of Brazil, Grahn (1992) mentioned A. harlani in the Silurian. However, his report is only a collection of published reports and cannot be evaluated on its own merit. It is included in the counts of the biostratigraphic and geographic records herein because several of the source reports were not otherwise included. Grove (1960) reported Harlania in the Silurian of Chad, from the sandstones of Tibesti. He did not provide figures or a description, and wrote that even the assigned age of Silurian was uncertain (but probable). This report is an unverifiable record, as it lacks figures or written description. Gubler et a1. (1966) reported Harlania from the Ordovician of Algeria. Their figure shows regular annulations, a simple form, and multiple occurrences of pseudo- branching in high relief. The rather straight and independent shapes are reminiscent of A. brongniartii, but the traces are much closer together than is typical for that ichnospecies, and at least one group of traces may be coalescing into a bundle. The authors’ description stated that bundles did occur, and that a single individual could be traced for over 1 m in length. The fossils conform to A. alleghaniensis based on the multitude of characters presented above from both the figure and the written description. Janvier and Melo (1988) noted A. harlani from the Silurian-Devonian Trombetas Formation of Brazil. Their report was concerned mostly with fossil acanthodians, but the authors found specimens of Arthrophycus in certain portions of 65 the field sites as well. Lacking figures or descriptions of the traces, this record is unverifiable. Kumpulainen et al. (2006) found A. alleghaniensis in the Adi MaEkheno Member of the Adigrat Formation of Eritrea. The authors noted the presence of confirmable Arthrophycus in strata from the Ordovician to the Devonian in different parts of the world, but they dated the occurrence in Eritrea as Ordovician only. Regardless of potential problems in dating the fossils, the traces themselves displayed regular annulations, subquadrate cross- sections, pseudo-branching, simple shape, and mass aggregation of individuals. The authors also noted an occasional faint median groove and some palmate fans. These specimens conform to the definition of A. alleghaniensis, based on fi'om the many features common to Arthrophycus and observable in the excellent figures. Lesley (1889) included A. harlani from the Silurian White Medina of Pennsylvania in his encyclopedia of fossils. Oddly, he included a separate, shorter entry for Harlania halli, calling A. harlani a junior synonym, but contradicted this synonymy by placing most of the information and figures under the entry for A. harlani, which he regarded as a fossil seaweed. His specimens display the regular annulations, simple form, and frequent palmate fans and narrower bundles common to ‘ Arthrophycus, so they conform to the standard concept of Arthrophycus. Lessertisseur (1955) discussed Harlania (using both H. alleghaniensis and H. harlani on the same page) in an early support of Sarle’s (1906) suggestion of a burrowing “program” to explain the traces. Lessertisseur’s work is encyclopedic in nature, discussing the ichnogenus in toto rather than any particular ichnospecies or 66 specific occurrence. His figured specimens show regular annulations, a simple shape, and palmate branching, and conform to Arthrophycus. Lifian (1984) described A. alleghaniensis from the Cambrian Julia Member of the Torrearboles Formation of Spain. The single figure shows only a few traces (perhaps three — the image is indistinct) with regular annulations and some branching; the description mentions an intermittent median groove and a sub-rounded cross- section. These traces are among the smallest described as A. alleghaniensis (2.5-4.0 mm wide), but are within the size range for the ichnogenus. These specimens conform to the standard concept of A. alleghaniensis, although the paucity of preserved traces on the slab is problematic, as any Arthrophycus ichnospecies is typically a gregarious trace. Mangano et al. (2005a) were also concerned because of the fragmentary nature of the material, but suggested that the specimen could be included in A. minimus instead of A. alleghaniensis, based primarily on the lack of bundles necessary for designation as A. alleghaniensis. Metz (1998) discussed A. alleghaniensis fiom the Silurian Shawangunk F orrnation of New Jersey. His figured specimen shows regular annulations and a simple shape, with very few preserved traces. However, the fossils’ very low relief could indicate erosion of pre-existing layers of other Arthrophycus traces; the faint outlines of traces in the comer of the figure indicated that other Arthrophycus individuals lie below the surface. The text indicated that the author found 63 specimens on only 6 slabs. In his description, Metz noted constant diameter, an occasional faint median groove, and structure-less internal fill. Lack of bundles or palmate fans may indicate an assignment of A. brongniartii rather than A. 67 alleghaniensis, but the fossils conform to Arthrophycus regardless of ichnospecies designation. Metz (2006) also reported A. alleghaniensis from the Silurian Lizard Creek Member of the Shawangunk Formation of New Jersey. However, this report was only included in a meeting abstract, showed no figures or details, and therefore is unverifiable. Miller et al. (2009) documented a remarkable occurrence of A. alleghaniensis from the Silurian Tuscarora Formation of Virginia. The preserved slab of sandstone is 30 cm thick on average, with a surface area of 3.4 m2 across the main surface, with hundreds of specimens of Arthrophycus preserved across the bedding. Though the photographs are blurry, these specimens show regular annulations, retrusive spreiten, subquadrate cross-section, and both pseudo-branching and palmate fans. There are also some rather unusual orientations, including loops of the same type as in a similarly-sized slab at Yale Peabody Museum (Figure 2.18). The authors interpreted the orientations as a result of avoidance behavior on the part of the trace-makers, perhaps using a chemical signal. The traces conform to Arthrophycus based on the list of characters and on the fact that Seilacher reviewed the manuscript and found no problems with the assignment. Moneda (1963) used both the names A. alleghaniensis and A. harlani (in the same sentence) to describe fossils from Argentina. The author gave no stratigraphic position, description of the fossils, or figures, so the report cannot be confirmed. Moreira et al. (1998) recorded A. aff. alleghaniensis from the Silurian to (?) Early Devonian of Brazil. The authors noted some differences between their 68 specimens and A. alleghaniensis, notably smaller size and in the annulation size and spacing. However, these may reflect only small changes in the trace-maker and are not enough for a different ichnospecies designation. Without a detailed description or any figures, this record cannot be confirmed. Moreira and Borghi (1999) identified A. alleghaniensis from the Silurian Furnas Formation of Brazil. The ichnospecies is mentioned only as a part of the Cruziana ichnofacies found in the study, without further description, and the single photograph showing Arthrophycus is too blurry to see well. Because these authors have been reliable in the past, this occurrence was considered unverifiable because it is neither a good specimen nor is it truly questionable. Neto de Carvalho et al. (2003) found A. alleghaniensis in the Ordovician Armorican Quartzite of Portugal. The authors observed subcircular cross-section, regular annulations, constant diameter, and near-vertical spreiten. They emended the ichnospecies diagnosis as “burrows in tridimensional bundles with protrusive or retrusive backfill structures, which may ramify from a single point in a palmate fashion or bend asymmetrically” (English theirs). The included photograph is too blurry for confident assignment, but the description is very promising, so these specimens conform to the standard concept of A. alleghaniensis. Nogueira et al. (1999) described specimens of A. alleghaniensis from the Lower Silurian Nharnunda Formation of Brazil. The specimens were very large (2-6 cm wide, up to l m in length) and possessed large regular annulations (but not large relative to the width of the traces), a subquadrate to elliptical cross-section, and a median groove. The figures were too dark to see any of the described features, so the 69 large reported size cannot be confirmed. However, based on the strength of the description, the specimens conform to A. alleghaniensis. Pickerill et al. (1984) collected A. alleghaniensis from the Ordovician Armorican Quartzite of Spain. The traces were subquadrate in cross-section, with regular annulations and an occasional faint median groove. No branching was observable, but the single figure included depicted a rather small piece of rock with traces very close to one another; a more complete or larger sample might have exhibited better orientations. These traces conform to Arthrophycus based on the annulations, groove, and cross-section. Poiré et a1. (2003) described A. alleghaniensis from the Cambrian-Ordovician Balcarce Formation of Argentina. Their figure shows specimens with regular annulations, a median groove, simple shape, dense aggregations, and both bundles and relatively independent individuals, and the description concurs. These specimens conform to A. alleghaniensis based on those characters. Prouty and Swartz (1923) described A. alleghaniensis from the Silurian Tuscarora Formation of Maryland. They illustrated both bundled (A. alleghaniensis) and independent (A. brongniartii) forms, placed Harlan’s two species in synonymy, and observed annulations, occasional median groove, subcylindrical cross-section, simple shape, and lack of tapering at the ends. These specimens conform to Arthrophycus based on the characters listed, though divided between A. alleghaniensis and A. brongniartii. Romano (1991) noted A. alleghaniensis in the Ordovician Armorican Formation of Spain and Portugal. The paper did not include a written description or 70 figure, but only noted Arthrophycus as present as part of the Cruziana ichnofacies. Although the small sketch of Arthrophycus is very promising (it includes regular annulations, a faint median groove, and some branching of finger-like shapes), the record cannot be verified. Schuchert (1916) noted “A. alleghaniense” in the Silurian of New York, Pennsylvania, and New Jersey. He listed the ichnospecies as common in the Silurian formations of those states, but his only figure is at much too large a scale to see details. The record cannot be confirmed because it lacked a written description or figured specimens. Seilacher (1997) included A. alleghaniensis from the Silurian of Libya in his book Fossil Art. The figured specimen displays regular annulations, simple shape, and many bundles among a mass aggregate of traces. These specimens conform to A. alleghaniensis based on those traits. Seilacher (2000) also described A. alleghaniensis fi'om the Ordovician-Silurian of formations around the world. He emended the ichnospecies diagnosis to “Arthropycid (sic) burrows, 5-15 mm in diameter, which explore the sediment mainly in a horizontal fashion.” Curiously, he did not invoke any branching pattern to distinguish the ichnospecies from A. Iinearis, which was defined in the same paper. However, his figures of A. alleghaniensis do show prominent palmate forms, along with typical regular annulations, finger-like shape, and retrusive spreiten. These specimens conform to A. alleghaniensis based on the characters in the figured specimens. 71 Seilacher et al. (2003) observed A. alleghaniensis in the Ordovician-Silurian Balcarce Formation of Argentina. The authors did not include a detailed description, but their figured specimens show regular annulations, a simple form, and many instances of bundling. These specimens conform to A. alleghaniensis based on the characters in the figured specimens. Seilacher (2007a) included A. alleghaniensis in his arthrophycid group. The figures display regular annulations, a simple form, bundles and palmate forms, and faint median grooves. The specimens conform to A. alleghaniensis based on those characters. Selley (1970) noted Harlania from the Ordovician Um Sahm Formation of Jordan. He did not include a description, but the illustrated specimens possess regular annulations, simple forms in high relief, and both pseudo-branching and bundling. These specimens conform to Arthrophycus based on those features. Selley (1972) mentioned Harlania in the Ordovician Graptolite Sandstone of Jordan. He did not include a description or any pictures of the fossils, but only mentioned them as being in the informal “Harlania Shale Member” and included the ichnogenus in his stratigraphic coltunns. Without any further evidence, the record remains unverifiable. Shimer and Schrock (1944) mentioned A. alleghaniensis and E harlani in the Silurian Tuscarora and Clinch Formations of eastern North America. The authors described the ichnospecies as “simple or apparently branching, rounded or subangular ridges with median groove and closely set transverse grooves,” which includes the most important and common characters of Arthrophycus. The drawings included were 72 some of Hall’s originals, so this is not a newly reported occurrence but merely an encyclopedic entry. Taylor (1834) mentioned F. alleghaniensis from the Silurian of Pennsylvania, near the area where Harlan’s original specimens may have originated. Taylor’s sketch reveals regular annulations, simple shape, pseudo-branching, and palmate bundles; his specimens conform to Arthrophycus based on those characters. Taylor (1835) again mentioned F. alleghaniensis from the Silurian of Pennsylvania, in different sites than in his earlier (1834) report. However, this time Taylor did not include a detailed description or figure, his report cannot be confirmed. Turner and Benton (1983) reported A. alleghaniensis fi'om the Cambrian- Ordovician, Silurian, and Early Devonian in several formations of Libya. The fossils have regular annulations, simple form, and intersect each other at all angles in high relief. The authors also wrote that the specimens had a circular to elliptical cross- section, and that bundles do exist. These specimens conform to A. alleghaniensis based on the characters in the figure and those mentioned in the description. Turner et al. (2005) noted Harlania in the Late Ordovician Tubeiliyat Formation of Jordan. The authors did not include a description of the fossils, and the single figure purported to show Harlania is very small and includes numerous brachiopods. The brachiopods are visible, but the trace fossils are not discernible, so this report remains questionable. Wolfart (1961) reported A. alleghaniensis from the Devonian of Paraguay. He described the fossils as subcylindrical, branched, with a constant diameter, annulations, and possibly a median groove (translation unclear). However, the single figure is 73 blurry and entirely unconvincing as Arthrophycus, so this record is hesitantly classified as questionable. The description is fine, but the figure remains problematic. Finally, Young (1955) observed A. alleghaniensis in the Silurian Keefer Sandstone of Virginia. He considered Arthrophycus to be a possible “guide fossil,” but suggested that the stratigraphic definition not be limited to the Medinan (Lower Silurian). The figured specimen shows regular annulations, simple form, and a typical branching pattern, so these specimens conform to A. alleghaniensis. In addition to the above accounts, some specimens reported only as Arthrophycus isp. may actually be A. alleghaniensis. These include Abel (1935), Becker and Donn (1952), Pfltiger (1999), and Seilacher and Alidou (1988, Fig. 1e and 1’10 and are discussed below. A. harlani and Hfiarlania halli A. harlani Conrad 1838 was, as mentioned in the history of Arthrophycus, erected by Conrad in 1838 as F ucoides harlani in place of F. brongniartii with no explanation. Hall (1852) then retained Conrad’s specific name when he renamed the genus Arthrophycus, but there was no reason to drop Harlan’s original name of alleghaniensis for the first (and for Hall, only) ichnospecies (James 1893a). As there are no differences between the two ichnospecies, A. harlani is a junior synonym to A. brongniartii. The name A. harlani does not come up often in the literature; only Schimper and Schenk (1879-1890) Lesley (1889), J anvier and Melo (1988), and Grahn (1992) have used that name since Hall (1852), and all of these authors used the name A. harlani in a sense that certainly meant A. alleghaniensis. However, labels persist through museum catalogs: the Yale Peabody Museum, the American Museum of 74 Natural History, the New York State Museum, the Science Museum of Buffalo, and the Paleontological Research Instititution among the museums visited for this study used the name A. harlani. Harlania halli Goppert 1852 is a junior synonym to A. alleghaniensis, as determined by James (1893a) and by popular consent in the literature. All specimens originally designated in published reports as either A. harlani or H. halli are included under the discussion of A. alleghaniensis; thus, they need not be critiqued further. A. bronggiartii A. brongniartii Harlan 1831, named less than a year after A. alleghaniensis, is perceived as largely forgotten in the literature. However, this is not entirely the case. The first description of the ichnospecies was by Harlan as F. brongniartii (1832) and featured a written description but did not include a figure. Harlan described his new species as “F ronde elongata, sub-quadrangularis, canliculata, transverse rugosa; ramulis inequalis, sparsis, remotis, compressis, rugatis, recurvis, nudis,” which translates to “frond elongate, sub-quadrangular, channeled, transversely ribbed, unequal branches, sparse, distant (from each other), compressed, wrinkled (ribbed), recurved, naked.” The important features of A. brongniartii that place it in Arthrophycus are the subquadrate cross-section, annulations, and branching, but A. brongniartii is distinCt from A. alleghaniensis in the relatively independent or “distant” nature of the individuals and in the lack of bundles, having only pseudo-branching. Taylor (1835) provided the next mention of the ichnospecies and what may be the first figured specimen (Figure 3.2) and Harlan referred to the name again in his 1836 paper. Conrad (1837) noted F. brongniartia (spelling his) as particularly 75 abundant in the Medina of New York. Hall (1852) and Goppert (1852) placed F brongniartii in synonymy with F. alleghaniensis in Arthrophycus and Harlania, respectively, without further explanation, and James (1893a) mentioned both forms in his history of the ichnogenus but considered them to be forms of the same ichnospecies. Prouty and Swartz (1923) mentioned A. brongniartii but included it within A. alleghaniensis. The figures of Prouty and Swartz (1923) clearly indicate which specimens they considered to be the A. brongniartii form, but Hall (1852) and Goppert (1852) were not so precise. The specimens of both Taylor (1835) and Prouty and Swartz (1923) are accepted as A. brongniartii for their annulations, median groove, simple shape, and independent form. Conrad (1 83 7) did not include a description or figure and Hall (1852) and Goppert (1852) were unclear, so the specimens of those three authors cannot be assessed. After 1923, the name A. brongniartii was not mentioned until Rindsberg and Martin (2003), who found abundant specimens of the ichnospecies in the Silurian Red Mountain Formation of Alabama. Their specimens possess annulations and independent orientations, as well as retrusive spreiten. A new feature is the “stellate” shape of multiple traces coming together at wide angles (palmate forms have acute angles); this may be considered a compound trace. Although Mangano et al. (2005a) had some doubts as to the assignment of these traces, these traces conform to A. brongniartii for their annulations, orientations, and retrusive spreiten. Kumpulainen et al. (2006) have also used the name A. brongniartii for specimens from the Adi MaEkheno Member of the Adigrat Formation of Eritrea. The authors were not certain of their assignment, but their figured specimens share the 76 annulations and independent nature of A. brongniartii, and even have retrusive spreiten. Their specimens are certainly worthy of inclusion in A. brongniartii. The eighty-year hiatus between 1923 and 2003 does not violate the rules of the ICZN, otherwise few fossils would retain their original names. However, this lapse may explain why Seilacher (1997, 2000) introduced the name A. linearis, which is almost certainly a synonym of A. brongniartii (Rindsberg and Martin 2003). Because of the eighty-year gap and the many authors’ combinations of A. alleghaniensis and A. brongniartii, many specimens identified as A. alleghaniensis may actually be A. brongniartii. The main morphological difference between the two ichnospecies is in the morphology of the bundling: A. alleghaniensis is more typically bundled and A. brongniartii has individual branches that do not bundle but are more independent of one another. Specimens of A. brongniartii may appear to branch, but that is probably a result of over-crossing and not true branching. In addition to those specimens that the authors originally described as A. brongniartii, there are a number of A. alleghaniensis specimens that might more properly be assigned to A. brongniartii. These include specimens figured in Bender (1963 and 1968), Borghi et al. (1996), Borrello (1966, only Plates 111, V, VI, and XI), Burjack and Popp (1981), Downey (1980, only Fig. 6), Metz (1998), and Poiré et al. (2003, only Fig. 5C and 5D). Moreover, some specimens originally designated only as Arthrophycus isp. might be more properly assigned to A. brongniartii. These specimens include those in Durand (1985), Pemberton and Risk (1982), and Seilacher and Alidou (1988, Fig. 1d), and are discussed below. 77 The actual stratigraphic and geographic ranges for A. brongniartii may include some reports of specimens that were under the name A. alleghaniensis. However, considering only those described by the authors as A. brongniartii, the stratigraphic range is Ordovician to Devonian and the geographic range includes the country Eritrea and the American states New York, Pennsylvania, and Alabama. The specimens originally described as A. linearis by other authors (e. g. Femandes et al. 2002, Acefiolaza and Acefiolaza 2003) extend the geographic range of A. brongniartii to Argentina, Brazil, Portugal, Jordan, and to additional parts of North Afiica, and extend the stratigraphic range into the Cambrian. A. linearis A. linearis Seilacher 1997 was first named by Seilacher in his book Fossil Art, in the caption for a small sketch (Figure 3.3), but he did not formally describe A. linearis. From this sketch, it can be seen that the traces have regular annulations and some kind of median structure, and that the traces are independent of one another. Seilacher and Alidou (1988) mentioned Arthrophycus specimens from the Kandi Sandstone of Benin and in one of the figures, the authors referred informally to a linear form of Arthrophycus. However, they did not describe a new ichnotaxon, and were apparently unaware of the previous name A. brongniartii. Seilacher (2000) subsequently described A. linearis more fully, with more images (Figure 3.4) and descriptions, and divided the ichnospecies into two subspecies based on two different modes of backfill (protrusive and retrusive). He reported the stratigraphic range of the ichnospecies to be Ordovican to Silurian and the geographic range to include North Afiica, North America, and Jordan. 78 A. linearis possesses the standard features of Arthrophycus: annulations, a median groove, subquadrate cross-section, and general shape. As already noted, individual traces tend to weave more independently of one another than in A. alleghaniensis, showing only pseudo-branching as they cross over one another and never coalescing into bundles or palmate fans. Both protrusive and retrusive forms of backfill are present (Seilacher 2000). Given these characteristics, the inclusion of A. linearis in Arthrophycus is supported. Femandes et al. (2002) referred to A. linearis in their Guide to Ichnofossils of Invertebrates of Bra_zr_'1, though without an image, so the report could not be evaluated. Seilacher et al. (2002) mentioned A. linearis in Ordovician Hawaz Formation sandstones of Libya. The authors did not include a detailed description of the specimens, and the only photograph of Arthrophycus is labeled as A. simplex, so the report could not be evaluated. Seilacher et al. (2003) mentioned A. linearis in the Balcarce Formation of Argentina in a study that focused on correlating rocks based on their trace fossils (chiefly Arthrophycus and Cruziana). There is no description and only one figure, which is dark and indistinct. However, given that one of the authors of the paper is the same person who erected the ichnospecies, there is no reason to doubt the assignment of the specimens. Acefiolaza and Acefiolaza (2003) found A. linearis in the Cambrian-Ordovician Balcarce Formation of Argentina. The authors did not include a description in the text, but the small figure of A. linearis appears to conform to A. linearis. The specimens are 79 armulated and show both the characteristic shape of Arthrophycus and the independence of A. linearis. Neto de Carvalho et al. (2003) described specimens of A. linearis from the Armorican Quartzite of Portugal in detail. They noted the independent shape of the traces, a lack of true branching, and the two types of backfill that Seilacher (2000) had described. Unfortunately, the quality of most of the figures is rather poor, and most features are indistinct, but there are enough features there to identify the specimens as A. linearis, especially when coupled with the description. The clearest of the figures is the most intriguing, for it shows a circular three-dimensional spiral that may indicate circling behavior similar to that of Cruziana semiplicata. This specimen deserves further study, but such will not be attempted here. No one except Seilacher (2007a) has described A. linearis since Rindsberg and Martin (2003) pointed out its synonymy with A. brongniartii. A. montalto A. montalto Simpson 1888 was referred to by Lesley (1889), who included a sketch of a specimen by Simpson (Figure 3.5) but did not provide a detailed description or full bibliographic information. The original specimen was found, in a scenario eerily similar to Harlan’s (1831) find of A. alleghaniensis in a tavern stone, in the wall of a sawmill in Pennsylvania. It allegedly came from a quarry of sandstone above the Cambrian Potsdam Formation, but the quarry did not yield any more specimens. The only description of A. montalto Simpson 1888 available for study is Lesley (1889), which reproduces Simpson’s original figure. The specimen illustrated, presumably of the primary type material, appears similar to A. brongniartii with 80 slightly curving, rather independent individuals with apparent branching by overcrossing, and regular annulations. A line down the middle of some portions is also present. Lesley (1889) compared the specimens to crinoid stems, which they may indeed be, or they may be specimens of A. brongniartii. Alpert (1975) mentioned their similarity to PIanoIites virgatus, but did not place Lesley’s specimens in that ichnospecies; this possibility may merit further investigation. No other authors have mentioned A. montalto except in lists of ichnospecies in Arthrophycus. There is not enough evidence to separate this material from A. brongniartii. A. siluricus The origins of A. siluricus Schimper 1890 remain a mystery. In their volume of Zittel’s encyclopedic work Handbook of Paleontology, Schimper and Schenk (1879- 1890) mentioned A. harlani as an algal form and then noted A. siluricus. Their “description” was one sentence long and is translated here: “A second thinner and shorter species (A. siluricus Sch.) occurs in the lower Silurian [Cambrian] formations from which I possess the same from Sardinia [Italy], where they were greenish micaceous shale heaps is the same as the preceding species in the Medina sandstone.” The authors did not describe the ichnospecies any firrther, did not include a figure, and did not include any reason for the new name. The fossil presumably had some kind of similarity to A. alleghaniensis, in order to justify the placement in Arthrophycus, but this supposed similarity is unknown. Schiller (1930) listed A. siluricus in a list of synonyms of A. alleghaniensis, perhaps indicating that he thought A. siluricus was not a valid ichnospecies by itself. Because of the lack of description and figure, A. siluricus is designated a nomen nudum. 81 A. elegans A. elegans Herzer 1901 has not appeared in print since its original description, with the exception of Rindsberg and Martin (2003) who included it only in a list of Arthrophycus ichnospecies. Herzer (1901) described his new ichnospecies as similar to A. alleghaniensis, but denser in aggregate, slimmer, and with deeper “firrrows,” apparently referring to transverse markings rather than to a median groove. Some sections of his described slab preserve groups of individual traces that run parallel to one another, but this is not part of the larger pattern. The delicate nature of A. elegans is also seen in the size of the reported specimens (1/16 to 3/16 of an inch in width) and in the dense transverse markings. Herzer did not include a photograph, but made a sketch of the specimen (Figure 3.6). The slightly-curving shape, annulations, and consistent width are compatible with Arthrophycus and the lengths of parallel behavior with the shared walls are intriguing, but there is not enough evidence to consider A. elegans as a separate ichnospecies from A. brongniartii. A. elegans is judged as a possible synonym of A. brongniartii, but the report conforms to Arthrophycus for the purpose of biostratigraphy and biogeography. Herzer was not specific about the stratigraphic position of A. elegans, referring it only to “Coal Measures,” and Rindsberg and Martin (2003) listed the ichnospecies as Carboniferous in age. The ichnospecies has been reported only from Ohio, USA. A. corrugatus A. corrugatus (F ritsch 1908) was originally described as Radix corrugatus. Mikulas (1992) referred Fritsch’s description and figure to Arthrophycus. No other 82 discussions of the ichnospecies are known; it has only been included in lists of Arthrophycus ichnospecies. Fritsch described the ichnospecies as branching and knotty, with transverse wrinkles, but the specimen appears rather wavy with an inconsistent diameter and irregular rings, lacks the median groove, and is a singleton. Fritsch (1908) also wrote that the specimen could be similar to a stolon (a stem or a runner) belonging to a plant such as Lepidotruncus, which was found in the same layer. The main text referred to F ritsch’s original figure (Figure 3.7) as R. corrugatus, but the caption for the figure is for Radicites rugosus, with no explanation for the apparent discrepancy. Mikulas (1992) figured the same specimen as F ritsch (1908) and mentioned an irregular swelling of the trace and irregularly-developed annulations along with an oval cross-section, which are not characteristic of Arthrophycus. Therefore, this ichnospecies assignment to Arthrophycus is questionable. Uchman (1998) questioned the placement into Arthrophycus and Méngano et al. (20053) considered the name a nomen nudum, without explaining why. Radix corrugatus cannot be considered a nomen nudum because it has both a verbal description and an illustration, but Radicites rugosus (Fritsch 1908, plate 6 figure 8) lacks a verbal description. Radicites rugosus appears to be both an objective synonym and a nomen nullum of Radix corrugatus. This ichnospecies has only been reported fiom the Ordovician and Silurian quartzite band of “Repor d5” of the Czech Republic. A. flabelliformis A. flabelliformis Hundt 1940 was named in a paper primarily devoted to a discussion of the new ichnospecies of Keckia and Palaeonereis, leaving very little 83 room for a description of A. flabelliformis. Hundt described A. flabelliformis as somewhat similar to A. alleghaniensis, with a fan-shaped structure of tubes. His figure of “Arthropkycus [sic] flabelliformis” (Figure 3.8) is rather dark and unrevealing, with a different fan shape from the palmate form of A. alleghaniensis, and no other descriptive information is provided. An attempt to contact two professors at the University of Koln with regard to specimens of A. flabelliformis received no reply. Hundt’s specimens came fiom the Ordovician upper quartzite of Germany. The only other mention of A. flabelliformis comes from Benton (1982), who referred Hundt’s (1940) specimen to Dictyodora zimmermanni Hundt, as part of the vertical wall of D. zimmermanni. Hundt (1940) had compared his specimen to Dictyodora manni, but found insufficient similarity. Benton’s figure showing the vertical wall of D. zimmermanni does show “rippled” structures that Hundt (1940) could have interpreted as individuals of Arthrophycus converging into a fan (Benton 1982). On the basis of Benton’s suggestion and the lack of positive evidence that Hundt’s specimens belong in Arthrophycus, there is no reason to include A. flabellrformis in the ichnogenus. AMI A. krebsi Hundt 1941, was first described as “striated tubes” from the Ordovician quartzite of Germany. Hundt recognized the “sweeping” nature of the structure and suggested that it resulted fi'om a burrowing method similar to that suggested by Sarle (1906a) for A. alleghanierrsis, but gave no other details about the ichnospecies. The single photograph (Figure 3.9), is far too dark and provides no diagnostic criteria. Attempts to track down the specimens at the University of Koln 84 have not yielded results, and Hundt was the only person to describe or recognize A. krebsi. Without any positive evidence that this is Arthrophycus, these specimens are removed from the ichnogenus and there is insufficient support for the ichnospecies concept. A. minoriceniig A. minoricensis Bourrouilh 197 3 was originally named in an unpublished PhD thesis from France. This does not satisfy the current requirements for publication distribution for the ICZN, but in 1973, those rules did not apply to trace fossils. Since then, this ichnospecies has been redescribed twice, by Orr (1994) and Llompart and Wieczorek (1997). Femandes and Borghi (1996) also included a small discussion of the ichnospecies in their paper, without describing specimens. There is some disagreement as to the spelling of the name: Orr (1994) and Llompart and Wieczorek (1997) used A. minoricensis, but Femandes and Borghi (1996) referred to A. minorcensis. According to Orr (1994), Bourrouilh may have originally described the ichnospecies as Harlania minoricensis. A. minoricensis has been reported only fiom the Carboniferous of the Balearic Islands, specifically from the island of Menorca (or Minorca) (Orr 1994 and Llompart and Wieczorek 1997). In his 1994 paper, Orr also described A. minoricensis as associated with Nereites. His specimens have elliptical cross-sections, no inner fill, and have either “radial” or “arborescent” (branching) forms (Figure 3.10a). The radial forms are similar to the palmate orientation of many A. alleghaniensis specimens, but the palm structure is much larger and the individuals are not as condensed into the palmate form as in A. alleghaniensis. Orr did report fine striations, but they may be only on the 85 external wall and the figures are not clear enough or large enough to make this determination. Llompart and Wieczorek (1997) did not include a formal description of their Arthrophycus specimens, and their figure (Figure 3.10b) is not convincing. The specimens cover the rock densely and evenly, with no apparent overlap or bundling, which seldom, if ever, occurs in Arthrophycus. The traces do have a constant width and a slightly weaving shape, but no median groove or annulations are apparent and it is not possible to determine the cross-section. Llompart and Wieczorek (1997) described A. minoricensis from the Carboniferous Cuhn siliciclastic sequence of Minorca Island, the same locality studied by Bourrouilh (1973). They described specimens from the Nereites ichnofacies of a turbidite, which includes ichnogenera such as Palaeodictyon and is generally understood to be a deep-water ichnofacies, whereas Arthrophycus is usually placed in the Cruziana ichnofacies, a shallow-water ichnofacies (Seilacher 1964a, Bromley 1990). Llompart and Wieczorek suggest that the presence of the shallow-water SkoIithos ichnofacies directly above their specimens indicated a difference in energy, and not necessarily water depth, but in their conclusion the authors appear to have decided on the deep-water interpretation. On the basis of the above characters, there is no reason to include the specimens of either of the above works in Arthrophycus. Uchman (1998) placed the inclusion of Orr’s specimens in Arthrophycus under question and described A. minoricensis as a nomen nudum for the insufficient diagnosis, but did not mention Bourrouilh’s specimens specifically (he had probably not seen them). He also did not mention the 86 work of Llompart and Wieczorek (1997), but it is likely that the publication dates made this impossible. Mangano et al. (2005a) agreed that A. minoricensis should be considered a nomen nudum. A. annulatus A. annulatus KsiaZkiewicz 1977 was the first of the three new ichnospecies of Arthrophycus defined by KsiaZkiewicz. No one else has used the name since to describe new occurrences. Ksiazkiewicz (1977) described A. annulatus as occurring in aggregates with occasional branching (probably pseudo-branching); the individuals are mostly straight with a circular or oval cross-section. The specimens have wide annulations (2.0-2.5 mm for 1 cm of width) and no median groove (Figure 3.11). The type specimens are Lower Eocene Ciezkowice Sandstone, but Ksiazkiewicz (1977) reported other, narrower specimens from the Cretaceous Inoceramian Beds. Both are from Poland. Ksiazkiewicz (1977) discussed differences between A. annulatus and Ophiomorpha, but Mangano et al. (2005a) suggested that A. annulatus be placed there and Uchman (1998) suggested that A. annulatus should be placed in Ophiomorpha under the specific name 0. annulata (Ksiazkiewicz). The case to include A. annulatus within Arthrophycus is not particularly strong, based on the lack of positive evidence. A. strictus A. strictus Ksiazkiewicz 1977 was the second of the three new ichnospecies of Arthrophycus described by KsiaZkiewicz. In her specimens of A. strictus, KsiaZkiewicz (1977) noted dense striations, some branching, and mostly straight and small (1.5-2.0 mm wide) burrows in the Cretaceous Lgota and Ropianka Beds of 87 Poland (Figure 3.12). She also noted another variety of Arthrophycus, placed in the same ichnospecies, that was larger (2.5 mm wide), with more pronounced transverse markings, a point at one end, and a radial arrangement. Neither variety of A. strictus had a median groove and was three times more narrow than specimens of A. alleghaniensis. Paczesna (1996) provided a short description of material similar to the holotype form of A. strictus. The specimens are straight traces with annulations, but they taper at the ends and are not very common on the sample of rock — the sole figure shows only one branched structure. This specimen may not belong in Arthrophycus because of the tapered ends and small number of traces figured. Uchman (1998) removed Paczesna’s specimen from Arthrophycus. Mangano er al. (2005a), however, tentatively suggested placing the specimens in A. minimus. Uchman (1998) analyzed the Ksiazkiewicz specimens of A. strictus further. He noticed differences even within her second morphotype and assigned some of those to A. tenuis, but accepted the holotype specimen as Arthrophycus. Mangano et al. (2005a) rejected the holotype specimen as Arthrophycus because of the cylindrical cross-section and the simplicity of the general shape of the trace. A. strictus (holotype) is tentatively considered as Arthrophycus herein, but Ksiazkiewicz’s second morphotype does not conform to the ichnogenus because the radial arrangement and pointed terminus do not resemble the blind endings and general form of Arthrophycus. A. dzulznskii A. dzulynskii Ksiazkiewicz 1977 is the third new ichnospecies of Arthrophycus that KsiaZkiewicz designated from Poland. She was unsure of the ichnogeneric 88 assignment, noted that the straight specimens possessed transverse ribs (Figure 3.13), but also observed an odd curvature of the ribs and thought that there was a difference between the ventral and dorsal sides. The specimen is a singleton, with a depression in the center and a knob at one end, neither of which is characteristic of Arthrophycus, and the annulations are not consistent in size, but narrow toward the outer edges of the trace. No other authors have described occurrences of A. dzulynskii, so it remains known only from the Oligocene Krosno Beds of Poland. KsiaZkiewicz (1977) placed her specimens in Arthrophycus due to the presence of the annulations and the general shape of the trace, but also suggested that the specimens could be placed in Climactichnites in view of the ending knobs. Uchman (1998) suggested that specimens of A. dzulynskii be placed in Protovirgularia under the specific name P. dzulynskii (KsiaZkiewicz), and Mangano et al. (2005a) agreed. A. dzulynskii is rejected herein as an ichnospecies of Arthrophycus due to the single occurrence, central depression, terminal knob, and curvature of the ribs, but which, if either, of these other assignments is preferable has not been determined. W13 A. tenuis (KsiaZkiewicz 1977) was first described by Ksiazkiewicz as Sabularia tenuis, fi'om the Oligocene Krosno Beds of Poland. Her original description of S. tenuis included delicately armulated burrows, pseudo-branched, with one or both ends pointed (Figure 3.14). She noted the very small size (0.5 to 0.7 mm wide) of the short and straight burrows, which occur in large numbers on the slab. The specimens have no median groove and one or both ends of each individual trace are pointed. Uchman 89 (1998) subsequently assigned the specimens to Arthrophycus under the ichnospecies name A. tenuis (Ksiazkiewicz). Uchman (1998) has since referred other specimens to A. tenuis. Alexandrescu and Brustur (1984) described specimens of S. tenuis from the Oligocene-Miocene Tarcau unit of Romania that are slightly wider (1-2 mm) and-occurred in dense accumulations, but cross one another in only a few places, are pointed on one end, and do not have transverse annulations. Uchman (1998) suggested that such fine markings might not always be preserved and that thus the absence of annulations was not important, and placed the specimens in A. tenuis. Uchman (1999) also placed traces from a turbiditic sequence from the Cretaceous Flysch-Gault Quartzitseries of Germany into A. tenuis. These specimens show some branching and are similar in size to Ksiazkiewicz’s material (0.6-0.7 mm wide), but do not have transverse annulations. They rather resemble impressions of small twigs. Uchman and Demircan (1999) described similar specimens from a deep- sea fan fiinge from the Miocene Cingoz Formation of Turkey, but were unsure of their placement in Arthrophycus. The specimens in the figure resemble those in the figure in Uchman (1999), except for one intriguing larger individual. This specimen was not specifically discussed in the paper. Though interpreted as a deeper water formation, the authors noted seventeen species of three different ichnofacies (Skolithos, Cruziana, and Nereites). All of the specimens in A. tenuis should be removed from Arthrophycus. The primary types are the only ones with annulations, though other specimens assigned to this ichnospecies may have had annulations that were subsequently eroded. More 90 problematic are the pointed ends of the traces, tiny size, and lack of median groove or any other character common in Arthrophycus, and few other specimens of Arthrophycus are reported from deep water sequences. Mangano et al. (2005a) also did not consider the holotype of S. tenuis as belonging to Arthrophycus, citing the odd branching pattern and typical lack of annulations. It is therefore suggested that Sabularia tenuis be excluded fiom Arthrophycus, and that the specimens that Uchman referred to A. tenuis return to S. tenuis. Uchman (1998) proposed that the ichnogenus Sabularia is a subjective synonym of Ophiomorpha, but this proposal is not accepted herein. The original specimens of A. tenuis were known fiom the Oligocene of Poland (Ksiazkiewicz 1977); Alexandrescu and Brustur (1984) noted S. tenuis in the Oligocene to Miocene of Romania. Uchman (1999) noted A. tenuis in the Cretaceous of Germany, while Uchman and Demircan (1999) referred Miocene material from Turkey to A. tenuis. The entire reported range of the ichnospecies is thus Cretaceous to Miocene, in parts of Turkey and eastern Europe. A. giongghusiensis A. qiongzhusiensis Luo 1994 was the first ichnospecies of Arthrophycus to be described from China, but no new occurrences of the ichnospecies have been recognized by any other authors since. Luo et a1. (1994) described A. qiongzhusiensis as branching, slightly curved, with both transverse ridges and fine striae in between the ridges, and a circular cross-section. The median groove is not present, and the figures show very few individuals. Neither of the two figures (Figure 3.15) included are of high enough resolution to reveal these finer striae and my efforts to contact the authors 91 were fruitless. The ichnospecies is known only fiom the Cambrian Yu’anshan Member of the Chiungchussu Formation of China. The placement of these specimens in Arthrophycus is problematic because of the small number of individuals, the very small size (1-2 mm wide), and the “wriggly” appearance of the traces. This last observation is something more than just a curving shape, but something that is not straight either. Mangano et al. (2005a) questioned the placement of A. qiongzhusiensis in Arthrophycus because of the circular cross-section and poorly-defined annulations. Those authors also found the specimens to be too isolated from any others to be Arthrophycus. Uchman (1998) questioned the placement of the specimens in Arthrophycus and suggested that they were reminiscent of “internal sediment pa ” of T orrowangea Webby, but Mangano et al. (2005a) considered Torrowangea to be quite different from Arthrophycus (they did not address the possible resemblance of A. qiongzhusiensis and T orrowangea specifically). A. tarimensis A. tarimensis Yang 1994 was also described from China and is known only from the Ordovician Ulikeztag Formation. Yang (1994) described the new ichnospecies as somewhat curved, with annulations, but with a rounded cross-section and no median groove, and the traces are rather small (3-4 mm in width). The individuals are in a dense aggregate and truncate each other, but the only figure included in the paper (Fig. 3.16) is too blurred, small, and dark to reveal details. Efforts to contact the author and locate the specimen were unsuccessful. There is not enough evidence to include the specimens in Arthrophycus or to accept this ichnospecies. 92 A. hunanensis A. hunanensis Zhang et Wang 1996 is a third ichnospecies of Arthrophycus described from China, and is currently limited to the Silurian Xiaoheba Formation. Zhang and Wang (1996) described the new ichnospecies as somewhat curved, occasionally overlapping but not branching, on the surface of the beds, and with annulations (Figure 3.17). The specimens have a similar width range (8-10 mm) compared to the typical size range of Arthrophycus and have a rounded or elliptical cross-section. No other authors have reported new specimens of A. hunanensis. The placement of this ichnospecies in Arthrophycus is problematic because of the rippled appearance of the traces, particularly the one on the left in the figure. However, Mangano et al. (2005a) and Uchman (1998) both considered the possibility that these specimens were synonymous with A. alleghaniensis. A. unilateralis A. unilateralis Seilacher 1997 was first mentioned by Seilacher in his book Fossil Art. It is featured in the figure caption of a small sketch, without firrther description (Figure 3.3). In his 2000 work, Seilacher included the name in the synonymy of his ichnospecies A. Iateralis, making the name A. unilateralis a nomen nullem and an objective synonym of A. Iateralis. However, the name A. unilateralis has priority over the name A. Iateralis by three years. A. Iateralis A. lateralis Seilacher 2000 incorporated the sketch that his earlier work (1997) had labeled as A. unilateralis. In the description, A. Iateralis possesses the typical subquadrate cross-section, annulations, median groove, and general shape of 93 Arthrophycus. The new ichnospecies is rather similar to A. alleghaniensis, but the individuals in the dense bundles “bend only to one side” (Figure 3.18) (Seilacher 2000). Seilacher has also remarked upon the backfill, which is horizontal as Opposed to vertical, revealing how the trace-maker had burrowed. In the work of Femandes et al. (2002), the authors observed the asymmetrical fan shape of A. lateralis and also noted that the traces were somewhat smaller than those of A. alleghaniensis. The specimens of both authors conform to Arthrophycus, as they share the important characters that are diagnostic of the ichnogenus. The stratigraphic range of A. lateralis is Ordovician to (Early?) Devonian, with a geographic range that currently includes only Brazil and Libya (Seilacher 2007a, Femandes et al. 2002). A. lateralis is recognized herein as an ichnospecies of Arthrophycus, but it may more properly be a subspecies of A. alleghaniensis. Neto de Carvalho et al. (2003) supported the latter view, including A. lateralis in their synonymy for A. alleghaniensis; their report may extend the geographic distribution of A. Iateralis to Portugal, but the figure is too blurry for confident assignment. Mangano et al. (2005a) considered A. lateralis to be a valid ichnospecies, but noted that polished sections are essential to reveal the burrowing program. If A. lateralis is indeed a valid ichnospecies of Arthrophycus, and if the main diagnostic character (without sectioning) is the asymmetrical palmate form, then it may be possible to extend the geographic range further. Specimens at the Yale Peabody Museum (Grimsby Sandstone, Figure 3.19a), the Science Museum at Buffalo (not illustrated), and Paleontological Research Institution (Tuscarora Sandstone, Figure 94 3.1%) also appear to show this asymmetry, though the Buffalo specimen is palmate in the vertical dimension rather than horizontal. If these specimens fit the concept of A. lateralis, then they would extend the range of the ichnospecies to the Silurian of New York. If the ‘Wertical” Buffalo specimen is A. lateralis, then perhaps the odd stacked specimen in Legg (1985) is also A. lateralis, which would extend the range of A. lateralis to the Cambrian of Spain. A. simplex A. simplex Seilacher 2002 is something of a mystery (Figure 3.20). It was first mentioned in a figure caption by Seilacher et al. in 2002 (their Fig. 4); the text of the paper reported A. linearis in the Hawaz Formation (Middle Ordovician) and A. simplex in the same formation, but in the figure caption only. The only other authors to report the ichnospecies are Konate et al. (2003). In their paper, they referred to “Arthrophycus simplex or Harlania-type galleries (linear)” in the Late Ordovician Ka Member of the Kandi Formation in a figure caption and used the terms Arthrophycus or Harlania in both the Ka and Kb Members of the Kandi Formation in the text, with no specific names or detailed discussion. A. simplex thus has no formal description, though it resembles A. brongniartii from the two figures in which it is noted. The name A. simplex may have been included in the first paper in error and then copied into the paper by Konate et al. If that is the case, the name appears to be a nomen nullum, and an objective synonym of A. linearis. The ichnospecies is reported fiom the Middle Ordovician of Libya (Seilacher et al. 2002) and the Late Ordovician of Niger and Benin (Konate et al. 2003). 95 A. minimus A. minimus Mangano 2005 was first reported from the Cambrian Alfarcito Member of the Santa Rosita Formation of Argentina. Some of the same authors reported a new occurrence of the ichnospecies later that year (Mangano et al. 2005b) but A. minimus has not been noted since. Mangano et al. (2005a) described A. minimus as small, long, and narrow structures with annulations and a median groove, with a subquadrate cross-section and a slightly curving shape. The figures reveal dense accumulations and pseudo-branching but do not have the palmate forms typical of A. alleghaniensis (Figure 3.21) (Méngano et al. 2005b). The specimens of A. minimus are small (1.8-4.8 mm in width), but are within the typical size range of A. alleghaniensis and other ichnospecies of Arthrophycus (Tables 2.3 and 2.4). A. minimus has all the hallmarks of Arthrophycus, even including retrusive spreiten as backfill. The inclusion of the A. minimus specimens in Arthrophycus is accepted herein on the basis of the many shared characters. Seilacher (2007a) suggested that the specimens be included in Phycodes, based partly on the small size of both Phycodes and A. minimus, but typical Phycodes morphology is more bundled than the independent structures seen in A. minimus. The only remaining question regarding this ichnospecies is whether the specimens of A. minimus are too similar to A. alleghaniensis or A. brongniartii to warrant a new ichnospecies name. Mangano et al. addressed this question in their 2005(a) paper, noting that their specimens were exceptionally straight, with hardly any curving at all, and that the annulations were less evident. The size of the specimens is notably small, and some workers do support size as a criterion for dividing 96 ichnospecies (e. g. Magwood 1992, Pickerill 1994). Small size may also reflect a response to stress in the environment (MacEachem et al. 2005); it is reasonable to accept this difference in body dimension as an ichnospecific character. Although the size range of A. minimus is actually within the lower bounds reported for A. alleghaniensis (Table 2.3), the smallest on the list (Lifian 1984) may actually be A. minimus (as suggested by Mangano et al. 2005a), leaving only the report of Metz 1998, which overlaps only slightly. If one accepts size as an acceptable ichnospecific criterion, then A. minimus is distinct from the ranges of A. alleghaniensis and A. brongniartii. A. minimus is therefore accepted as a valid ichnospecies of Arthrophycus based mainly on the unusually straight form and the small size. A. parallelus A. parallelus Brandt et al. 2010 was first described from float material determined to have come from the Grand River Formation of Michigan. As it is the most-recently described ichnospecies of Arthrophycus, there has not been sufficient time for a reaction in the literature. The notable features of A. parallelus include the regular annulations, median groove, pseudo-branching, small size (3.5-4.5 mm wide with relatively short lengths), and predominantly parallel orientation of the traces (Figure 2.20) (Brandt et al. 2010). Currently, the ichnospecies is known only fiom the Late Pennsylvanian of Michigan. A. parallelus resembles A. minimus in its small size, but A. parallelus has a more restricted size range. A. parallelus also has annulations that are more pronounced, much wider (0.9-1.9 mm), and are clearly visible to the unaided eye, in contrast to the annulations of A. minimus that are “most visible under magnification” 97 (Méngano et al. 2005a). The A. parallelus material conforms to Arthrophycus based on the annulations, median groove, and general structure of the specimens. It is a valid new ichnospecies on the basis of the small size, unusual behavior producing parallel traces, and the coplanar nature of the traces. Another specimen from the Devonian of New York may be a new occurrence of A. parallelus (Figure 3.22). These traces share the characteristics of A. parallelus: small size, annulations and a faint groove present, straight shape, and parallel orientation. The specimen, which was discovered in the collections of the New York State Museum, was in a cabinet labeled only “Devonian of New York” with no other data. The stratigraphic range of A. parallelus is therefore extended to the Devonian and its geographic range to New York. Many specimens have been assigned simply to Arthrophycus isp., with no further designation. Often, even the original author is not confident of placement in the ichnogenus. As Méngano et al. (2005a) also noted, many of these authors seem to have placed specimens in Arthrophycus purely on the basis of annulations or other transverse markings. Abel (1935) mostly discussed A. alleghaniensis that others had found in the Medina Sandstone of New York (for this reason, it was not included in the above discussion of A. alleghaniensis). However, he did mention an intriguing Arthrophycus isp. specimen from upper Austria that may have come from the upper Oligocene (“oberoligozanen”). Abel marked the date with a question mark, but did not explain 98 the reason for the uncertainty. The specimen shows the regular annulations and palmate structure of A. alleghaniensis. Acefiolaza and Acefiolaza (2003) mentioned Arthrophycus isp. from the Cambrian-Ordovician Balcarce Formation of Argentina. The figure for Arthrophycus isp. is similar to a figure of A. linearis in the same paper, but appears more weathered and indistinct. This specimen conforms to A rthrophycus for the regular annulations, simple shape, and possible bundle toward the bottom of the sample in the figure. Aceflolaza and Heredia (2008) noted Arthrophycus in the Ordovician of Argentina. The authors mentioned the ichnogenus in the context of ichnostratigraphy, but did not appear to use it in that manner (they used Cruziana for the ichnostratigraphy). As there was no description or figure of Arthrophycus, the record remains unverifiable. Akpan and Nyong (1987) mentioned Arthrophycus from the Cretaceous Eze- Aku Formation of Nigeria. The traces are described as “common” in the rocks, but the one figured (and presumably the others) is a single unbranched armulated trace. Even the authors questioned the placement of their specimens in Arthrophycus, and the placement is questioned herein as well. The subject of the figure resembles a fossil polychaete more than it does a trace fossil. Alpert (1975) mentioned Arthrophycus in the Cambrian of California. He noted the ichnogenus only in a list without further discussion or any figures, so the record cannot be confirmed. Alpert (1977) again mentioned Arthrophycus in the Cambrian of California. In this paper, he reported Arthrophycus from several formations of Early Cambrian age 99 and placed it in a group of trace fossils that he found to be “indicative of early Cambrian age.” He did not include a detailed description or figures, so this record cannot be confirmed either. Banerjee (1982) referred specimens from the Cretaceous Eze-Aku Formation of Nigeria to Arthrophycus. The author described the specimens as tapered, with transverse markings, and about the right size for Arthrophycus (1 cm wide, 10 cm long). The paper does not include a photograph of a specimen, only a sketch, which shows a single trace that is pointed on one end, runs at an angle to bedding, has irregular annulations, and has some kind of scratch-like marks along the outer margin. This specimen should be removed from Arthrophycus based on the pointed end, irregular annulations, and non-diagnostic outer markings. Becker and Donn (1952) reported Arthrophycus from the Silurian Tuscarora Formation of Pennsylvania. Their figure shows a large slab with many traces, most of which form narrow bundles with regular annulations and simple shape of individuals. Becker and Donn interpreted the trace as the remains of a fossil plant, which in itself does not invalidate their assignment of the specimen to Arthrophycus, as trace “producer” is not a valid ichnotaxabase (Bertling et al. 2006). This report conforms to Arthrophycus and it is further suggested that the specimens be placed under A. alleghaniensis. Bhargava et al. (1984) noted Arthrophycus from the Silurian Manchap Formation of India. The authors discussed the ichnospecies but did not describe any of its features, other than the presence of annulations and a “gently meandering” form. The sole figure shows a single fossil with a somewhat wavy form and transverse 100 markings. This specimen is excluded from Arthrophycus for the lack of sufficient . evidence to place the fossil in the ichnogenus. Bhargava and Bassi (1988) observed Arthrophycus in the Silurian Takche Formation of India. Their description mentions branching and bundling, along with annulations and a sub-circular cross-section, but the figure is too grainy to reveal any features of the specimens. Therefore, this report is considered questionable. Bjerstedt (1987) found an Arthrophycus specimen from the Devonian- Mississippian Oswayo Member of the Price Formation of West Virginia. The description included a simple shape with some curve and annulations, without a median groove. The single figure is small and not very convincing, as it shows only one definite specimen. However, after examining this specimen at the Cleveland Museum of Natural History (Figure 3.23), the specimen may conform to Arthrophycus because there is more than one trace present, and all have faint annulations and the general shape of Arthrophycus. Chiplonkar and Ghare (1975) reported Arthrophycus fiom the Bagh Beds of India and referred to Maberry (1971, questioned by several authors including myself) for their comparison with Arthrophycus. The authors never specifically mentioned an age for their formation, but Ghare and Kulkami (1986) indicated that a previous report from the Bagh Beds of India was Cretaceous in age, and their reference list included the paper by Chiplonkar and Ghare. The authors’ choice of Maberry’s Cretaceous specimens for comparison also points to a Cretaceous age for the Indian fossils. The Indian specimens were variously described as “round, elliptical or flattened in cross section” and tapered toward the end. Annulations, median groove, interior structure, 101 branching, and other characteristic features of Arthrophycus were not mentioned and the single figure is too dark to discern anything. These specimens are therefore excluded from Arthrophycus, as Uchman (1998) and Mangano et al. (2005a) also suggested. Correia Perdigao (1964) mentioned Arthrophycus from the Ordovician of Portugal. The author listed Arthrophycus among a group of questionable specimens and noted that the putative Arthrophycus was even more questionable. The specimen resembled Arthrophycus, but without ornamentation; the meaning of this is unclear. The single figure is grainy and shows only a linear feature at high relief. There is no positive evidence for this report, so it is recorded as questionable. Cotter (1983) reported Arthrophycus from the Silurian Tuscarora Formation of Pennsylvania. He did not include a description, but the figure shows specimens with regular annulations, simple form, and both the independent pattern of A. brongniartii and the bundling of A. alleghaniensis. The specimens conform to Arthrophycus, probably as a gradation between those two ichnospecies or as an example of co- occurrence. Crimes (1981) recorded Arthrophycus from the Silurian-Devonian of North Africa. He mentioned that the ichnogenus is indicative of shallow marine sequences and is cosmopolitan in its geographic distribution, but did not discuss specific occurrences, so his report is unverifiable. De Alvarenga et al. (1998) reported Arthrophycus from the Silurian Vila Maria Formation of Brazil. The ichnogenus is mentioned only once in the text and in a figure 102 of stratigraphic correlations. In the absence of pictures and specimen descriptions, the record cannot be confirmed. Duimovich (1963) noted Arthrophycus in Argentina. He did not provide a date for his specimens and recorded only that Arthrophycus was present with Cruziana, with no figures or description, so the record cannot be confirmed. Durand (1985) reported Arthrophycus fiom the Ordovician Armorican Sandstone of France. He described his fossils as cylindrical to subcylindrical, with some branching or overlap of gently-curving shapes, and in the two figures, some annulations are visible as well. Durand suggested that the fossils might be A. harlani, but that many were too weathered for a confident ichnospecific designation. The size is rather large (2-3 cm wide), but it is suggested herein that the specimens be included in Arthrophycus for their annulations and general shape, and in A. brongniartii for their rather independent orientation and lack of bundles or palmate fans. Dutta and Chaudhuri (2005) observed Arthrophycus from the Permian of India. Their paper did not give an age for the fossils, but the Manendragarh Beds of the Talchir Formation are listed as Permian by Ghosh (2003) (and Maejima et al. 2001 list the entire formation as Carboniferous-Permian). The description included a circular cross-section, annulations, the occasional median groove, possible retrusive spreiten, and a slightly curved shape. Some annulations are at an angle rather than parallel, and some seem to occur in pairs. The figure is small and blurred, but on request, Dutta and Chaudhuri supplied a better photograph. This photograph shows several traces in one sample, with rather odd scratch-like transverse markings. The specimens do not have 103 the ring-like appearance of the more typical “annulations” and the markings seem to crisscross in some places. These specimens do not conform to Arthrophycus. Eagar et al. (1985) reported Arthrophycus from the Carboniferous Fletcher Bank Grit of England. The authors noted a subquadrate cross-section, tapered ends, annulations, a median groove, some branching, and a rather small size (3-12 mm wide, with a mode of 7 mm). The tapered ends are odd, as is the angle of branching (80°). Mangano et al. (2005a) were hesitant to accept the report because of the faint annulations and odd branching angle, but the traces are tentatively placed in Arthrophycus based on the strength of the other characters. Frey (1970) noted Arthrophycus-like burrows fi'om the Cretaceous Fort Hays Limestone Member of the Niobrara Chalk of Kansas. The author described the fossils as simple, sometimes branched, cylindrical, and with annulations in some specimens. One of the two figures is at too large of a scale to show details; the other does not show enough. The figured structure is single and may have annulations, but nothing else about it is reminiscent of Arthrophycus. It is therefore recommended, with support fiom Uchman (1998) and Mangano et al. (2005a), that this material be removed from the ichnogenus. Frey (1972) also noted Arthrophycus-like burrows from the Cretaceous Fort Hays Limestone Member of the Niobrara Chalk of Kansas. Arthrophycus is included only in a list on one page and in a block diagram (originally from Frey and Howard 1970). The diagram shows a single branched structure with regular annulations, but the orientation to the substrate is odd and there is not enough evidence to place the 104 specimen in Arthrophycus. Uchman (1998) and Mangano et al. (2005a) also disputed the report. Frey and Howard (1970) noted Arthrophycus from the Cretaceous Fort Hays Limestone Member of the Niobrara Chalk of Kansas and the Star Point Formation of Utah. All of the traces from Utah were generally better preserved than those fi'om Kansas, but the block diagram of the alleged Arthrophycus specimen (originally in Howard 1966) shows only a single branch with annulations. The description of the Utah specimen indicates that a median groove is present in some specimens, and that the branching point was slightly enlarged. The block diagram of the Kansas trace shows annulations and an odd bulge at the branching point, but no other features of Arthrophycus. The description for that specimen, which even the authors question, also indicates that the diameter of the trace was inconsistent. It is recommended herein that both of these reports be removed from the ichnogenus for their odd bulges at the nodes and for the lack of enough evidence to designate them as Arthrophycus. Most other workers support this view (e.g. Hantzschel 1975, Uchman 1998) and Mangano et al. (2005a) suggested that this and similar specimens belong in Thalassinoides. Ghare and Kulkarni (1985) reported Arthrophycus from the Jurassic Khadir Formation of India and are the only authors to report Arthrophycus from the Jurassic. Their description noted an oval cross-section, tapered ends, annulations, and a curved shape without branches. They concluded that the difference in cross-section was due to differences in the consistency of the substrate and was therefore not a taxonomic difference or concern. The single figure shows a few blurred traces with the described simple shape and annulations, along with some possible pseudo-branching. This 105 material should be removed from Arthrophycus based on the lack of enough positive evidence, as Mangano et al. (2005a) also suggested. Gong Yirning (1999) described specimens of possible Arthrophycus from the Carboniferous J ulideneng Formation of China. The author described the specimens as armulated and unbranched, with a faint median groove and subquadrate cross-section. The specimens were uncommon in the formation and not well preserved, and the poor quality of the figure prevents confidence in the ichnogenus assignment. This record is tentatively classified as Arthrophycus based mostly on the strength of the description. Greb and Chesnut (1994) discussed possible Arthrophycus from the Pennsylvanian Breathitt Formation of Kentucky. The authors did not include a description and questioned the taxonomic placement themselves. The sole figure of “Arthrophycos” shows a slightly curving tube-like structure with relatively thick transverse ribs, an inconsistent diameter, and no branches. This material should be removed fi'om Arthrophycus based on the inconsistent diameter and the thickness of the annulations, as Mangano et al. (2005a) also suggested. Howard (1966) reported Arthrophycus fi'om the Cretaceous sandstones of Utah. His description and block diagram present a “segmented trace” with occasional branches with a bulge at the point of branching, a round or oval cross-section, and sometimes a median groove. This material should be removed from the ichnogenus because of its odd bulge and branching pattern, and most other workers agree with this position (e.g. Hantzschel 1975, Mangano et al. 2005a). Kern (1978) observed possible Arthrophycus specimens from the Cretaceous (Sievering Formation) to Eocene (Greifenstein Sandstone) flysch of Austria. He 106 described the fossils as mostly straight cylindrical tubes with annulations, though with a small diameter (1-6 mm); the description mentions branching but the traces are pseudo-branched in the sketch. Mangano et al. (2005a) did not accept this report; Uchman (1998) expressed doubt that it belonged in Arthmphycus. KsiaZkiewicz (1970) mentioned Arthrophycus from the Cretaceous flysch of Poland. She described the specimens as straight and unbranched, with annulations. The tiny sketch shows transverse markings on a very simple structure that does not have nearly enough evidence for an assignment to Arthrophycus. Uchman (1998) placed the specimen in A. strictus, but it is recommended herein that the material be removed fiom Arthrophycus. In a very intriguing report, Laird (1981) noted Arthrophycus from the Ordovician Camp Ridge Quartzite of Antarctica. Unfortunately, this mention was merely a note in a list of ichnogenera that included Daedalus, and the author did not provide any further information, so the record cannot be confirmed. Legg (1985) reported Arthrophycus from the Cambrian Oville Formation of Spain. He included annulations and a faint median groove in the description and noted possible indications of spreiten. The figure shows noticeable annulations that angle toward the middle of the trace, a feature not previously described for Arthrophycus, and the median groove is not apparent. This fossil is in much higher relief than normal for Arthrophycus and may be a vertical stack of traces. Mangano et al. (2005a) questioned the assignment of these specimens because of their vertical nature and suggested further analysis to determine if the specimens belong to Arthrophycus at all. 107 The assignment was classified as questionable herein because of the isolation, atypical annulations, and verticality. Li (1993) mentioned two unnamed Arthrophycus types (A and B) from the Ordovician Gongwusu Formation of China. The author described the traces as branching, with annulations and a simple shape (“twig-shaped”), but these characteristics are not necessarily evident in the photographs. Specimen A is a singleton, is irregular in diameter, does not branch, and annulations are only faintly present. Specimen B shows the annulations more clearly and has branches, but the angle of branching (possibly true branching rather than pseudo-branching) is unusual for Arthrophycus. Mangano et al. (2005a) questioned the placement of the specimens in Arthrophycus based on their indistinct morphology and Uchman (1998) questioned them both, particularly Specimen B. Neither specimen belongs in Arthrophycus, based on the many problems with Specimen A and the unusual branching in Specimen B. Lin et al. (1986) mentioned Arthrophycus in the latest Precambrian Gaojiashan Formation of China. Their description included a circular cross-section, annulations, and orientation of the traces as parallel to the bedding. Of the two included figures, one is an odd J -shaped trace with an uneven diameter and is therefore not Arthrophycus. The other figure is a single trace with a rather wavy shape but more consistent diameter, but the annulations are somewhat lumpy and irregular. It is not quite enough for a confident designation as Arthrophycus. Lopez and Roy (2002) noted Arthrophycus in the Silurian Spragueville Formation of Maine. However, this report was an abstract from a meeting, in which 108 Arthrophycus is part of a list of ichnospecies found in the Cruziana ichnofacies in this stratigraphic unit, so the record remains unverifiable. Maberry (1971) described Arthrophycus from the Cretaceous Blackhawk Formation of Utah. His description included annulations, branching with a bulge at the node, and an oval cross-section. However, the two figures do not depict anything that resembles Arthrophycus. The photographed specimen seems to run perpendicular to the bedding and is a singleton with an inconsistent diameter. The sketched specimen is also vertical rather than horizontal and has irregular walls and annulations; it resembles a badly-stacked column of coins. It is recommended that this material be removed from Arthrophycus, as did Uchman (1998) and Mangano et al. (2005a). Manca (1986) discussed Arthrophycus in the Cambrian Carnpanario Formation of Argentina. The specimens are short, mostly straight cylindrical tubes with annulations. In the figure, the traces are quite short and rather lumpy, so the occurrence was classified as questionable herein. Mergl and Massa (2000) mentioned Arthrophycus in the Silurian Akakus Formation of the Djado Basin (SW Libya or NE Niger). Arthrophycus is “characteristic” of the Silurian Akakus F orrnation there, but is not found in the overlying and otherwise similar beds. The authors did not provide a description or figure, therefore their record cannot be confirmed. Moore (1933) mentioned Arthrophycus as an index fossil of the Silurian of the Appalachian region of the United States. He did not include a description, but the sketch looks like the typical Arthrophycus with its annulations, median groove, simple shape, and pseudo-branching. 109 Mukherj ee et al. (1987) reported Arthrophycus in the Proterozoic Gulcheru Quartzite of India. The authors described their fossils as simple, horizontal, and both branched or not branched. The single figure shows structures that are labeled only as lebensspuren and does not show enough detail for an ichnogeneric assignment. The authors referred to Pettijohn and Potter (1964) for their assignment, but Kulkarni and Borkar (2002) stated that the structures in Pettijohn and Potter’s (1964) work were inorganic sedimentary structures and thus refuted the Mukherj ee et a1 . traces. It is recommended herein that the traces be removed fiom Arthrophycus based on the absence of positive evidence to place them there. Peeples et al. (1997) noted Arthrophycus in the Middle Ordovician Swan Peak Sandstone of Idaho. They indicated that Arthrophycus was abundant in their study, but the report is only a meeting abstract and so could not be confirmed. Pemberton and Risk (1982) mentioned Arthrophycus from the Silurian Thorald Sandstone of New York and Ontario. They described typical Arthrophycus specimens with subquadrate cross-section, faint median groove, annulations, and branching. The authors included a simple sketch rather than a figure, but their specimens conform to Arthrophycus (possibly A. brongniartii) based on the description. Perez and Salazar (1978) observed Arthrophycus in the Cretaceous Dura Formation of Colombia. The authors described their specimens as “ringed, sometimes forked, folded” and as part of the Cruziana ichnofacies, but the plate that they included is small and blurry. It depicts a single, irregular, ridged structure that does not resemble Arthrophycus. 110 Pettijohn and Potter (1964) mentioned Arthrophycus from the Silurian Tuscarora Quartzite of Pennsylvania. The single plate that shows Arthrophycus depicts a network of branched traces without much structure or any other discernible features. The traces could have been weathered enough to erase all structure, or the “traces” could actually be inorganic sedimentary structures, as was proposed by Kulkarni and Borkar (2002). The traces cannot be accepted as Arthrophycus because of the lack of annulations, median groove, appropriate cross-sectional shape, or other features common to Arthrophycus. Pflr’iger (1999) reported Arthrophycus from the Silurian Acacus Sandstone of Libya. He did not describe the occurrence in detail, but the figured specimens have regular annulations, simple structure, and many instances of bundling. This record conforms to Arthrophycus based on those characters and it is furthermore suggested that the specimen be designated as A. alleghaniensis. Pickerill et al. (1991) noted Arthrophycus from the Cambrian-Ordovician Beach Formation of Newfoundland. The authors noted that the fossil was straight and simple in shape with annulations but no branches or intersections, as only one trace was present. The figure emphasizes the poor preservation and lack of very many features. The lack of diagnostic features and presence of only a single trace means that this report should be termed merely Arthrophycus-like. Poiré et al. (2003) mentioned both A. alleghaniensis and Arthrophycus isp. from the Cambrian-Ordovician Balcarce Formation of Argentina. They did not describe the two separately or give any reasons for listing more than one, and included figures only of A. alleghaniensis, so their claim of any specimens that may be different 111 from A. alleghaniensis cannot be evaluated. The report itself is classified as accepted based on the A. alleghaniensis specimens. Romano (1991) also noticed both A. alleghaniensis and Arthrophycus isp. in the same paper, from the Ordovician Armorican. Formation of Spain and Portugal. He did not include descriptions or figures of either Arthrophycus, so it cannot be determined what he perceived as different about some of his specimens. His claim therefore cannot be evaluated. Roniewicz and Pienkowski (1977) reported Arthrophycus from the Eocene- Oligocene Podhale Flysch of Poland. However, this was only in a list of “transversely ornamented knobbly and ridge-like forms,” so the validity of this claim cannot be assessed. Sarle (1906) referred to many specimens of Arthrophycus in the Silurian Medina Sandstone of New York. However, his focus was on determining the burrowing programs of Arthrophycus and Daedalus, not on documenting a particular occurrence. Schiller (1930) discussed an occurrence of Arthrophycus fi'om the Cambrian of Argentina. However, as already discussed above, he found the specimens to be tectonic in origin and removed them fi'om Arthrophycus, and his interpretation is accepted herein. Seilacher and Alidou (1988) described Arthrophycus specimens from the Ordovician—Silurian Kandi Formation of Benin. The authors mentioned three different forms of Arthrophycus: linear, palmate, and deeply-scooping palmate. They did not describe these forms fully, but it is logical to presume from the photographs and 112 drawings that the linear form is A. brongniartii and that one or both of the palmate forms are A. alleghaniensis. All the figured specimens have annulations, simple shape, and branching, so they conform to Arthrophycus. Silva (1951) mentioned “Arthrophycis” in the Silurian of Brazil. The paper did not include a description or figure, so the record cannot be evaluated. Stanley and Feldman (1998) found possible Arthrophycus specimens in the Ordovician Deadwood Formation of South Dakota. The authors claimed that the specimens were bilobate and armulated, but not branched, and the single figure depicts a single specimen with no apparent annulations or groove. The authors questioned the placement of the specimens in Arthrophycus, noting that their material was fiagmentary. Mangano et al. (2005a) rightly removed these specimens from Arthrophycus. Terrell (1972) mentioned a possible Arthrophycus specimen from the Permian Elephant Canyon Formation of Utah. He did not include a figure or describe the structure specifically, but noted only that it was similar to Arthrophycus. The Utah specimens described by Howard (1966), Frey and Howard (1970), and Maberry (1971) did not conform to Arthrophycus, Terrell’s (1972) specimen to be doubted herein, but without a figure or description, the record cannot be confirmed. Wagner (1978) noted Arthrophycus from the Silurian of Tennessee and Pennsylvania. The report was only a meeting abstract, so it cannot be confirmed, but at least one of the two formations cited in the paper (the Tuscarora Sandstone of Pennsylvania) has produced Arthrophycus in the past. 113 Webby (1977) observed Arthrophycus in the Cambrian-Ordovician quartz-rich clastics of New South Wales, Australia. As this is the only report of Arthrophycus fiom Australia, it is rather intriguing, but it is a meeting abstract and thus is not part of the peer-reviewed literature and so cannot be confirmed. Wolfart (1981) mentioned Arthrophycus in the Silurian “Worm Burrows Sandstone” of Jordan. He did not discuss or figure any specimens of Arthrophycus, so his claim cannot be evaluated. Yang et al. (1996) mentioned Arthrophycus in the Late Triassic Zhuwo Formation of China. They are the only authors to report the ichnogenus from the Triassic, but they did not include a description. The single figure is far to dark to reveal any details, and the traces come fi'om black silts that may be deep marine flysch, which is highly unusual for Arthrophycus. There is no positive evidence to include this material in Arthrophycus, so it must be recommended that the material be removed from the ichnogenus, as did Mangano et al. (2005a). Taxonomic Conclusions After reviewing the publications, five ichnospecies were accepted herein as valid Arthrophycus: A. alleghaniensis, A. brongniartii (= A. linearis), A. lateralis, A. minimus, and A. parallelus. This evaluation was based on the major morphologic characters including annulations, simple form, branches, bundles, shape of cross- section, median groove, and internal structures. Table 3.2 summarizes these assessments of ichnospecies and provides some suggestions of synonyms from other authors. 114 The review of Arthrophycus literature revealed five characters in addition to those given by the authors of the primary literature on Arthrophycus (Harlan 1831, Hall 1852, Goppert 1852, Hantzschel 1975) in assigning specimens to Arthrophycus. The review also documented variability in some characters, such as the inconsistency of the median groove, definitions of annulations, presence of bundles and/or branches, and the question of singletons, and inconsistency in applying diagnostic characters (i.e. the ones listed by the primary authors). With so many variable characters, the question arose as to how many of them are necessary for designation as Arthrophycus, and if a lack of a single character might be enough for rejection of a specimen or proposed ichnospecies. Condensing so many characters for so many taxa is difficult to do qualitatively, but a quantitative approach may help to reduce all of the information in an objective manner. These issues and the plethora of both specimens and ichnospecies referred to Arthrophycus make precise definitions of Arthrophycus and its diagnostic characters difficult, and highlight the desirability of applying more objective, reproducible techniques in making taxonomic decisions. 115 Table 3.1: Published ichnospecies of Arthrophycus. First Reported Reported Ichnospecies Date Author Time Place Rock Type alleghaniensis 1831 Harlan Sil New York ss1 brongniartii 1832 Harlan Sil New York ss hariani 1838 Conrad Sil New York ss montalto 1888 Simpson Camb Pennsylvania ss siluricus 1879/1890 Schimper Camb Italy sh2 elegans 1901 Herzer Penn Ohio ss conugatus 1908 Fritsch Sil Czech Rep. qztit3 flabellifonnis 1940 Hundt Ordo Germany qztit krebsi 1941 Hundt Ordo Germany qztit? minor(i)censis 1973 Bourrouilh Carb? Minorca ls.? sand-mud? annulatus 1977 KsiaZkiewicz Cret-E0 Poland 55 strictus 1977 Ksia2kiewicz Cret Poland 55 dzulynskii 1977 KsiaZkiewicz Oligocene Poland ss tenuis 1977 KsiaZkiewicz Oligocene Poland 55 qiongzhusiensis 1994 Luo Camb China sh, ss, siltss4 tarimensis 1994 Yang Ordo China not spec. hunanensis 1996 Zhang Sil China ss linearis 1997 Seilacher Ordo-Sil several 33? unilateralis 1997 Seilacher Sil? Libya? ss? Iateralis 2000 Seilacher Ordo-Sil Libya ss simplex 2002 Seilacher Ordo Libya ss minimus 2005 Mangano Camb Argentina mudsss. SS parallelus 2010 Brandt Carb Michigan 53 Arthrophycus 1852 Hall Sil New York 38 (Harlania) 1852 Goeppert Sil New York 53 ' ss = sandstone 2 sh = shale 3 qztit = quartzite " siltss = siltstone 5 mudss = mudstone 116 8838...? 8.8.. 8 80> 382 _.m .8880 82 8.88:. .88 8 93> 3.2 _.m .3188 8.88.2.8... 88 8 58.8.... 28 .8328 8.9.88 coco mm .83.: «5.59.8. 9:80 0:355. woow 9.8.5:. €8.58... .< 8 gen: 23 mm m3... 020 .m..om..o.w woow x2958 .88 8 93: 8.8.0 889.8 88 2.898. 8.388. .< E93: 28:6: «mm 8.3... e.w 528:3 32 m...Eo.m...§ ..Em..:m:2c .< L0...... «mm .993 .5690 .m...om..om 52 8.58:... 8.. 8 8.8 .8 88w 82 8.8888 can 8on .0: 2.20 85 98> 39 @8858. 39.8.26... 8 23 m8...» .3 .5 mch 950 03 3? 8.6583833 8.35.. $8.38 own mm canon. 8589.0 8.32%.23. RE. 8.322 8.28:8 8839.8.en. 8.. 8 Been. 888.6 8.32.8.3. tee 8.888 can mm vegan. .90 8.32.8.3. K? 3.8.58 8.83:8 38026qu m 23 mm Ewan. 3-8.0 3.26333. R? «38.32% Eng: :28: on: E65» 8....QO 350 55258 as? e.w:ooeeoSE 8.. 9.8 28:80 8.0 .82. :8. .88. EamEmEEfi .Q nan ENG 885.80 020 .95... ova? 8.25.338. 88.88. a. 8.. s8 .8”. :88 _.m 82E 82 8.83.8 8.8533828 .< m can mm 0.50 Sea 880.... 32. 3828 E88 :28: 8 as. 980 .8528 8a :82 88.88 3.89.3 3398?. 8 won an £52853. nEmo comaE.m mm? 2.822: 8.65.5828 .< 5.5... mm fo> .582 _.w 89:00 82. .Emtmc 88 8 80> 282 __m 88: 8.: 88.85.. .888 8 8> .582. __m 88... 82 e.e=e.eee8..e Exec—Sm 331mm?» EoEoeoeu< 2:... zoom can... 2...... 35.3 «2.... 83 8.02.853. 8:89... 8:83. .eaeée‘exxtwwo 8.02.853. @88er some Mo was; 05 $58me 358888.. 9 0230 ”Nm 2an 117 Figure 3.1: Drawings of F ucoides alleghaniensis from Harlan’s original work (1831). Silurian of Pennsylvania, found in a tavern wall. No scale given. Figure 3.2: The first known drawing of E brongniartii, fi'om Taylor (1835). Silurian of Lewiston, PA. No scale given. 118 Arthrophycus Figure 3.3: Seilacher’s original drawings of A. linearis (top left) and A. unilateralis (bottom) (1997). No locality information or scale given. Figure 3.4: A. linearis in Seilacher (2000), along with a figure of wrinkles on that specimen. Silurian of Rochester, NY, Medina Sandstone. Scales represent 3 cm for (a), 10 mm for (b). 119 Figure 3.5: A. montalto from Lesley (1889). Cambrian of Pennsylvania, found in a sawmill wall. No scale given. Figure 3.6: A. elegans from Herzer (1901). Carboniferous of Marietta, OH. Thinner traces are approximately 1/16” wide, wider traces are approximately 3/16” wide. 120 Figure 3.7: A. corrugatus (as Radicites rugosus) from Fritsch (1908). Silurian of Czech Republic. No scale given. Figure 3.8: A. flabellzformis from Hundt (1940). Ordovician of Wuenschendorf ad Elster, Germany. No scale given. 121 Figure 3.9: A. krebsi (arrow with base line) from Hundt (1941). Ordovician of Wuenschendorf ad Elster, Germany. No scale given. 122 Figure 3.10a: A. minoricensis from Orr (1994). Lower Carboniferous of Menorca Island. Traces are approximately 1 cm wide. Figure 3.1%: A. minoricensis from Llompart and Wieczorek (1995). Carboniferous of Minorca Island, Culm siliciclastic sequence. No scale given. 123 , Figure 3.11: A. annulatus from Ksiaikiewicz (1977). Lower Eocene of Poland Ciezkowice Sandstone (8) and Turonian, Inoceramian Beds (9). Traces are approximately 1 cm wide. Lgota Beds 3 Figure 3.12: A. strictus from Ksiazkiewicz (1977). Cretaceous of Poland (1 l), Ropianka Beds (12). Traces are 1.5 to 2.0 mm wide. 124 Figure 3.13: A. dzulynskii from Ksia'zkiewicz (1977). Oligocene of Poland, Krosno Beds. Traces are approximately 1 cm wide. Figure 3.14: A. tenuis (as Sabularia tenuis) from KsiaZkiewicz (1977). Oligocene of Poland, Krosno Beds. Traces are 0.5 to 1.0 mm wide. 125 Figure 3.15a: A. qiongzhusiensis from Luo et al. (1994). Early Cambrian of China, Yu’anshan Member of the Chiungchussu Formation. It was not clear from the paper which structure is Arthrophycus; the most likely candidate is circled. Figure 3.15b: A. qiongzhusiensis, also from Luo et al. (1994). Early Cambrian of China, Yu’anshan Member of the Chiungchussu Formation. No scale given. 126 Figure 3.16: A. tarimensis from Yang (1994). Lower Ordovician of the Tarim Basin, China, Ulikeztag Formation. No scale given. Figure 3.17: A. hunanensis from Zhang et al. (1996). Silurian of China, Xiaoheba Formation. Traces are approximately 1 cm wide. 127 Figure 3.18a: A. Iateralis from Seilacher (2000). Lower Silurian of Libya, Akakus Sandstone. Scale bar in corner represents 3 cm. x aw 188.33%? '4 ‘ . _- a. 3“ ” arifl‘é’fi )3» ‘ ’ 89%»? %g%%%§ééfié , ,1 fig w ’ h V Arthrophycus/ateralis w%§%a§ejgfi L. Silurian, Libya (protrusive) Figure 3.18b: Sketch of A. lateralis fi'om Seilacher (2000). 128 a" 3 Figure 3.19a: Loop of Arthrophycus showing asymmetry of the bundling toward the bottom of the photograph. Silurian of Rochester. NY, Grimsby Sandstone. YPM207001. Scale bar represents 2 cm. Illlll Figure 3.1%: Asymmetrical bundle in an Arthrophycus specimen, labeled as A. harlani but possibly representing A. lateralis. Silurian of Tyrone, PA, Tuscarora Sandstone. PRI, unnumbered specimen. Scale bar is in cm. 129 "*» ya Figure 3.20: A. simplex (arrow) from Seilacher et al. (2002), with specimen of Cruziana goldfirssi. Ordovician of the Kufra Basin, Libya, Hawaz Formation. Figure 3.21: A. minimus from Mangano et al. (2005a). Upper Cambrian of Argentina, Alfarcito Member of the Santa Rosita Formation. Scale bars represent 1 cm. 130 Figure 3.22: Unnumbered specimen referred herein to A. parallelus, from the New York State Museum. Devonian of New York, possibly Hamilton Group. Scale bar represents 1 cm. Figure 3.23: Arthrophycus isp. Late Devonian to Early Mississippian, Rowlesburg, WV, Oswayo Member of the Price Formation. CMNH8340. Scale bar represents 1 cm. Specimen is the same as that figured in Bjerstedt (1987). 131 Chapter IV. Definition of Arthrophycus, Part III.: numerical ichnology Previous Use in Ichnology Although most techniques for evaluating ichnotaxa are wholly or partially qualitative, some quantitative methods have been used. In a study of vertebrate trackways, Demathieu and Demathieu (2002) used a system of measured characters (e. g. width of digit, stride and pace lengths) to analyze different ichnospecies and estimate potential for overlap in the definitions of the ichnotaxa. Lockley (1998) also attempted to evaluate the use of measurements for vertebrate trackways but met with mixed results and difficulty of standardization, while Trewin (1994) attempted to standardize arthropod trackway identification with character states coded as numerical data. In Arthrophycus, the only attempt at quantitative data comes from Neto de Carvalho et al. (2003), who published a single small graph comparing diameter measurements of A. brongniartii (as A. linearis) and A. alleghaniensis specimens. No other quantitative attempts in ichnotaxonomy are known. Basic Methods In the statistical program PAST (Hammer et a1. 2001), a simple clustering diagram, or dendogram, is a way of showing hierarchical distance relationships among groups. Taxa that are more similar to each other will group together, with more- dissimilar taxa placed farther away. As with Cladistics, all nodes in the dendogram can be rotated; left and right placements have no particular meaning. The first step is the creation of a data matrix of numerically coded character states for each character. In the data matrix, OTUs (operational taxonomic units, in this case equal to ichnospecies with some exceptions) are in rows and characters are in 132 columns. Each character has a number of different character states (e.g. groove: absence and presence), which are each designated by a number (e.g. absence of a groove = 0, presence = l). PAST then uses these data to construct the dendogram or cladogram. There are three different algorithms available in PAST for cluster diagrams, and twenty-two different similarity measures, not all of which can accept more than two states per character. Ward’s method was the algorithm chosen for this study, because it is suitable for studying morphological data and permits the use of multistate characters. Ward’s method assumes Euclidean distance as the similarity measure, so no additional decision among the similarity measures is necessary. The characters chosen were based on those that are common in the literature describing Arthrophycus. These characters are median groove, shape of annulations, shape of cross-section, proximity to other traces, overall shape, palm shape, radial shape, branching pattern, parallel behavior, orientation to the substrate, internal structure, wrinkles, shape of the ends of the trace, smoothness of the walls, and width. This list includes the common characters from Harlan (1831), Hall (1852), Goppert (1852), and Hantzschel (1975); the others are those that recur in published descriptions and that were noted during examination of specimens during this study. In order to code the character states for each ichnospecies, the original description and figure (drawing or photograph) were examined wherever possible. If the image and the specimen did not agree, the character state was coded based on the image or one of the intermediate states for the character was chosen, if intermediate states were available and not contradicted by the image, because intermediate states 133 minimize the degree of uncertainty. In cases with more than one report of the ichnogenus (widespread for A. alleghaniensis, common in A. brongniartiiflinearis), all available photographs and written descriptions were inspected. If the specimens were correctly diagnosed in these descriptions, then the codings from both should match. This was usually the case; the only notable exceptions were the measurements for width, as detailed below. In the case of A. alleghaniensis, it is possible that some specimens designated as A. alleghaniensis should be A. brongniartii, so Harlan’s (1831) sketch and/or a simple majority of all available illustrated material were used in case of any discrepancies. The entire data matrix may be found in Appendix III. Character States Annulations were initially coded as either present or absent. All of the proposed ichnospecies of Arthrophycus have annulations present, so this was an uninformative character. However, on closer inspection, differences in shapes of the annulations were apparent and this character became more interesting. The states for annulations are now coded as ring-like (0), concavo-convex (1), chevron-like (2), or biconvex (3) (Figure 4.1). The median groove, internal structure, and _wLink1_es_ are all coded as either present (1) or absent (0). Cross-section is the second multistate character. The three states are subquadrate (0), cylindrical (1), and elliptical or oval (2). Closeness is an attempt to code how gregarious the traces are. The traces may be singletons (O), few traces (l), somewhat close (2), or dense (3). 134 Overall shape was difficult to code because the states grade into one another. The burrows may be predominantly curving (0), predominantly straight (2), or intermediate between the two (1). For this character in particular, both the published descriptions and all available figures were used to assess the shape as accurately as possible. Palm sham refers to traces that form bundles. The states are: not bundled (O), asymmetrically palmate (l), and symmetrically palmate (2). Radial shap_e refers to traces that form a radial pattern from a single point, rather than a bundle with an axis. The states are not radial (O) or radial (l ). Branching pattern was difficult to code because the degree of branching of traces can be difficult to quantify. For this character, the published descriptions were the primary sources of information (“extensive” means whatever the original authors meant), as well as the images. For the purposes of this coding, pseudo-branching and true branching are regarded as the same character state. The states are: no branching (0), some branching (1), and extensive branching (2). Parallel behavior is coded as not parallel (0) or traces predominately parallel to one another (1 ). Planar relationships refer to the orientation of the traces with respect to one another in three dimensions. They are either in a single plane (0) or in several (1). End_srefer to the nature of the termini of the traces. They are termed blunt (0), pointed (l), or knobby (2). Traces that are bundled are still considered to have blunt ends unless the unbundled end is pointed. Traces in which one or both ends are not 135 observable because the trace goes off the edge of the sample or descends beneath the surface were coded as blunt (0) because it is the most common character state. Diameter refers to the variability of the diameter: the trace is either largely variable along its length (0) or consistent (1). As with the trace terminus, bundled individuals are not examined at the bundled end and traces that extend beyond the slab are examined only where observable. The width of the ends of the trace, where applicable, is part of the “end” character above and is not included here. Smog refers to the condition of the surface of the sides of the trace: smooth (0) or rough and irregular (1). Only one OTU was coded as exhibiting character state 1. As noted above, _w_id_th was the only notable character for which there was variation among the reports of the different ichnospecies. All reported width measurements for each ichnospecies were recorded and a range of measurements for each ichnospecies was compiled. Table 4.1 is a diagram of these ranges and is divided into three size classes: 0-5 mm (0), 5-10 mm (1), and greater than 10 mm (2). All of the ichnospecies with width measurements fell nicely into these three categories, with three exceptions: A. alleghaniensis, A. brongniartii, and A. linearis. A. brongniartii has a reported range of 5-14.1 mm, with the greater sizes being more typical, so I used an approximate midpoint of 11. A. alleghaniensis has a very large reported range of width measurements, but most of them fall between 5 and 15, with a larger tail toward the higher end, so a midpoint of 11 was suitable here. A. linearis also has a large reported range (mostly from Neto de Carvalho et al. 2003), but nearly all of its width measurements are over 10 mm. 136 Missing Data As is ofien the case in paleontology, there are a number of instances of missing data in the data matrix, despite efforts to be complete. Some of the illustrations available were not sufficiently clear for confident assignment of a particular character’s state, while many of the written descriptions also did not provide enough information. Because PAST cannot run the cluster analysis with missing information, the state that would minimize error for each empty cell was chosen.6 For characters that are either present or absent, “absent” may be assumed to mean “absent or unobservable.” However, some ichnospecies had excessive missing data. A. siluricus, which had no available drawing or photograph and essentially no description (Schimper and Shenk 1879-1890), did not fmnish any data at all. A. krebsi (Hundt 1941) and A. flabelliformis (1940) both had photographs and some description, but far too little to fill in the data matrix with any confidence. These three ichnospecies are therefore eliminated from the analysis. A. unilateralis and A. harlani are also absent from the analysis, because their synonymy with other ichnospecies of Arthrophycus has been clearly established and accepted, and thus their inclusion was not necessary. Conversely, one ichnospecies had to be divided into three OTUs for this analysis. A. strictus has been described in two reports, that of KsiaZkiewicz (1977) in the Cretaceous and Paczesna (1996) in the Cambrian. Such a broad stratigraphic range, with no occurrences in between, suggested that the two should be considered 6 Groove: A. minoricensis = 0. It is better to infer absence than presence. Cross-section: A. montalto, A. elegans, A. corrugatus = l. Cylindrical is the closest to an intermediate state for this character, reducing potential error either way. Shag: A. strictus (second KsiaZkiewicz type) = 2. Ksialkiewicz noted that the second type was similar to the first type she described, which is coded as 2 herein. Width: A. montalto = l. A. montalto had no size descripton and the figure had no scale. Character state 1 was chosen because it is the width most often found in “typical” Arthrophym specimens and because 1 is the intermediate state. 137 separately; if indeed similar, they should plot together on the dendogram. In addition, Ksiazkiewicz described two similar morphotypes of A. strictus in her original description (1977). They were of different sizes and had different types of branching, so I chose to code them separately (Uchman (1998) also supported the idea of two different morphotypes in Ksiazkiewicz’s work). Again, if truly similar in spite of those differences, the morphotypes (designated strictus K1 and strictus K2) should plot together. Cluster Analysis Results The cluster diagram produced by PAST is in Figure 4.2. The numbers along the y-axis represent the Euclidean distances between clusters; higher numbers indicate groups of ichnospecies that are not very similar to one another, while lower numbers indicate more similar groupings. Depending on the level of similarity desired, the cluster diagram can be divided into several groups. The cluster on the left side is composed of only six ichnospecies: A. alleghaniensis, A. brongniartii, A. linearis, A. lateralis, A. minimus, and A. parallelus, and the rest of the ichnospecies fall into a separate cluster. It is striking that these six ichnospecies should cluster together, and be so distant from the other ichnospecies, because they are all ichnospecies previously supported as conforming to Arthrophycus in the qualitative section. Within the cluster of six ichnospecies, A. linearis and A. brongniartii plot very closely together, which would be expected if they are indeed synonyms as Rindsberg and Martin (2003) suggested. In fact, the only difference in the coding between A. linearis and A. brongniartii is that the branching pattern of A. linearis is “some 138 branching” and that of A. brongniarti i is “no branching” — these are probably quite similar orientations and possibly an artifact of the illustrations. The analysis therefore supports the synonymy of A. linearis with A. brongniartii. Also within the same cluster, A. lateralis and A. alleghaniensis clustered more distantly, which was a surprise because these two ichnospecies appeared very similar except for the direction of their palmate forms. However, A. lateralis has less curvature, which is probably related to the difference in the shape of the palm fan. As A. lateralis was named only within the last decade, a lack of study and few occurrences could explain the apparent absence of wrinkles and the narrower range of widths (which is, however, within the lower bounds of A. alleghaniensis). A. lateralis is most similar to A. minimus, another unanticipated result, as A. minimus is more similar to A. brongniartii and A. linearis in its shape. A. parallelus, similar to A. minimus in size, differs in other characters (e.g. shape of annulations, parallel behavior), but still plots with the other five ichnotaxa determined previously to be Arthrophycus. The remaining putative ichnospecies of Arthrophycus all plot into one large cluster on the right of Figure 2. Therefore, none of these is considered to be Arthrophycus: they should be removed fi'om the ichnogenus and, where applicable, be placed in other ichnogenera as other authors have suggested (see Chapter [11). All three of the morphotypes of A. strictus fell into the same sub-cluster of this analysis, indicating that they may be the same morphotype. The position of A. simplex within the outer cluster was unexpected. The conclusion from the qualitative analysis was that A. simplex was actually A. brongniartii (=A. linearis), because A. linearis was the only ichnospecies of 139 Arthrophycus mentioned in the text of the initial paper that mentioned A. simplex (Seilacher et al. 2002). It is possible that some of the coding choices for A. simplex were in error because there were so few photographic images and no published written descriptions to draw from. Width may not be a reliable character, due to the large ranges and overlaps in sizes of ichnospecies, and because size is not usually considered a valid ichnotaxobase (Bertling 2007). To determine the impact of width, the analysis was run with the same parameters as before, but without the width data (Figure 4.3). The six taxa that were supported as belonging to Arthrophycus are still together, but some of the internal relationships changed: A. parallelus and A. alleghaniensis were each outsiders to the subgroup that includes the other four OTUs. The positions of most of the OTUs within the larger cluster changed, preserving few of the previous sub-clusterings. Notably, the three morphotypes of A. strictus no longer fell in the same sub-cluster: the Cambrian specimens described by Paczesna (1996) are more similar to the holotype of A. strictus (strictus Kl) than the two Cretaceous morphotypes described by Ksiazkiewicz (1977) are to each other. In order to discover which characters the software used to group taxa, PAST was used to make a two-way cluster diagram, with OTUs on one side (Q-mode analysis) and characters on the other (R-mode analysis) (Figure 4.4). The data matrix between the two dendograms indicates the character state for each OTU for each character, and non-zero numbers are indicative of an impact on the groupings. For the cluster containing the six Arthrophycus ichnotaxa, there is no distinct cluster in the R- mode analysis of important characters; the contributing characters are scattered in the 140 dendogram. The characters contributing to the group of six ichnotaxa are diameter, shape, shape of annulations, width, internal structure, plane, median groove, and closeness as the most important. These eight characters should therefore be a major part of any definition of Arthrophycus. Principal Coordinates Analysis (PCO) The PAST software can also produce a “map” of the character data to identify patterns in how the OTUs are grouped together. Principal coordinates analysis reduces the data to the primary axes of variation, which are plotted in two-dimensional space (Legendre and Legendre 1983). As in the cluster analysis, closely-related taxa should cluster closer together in the plot space. For consistency, the Euclidean similarity index was used in the principal coordinates analysis as well as in the cluster analyses. The first three axes were the most important, comprising 58.545%, 20.852%, and 10.649% of the variance, in descending order. The OTUs conformable to Arthrophycus are represented by the + symbols and the other OTUs are represented by dots. Figures 4.5 and 4.6 show the results of this analysis with coordinates 1 and 2 and coordinates 1 and 3, respectively. The groups identified earlier in the cluster diagrams are still present as entities in both PCO diagrams, as they group together and do not overlap, though one of the Arthrophycus points (A. parallelus) plots further toward the middle than the others do. Coordinate 1 provides the separation of the clusters in both diagrams, while Coordinate 3 separates them to some degree but Coordinate 2 does not at all. As Coordinate 1 has most of the weight (58.545%), it is the most important axis. PCO analysis thus supports the results of the cluster analysis 141 and the conclusion that the “true Arthrophycus” cluster is real and distinct from the other OTUs. Cladistic Analysis A cladistic analysis requires the choice of an outgroup. It is possible to take the oldest member of the ingroup and code it as the outgroup, but there are several potential problems with this. One is that some doubt remains over which, if any, of the oldest ichnospecies of Arthrophycus are true representatives of the ichnogenus: A. qiongzhusiensis, A. strictus, and A. minimus are all Cambrian, and the assignments of the first two are particularly questionable. Another problem is that preconceived ideas of the evolution of Arthrophycus may not be consistent with the oldest representatives known as being necessarily the most “primitive,” meaning that they should belong toward the base of the cladogram. The other possible outgroup is a non-Arthrophycus ichnospecies from an ichnogenus that is thought (e. g. Seilacher 2000) to be related to Arthrophycus, such as Daedalus. Many of the characters used to describe Arthrophycus are applicable to Daedalus, but there is still the problem of which ichnospecies of Daedalus, or any other proposed ichnogenus, to code for use as an outgroup. The ichnospecies of Daedalus are different from each other (Durand 1985), just as the species of Arthrophycus are, and thus the selection of any particular ichnospecies of Daedalus could greatly change the results of the cladogram generated. By its very nature, the selection of an outgroup will polarize the character states of each character, since whatever state is present in the outgroup will be presumed to be primitive. 142 A cladistic analysis was performed using PAST to test the efficacy of the technique with ichnological data. The Cambrian occurrence of A. strictus (strictus P) was chosen as the outgroup because it was one of the Cambrian reports of Arthrophycus, although its conformity to Arthrophycus is questionable. The first appearance datum (FAD) and last appearance datum (LAD) for each OTU were also included in order to calculate the stratigraphic consistency index (SCI) for the cladograms, using the same midpoint ages as for the size graph in Chapter 11 (Figure 2.24). The heuristic algorithm was used for all of the cladograms because there are too many OTUs for the other algorithms (even branch and bound works better for fewer than fifteen OTUs). The tree bisection and reconnection (TBR) rearrangement was used because it can usually find shorter trees than can the other heuristic options (Hammer and Harper 2005). The Fitch optimization, which allows the sequence of characters to be reversible and unordered (all steps have equal length in any direction: 192 = 092 = 291 and so on), involves the fewest assumptions and is therefore preferable (Hammer and Harper 2005). The first cladistic analysis produced three most parsimonious trees (Figures 4.7 and 4.8). The same cladistic analysis with a different Cambrian ichnospecies, A. qiongzhusiensis, as the outgroup, produced nineteen most parsimonious trees (Figure 4.9). The cladograms revealed a similar but different story from the cluster diagrams, which is due to the different assumptions present in the cladistic techniques, chiefly 143 that of an outgroup, which forces one OTU to be on its own and unconnected with any others; none of the OTUs were isolated in the cluster diagrams. There were some parallels between the phenetic and cladistic analyses. With the Cambrian A. strictus as the outgroup (Figures 4.7 and 4.8), the six verified Arthrophycus ichnotaxa remain together, though not in the same order, and the other OTUs group away from them. This provides further support for the idea that this cluster represents a real grouping of ichnospecies that properly belong in Arthrophycus. The remaining OTUs have sub-groupings different from either of the cluster diagrams. The length of the cladogram in Figure 4.7 is 49, which means that 49 steps were necessary to map all the character states. The cladograrn’s ensemble consistency index (CI) is 0.5417, which means that the levels of homoplasy in its characters are not too high (45.83%, or about half). The ensemble retention index (RI) is 0.6857, which is also a fairly good indicator of lack of homoplasy. The stratigraphic consistency index (SCI) is 0.3889 for the third tree, which implies many gaps and ghost ranges. This last is obvious even when considering only the three Cambrian occurrences: one is the outgroup and the other two are embedded in “crown” sections. In the other cladistic analysis, A. qiongzhusiensis was used as the outgroup, which resulted in nineteen most parsimonious trees with a length of 49 (Figure 4.9 contains Tree #1). These trees have an ensemble CI of 0.5306, and an RI of 0.6761 , which are all comparable to those of the first cladogram with A. strictus as the outgroup. The SCI is 0.4444, indicating that this cladogram fits the stratigraphic record slightly better than does the one above. 144 The Arthrophycus cluster is still one unit in the middle left of the cladogram, forming a monophyletic group. The other ichnotaxa again have changed their sub- groupings and do not form a monophyletic group. However, it should be noted that as a result of the use of the heuristic algorithm, both of these cladistic analyses can be run many times and yield slightly different results of the same tree length. Potential Errors As with the qualitative data, there are a number of possible errors arising from preservation of the trace fossils. The characters were coded based on the available images and descriptions, but if features were present at the time of the trace formation and subsequently eroded, some characters may be coded incorrectly. Some of the characters are gradational, without clear division between their states. For example, “shape” includes curving, intermediate, and straight. As noted above, the measurement of width involves some overlap of categories and two of the ichnospecies have large reported ranges. In all cases of uncertainty, the codings chosen were those with the greatest potential for minimizing error. Codings also depend on the accuracy of the translations of foreign-language descriptions. Three of the ichnospecies were originally described in German, but were excluded based on insufficient information. A. corrugatus was originally described in French, but a subsequent report was in English. Harlan’s original descriptions of F. alleghaniensis and F. brongniartii were in Latin, but as there are numerous other reports of both ichnospecies, that translation should not be a problem. However, three new ichnospecies were described only in Chinese, and the coding for those ichnospecies (A. qiongzhusiensis, A. tarimensis, and A. hunanensis) rests largely on the 145 translated descriptions because of the poor quality of the figures. Finally, A. minorcensis was initially described in French, in an unpublished PhD thesis, so the papers by Orr (1994) and Llompart and Wieczorek (1997) were used to code the character states. It is entirely possible that some of the specimens assigned to Arthrophycus may not be Arthrophycus at all. Apart from A. alleghaniensis and A. brongniartii, originally assigned to F ucoides (Harlan 1831 and 1832), A. corrugatus and A. tenuis were originally assigned to other ichnogenera, and most of the others have been questioned in the previous chapter. One of the reasons in using a phenetic analysis here is to utilize quantitative techniques to sort out misidentifications of this sort. There are also many specimens of Arthrophycus identified only to the ichnogenus level. These specimens are not included in the main analysis, but it may be possible to incorporate some of them in future work, in an effort to find which, if any, of the established ichnospecies an unknown specimen most closely resembles. Numerical Taxonomy Conclusions The cluster analyses, PCO diagrams, and cladistic analyses converge on a consistent classification of Arthrophycus ichnospecies: A. alleghaniensis, A. brongniartii (= A. linearis), A. Iateralis, A. minimus, and A. parallelus. The numerical analyses supported the tentative conclusions of the qualitative analysis (Chapter III). The qualitative analysis had left the placement of a few ichnospecies unresolved: the placement of A. strictus and of some ichnospecies considered to be possible junior synonyms of A. alleghaniensis (A. hunanensis) or A. brongniartii (A. montalto, A. elegans, and A. simplex). Based on the cluster diagrams, none of these 146 ichnospecies is actually Arthrophycus and none are closely related to one another. These placements could be real, or may be artifacts of the absence of information. The hypothesis that A. simplex was mislabeled and is actually A. brongniartii may be correct, because the coding differences could have been a result of insufficient illustration or lack of other reports of the ichnospecies. The clusters of non-Arthrophycus OTUs do not represent real taxonomic entities. Although the OTUs within it plotted together in the PCO graphs, the internal relationships did not remain the same throughout the two cluster diagrams or the cladograms. As indicated in Chapter III, other ichnogeneric assignments have been proposed for many of these ichnospecies, so it is likely that they do not represent a single cohesive group. These ichnospecies are simply taxa that have mistakenly been placed in Arthrophycus and do not necessarily bear any relationship to one another. The cluster diagrams, cladograms, and principal coordinates analyses tend to support both one another and the qualitative observations, but there is an important caveat: the coded characters are all based on qualitative observations; even width is broken into three qualitative states. The analyses all support one another, but this may be largely because they were already linked. However, through precise definitions of characters, the quantitative analyses offered a degree of objectivity not available in qualitative descriptions. The difference is that the treatment is rigorously systematic and the subjectivity inherent in qualitative treatments is controlled. Phenetics should prove to be a useful tool in studies of trace fossil taxonomy. Cluster diagrams provide a more concise visual summary of the data than does a list of features, and PCO compresses many dimensions into a few simple plots. In the case of 147 Arthrophycus, the cluster diagrams have helped to show that nearly all of the ichnospecies assigned to the ichnogenus were placed there in error. Cladistic analysis proved to be less useful, but may work well for a small sample size that will allow the use of the branch-and-bound algorithm. An Application of the Method The numerical analyses used herein have potential for evaluating new ichnospecies or specimens that could be Arthrophycus. The methods proved useful in two additional trials: 1) The unlabelled specimen fi'om the New York State Museum at Albany, similar in size, shape, and orientation to A. parallelus, was added to the cluster analysis as a new OTU. The character states were all identical to those for A. parallelus, supporting the hypothesis. 2) The fossil reported by Kern (1978) represents a report whose placement in Arthrophycus was uncertain in the qualitative analysis. Therefore, this report was coded as a new OTU and included in separate cluster and principal coordinates analyses. Kem’s report plotted well within the non-Arthrophycus cluster in both analyses, so this occurrence is rejected as Arthrophycus. 148 mime... «5.2.qu 0.70. _. w:E..EE .96 erEE . _. Tn £6.65 .NNrw mtmmt: ., . .... . .3. E8588 I .0 Ta mmmtmtmtnc "1 .. .Ym m..m=mE.=a. r .. .... N; 6.6.8.6383... .. ... ofmd @522 H.HHUMHHH . r 79 $2528 w- .. .3 3:88.58 ... «-3 5.8.0.8 . t..- O_. hamsttm film or m..m:oo€..o:..E .385. m..:to.....oa§ N73 negate... .... .. . .. n 310.? .3303 9.6.5.? 2.8.5.... .. . . xi : r i z —..vFro.m .EKWEGQEG . ll 88 8888...... .mm vm mm mm rm om 2 2 t. 9.2 3 9 Ne : or m w n m m v.m...Nw ES 52; 022. 85.0. 149 £58.88 888.8868 on... 8.8:. MN one 28 55305.0. 85 no.“ 2:23 we emcee 2.8.90 208.. 65 880%.: 8.8 :888 2:. .ownfi ombu— bo> a £232.: .V 28 .....tufimgaa .V £8222§m§e .Vmo 88 05 a. do 895 we amigo an 88:5. M59 E $.85 .37. “wows. 2.: mm 6:3 98 .BEBEBE m. 5on .38 82.888 05 88888. 32.0% 8289.250. mauémeétwmo 0mg cum .3. 038. A Figure 4.1: Different types of annulations observed in ichnospecies of Arthrophycus. A) rings; B) concave-convex; C) chevrons; D) biconvex. 150 $295.5 IIJ 35.35550 Illu— .~v_.»a§m J .3365» IL Exec...” x295» III. «_mcocmcas IL mamastoo 353:5 u||||._ 2.858 III. 382:8 $238335 3.0223 3.2.5:. £329 «.505 II. .5885 ..L 3:95:30 1 o 1.64 3.2- 4.8- . 4” 6 3:530 9.6+ 11.2- Figure 4.2: Cluster diagram of twenty OTUs of Arthrophycus. 151 __xmczfiu .Nwiaosm @385 £59.80 macwcmcsc xo_aE_m . 5:393 3535 ofiEoE micmfiafimcofi mace. $236.05.: mcmmgo 220629 3.55:. 3239 2.39... .5585 322532.» 3.223 °1 1 2. 2.4— 3 6— WJ WL @1 MPH IL . 8. 4 _ 6 353.0 7.2-1 8.4-1 9.64 10.8- Figure 4.3: Cluster diagram of twenty OTUs of Arthrophycus, without the width character. 152 "NMVWCFWO‘ O—‘NMV‘A‘O o—n—o—uv—n—o—c—n NFv-OOOOOOOOOOOOC‘O NFv-vv-OOOOOOOOOOP OFs-NOOOFOPOOOOOP FFFNOOOOOOOOOOOP PFONOOOOOvoooos-P OPOFOFOOOOOOOOOP OOPv-OFFOOOOOOOOP OOFNONOOOOOOOOOO PFFNPNOOOOOOOOv-v- PFPFOFOOOOOOOOFN OPPNPNCOONOOOOOO OFNVOOOOOv—OOOOOM CFPNOOOOs-OOOCOOC') OPNOCNOFOOOOOOON Pv-ONNooos-FOOOOFG) FPONMOOOOOOOvv-FM OV-Ov-C‘OPOOOOFOv-u—v-OO v-v-Os-(ONOOOOOF'r-v-v-m Ov-OVCONOOOOOFFFs-m Os-OOMNOOOONFV-FFCO a1 br li la mi pamrel te dz moancohu si sPsKl sK2qi 153 ILEHJ||LIJILEJEEJLLJII te which ica (1 characters are most important for defining clusters of taxa. The vertical line divides the numbers in the ma 'x into those for the Arthrophycus cluster on the left and all others on the right. See key on next page for abbrevi tions. X] the data ma ’ 1n f characters with Arthrophycus OTUs. Non-zero numbers iagram o Two-way cluster d Figure 4.4 Key for Figure 4.4 912 al br li la mi pa mr e1 te dz mo an co hu si sP sKl 8K2 qi ta alleghaniensis brongniartii linearis lateralis minimus parallelus minoricensis elegans tenuis dzulynskii montalto annulatus corrugatus hunanensis simplex strictus(P) strictus(Kl ) strictus(K2) qiongzhusiensis tarimensis Sham 1 Branching 2 Diameter 3 Cross-section 4 Shape 5 Annulations 6 Width 7 Smooth 8 Radial 9 Parallel 10 Ends 1 l Palmate 1 2 Wrinkles 13 Internal 14 Plane 1 5 Groove 16 Closeness 154 Coordinate 2 Figure 4.5: PCO graph, coordinates l and 2. The cluster of six is represented by the + symbols; all other OTUs are represented by the dots. 155 Coordinate 3 i I T Coordinate l Figure 4.6: PCO graph, coordinates 1 and 3. The cluster of six is represented by the + symbols; all other OTUs are represented by the dots. The dot near the top of the graph represents A. minoricensis; this point was not separate fi'om the others in the previous graph. 156 SH|9||BJBd .snwgugw sumam sueauu ggu‘egufiumq sysuagueufiaue smalnuue smefinuoo unsunnzp (msnmms suebqa sgnua) sgsuao(g)Jougu1 sgsuamue; (DUSNOMS ouewow quww nsuaueunu sgsuagsnuzfiuogb unsnnnns 157 Figure 4.7: Cladogram (third of three most parsimonious trees) of twenty OTUs of Arthrophycus, using A. strictus (P) as the outgroup. snfimeied snwunw gnlmu5u01q nsdmueqfiaue sueunel mieaml oueuuaw smelnuue (zx)smoms usuawue; nsuaouhougw qnua; suefima nsumsnuzfiumb xmduus qsuaueunu sniefinuoo unsumnzp (msnmms unsnunns 158 Figure 4.8: Strict consensus cladogram using A. strictus (P) as the outgroup. nnue; sgsuaoohougw suefiqa nsuaueunu quuus unsunnzp smefinuoo (Lx)smoms oneiuow smelnuue (d)smoms snlaueied 8u81618| meu6u01q sueam| sgsuagueufiaue snwuuw (msmoms sgsuawue; msumsnuzfiumb 159 Figure 4.9: Cladogram (first of nineteen most parsimonious trees) of twenty OTUs of Arthrophycus, using A. qiongzhusiensis as the outgroup. Chapter V. Systematic Ichnotaxonomy Ichnogenus Anhrophycus Hall, 1852 Generic Synonymy Arthrophycus Hall, 1852 [nomen protectum Rindsberg and Martin 2003] F ucoides Brongniart, 1822 [partim] Encrinus Andreae 1764 [partim] Crinosoma de Castelnau, 1843 [nomen oblitum] Harlania Goppert, 1852 Synonymy 1831 1832 1832 non 1834 1834 1835 ? 1837 ? 1838 ? 1839 1843 1852 1889 non 1889 non 1890 F ucoides alleghaniensis — Harlan Encrinus giganteus — Eaton F ucoides brongniartii — Harlan F ucoides brongniartii — Mantell F ucoides alleghaniensis — Taylor F ucoides alleghaniensis and F ucoides brongniartii — Taylor F ucoides alleghaniensis — Conrad F ucoides alleghaniensis —— Conrad F ucoides alleghaniensis — Conrad Crinosoma antiqua - de Castelnau Harlania halli — depert Arthrophycus harlani — Lesley Arthrophycus montalto — Lesley Arthrophycus siluricus n. sp. - Schimper 160 non non non non non non 1901 1908 1916 1923 1925 1933 1934 1935 l 940 1941 1944 1951 1952 1955 1955 1958 1960 1961 1963 l 963 l 963 1964 Arthrophycus elegans n. isp. — Herzer Radix corrugatus and Radicites rugosa - Fritsch Arthrophycus alleghaniense —— Schuchert Arthrophycus alleghaniensis and Arthrophycus brongniartii — Prouty and Swartz Harlania — Fritel Arthrophycus — Moore Harlania - Dalloni Arthrophycus — Abel Arthrophycusflabelliformis n. isp. — Hundt Arthrophycus la'ebsi n. isp. — Hundt Arthrophycus alleghaniensis and F ucoides harlani — Shimer and Shrock Arthrophycis — Silva Arthrophycus — Becker and Donn Harlania — Lessertisseur Arthrophycus alleghaniensis — Young Arthrophycus alleghaniensis — Fenton and Fenton Harlania — Grove Arthrophycus alleghaniensis — Wolfart Harlania alleghaniensis — Bender Arthrophycus - Duimovich Arthrophycus alleghaniensis and Arthrophycus harlani — Moneda ?A rthrophycus - Correia Perdigao 161 non non non non non non non non non non non non 1964 1 966 1 966 1 966 1968 l 970 1970 l 970 1 970 1971 1972 1972 1972 1975 1975 1977 1977 l 977 1977 1977 1 977 1977 1977 Arthrophycus — Pettijohn and Potter Arthrophycus alleghaniensis - Borrello Harlania — Gubler et a1. Arthrophycus - Howard Harlania alleghaniensis -— Bender Arthrophycus - Frey Arthrophycus —- Frey and Howard Arthrophycus — KsiaZkiewicz Harlania — Selley Arthrophycus — Maberry Arthrophycus — Frey Harlania — Selley Arthrophycus — Terrell Arthrophycus — Alpert Arthrophycus — Chiplonkar and Ghare Arthrophycus — Alpert Arthrophycus alleghaniensis — Baldwin Arthrophycus annulatus n. isp. — Ksiazkiewicz Arthrophycus strictus n. isp. — Ksiazkiewicz ?Arthrophycus dzulynskii n. isp. — KsiaZkiewicz Sabularia tenuis n. isp. — Ksiazkiewicz Arthrophycus — Roniewicz and Pienkowski Arthrophycus — Webby 162 non non non non non non non non non non 1978 1978 1978 1980 1981 1981 1981 1981 1982 1982 1983 1983 1984 1984 1984 l 984 1985 1985 1985 1985 1986 1986 1987 Arthrophycus — Kern Arthrophycus — Perez and Salazar Arthrophycus - Wagner Arthrophycus alleghaniensis - Downey Arthrophycus alleghaniensis —- Burjack and Popp Arthrophycus — Crimes Arthrophycus — Laird Arthrophycus — Wolfart Arthrophycus —- Banerj ee Arthrophycus — Pemberton and Risk Arthrophycus — Cotter Arthrophycus alleghaniensis — Turner and Benton Sabularia tenuis - Alexandrescu and Brustur Arthrophycus - Bhargava et al. Arthrophycus alleghaniensis — Lifién Arthrophycus alleghaniensis — Pickerill et al. Arthrophycus — Durand Arthrophycus — Eagar et al. Arthrophycus - Ghare and Kulkarni Arthrophycus — Le gg Arthrophycus — Lin et al. Arthrophycus — Manca ?A rthrophycus — Akpan and Nyong 163 non non non non non non non non non non non non 1987 1987 1988 1988 1988 1991 1991 1 992 1 993 1 994 1 994 1 994 l 994 1995 1 996 1 996 1996 1 996 l 996 1 997 1997 1 997 Arthrophycus — Bj erstedt Arthrophycus — Mukherjee et al. Arthrophycus — Bhargava and Bassi Arthrophycus harlani — Janvier and Melo Arthrophycus — Seilacher and Alidou Arthrophycus — Pickerill et al. Arthrophycus alleghaniensis and Arthrophycus isp. — Romano Arthrophycus corrugatus — Mikulas Arthrophycus A and B — Li Arthrophycus — Greb and Chesnut Arthrophycus qiongzhusiensis n. isp. — Luo Arthrophycus minoricensis — Orr Arthrophycus tarimensis n. isp. — Yang Arthrophycus alleghaniensis — Femandes et al. Arthrophycus alleghaniensis — Borghi et al. Arthrophycus alleghaniensis — Femandes and Borghi Arthrophycus strictus — Paczesna Arthrophycus — Yang et a1. Arthrophycus hunanensis n. isp. — Zhang and Wang Arthrophycus minoricensis — Llompart and Wieczorek Arthrophycus — Peeples et‘ a1. Arthrophycus alleghaniensis, Arthrophycus linearis, and Arthrophycus unilateralis — Seilacher 164 non non non non 1998 1998 1998 1998 1999 l 999 1 999 1 999 l 999 1 999 1 999 2000 2000 2000 2001 2002 2002 2002 2003 2003 2003 Arthrophycus — de Alvarenga et al. Arthrophycus alleghaniensis — Metz Arthrophycus aff. alleghaniensis — Moreira et al. Arthrophycus - Stanley and Feldman Arthrophycus alleghaniensis — Femandes Arthrophycus — Gong Yiming Arthrophycus alleghaniensis —- Moreira et al. Arthrophycus alleghaniensis — Nogueira et al. Arthrophycus — Pfliiger Arthrophycus tenuis — Uchman Arthrophycus tenuis — Uchman and Demircan Arthrophycus alleghaniensis — F emandes et al. Arthrophycus — Mergl and Massa Arthrophycus alleghaniensis, Arthrophycus linearis, and A. Iateralis — Seilacher Arthrophycus alleghaniensis — Femandes Arthrophycus alleghaniensis and A. Iateralis — Femandes et al. Arthrophycus — Lopez and Roy Arthrophycus simplex - Seilacher et al. Arthrophycus alleghaniensis, Arthrophycus linearis, and Arthrophycus isp. — Acefiolaza and Acefiolaza Arthrophycus alleghaniensis and A. linearis — Neto de Carvalho et al. Arthrophycus alleghaniensis and Arthrophycus isp. — Poiré et al. 165 non 2003 2003 2005 2005 2005 2005 2006 2006 2007 2008 2009 2010 Arthrophycus brongniartii — Rindsberg and Martin Arthrophycus alleghaniensis and Arthrophycus linearis — Seilacher et al. Arthrophycus —— Dutta and Chaudhuri Harlania — Eschard et al. Arthrophycus minimus n. isp. — Méngano et al. Harlania — Turner et al. Arthrophycus alleghaniensis and Arthrophycus brongniartii — Kumpulainen et al. Arthrophycus alleghaniensis — Metz Arthrophycus alleghaniensis — Seilacher Arthrophycus — Acefiolaza and Heredia Arthrophycus alleghaniensis — Miller et al. Arthrophycus parallelus — Brandt et al. Type Ichnospecies A. harlani Conrad, 1838, by original monotypy. F. harlani Conrad, 1838, which was a junior synonym of F. brongniartii Harlan, 1832, as judged by James (1893a) and Rindsberg and Martin (2003), was reassigned as A. harlani. Hall (1852) considered both F. alleghaniensis and F. brongniartii as synonyms of A. harlani. Original Diagnoses Encrinus giganteus Eaton, 1832: “Branching, red or grey: often compressed, whirls uniform and generally obscure: branches of great length; mostly lying in the direction of the layers, or nearly so.” 166 Crinosoma de Castelnau, 1843: “Ce corps est tellement diflerent de tous les crinoi'des connus, que, malgre' le mauvais état de conservation de I’échantillon, j ’ai cru qu ’il était nécessaire d ’en former un genre distinct.” (This body is so different fiom all crinoids known that, despite the poor condition of the sample, I thought it was necessary to form a separate genus.) Arthrophycus Hall, 1852: “Stems simple or branching, rounded or subangular, flexuous, ascending, transversely marked by ridges or articulations. The species of this genus yet known consists either of simple elongated stems of nearly equal dimensions throughout, or those which divide near the root into several branches and afterwards remain simple.” Harlania Goppert, 1852: (F rons coriacea simplex cespitose aggregata vel dichotoma, rami in statu iuniori longitudinaliter sulcati; rami adultiores subcylindrici interrupte transversim elevato-striati.” (Straplike simple turflike aggregate or dichotomous branch, in younger states longitudinally sulcate, branches of adults subcylindrical interrupted transversely by elevated ridges.) Emended Diagnosis Dense accumulations of subhorizontal traces with simple smooth-sided form, terminating blindly, annulations typically biconvex, commonly branched and/or bundled into narrow or palmate fans, subquadrate in cross-section, median groove often present, filling chevron-shaped, diameter consistent in individual traces. Discussion Chapters 11 to IV 167 Included Ichnospecies A. alleghaniensis (Harlan, 1831) A. brongniartii Harlan, 1832 (= A. linearis Seilacher, 1997) A. lateralis Seilacher, 2000 A. minimus Mangano et al., 2005 A. parallelus Brandt et al., 2010 Stratigraphic and Geographic Range Cambrian of Argentina, Spain, and Libya Ordovician of Argentina, Brazil, Algeria, Libya, Eritrea, Benin, Niger, Jordan, France, Portugal, and Spain Silurian of Argentina, Brazil, United States, Canada, Chad, Libya, Eritrea, Benin, Niger, Jordan Devonian of United States, Brazil, Eritrea, and Libya Carboniferous of United States, China, and England 168 Chapter VI. Arthrophycus in time and space Biostratigraphy Dating sedimentary rocks is ofien difficult to attempt. Lacking radiometric dates, sedimentary rocks must be dated by relative methods, often using fossils that are found within them and correlating with known dates established elsewhere. Index fossils, or those that are relatively short-lived in geologic time, make good tools for this work, and can be used to date formations in which they are found, and many such examples exist, comprising whole books (Shimer and Shrock 1944). As a fossil that might be limited in time to the Ordovician and Silurian (Seilacher 2000), Arthrophycus is a possible correlative fossil. The concept of index fossils is not limited to body fossils; trace fossils have also been used in this manner. Such papers are unusual because most trace fossil have very long geologic time ranges, but trace fossils are potentially useful for the many strata that do not preserve body fossils (Seilacher 1970). Crimes (1968) studied two ichnospecies of Cruziana (C. semiplicata and C. fitrcifera) in Wales and used them as index fossils to date two formations as upper Cambrian and lower Ordovician, respectively. Seilacher (1970) used many more ichnospecies of Cruziana to establish a global ichnostratigraphy. More recently, Acefiolaza and Heredia (2008) used Cruziana to date two Ordovician formations in Argentina. A number of researchers have viewed Arthrophycus as limited in time. Conrad (1839) was the first of these, and several others (e.g. Moore 1933, Shimer and Shrock 1944) described Arthrophycus as an index fossil to the Ordovician and Silurian. Young (195 5) wrote that Arthrophycus was easy to recognize, a necessary attribute of a 169 good index fossil, but that it was not as strati graphically restricted as previous authors thought. Silva (1951) dated a formation as Silurian rather than Carboniferous, based largely on the presence of Arthrophycus. Femandes and Borghi (1996) considered their formation to be Silurian based on the presence of A. alleghaniensis. Seilacher et al. (2003) used Cruziana, A. alleghaniensis, and Gyrochorte to correlate sandstones in Argentina and North Africa, concluding that the Argentinean formations were younger than first supposed (Ordovician-Silurian rather than Cambrian-Ordovician). Most recently, Kumpulainen et al. (2006) noted a number of trace fossil genera in their formation in Eritrea and dated it as Late Ordovician, based mainly on the presence of A. alleghaniensis and A. brongniartii. There are problems with using Arthrophycus in ichnostratigraphy. Kumpulainen et al. (2006) provided a table of A. alleghaniensis occurrences that were Cambrian to Devonian in age, but then used the ichnospecies as evidence of a Late Ordovician (or Early Silurian) age. If the authors used other data to aid them in their decision, they did not make such evidence clear. Femandes and Borghi (1996) wrote that A. alleghaniensis was not exclusive to the Silurian and should therefore be excluded as an indicator of Silurian age, but then proceeded to date their formation as Silurian based on that evidence. Crimes (1981) considered even an Ordovician- Silurian range to be too wide for use in ichnostratigraphy, and Dalloni (1934) wrote that Arthrophycus had a geographic range that was too wide for meaningful stratigraphic correlation. 170 Many researchers (e.g. Turner and Benton 1983, Femandes 2001) suggest a Cambrian-Devonian age for Arthrophycus. Still, Seilacher (2000, 2007a) and other authors (e. g. Mangano et al. 20053, Kumpulainen et al. 2006) hold to the idea that Arthrophycus has a restricted geologic range and is therefore useful for ichnostratigraphy. Some authors reject any report outside of the Ordovician-Silurian range as Arthrophycus, even though most ichnologists reject geologic time as an ichnotaxobase (Bertling et al. 2006, Bertling 2007). Ideally, reports of Arthrophycus would be verified solely on morphologic characters, independent of stratigraphic or geographic occurrence. Table 6.1 is an indication of the reported geologic range for each ichnospecies of Arthrophycus, with the number of reports that place the ichnospecies into each time period. These data were used to construct a number of histograms in order to assess the utility of Arthrophycus as an index fossil. Figure 6.1 is a histogram of all reports of Arthrophycus ichnospecies (or “Arthrophycus isp.”) reports through time. In order to make this figure and all of the following histograms for geologic time, the reports of Arthrophycus were dated as their authors reported, without judgment as to the accuracy of the dating by those authors. Reports that included more than one time period were counted for both periods (e.g. “Ordovician-Silurian” is counted twice, once for each period, “Ordovician-Devonian” is counted thrice) and the x-axis is non-numerical. Any reports that only discussed previous occurrences of Arthrophycus (e. g. Moore 1933, Femandes 2001) or that were additional reports on the same occurrence (e. g. Frey 1972) were not included in this census in order to reduce noise. In a few cases, a complicated report was divided into a 171 number of occurrences to minimize compressing too much information into a single data point (e.g. Lesley 1889, KsiaZkiewicz 1977). Figure 6.2 provides the results of the geologic range analysis, in which the reports of Arthrophycus are divided into the categories of “true” Arthrophycus, unverifiable reports, and non-A rthrophycus reports based on the earlier taxonomic assessments. Figure 6.1 shows Arthrophycus appearing scattered throughout the Paleozoic, with a peak toward the Ordovician-Silurian and only 19 occurrences in the Mesozoic and Cenozoic. However, in Figure 6.2, it becomes evident that this apparent range is distorted by the inclusion of questionable assignments to Arthrophycus. Figure 6.2 shows that Arthrophycus occurrences peak in the Silurian, with more confirmed occurrences than any other time period even has reports. The Ordovician ranks second, with other Paleozoic periods making small additional contributions. This “corrected” distribution is even more apparent if only confirmed reports are shown (Figure 6.3) or both confirmed and unverifiable reports together (Figure 6.4). These histograms (Figures 6.1 to 6.4) are for the whole ichnogenus, including all reported ichnospecies of Arthrophycus. Because most of those ichnospecies have been eliminated from Arthrophycus, it is possible that those invalid ichnospecies were distorting the apparent stratigraphic range of the ichnogenus. To analyze this question, a smaller data set, composed of reports of only those ichnospecies herein considered valid ichnospecies of Arthrophycus was compiled (Figure 6.5). This reduced data set did not produce markedly different results, as eliminating invalid ichnospecies removed only 15 reports from the dataset. 172 To evaluate the effect that reportsof “A rthrophycus isp.” may have had on the distortion of the stratigraphic range of the Arthrophycus ichnogenus, a new histogram with only reports of valid ichnospecies with definite ichnospecific assignments (also including those few reports assigned previously herein to either A. alleghaniensis or A. brongniartii, such as Durand 1985 and Pfliiger 1999) was constructed (Figure 6.6). This action reduced the total number of Arthrophycus records by 38 and resulted in a different distribution (Figure 6.6). Only three non-A rthrophycus results remained and the histogram shows a very high peak in the Silurian, overshadowing its nearest competitor, the Ordovician. As with Figure 6.2, the number of confirmed reports in the Silurian is greater than that for the total of any other period, but in this histogram that difference is much more apparent. The dominance of valid Arthrophycus reports from the Silurian, and to a lesser extent, the Ordovician, is real. However, this dominance of Silurian (and Ordovician) reports might have been a bibliographic artifact of the dominance of the Medina and Tuscarora Sandstones in the Arthrophycus literature: perhaps a plethora of reports from those formations overwhelmed single reports from other Paleozoic formations. Figure 6.7 is a histogram of the occurrence of valid Arthrophycus through time in which each different Arthrophycus-bearing formation is counted once. This graph shows that even after removing the bibliographic effect of multiple reports from a single formation, most Arthrophycus reports still come fi'om Silurian strata. There are confirmed occurrences of Arthrophycus in the Paleozoic outside of the Ordovician-Silurian window, in both the unreduced and reduced time analyses (Figures 6.1 to 6.6). However, these reports are not as straightforward as they may 173 appear. In the unreduced analysis (Figure 6.2), there are four confirmed or unverifiable Cambrian reports, four confirmed or unverifiable Cambrian-Ordovician reports, and one confirmed Cambrian-Devonian report. In most of these multiple-period reports, there is no way to determine whether the Arthrophycus fossils actually occur in both time periods: they may be fiom the top of the formation, and thus Early Ordovician, or from the bottom of the formation and thus Late Cambrian. Of the remaining four confirmed or unverifiable reports that are Cambrian only, one is unverifiable (and was questioned by Mangano et al. 2005a) and the other three are likely A. minimus. The Devonian presents a similar case of unclear dating. Three of the reports of Arthrophycus fiom Brazil are listed as Ordovician to Devonian (possibly only to Early Devonian). Four other reports give a range of Silurian-Devonian, but three of these are unverifiable. Turner and Benton (1983) had rather vague dating, reporting their fossils from the Cambrian to the Devonian. Mangano et al. (2005a) considered this dating to be suspicious and unconfirmed, but the Devonian portion of the range is accepted herein. The only other reports of Devonian Arthrophycus are that of Wolfart (1961), assessed here as questionable, and Bjerstedt (1987), which is hesitantly judged conformable to Arthrophycus after examining the actual specimen. Only a few post-Devonian reports of Arthrophycus identify the fossils to the ichnospecies level. Of those, only the reports of Brandt et al. (2010) and Herzer (1901), both from the Pennsylvanian, may be conformable to Arthrophycus. Abel (1935) also reported a specimen of A. alleghaniensis from the Oligocene, but this occurrence was questionable. No other reports of Arthrophycus from post-Paleozoic rocks were judged herein as acceptable. 174 The specific behavior represented by Arthrophycus may have continued into the Devonian and beyond, but its grip was tenuous, judging by the low number of reports. Five of the nine Cambrian occurrences may not actually be Cambrian, so the Ordovician and Silurian remain as the peak time periods for Arthrophycus. A. minimus is the only Cambrian ichnospecies accepted herein, and A. parallelus the only valid Carboniferous ichnospecies. However, even when considering only A. alleghaniensis, an Ordovician-Silurian time range is still rather broad to use for stratigraphic purposes. It is suggested that Arthrophycus only be used as supplementary evidence, along with a more reliable indicator such as ichnospecies of Cruziana. A. lateralis still has ichnostratigraphic potential, as it is known only from the Silurian, but this ichnospecies was named only ten years ago (Seilacher 2000). A. lateralis may yet be found in a wider range of geologic time, and given its rather large geographic range, it should not be used in biostratigraphy. Biogeography Arthrophycus has been reported on all continents and twenty-seven countries, including sixteen U. S. states, but not all of these reports are judged herein as valid. Figure 6.8 is a histogram showing the geographic distribution of Arthrophycus, using the same methods employed to make the stratigraphic histograms. For the sake of simplicity, the regions are reduced to the seven continents, with an additional category for the countries in the Middle East (Jordan and Turkey). As Figure 6.8 shows, Arthrophycus appears to be scattered evenly across the world, with many occurrences in South America, Asia, Afi'ica, and Europe, and only a slightly higher peak for North America. However, in a histogram showing reports 175 divided into the three taxonomic categories of conformable to Arthrophycus, unverifiable, and not conformable to Arthrophycus (Figure 6.9), North and South America emerge as the prominent regions of confirmed Arthrophycus finds, followed by Afiica. In a histogram limited to confirmed records (Figure 6.10), North America has the highest peak, as might be expected given that the first records of Arthrophycus came from that continent. Even with the questionable records of Arthrophycus removed from the dataset, the ichnogenus still shows a wide geographical distribution. There are confirmed reports from five continents plus the Middle East, including seventeen countries and eight U. S. states. If unverifiable reports are also accepted, two continents, two more countries, and five states are added to the count. There are some notable gaps in the reported geographic range of Arthrophycus. There are many good records fiom Africa, comprising six countries (plus a seventh indeterminate record), but all of them are from North Afiica; no records of Arthrophycus, valid or invalid, have come fi'om sub—Saharan Africa. There is also only one valid report of Arthrophycus from Asia (China); the other twelve are judged questionable. Seven of the Asian reports were fi'om China (and three of those proposed new ichnospecies of Arthrophycus) while the other six Asian reports were fi'om India. Russia is also missing from the survey, but one author did provide a tantalizing mention of the country. Dalloni (1934) mentioned Arthrophycus (as Harlania) in a list of places that the ichnogenus had been found in Silurian rocks, but he did not include a reference for that observation, preventing further investigation. 176 Paleobiogeography Kumpulainen et al. (2006) suggested that A. alleghaniensis originated in the seas surrounding the southern continents in the Ordovician Period and then moved north in the Silurian. Figure 6.11 is a plot of Arthrophycus distribution in time and space. As in the stratigraphic histograms (Figures 6.1 to 6.10), the repetitive reports or discussion papers were not included, and the reports were categorized as conformable to Arthrophycus, unverifiable, and not conformable to Arthrophycus. The geographical regions are again the seven continents and the Middle East, and each time period was assigned a numerical code for the y-axis (Table 6.2) rather than using actual ages, because using the numerical ages produced a graph with points much too crowded to be useful. Circle size at each point in Figures 6.11 to 6.13 represents the number of reports at that particular space and time. Figure 6.11 shows a scatter of Arthrophycus reports across time and space. The confirmed records cluster in the Paleozoic and in Afiica, South America, and North America, as shown in the histograms. However, there is no concentration of early records in the southern regions: the earliest confirmed records are in South America, but Europe has four Cambrian and Ordovician reports, the Middle East has at least one in the Ordovician, and North America has two early occurrences that are indeterminate. I then investigated the question of individual ichnospecies of Arthrophycus being limited to the southern regions in the Ordovician, with the northern specimens in the Silurian being from different ichnospecies. Figure 6.12 shows the time and space data as in Figure 6.11, but divided into ichnospecies. For this graph, all questionable ichnospecies and records were removed, as well as any “Arthrophycus isp.” reports. 177 Reports of Harlania are included under A. alleghaniensis, but A. linearis and A. brongniartii were kept separate for clarity. In cases for which there were more than one ichnospecies in the same time and place (e. g. the Silurian of Africa and the Ordovician-Silurian of South America), the circles were offset slightly so that all ichnospecies became visible. Figure 6.12 shows evidence of early southern occurrences. The only Cambrian occurrences are A. minimus and A. alleghaniensis, both in South America. Arthrophycus is reported from “Cambrian-Ordovician” of Europe and Ordovician strata of Europe and the Middle East. A. alleghaniensis and A. brongniartii (and A. linearis) are widely distributed over time and space and thus are less-sensitive indicators of expansion. Only A. lateralis is currently restricted to the Silurian of the southern continents. Figure 6.13 is a bubble chart with the confirmed and unverifiable “A rthrophycus isp.” reports included. There are only four confirmed reports of Arthrophycus isp., all Silurian or later, but the unverifiable points are scattered over time and space. Most of the reports from the Ordovician and earlier are from the southern regions and were on the fringes of the glaciers flowing over Africa; at that time the southern continents were centered on the south pole, located in western Africa (Figures 6.14 to 6.16). The Arthrophycus trace-maker, therefore, may have been a cold-loving organism. 178 Facies Control of Arthrophycus Distribution Arthrophycus is typically found on the underside of beds, preserved as convex hyporelief. This is interpreted as infill material that preserved a relic of the original trace, which was made in the beds below and subsequently eroded (Seilacher 1964b, ‘ Martinsson 1970). Most Arthrophycus specimens are found preserved in sandstone, or occasionally shale (e.g. Bjerstedt 1987); approximately 73% of the reported occurrences reviewed herein are in sandstone, with 10% of the reports not recording lithology (see Appendix 1). None of the specimens conformable to Arthrophycus come from carbonate rocks; all are from siliciclastic materials. In the ethological classification introduced by Seilacher (1953), Arthrophycus is usually placed in fodinichnia, or feeding traces (e.g. Borghi et a1. 1996). As the schemes have some overlap, it is also possible to consider Arthrophycus to be one of the repichnia, or crawling traces. Seilacher’s (1964a) introduction of the ichnofacies scheme for trace fossils also applies to Arthrophycus. Most authors (e. g. Femandes et al. 1995) place Arthrophycus in the Cruziana ichnofacies, which means that Arthrophycus formed in the relatively shallow water of the shelf environment, with medium energy, and typically in sand or silt (Bromley 1990). This area was between the daily and storm wavebases and the fossils found there are typically diverse in both the number of ichnospecies and in the number of ethological categories, but little deep burrowing is present (Bromley 1990). Judging by the previous reports of Arthrophycus, it is likely that the best place to search for Arthrophycus is in Paleozoic sandstones from nearshore environments, in the Cruziana ichnofacies. Strikingly, Arthrophycus is a very facies-dependent fossil, 179 so elastic sequences are crucial. Nearly all of the papers that report valid Arthrophycus agree on a nearshore environment, with two minor exceptions. The first is a personal communication from Alidou to Dutta and Chaudhuri (2005), in which Alidou intimated that he knew of a continental Arthrophycus from West Afiica. In the second, Nogueira et al. (1999) reported that Cotter (1983) had interpreted the Tuscarora Formation, which contains abundant Arthrophycus, as being from a braided river environment. However, the actual paper by Cotter (1983) reported braided rivers in parts of the Tuscarora, with the typical nearshore environment for the portions of the formation that contained Arthrophycus. The affinity for elastic facies may help explain the northward expansion of Arthrophycus fiom Gondwana to the northern continents as described above. In the Early Silurian, the Iapetus Ocean began to close as continents moved closer together; this may have allowed the maker of Arthrophycus to expand its range to the northern continents. As Arthrophycus seems to have been a trace fossil of largely deltaic and other shallow-water environments (e. g. Cotter 1983), the trace-maker may not have been able to cross the deeper oceanic waters of the Iapetus until this ocean became narrower and presumably shallower. 180 181 F E8 .290 28 32.88 N N 0800 0:00:05. moom 008.55 N F N __0620 06000 88 8.050 N F N N >00 006900 88 0.0.20. .020 F F 20 06.200 32 0.820%: N m 0 F 5.980 06900 32 0:85. F F __0 BEN 80F 08882 F F 8.0 05> 80F 00888. F F 0800 03 30F 0.020.033008 0 2.2.0.0 0030:0000. NSF 0.520. F 09.0 5300....me NSF .2258 N F .20 8.35wa NSF 050.50 .980 F 00.0.0 8.3000000. NSF 050355 0 98 559.80 2.2 028.855 F F 8.0 .25: FEF .080. F F 220 65: 049 058308.. N F F Ea... some“. 08F 02888 00.0 F Eon. Foam... FomF 3000.0 F F __0 85200 80:28 035.0 F F 9:00 88.50 0me 9.0.85 0 F 0 F >00 :08: N02 0:08:20 .005 mm. 0 00 0F 0 >00 8:0: 58 0.2200020 .0500 Egon-om .550 02n— anO 2:. _...P gon— EGO >00 .5 09.0 9:00 09.51 ..O:«=< 3CD 02009005.»... 0 Fe: .0? 530000: .3 3.5000 3.808 5:300 ..0050.. 0...... .30..§.m-§.0.>0.000 .m.0v 00.000 000 :05 0008 05% 0.800.— 0.w0.m 0 .F. 000500 000 0.00.00 ..0 0000000 300 05 F0 .88 05 .0000 Foe >08 3.8000 ..0 000807. 00.02.0053. .3 08: swag: magmeétwwo 000005000 ”.0 0.00... Table 6.2: Coding for geologic periods and regions used in the bubble charts. Time Place Code Time Code Place 1 Proterozoic 1 Africa 2 Cambrian 2 Antarctica 3 Cambrian—Ordovician 3 Australia 4 Ordovician 4 S. America 5 Ordovician-Silurian 5 Asia 6 Silurian 6 Europe 7 Silurian-Devonian 7 Mid East 8 Devonian 8 N. America 9 Carboniferous 10 Permian 1 1 Triassic 12 Jurassic 13 Cretaceous 14 Cremceous-PreQuaternary 15 PreQuatemary All Reports of Arthrophycus Through Geologic Time # of reports «0906\5‘t06‘4‘ééo ‘\?’c)’bé\()p Q QQQQ‘ 3 dqfle’ Figure 6.1: Histogram of reported occurrences of Arthrophycus through time. 182 Categorized Occurrences Through Time 60 50 4O - v CI Not Arthro a ?Arthro j 1 I Arthro V 7‘ 30 # of reports 20 i 10 l o '\ ' k R. l (waxy 00° 0‘ 00‘ 00° Q°§ «0 3° 0‘0 00 Figure 6.2: Histogram of occurrences of all Arthrophycus reports through time, divided into the three taxonomic assessment categories. 183 Confirmed Arthrophycus Occurrences 40 l 35 30 25* # of reports N O 10 5 0 F L 1 o o ‘\ ~\ *0 ‘0 x \ Q‘é‘é 0’5“ 0‘50 9 0° 0”} Q‘b‘éx ’\ 3° 0‘0 Q‘°O ' Figure 6.3: Histogram of occurrences of Arthrophycus reports through time, confirmed reports only. 184 Confirmed and Unverifiable Occurrences 60 El ?Arthro I Arthro 1 # of reports ‘0 o \\ 4 o ‘0 x x 0. 0 0‘0‘ Qfl Figure 6.4: Histogram of occurrences of Arthrophycus reports in time, continued and unverifiable reports only. 185 Valid lspp Through Time, + "A . isp" 60 50 ‘ l l 1DNot Arthro‘. l ‘E?Arthro .l l [IArthro ‘ 40 30 # of reports 20 10 o — - . = — . o '\ 4 ‘0 x x l_ H, . n, W ,, ,7, W. , . W, Figure 6.5: Histogram of occurrences of valid ichnospecies of Arthrophycus through time, including “Arthrophycus isp.,” broken into the three taxonomic assessment categories. 186 Arthrophycus Through Time, Valid lspp. only a Not Arthro, E ?Arthro . i I Arthro # of reports l -\ A Q X N. : , n ,, E Figure 6.6: Histogram of occurrences of valid ichnospecies of Arthrophycus through time, Bit including “Arthrophycus isp.,” broken into the three assessment groups. Arthrophycus -bearing Formations . l 20 7 , ,_ g 15 F 7 ‘ o . a l N . ,E_1o . . rrrrrrrrrrrrrrrrrrr . i .9 1 '0- ‘: 1 .2 s 0‘ f on. . I l- Camb Ordo Sil Dev Carb Time l 7,,J Figure 6.7: Histogram of valid Arthrophycus-bearing formations throughout geologic time. 187 Reported Geographic Distribution of Arthrophycus 40 f ‘ l t 35 , T E, 30 l ‘ l a 33 l , ‘ g 15 l l * 10 ‘ l l 3 5 ‘ 0 Figure 6.8: Histogram of all Arthrophycus occurrences, by geographic region. Categorized Geographic Occurrences g 3 l0 Not Arthrol‘ 3 55555 .IQ?Arthro '6 ilArthro Figure 6.9: Histogram of all Arthrophycus occurrences, by geographic region, divided into the three taxonomic assessment categories. 188 20 , .3 15 ; o l g 10 l 0 - l _ _ K ' : fissssygg @ 09 ‘0 \§\ l v9 F e- a? I Figure 6.10: Histogram of regions reported for all Arthrophycus occurrences, confirmed occurrences only. 189 800025802 00:23: 8 5x :00 N0 030,—. 00m 000% 0:0 0:5 8 F800 F05 0 08000: .Fo 000.80: 08 0000208 2080 05 .8 050 05 05 $8009 FF 00:05 8 00:00 080. 05 00:08 20:0 :000 .8 0500 08:00 2:. .:0t0=m 0:0. .8 200:: 05 00008 0:: 03:08.8: 000:: 08 0:0 8060605 05 .8 2008 05 8:08 0:: 0080000: 0032 06 0:23: 80:80: 0:0 E08000 0:0. 00030 0:: F0000?» 2:. 80008 0:0 00000000 .3 m0t0w0F00 88 00230 080 000% 0:0 080 00>0 8:000: 0008060020. mo 80:0 0385 ”2.0 08wE all l :l- l uCOEar—OU w x. o m 0 m N F 95:. o 28$. o m n . O 0025 0:0 08:. :. uaoxcqocct< 190 80:00:80.0: 0:080: 0: >00: :00 N0 0500. 00m .8000. 0:0 08: 8 8:00 00:: :0 8:000: 00 3:80: 0:: 00:00:08 20:0 0:: 00 00:0 05 0:0 08000: F: :03? 8 00:00 08: 0:: 00:08 20:0 :000 .00 8:00 0:88 0:0. 83:35 0:: 00 2008 0:: 3:08 0:: 0080005: :000: 0:: 0:0 40063005 0:: 00 0:008: 0:: 8:08 0:: 388:0: :030: 0:: 0:009: 80:20: 0:0 80:88 0:: 80:30 0:: _00::0> 0: 0. 020000086: 0: 0:0 00000 0:0 08: :0>0 3:000: e.gémexfixw 00000000 00 00:0 03:05 ”3.0 08mm”: 0:05:30 0 : 0 0 F. n N F 022.800 0 , 352:2 0 0:923 o 0000:: o .l. __tncmcem a 00090050220 . :03 .<.. 305:5 630000050. 3 .0085 0:0 08:. 191 80:00::800: 0:080: 0: 00: :0.: N0 2:00. 00m .0000: 0:0 08:: 8 8:00 :0:: :0 3:000: 00 3:80: 0:: 00:00:08 20:0 0:: .:0 0N0: 0:: 0:0 00:20: :: :03? 8 00:00 08:: 0:: 00:08 20:0 :000 .:o :800 00.800 0: 0. .:0::0:w 0:: .:0 0:000: 0:: 80:08 0:: 8:080: :000: 0:: 0:0 8063005 0:: .:0 0:008 0:: 3:08 0:: 8:088: :030: 0:: 00:080: 80:80: 0:0 80:80: 0:: 30:30 0:: :00::0> 0::. .8000: :00: 000A:00::::V: 8:00:08 0:0 020000080: .3 .8000 0:0 08: :0>0 0000.:00E2V 00:00000 :0 80:0 03:3: ”2 .0 000mm“: 302:: 00. O :80 3 o . 022900 0 a SEES 0 ,0 £6.55: 0 0000:: o , 5:23:05 0 ,, 20020050220 . :coEEoo @0 3 \m/omw E. 8 00 O ...02 .3. :03 -02000005.": 3 .0025 0:0 08:. 9W”. 192 Southern Europ , terrane — Continental \V— margins 60°S -> Ice flow Figure 6.14: Paleogeographic map of the southern continental positions during the Late Ordovician. Ice sheet is shaded and stars represent approximate locations of Arthrophycus specimens. From Turner et al. (2005). 193 ORDOVICIAN - 480 Ma PANTHALASSA Barentsia 6 m / mg 5 EQUATO lAPETU OCEAN Figure 6.15: Paleogeographic map of all the continental positions during the Early Ordovician. Arrows indicate plate movements and stars represent approximate locations of Arthrophycus specimens. From Scotese (2002). 194 SILURIAN - 440 Ma PANTHALASSA £3 darenfih SIBERIA EQUATOR 1; 5 ’ LAURENTI w “J . KAZAKHSTAN GONDWANA ‘ {7 o/ I Figure 6.16: Paleogeographic map of all the continental positions during the Early Silurian. Arrows indicate plate movements and stars represent approximate locations of Arthrophycus specimens. From Scotese (2002). 195 Conclusions Of twenty-three previously named ichnospecies of Arthrophycus, at least three names are duplicates of previously-named species (A. harlani, A. unilateralis, and A. linearis) and one (A. siluricus) is a name unassociated with any illustrations, descriptions, or known specimens. Of the remaining eighteen, only five ichnospecies possess sufficient diagnostic features to be assigned to Arthrophycus: A. alleghaniensis, A. brongniartii (= A. linearis), A. lateralis, A. minimus, and A. parallelus. The remaining putative Arthrophycus ichnospecies probably belong in other ichnogenera. The original diagnoses of Arthrophycus were detailed, but subsequent authors placed invalid specimens into the ichnogenus without due consideration for the characteristics set forth in the diagnoses. Other characters were not precisely defined in the original descriptions and thus were easily misinterpreted. The ichnogenus diagnosis is emended herein to “dense accumulations of subhorizontal traces with simple smooth-sided form, terminating blindly, annulations typically biconvex, commonly branched and/or bundled into narrow or palmate fans, subquadrate in cross- section, median groove ofien present, filling chevron-shaped, diameter consistent in individual traces.” The assertion that small size is characteristic of geologically older Arthrophycus (Mangano et al. 2005a) is countered by the occurrence of verified small Arthrophycus specimens in younger strata. Size differences among ichnotaxa may reflect environmental stress and nutrient availability; therefore, size should not be regarded as an ichnotaxabase. 196 A numerical phenetic approach using clustering techniques proved useful in clarifying taxonomic affinity of specimens assigned to Arthrophycus and has potential for application to other trace fossils. Cluster and principal coordinate analyses based on sixteen characters confirmed the qualitative taxonomic conclusions and forced closer examination and definition of the morphological characters proposed in Arthrophycus diagnoses. Trace fossils have numerous characters suitable for coding and these techniques should be transferable to other ichnogenera, as the resulting diagrams are very concise and may make connections clearer that were ambiguous in a more wordy form. Cladistic techniques can also be applied to trace fossils, with the caveat that relationships in the resulting cladograms reflect morphological rather than phylogenetic relationships. Arthrophycus is most abundant in Ordovician and Silurian strata, but has additional occurrences in the Cambrian, Devonian, and Carboniferous. The conclusion of previous workers that claims of Arthrophycus specimens in the Mesozoic and Cenozoic are erroneous, and that post-Devonian Paleozoic records should be treated with caution, is affirmed. However, most ichnospecies of Arthrophycus are not useful for biostratigraphic purposes, because they either have long geologic ranges or are known from only a few specimens (e. g. A. lateralis, A. parallelus). Arthrophycus is therefore not suitable for use as an index fossil. Arthrophycus has a worldwide distribution, present on five continents plus the Middle East, including seventeen countries and eight U. S. states. Notable gaps, such as sub-Saharan Africa and most of Asia, may be real, or may reflect the state of Paleozoic marine paleontology in many of those countries. 197 The hypothesis that Arthrophycus originated in the epicontinental seas of the southern continents and spread north through time (Kumpulainen et al. 2006) is supported. However, the timing of the expansion should be emended to Cambrian- Ordovician for the origin in the south, and Late Ordovician-Silurian for the expansion to the northern continents. Future work may include an investigation into Seilacher’s (2000, 2007a) idea regarding the greater taxonomic relationships of Arthrophycus, Daedalus, Phycodes, and the teichichnids. Additional OTUs can be added to the data matrix for the cluster diagrams and PCO plots to determine if additional specimens belong in Arthrophycus and the character list may be extended so that additional taxa (such as Daedalus) may be studied in the same context. The taxa rejected herein as Arthrophycus may be studied further to determine which existing ichnogenus each properly belongs in or if some of the ichnospecies are deserving of their own new ichnogenus, as suggested herein for Arthrophycus (Sabularia?) tenuis. The question of the maker of Arthrophycus is still unresolved and merits further work as well, as does the question of how dense Arthrophycus burrows that presumably were excavated in mud could maintain their shape and whether the Arthrophycus-maker used a chemical signal to avoid interference of separate Arthrophycus burrow systems as was suggested by Miller et al. (2009). 198 < 0:800:00 0:03»: 00:0 :::00_:w:m .8306 000: Starch N. 080020 02:0 :0 26:0 82 3:35: ... 22020 :30 =0 5:80 N8: 3322: ... 832020. 02:0 :0 :20: 0:: £52.: < :30 $00-85 £080 3058...: 88 20.22.300.000 < :52: 50-00 :08: $22.80: 88 2.22.3023 m< 0:800:00 mam $09005 £90m 00050800 000: 0:0:0::§M0:0 N. 0:800:00 0:05 am 86:00 00050800 30: figmfigmfifi < :22: 50.5 32380: 88 0.022300% N. 0:800:00 __Nflm >0Q-=m 00050500 000: 0:0:0:=§M0-0 < 0:900:00 .:0E< :toZ .m 0m :0::00 :0::00 wmm: 030033.000: ... 2222a... 2202 2:0 2303.... Baa: moo: sass: < 0:800:00 060200500 :m 00:32: 80: 0:0:0::0:M0-0 < 080009. 02:0 :0 :20: :2 0.525: m. 893:2. <> ...: .>z :0 08:50 on: acts: ..0 a 5008... :8.» 302 :0 0880 0:: 2.522: : ... amafim 25> 302 :0 22:8 5: 20.22200023 ...: < 2.22:0 :52: :0 :8: :85: :3: 0.522000% < 0:800:00 0::::0w:< 2mm..00:0 0:055 000: 0:30:20:M0-0 ... 820.020 :30 :0 22oz :08: 03: 0022:3003 < 0:800:00 50:00. 0m-00:0 :00:0m woo: 08:030.:M000 E < 0:300:00 308:. :m :00:0m M00: $303330: E < . 52:0 220-250 23%: :2 2022200003 < 0:58:00 0::::0w:< 00:00:80 0N0_0:00< 30—800»:V meow 0:0:0:2§M0-0 < 02:2 .052 .0 :0 03:80 E: .52: 3:55: < 20:53:09.0: :0 Ea: :0: 0.222003% .0 < 25005.0. 28> >52 :0 :0: Na: 3522: 0. 3...). E.22.: .5083 25:. N .203. : .853. 2:: as: .mzuémeétw .:0 0:000”: =< :00 0:5 08:00:00 : x:0:000< 199 <40- 0o<<<<<0-<0«0-0-<<<<<0-0-< <<<0- 0005008 0003000» 00060000 000000: 000303» 005300 00300000 00030000 005.300 000000000 00003 00060000 00800000 0:030:30 0:030:30 005.300 0090008 0300.003» 00800000 00030000 0006008 0:050:00 0:030:30 0.0030800:— 0033030000 .05. 082 .0 000:0.- 000:0:- 0:0-000:4:~ .:.:< .z .02 .50 an: _.2 :00 S2 0038: 00% 05:02 0:000w:< 00% 050m 0038: 050m 050m 00000w:< 0005, 00:02 302 0020.. 30 2 03% 300330000:— 00:00”.— 050.5 0m 0m 0m 00:0 00:0 5-200 5.80 gm 0m 00:0 0m 85-0.3.0 00:0 mm 00:0 0m $0-0m A0005 0m 0m 0w 000.0 0m $00530 50.5 0.800 0.0.3000 5.53m 020% 0:00:00 000500003:- 80:00:00 00.50 .0080 003 0.9.0000 202 803.0 0080 :0EEm >0=0m 00=0m :000000m :000000m :000300m 0000008 0:00:00 300:0 0000 00:00.05 300000 2 003200 00 0:0 2 3:00:02 «00:02 000002 0:02 502 502 00:5 00000000000 00—004 00000090002 :000-0:- mmwfi 00-3 0002 N03 000 2 moom ooom 0002 0—02 0002 mmofi meow vw02 0000 meow 0002 w00~ m002 000m woow w00_ vw0_ mm0~ 0ww~ ooom wwmfi 002000000000 ..m 00020000030: ..m 0::-20000030: .V ..ESS0 ..m 0:50:30 0:03:00: 0000000000000 0000:000M000 00000000000: :00000200300: 0000:000M000 0::-20:0:0M00: 0::-00000000000 000000000M0-0 000000000000: 0::-00000030: 0::-00:000M000 000-0000000000: 0002000000030000000 0::-20:200M00: 0::-00000000000 0::-00000000000 0020:000M000 0:03:00: 33:00 00000000300 0:00:00 00000000 .2 .000- 200 <11a<<? 3?: «BE m_§>_3==om acuwmz «8.8330 ~38sz 3.32 “magma «magma «39¢. «:20 bamboo 0:5 283m find 580 wdom nooNU mag—35$ x8? Bo Z 30> Bo Z «Edna? Saab: «chum tau—om «3qu 39335 988.. 935 3=z$om 3 3 3 «80 £50 £50 “20 85.35 85 3:0 3 85 SEA 3:0 85 3 3 23 3 3 3 $3.85 oméohU 3 >09 85 >oflun§0 33m 38m :95 wcomz §_o:oo< 335$ wad? £952 5.83m 55:03 32532 newcom 338 mm $6335 «$395 gov—8m oo bofim can? :32 592 «Na—053‘ «3353‘ _on< 3.23 655 8303 NOEQEEQ 323:2 cumin 8359 Swan: 3380 wugmuaa >395 comma—:agu 530233 w§o> tuba? Hos—PH 88:5 hwa~ wma~ vac— Nm¢~ Nwmfi nnm~ wha_ hwm~ meow moon mmafi woo“ ovafi Hoa~ who“ Nam“ woa_ mmw_ wa_ hmwfi meow . mmo_ woow whmfi mm¢~ _©m~ moon mwm_ 3m 3m 3m 3m 3m 3m 34 3_ 3m 3m 3m mmmzmauzg $23233 .4533» 5.3%:3 ”agate 33%;.28 .- Egfinmgxa E 3532383 K 5233303 E Examtmgga Exfizmzoxa mtcEmgS 3:33:23 $353338? autmmacxwmac Snatch ammzmmxgmfifi 3035380 .H da< 201 QGO-G¢0-20 :333 0:05:53 0:835: 85.63.: 383:8 :65 «2023:»: 522:3: 33:03:": :83 0:32 «:30 «:30 :Emm 8:255: 330m «3914. :83 308:0M «:20 :35: 3:5 .533 mamSWM :35 2:35 2:3 REE: «game mafim 88¢. .Z mag—mam: 3:38: «8:: 0:88: am =m 3:50 8:0 =m 08m 83 8:0 2:30 :85 3:0 8:0 :80 :5: 9:0 :2. :20 3:0 8:0 p.80 :5: 8:0 3:80 am >235 mm 8:0 920 :om 9.32 mom was: 550:0 gm 325:: 3:25 5:320 898830 22: 8.5532 0:82 3:02 8:32 93:2 5:0: :5 S wwoq ERA Notpoufiwmmvm 80M 3:30: 8:0 3:85 m:o© 98:0 5:: >8: xvi Ewmm SEQ 6:553 530:65 «30:92 on 88:0 :880 9%:me £8.50 522:2: baa: mmmfi coon cam: #5:: moon cum: mam: mam: flwafi ono_ whafi oca~ vao_ amofi owofi ova: mum: ona_ mwa_ moon mwafi moo_ woo: fiwa~ mwafi woo: who— :fl am: mm: :5 mm: mm: mm: m 2:: < am: am: :33:ch + am: am: am: am: mm: am: 9: mm: mm: mm: mm: mm: 33:3: :8: .9: mm: mm: mm: mm: am: am: 60:63:00 A .95 202 §§o© «:20 988—. 353$ 52 gum Each :83 ~80de 530m _mnfim Edam mfiEomE vicar 302 wan—om Ema—tom 4:QO «53%;. ufiagofisoz an: «wax/Ewan?“ «38200 View 302 :30 . 233 85 5-85 Em 85 85-288 3.85 $5.25 85 33th mm ozobgao mm 83 85 am 5.25 2:5 5 owsoém 85 85-3an 85-985 mm _mm «20 a 85 .3 B moucmfiom «Na—053‘ swam mag? 5353 822 3338: 323% Sam “atom .anmfim xma 583$ Snow—mom Snow—mom Snow—mom 0:390 on 802 mnflocou< “Begum moucwfiom 6.3: Q we; tam—03 333 .5:me __oboh 335m «3% Snow—mom 8:30m 23m NoEoEoM 0580M 86m __toV—oE gowns—.3 Eomufl Nahum notBEom 838m Noom coon naafi meow meow ooom Noam ~¢o_ woofi Ham— nnafi wha— th_ waa_ Hma~ ww¢~ emm_ coo~ nbm_ _om~ moom fiao_ aamfi v03 wnafi $3 32 @132: wing: 2:32.: mtcmzm 3.8%.: £583 £533 Bomb amm Qmm mam amm Qmm am— ..gafiesfi. (8 E mam mmm .333ch + 9mm amm mam amm mam Afinzfizufimzu 33:: E amm gmm amm mam .cosqzcg A dm< 203 <<<<§cca813. 88:00 820m «880% 8:5 8.20: 8.20: 85 8809802 Z 9.83 «838m 820 8232 «83880: 3:38 «8802 883 «8052 888.3 888.3. :85 88:85. 28 250032 80 omw=0 223:0 020 £50 880 03.5 8.80 €30 25.0 80m £50 £8 £30 £30 830 $8-220 8-85 :mobaom 8m3m <0 8 88:5 05,—. .3 E 880N083 820505» 8895 808.80 285 82850 Snow—mom 88.203 :8303 833088 38035082 was.» 8880.3 Nogoiflmmmm “onus—mom 880% 88828 0:4 355 >283 to 8880: £80.88“ 0:852 0:852 838883 Baum—Em Baa“ oao_ moa~ whafi vwofi voofi camfi unm_ meow meow cam :33 #008 ofiom aww~ voo~ mama muafi moon moon meow moom £5888; 8328 K0 338 8.528 ..m. 338 .m. 880528 .388. 880.28 833.3. Snatch .8 0835.3. 88.3% 8888:8603 83:38: 0:828: 8800:0285 88002085 8808x088 .8853 82¢sz 88808 8.88.: b.2085 388800 A .83 204 2.30% 353% m.v-m.m 60% 8: 08—6 385:6 m0» 0.502000% 530% 335% 3;; 60% 8: 08—6 283036 m0» “SENSE 0080005 60% you .on 60% Ho: 30.“ mwfiu 0: 2035.00. 60800:: 333% 4 Th 60% 8: 08—6 00% m0» £00003 6080005 033% .R-w 00% 0026 082865 m0» 02002.: 60800:: 3650536 .o _-w 60% 8: Rom mwat go: 0.00:0:35‘ wag». 303255 .0...” 60% Ho: 0026 $30 0: 0.220528 6080005 3050536 .N; 60% Ho: 30.“ x0>-o>06=o0 0: ”00200000000220 2.36% 653280: o. The 60% “o: 0026 mwcu o: 0.520» “$3.5m 30305—5 ._ 73 60% 8: :000—mam x0>-o>06:oo o: ..QEbSn 2.38% 3630336 e.w; 60% Ho: 30% mwct on 050.2% 238% 3030536 3 60% Sc 30.“ x0>-o>06:oo m0» 0:832:00 wfigo Hawaz—0 3 60% 8: 0020 330 60% Ho: 0320030055 230% 60% you 60% Ho: 60% “on 60% Ho: 60% 8: 60% B: 360$ wfigo 60% Ho: 60% 8: 60% 8: 0026 60% «o: 60% Ho: mmfigobagfix 330% 60% 8: .N _ A: 60% 8c :000—mam was: on 050M330 Ewfibm 60% 8: @410; 60% “on 08.0 mwnt o: 3&0? 60% Ho: 60% 8: 60% Ho: 60% Ho: 60% Ho: 60% Ho: 60% 8: 030.23wa 60800:: 60% “on 60% 8: 60% 8: 080m awn: m0» 83305 60800:: $0333 _ .v H -md 0.0% 0026 x0285 m0» ESEM§£ wfigo 0.033% om-m.m m0» 0026 x0385 m0» 03:033320 0925 55000.9” 3:5 82:35 08:00.20 gown—sag“ 959.0 830935—0— 52>? 003—93 3:83:30 8m $005 00068020 N: 59003qu 205 206 3328 388:8 8333 830:8 o: 8838 3:88: 038 0:, o: 88::8Q 3888 388:8 3:3 98:88 883333 8 038 o: o: .3855 3328 388:8 3:3 .88 3: .8838 o: o: o: 0: 838.8 3828 388:8 3:3 3.383 883333 0: o: o: .3388 83.83 58:8 388:8 3:3 __583 88332: 0: 038 o: o: 8.885 8:8 8:88:83 3:3 .88 3: ““8838 o: o: o: o: 88855: 58.8 388:8 3:3 .88 3: ““8838 o: 88:88 0: 0: 88283.28 88:8 388:8 3:3 .88 3: 8838 o: 98:88 0: 0: 8.3888888 8838 9:88:00 833: .88 3: .8838 o: o: o: o: 8.58: 58:8 388:8 3:3 .88 3: $2 8 o: o: o: 533:3: 58:8 388:8 333 .88 3: 8838 o: 8:8 8:8 0: 8:88» 5838 388:8 3:3 .88 3: 8838 0: 8:8 0: 0: 88358 583m 388:8 3:3 8388 o: 8838 o: o: 8:8: 0: 8383853 .88 3: .88 3: .88 3: .88 3: 8.8838 .88 3: .88 3: o: .883: .883 .88 3: 838888 .88 3: .88 3: .8838 o: o: o: 8838: 3583:8383 5838 8:88:83 3:3 .88 3: < \ Z 0: o: o: 0: 888.28 £83m 388:8 3:3 .88 3: 88838 8:88: o: o: o: 8888 .88 3: .88 3: .88 3: .88 3: .88 3: .88 3: .88 3: .88 3: .88 3: .8883 58:8 388:8 3:3 .88 3: 8838 0: 8:8 0: o: 83:85 5838 388:8 3:3 033:8 88383:: 0: 8:8 0: o: 383M283 3388 388:8 3:3 21.383 88333: 0: o: o: 8838: 8.3828383 58:5 88:85 8.8.— 38.8:— 883 _c=§amhwfl£u:§m 335— 8:33.— 8888:53— .8:338 d .8? : Coded Characters Appendix III x. Ann sac Int. ernk Width Ends Dlam Sm Bran Para Plano Shape Palm Radial Gro Close '89 1 alleghaniensis brongniartii montalto elegans corrugatus 0 minorwcensis annulatus 0 O strictus(K1 ) strictus(K2) dzulynskii tenuis 0 qiongzhusiensis tan‘mensis hunanensis stn’ctus(P) linearis 207 1 lateralis simplex minimus parallelus Literature Cited Abel, O. 1935. Vorzeitliche Lebensspuren. Fischer, Jena. 644 pp. [In German] Aceflolaza, G., and Heredia, S. 2008. The status of the Cruziana (trilobite trace fossil) stratigraphy in western Gondwana: the mixing of Lower and Upper Ordovician elements in the central Andean Basin of South America. In Rabano, I., Gozalo, R., and Garcia-Bellido, D. (eds.), Advances in trilobite research. Cuademos de Museo Geominero, no. 9. Instituto Geolégico y Minero de Espana, Madrid, pp. 13-17. Aceflolaza, G., and Acefiolaza, F. G. 2003. Ordovician trace fossils of Argentina. In Acefiolaza, F. G. (ed.), Aspects of the Ordovician System in Argentina. Serie Correlacion Geologica 16:177-194. Akpan, E. B. and Nyong, E. E. 1987. Trace fossils assemblage and depositional environment of Turonian calcareous sandstones in the southern Benue Trough, Nigeria. Journal of African Earth Sciences 6(2):]75-180. Alexandrescu, G., and Brustur, T. 1984. Ichnofaciesul cu Sabularia in Stratele de Vinetisu din Partea de Nord a Carpatilor Oriental. Dari de Seama ale Sedintelor, Institul Geologic si Geofisica. Paleontologia, 68 (for 1981):]7-22. [In Romanian] Alpert, S. P. 1975. Planolites and Skolithos fiom the White-Inyo Mountains, California. Journal of Paleontology 49:508-521. Alpert, S. P. 1977. Trace fossils and the basal Cambrian boundary. In Crimes, T. P., and Harper, J. C. (eds.), Trace fossils 2. Geological Journal Special Issue 9:1-8. Baldwin, C. T. 1977. the stratigraphy and facies associations of trace fossils in some Cambrian and Ordovician rocks of northwestern Spain. In Crimes, T. P., and Harper, J. C. (eds.), Trace fossils 2. Geological Journal Special Issue 9:9-40. Baldwin, C. T., and Strother, P. K. 2004. The internal structure and kinematics of production of the Paleozoic trace fossil Arthrophycus alleghaniensis and a possible non-trilobite tracemaker. Abstract Book Ichnia 2004, First International Congress on Ichnology, Trelaw: p. 17. Banerjee, I. 1982. Trace fossils of the bioturbate sandstone facies of the Eze-Aku Formation, Nigeria. Indian Journal of Earth Sciences 9(2):93-98. Bassler, R. S. 1915. Bibliographic index of American Ordovician and Silurian fossils. Vol. 1. Smithsonian Institution, Bulletin 92, 718 pp. 208 Becker, H. F., and Donn, W. 1952. A new occurrence and description of the fossil Arthrophycus. Science 1 15(2982):214-215. Bender, F. 1963. Stratigraphic dcr “Nubischen Sandsteine” in Sued-Jordanien. Geologische Jahrbuch 81:237-276. [In German] Bender, F. 1968. Geologic von Jordanien. In Martini, H. J. (ed.), Beitragc zur Regionalen Geologic der Erde, vol. 7, pp. 1-230. [In German] Benton, M. J. 1982. Dictyodora and associated trace fossils from the Palaeozoic of Thuringia. Lethaia 15(2)]15-132. Bertling, M. 2007. What’s in a name? Nomenclature, systematics, ichnotaxonomy. In Miller, W. (ed.), Trace fossils: concepts, problems, prospects. Elscvicr, Boston, pp. 81-91. Bertling, M., Braddy, S., Bromley, R., Demathieu, G., Genisc, J ., Mikulas, R., Nielsen, J., Nielsen, K., Rindsberg, A., Schlirf, M., and Uchman, A. 2006. Names for trace fossils: a uniform approach. Lethaia 39:279-286. Bhargava, O. N., Bassi, U. K., and Chopra, 8. 1984. Trace fossils from the Ordo- Silurian rocks of Kinnaur, Himachal Himalaya. Journal of the Geological Society of India, 25(3):175-186. Bhargava, O. N., and Bassi, U. K. 1988. Trace fossils from the Palaeozoic-Mesozoic sequence of Spiti-Kinnaur (Himachal Himalaya) with comments on palacoenvironmcntal control on their frequency. Journal of the Geological Society of India, 32(3):227-238. Bjerstedt, T. W. 1987. Latest Devonian-earliest Mississipian nearshore trace-fossil assemblages from West Virginia, Pennsylvania, and Maryland. Journal of Paleontology 61 :865-889. Blackith, R. and Reyment, R. 1971. Multivariate morphometrics. Academic Press, London, 412 pp. Borghi, L., Moreira, M. I. C., and Femandes, A. C. S. 1996. A ocorrcncia do icnogenero Arthrophycus Hall, 1852 cm Chapada dos Guirnaraes, Estado de Mato Grosso. Anais da Academia Brasileira de Ciencias 68(2)274-275. [In Portuguese] Borrello, A. V. 1966. Trazas, rcstos tubiforrncs y cucrpos fosilcs problematicos de la Formacion La Tinta, Sierras scptcntrionales — Provincia de Buenos Aircs. Paleontografia Bonacrense 5:1-42. [In Spanish] 209 Bourrouilh, R. 1973. Stratigraphic, scdimentologic ct tectonique dc l’ile dc Minorque et du Nord-Est dc Majorque (Baleares): Unpublished PhD thesis, Univ. Paris 822p. (not seen, trusting Orr 1994.) Brandt, D., Csonka, J ., McCoy, V. 2008. In search of a modern analog for Arthrophycus ichnogenus. Abstracts with Programs - Geological Society of America 40(6): 101 . Brandt, D., Seitz, M., McCoy, V., Csonka, J., Barringer, J., Holmquist, E., Kraig, S., Morgan, R., Myers, J ., and Paquette, L. 2010. A new ichnospecies of Arthrophycus from the Late Carboniferous (Pennsylvanian) of Michigan, USA. Ichnos 17:1-8. Brett, C. Personal communication. Bromley, R. G. 1990. Trace fossils: biology and taphonomy. Unwin Hyman, London, 280 pp. Burjack, M. I. A., and Popp, M. T. B. 1981. A ocorrcncia do icnogenero Arthrophycus no Paleozoico da Bacia do Parana. Pesquisas—Instituto dc Geociencias. Universidadc Federal do rio Grandc do sul 14:163-167. [In Portuguese] Chiplonkar, G. W., and Ghare, M. A. 1975. Some additional trace fossils fi'om the Bagh Beds. Bulletin of the Indian Geological Association 8:71-84. Conrad, T. A. 1837. First annual report on the geological survey of the third district of the State of New York. New York Geological Survey, Annual Report (l):155- 1 86. Conrad, T. A. 183 8. Report on the paleontological department of the survey [of New York]. New York Geological Survey, Annual Report 2: 107-1 19. Conrad, T. A. 1839. Second and third annual reports of the Paleontological Department of the Geological Survey, N.Y., pp. 57-66. Correia Perdigao, J. 1964. Sobrc a dcscoberta de Cruziana c Vexillum (=Daedalus) na colina dc cuncos (Mourao). Comunicacocs dos Servicos Geologicos de Portugal 48:161-163. [In Portuguese] Cotter, E. 1983. Shelf, paralic, and fluvial environments and custatic sea-level fluctuations in the origin of the Tuscarora Formation (Lower Silurian) of Central Pennsylvania. Journal of Sedimentary Petrology 53:25-49. Crimes, T. P. 1968. Cruziana, a stratigraphically useful trace fossil. Geological Magazine 105(4):360-364. 210 Jam—__— Crimes, T. P. 1981. Lower Palaeozoic trace fossils of Afi'ica. In Holland, C. H. (ed.), Lower Palaeozoic of the Middle East, eastern and southern Africa, and Antarctica: with essays on Lower Palaeozoic trace fossils of Africa and Lower Palaeozoic palaeoclimatology. Wiley, London, pp. 189-198. Dalloni, M. 1934. Ge'ologie. Mission au Tibesti (1930-1931). Mcmoires dc L’Académie des Sciences de L’Institut de France 61 :39-170. [In French] Dawkins, R. 1982. The extended phenotype: the long reach of the gene. Oxford University Press, Oxford, 336 pp. Dawson, J. W. 1864. On the fossils of the genus Rusophycus. Canadian Naturalist and Geologist, new series 1, pp. 363-367, 458. De Alvarenga, C. J. S., Guimaraes, E. M., Assine, M. L., Perinotto, J. A. dc J ., and Laranjcira, N. P. F. 1998. Seqtléncia Ordovicio-Siluriana e Devoniana no flanco nortc da Bacia do Parana. Anais da Academia Brasileira dc Ciencias 70(3):587-606. [In Portgucse] De Castelnau, F. 1843. Essai sur le Systéme Silurien dc l’Amérique Septentrionale. P. Bertrand, Paris; V‘. Levrault, Strasbourg, 56 pp., 27 pl. (not seen, trusting Rindsberg and Martin 2003) Demathieu, G. and Demathieu, P. 2002. Concerning the erection of ichnogenera and ichnospecies in vertebrate ichnotaxonomy. Ichnos 9:117-121. Downey, W. F., Jr. 1980. Decoding a fossil trail. Pennsylvania Geology 11(3):6-11. Duimovich, O. A. 1963. Observaciones geologicas en el Ccrro San Agustin, Balcarce, Buenos Aires. Revista de la Asociacion Geologica Argentina 18(1-2):108-109. [In Spanish] Durand, J. 1985. Les Grcs armoricain. Sedimentologie-traces fossils. Milieux dc depot. Centre Armoricain d’Etude structural des Sociles, Mcmoires ct Documents 3, 150 pp. [In French] Dutta, S., and Chaudhuri, S. 2005 (published 2008). Trace fossils in the Manendragarh Beds of the Lower Gondwana Talchir Formation and their significance. Indian Journal of Geology 77(1-4):23-31. Eagar, R. M. C., Baines, J. G., Collinson, J. D., Hardy, P. G., Okolo, S. A., and Pollard, J. E. 1985. Trace fossil assemblages and their occurrence in Silesian (Mid- Carboniferous) deltaic sediments of the Central Penninc Basin, England. SEPM Special Pub. 35, pp. 99-149. 211 Eaton, A. 1820. Index to the Geology of the Northern States. Troy, NY: W.S. Parker 21]. (not seen, trusting James 18933) Baton, A. 1832. Geological textbook, for aiding the study of North American geology, 2nd Ed. G. and C. and H. Carvill, New York, 134 pp. (not seen, trusting Rindsberg and Martin 2003) Eschard, R, Abdallah, H., Braik, F., and Desaubliaux, G. 2005. The Lower Paleozoic succession in the Tassili outcrops, Algeria: sedimentology and sequence stratigraphy. First Break 23 (Algerian feature), pp. 27-36. Fauchald, K. Personal communication. Fenton, C. L., and Fenton M. A. 1958. The Fossil Book. Doubleday & Company Incorporated, New York City, 482 pp. Femandes, A. C. S. 1999. Conteudo icnologico das forrnacoes do Ordoviciano- Dcvoniano da bacia do Parana, Brasil. Boletim do Muscu Nacional. Geologia 4621-12. [In Portuguese] Femandes, A. C. S. 2001. A paleoicnofauna brasileira dc artrépodes: estado atual de seu conhecirnento. Acta Geologica Leopoldensia 24(52/53):359-372. [In Portuguese] Femandes, A. C. S., Percira, E., and Bergarnaschi, S. 1995. A prescnca dc Arthrophycus na F ormacfio Vila Maria no municipio de Rondonopolis (MT). Anais Academia Brasileira dc Ciencias, Rio de Janeiro, 67(3):383-384. [In Portuguese] F emandes, A. C. S., and Borghi, L. 1996. Comentarios sobre o icnogenero Arthrophycus Hall, 1852 e seu registro nas bacias scdimentares brasileiras. Anais do Simposio Sul Americano do Siluro-Dcvoniano, Ponta Grossa, 131- 139. [In Portuguese] Femandes, A. C. S., Borghi, L., Moreira, M. I. C. 2000. Sobre a ocorrcncia do icnogenero Arthrophycus Hall, 1852 na Formacao furnas (Bacia do Parana). Boletim do Museu Nacional. Geologia 52: 1 -14. [In Portuguese] Femandes, A. C. S., Borghi, L., Carvalho, 1. dc S., de Abreu, C. J. 2002. Guia dos icnofosscis dc invertebrados do Brasil. Editora Interciéncia, Rio de Janeiro, 260 pp. [In Portuguese] Frey, R. W. 1970. Trace fossils of Fort Hays Limestone Member of Niobrara Chalk (Upper Cretaceous), west-central Kansas. University of Kansas Paleontological Contributions 53:141. 212 Frey, R. W. 1972. Palcoecology and depositional environment of Fort Hays Limestone Member, Niobrara Chalk (Upper Cretaceous), west-central Kansas. University of Kansas Paleontological Contributions 58:1-72. Frey, R. W. 1973. Concepts in the study of biogenic sedimentary structures. Journal of Sedimentary Petrology 43:6-19. Frey, R., and Howard, J. 1970. Comparison of upper Cretaceous ichnofaunas from siliceous sandstones and chalk. Western Interior region, USA. In Crimes, T. P., Harper, J. C. (eds.), Trace fossils. Geological Journal 3 (special issue), pp. 141-166. F ritel, P. H. 1925. Végétaux paléozo’r'ques ct organismes problématique dc L’Ouadofi. Bulletin de la Societe Géologique de France 25:33-48. [In French] F ritsch, A. 1908. Problematica Silurica. In Barrande, Systéme Silurien du centre de la Bohéme. Prague, 24 pp. [In French] F firsich, F. T. 1974. On Diplocraterion Torell 1870 and the significance of morphological features in vertical, spreiten-bearing, U-shaped trace fossils. Journal of Paleontology 48:952-962. Ghare, M. A., and Kulkarni, K. G. 1986. Jurassic ichnofauna of Kutch-II. Wagad Region. Biovigyanam 12(1):44-62. Ghosh, S. K. 2003. First record of marine bivalves from the Talchir Formation of the Satpura Gondwana Basin, India: palacobiogeographic implications. Gondwana Research (Gondwana newsletter section) 6(2):312-320. Gong Yiming. 1999. Flysch trace fossils from the Hercynian and Indosinian orogenic belts of northwestern China and their palacoenvironmcntal significance. Acta Geologica Sinica (English Edition), 73(4):384-394. Goppcrt, H. R. 1852. Fossilc flora des Ubergangsgebirgcs. Novorum Actorurn Acadcmiae Caesarae Leopoldino-Carolinac Naturae Curiosorum, Hallc 22(supplement) 1-298. [In German with Latin diagnosis] Grahn, Y. 1992. Revision of Silurian and Devonian strata of Brazil. Palynology 16:35-61. Greb, S. F ., and Chesnut, D. R., Jr. 1994. Paleoecology of an estuarine sequence in the Breathitt formation (Pennsylvanian), Central Appalachian Basin. Palaois 9:388-402. Grove, A. T. 1960. Geomorphology of the Tibesti Region with special reference to Western Tibesti. The Geographical Journal 126(1):18-27. 213 Gubler, Y., Bugnicourt, D., Faber, J ., Kubler, B., and Nyssen, R. 1966. Essai dc nomenclature et caractérisation des principalcs structures sédimentaires. Chambre Syndicale de la Recherche et de la Production du Petrole ct du Gaz nature]. Editions Technip, Paris, 291 pp. [In French] Hall, J. 1852. Palaeontology of New York, Vol. II. Containing descriptions of organic remains of the Lower Middle Division of the New-York system. Albany: Geological survey of New York 5. 362 pp., 85 pl. Hammer, 0., Harper, D. A. T., and Ryan, P. D. 2001. PAST: Paleontological Statistics software package for education and data analysis. Palaeontologia Electronica 4(1):9 pp. Hammer, 0. and Harper, D. 2005. Paleontological data analysis. Wiley-Blackwell, Oxford, 368 pp. Hantzschel, W. 1975. Treatise of Invertebrate Paleontology, Part W. Miscellanea. Supplement 1: Trace Fossils and Problematica. Lawrence, Kansas, University of Kansas and Geological Society of America. Harlan, R. 1831. Description of an extinct species of fossil vegetable of the family fucoids. Journal of the Academy of Natural Science Philadelphia 6:289-295. Harlan, R. 1832. On a new extinct fossil vegetable of the family Fucoides. Journal of the Academy of Natural Science Philadelphia, (ser. 1) 6:307-308. Harlan, R. 1836. Critical notes of various organic remains hitherto discovered in North America. Transactions of the Geological Society of Pennsylvania 1:46- 1 12. I-Ierzer, H. 1901. Six new species, including two new genera, of fossil plants. Ohio State Academy of Science, Ninth Annual Report, pp. 22-29. Howard, J. D. 1966. Characteristic trace fossils in the Upper Cretaceous sandstones of the Book Cliffs and Wasatch Plateau. Bulletin - Utah Geological and Mineralogical Survey, pp.35-53. Hundt, R. 1940. Neue Lebensspurcn aus dem ostthueringer Palaeozoicum (mit einem Bcitrag uebcr Nercietenfaehrten). chtr. Miner. Abt. B, 72210-216. [In German] Hundt, R. 1941. Das Mitteldeutsche Phycodesmeer. Gustav Fischer, Jena, 136 pp. [In German] 214 ICZN. 1999. International Code of Zoological Nomenclature, 4'h ed. International Trust for Zoological Nomenclature. Accessed 3 Sept. 2008. www.iczn.org/iczn/index.jsp. James, J. F. 1893a. Remarks on the genus Arthrophycus Hall. Journal of the Cincinnati Society of Natural History 16: 82-86. James, J. F. 1893b. Studies in problematic organisms: No. II — the genus Fucoides. Journal of the Cincinnati Society of Natural History 16: 62-81. Janvier, P., and Mclo, H. G. 1988. Acanthodian fish remains from the Upper Silurian or Lower Devonian of the Amazon Basin, Brazil. Palaeontology 31 :771-777. Keighley, D. G., and Pickerill, R. K. 1996. Small Cruziana, Rusophycus and related ichnotaxa from eastern Canada: the nomenclatural debate and systematic ichnology. Ichnos 4:261-285. Kern, J. P. 1978. Trails fi'om the Vienna woods: paleoenvironments and trace fossils of Cretaceous to Eocene flysch, Vienna, Austria. Palaeogeography, Palaeoclimatology, Palaeoecology 23 :23 1 -262. Konate, M., Guiraud, M., Lang, J ., and Yahaya, M. 2003. Sedimentation in the Kandi extensional basin (Benin and Niger): fluvial and marine deposits related to the Late Ordovician deglaciation in West Africa. Journal of African Earth Sciences 36:1 85-206. KsiaZkiewicz, M. 1970 . Ichnofossils of the Polish Carpathians. In Crimes, T. P., and Harper, J. C. (eds.), Trace fossils 3. Geological Journal Special Issue 283-322. Ksiazkiewicz, M. 1977. Trace fossils in the flysch of the Polish Carpathians Palaontologia Polonica 36, 208 pp. Kulkarni, K. G., and Borkar, V. D. 2002. Trace fossils and pseudofossils fiom the Proterozoic Cuddapah supergroup. Journal of the Geological Society of India 59:531-536. Kumpulainen, R. A., Uchman, A., Woldehaimanot, B., Kreuser, T., and Ghirmay, S. 2006. Trace fossil evidence from the Adigrat Sandstone for an Ordovician glaciation in Eritrea, NE Africa. Journal of African Earth Sciences 45:408-420. Laird, M. G. 1981. Lower Palaeozoic rocks of Antarctica. In Holland, C. H. (ed.), Lower Palaeozoic of the Middle East, eastern and southern Afi'ica, and Antarctica: with essays on Lower Palaeozoic trace fossils of Africa and Lower Palaeozoic palaeoclimatology. Wiley, London, pp. 257-314. 215 Legendre, L, and Legendre, P. 1983. Numerical ecology. Developments in environmental modelling 3. Elscvier Scientific Publishing Company, New York, 419 pp. Legg, I. C. 1985. Trace fossils from a Middle Cambrian deltaic sequence, North Spain. In biogenic structures: their use in interpreting depositional environments. H. A. Curran (cd.), Society of Economic Paleontologists and Mineralogists, Special Publication 35, pp. 151-165. Lesley, J. P. 1889. A dictionary of the fossils of Pennsylvania and neighboring states named in the reports and catalogues of the survey. Geological Survey of Pennsylvania, Report P4, Vol. I, pp. 37-41 and 272-273. Lessertisseur, J. 1955. Trace fossils d’activitc animalc ct leur signification paleobiologique. Societe Geologique de France, Memoire, nouvellc serie 74, pp. 1-150. [In French] Li, R. H. 1993. Trace fossils and ichnofacies of Middle Ordovician Gongwusu Formation, Zhuozishan, Inner Mongolia. Acta Palaeontologica Sinica 32:88- 104. [In Chinese with some English] Lin, 8., Zhang, Y., and Zhang, L. 1986. Body and trace fossils of Metazoa and algal macrofossils from the upper Sinian Gaojiashan Formation in southern Shaanxi. Shaanxi Dizhi 4(1):9-l 7. [In Chinese] Lifién, E. 1984. Los icnofosiles de la Formacion Torrearboles (Precambrico?- Cémbrico Inferior) en los alrededores de F uente de Cantos, Badajoz. Cuadernos do Laboratorio Xeologico dc Laxe 8:47-74. [In Spanish] Llompart, C., and Wieczorek, J. 1997. Trace fossil from Culm facies of Minorca Island. Proceedings of the XIII international congress on the Carboniferous and Permian, pages 99-101, 2 plates. Lockley, M. 1998. Philosophical perspectives on theropod track morphology: blending qualities and quantities in the science of ichnology. Gaia 15:279-3 00. Lopez, G. A., and Roy, D. C. 2002. Trace fossils from Early Silurian mudstones of the Spragueville Formation, northeast Maine, USA. Abstracts with Programs, Geological Society of America 34(1):59. Luo, H., Tao, Y., and Gao, S. 1994. Early Cambrian trace fossils near Kunming, Yunnan. Acta palaeontologica Sinica 33(6):676-685. Maberry, J. O. 1971 . Sedimentary features of the Blackhawk Formation (Cretaceous) in the Sunnyside District, Carbon County, Utah. U. S. Geological Survey Professional Papers 688:1-42. 216 MacEachern, J. A., Bann, K. L., Battacharya, J. P., and Howell, C. D, Jr. 2005. Ichnology of deltas, organism responses to the dynamic interplay of rivers, waves, storms, and tides. In Giosen, L., and Bhattacharya, J. P. (eds.), River Deltas: concepts, models, and examples. SEPM special Pub. 83. MacNaughton, R. B., and Pickerill, R. K. 2003. Taphonomy and the taxonomy of trace fossils: a commentary. Lethaia 36:66-70. Maejima, W., Das, R, Pandya, K. L., and Hayashi, M. 2001. Post-glacial sedimentation of Talchir Formation, Talchir Gondwana Basin, Orissa, India. Gondwana Research 4(4)694-695. Magwood, J. A. 1992. Ichnotaxonomy: a burrow by any other name. . .? In Maples, C. and West, R. (eds.) Trace fossils. In Culver, S. (ed.), Short courses in paleontology, number 5. The Paleontological Society, pp. 15-33. Manca, N. del V. 1986. Caractcrcs icnologicos de la Formacion Carnpanario (Cambrico Superior) en Salta y Jujuy. Ameghiniana 23(1-2):75-87. [In Spanish] Mangano, M. G., Cannona, N. B., Buatois, L. A., and Guinea, F. M. 2005(a). A new ichnospecies of Arthrophycus fi'om the Upper Cambrian-Lower Tremadocian of Northwest Argentina: Implications for the Arthrophycid lineage and potential in ichnostratigraphy. Ichnos 12:179-190. Mangano, M. G., Buatois, L. A., and Muniz Guinea, F. 2005(b). Ichnology of the Alfarcito Member (Santa Rosita Formation) of northwestern Argentina: animal-substrate interactions in a lower Paleozoic wave-dominated shallow sea. Ameghiniana 42(4):641-668. Mantell, G. 1834. Geology of the South East of England. Longrnan, Rees, Ormc, Brown, Green and Longrnan, London, 415 pp. (not seen, trusting Harlan 1836) Martinsson, A. 1970. Toponomy of trace fossils. In: Crimes, T. P., Harper, J. C. (eds.), Trace fossils. Geological Journal 3 (special issue), pp. 323-330. Mergl, M., and Massa, D. 2000. A Palaeontological Review of the Devonian and Carboniferous Succession of the Murzuq Basin and the Djado Sub-Basin. In Sola, M. A., and Worsley, D. (eds.), Geological exploration in Murzuq Basin. Elsevier, Amsterdam, pp. 41-88. Metz, R. 1998. Silurian trace fossils from marginal marine deposits, Lizard Creek member of the Shawangunk Formation, Delaware Water Gap, New Jersey. Northeastern Geology and Environmental Sciences 20(2):]01-116. 217 Metz, R. 2006. Trace fossil research in the Delaware Water Gap National Recreation Area, New Jersey. Abstracts with Programs, Geological Society of America 38(7)156. Mikulas, R. 1992. Trace fossils from the Kosov formation of the bohemian Upper Ordovician. Sbornik geologickych ved Paleontologie 32:9—54. Miller, S. A. 1877. The American Palaeozoic fossils: a catalogue of the genera and species. Cincinnati, Ohio, published by the author, 253 pp. Miller, W. 111, Webb, F. Jr., and Raymond, L. A. 2009. Clustering and morphologic variation in Arthrophycus alleghaniensis (Lower Silurian of Virginia, USA) as evidence of behavioral paleoecology. Neues J ahrbuch fuer Geologic und Palaeontologie Abhandlungcn 25 1 ( l ): 1 09-1 1 7. Minter, N., Braddy, S., and Davis, R. 2007. Between a rock and a hard place: arthropod trackways and ichnotaxonomy. Lethaia 40:365-3 75. Moneda, CF. 1963. El perfil geologico de la Albertina, Gral. Pucyrredon, provincial Buenos Aires. Revista de la Asociacion Geologica Argentina 18(1 -2):108. [In Spanish] Moore, R. C. 1933. Historical Geology. McGraw-Hill Book Company, Inc., New York, 673 pp. Moreira, M. I. C., Borghi, L., Femandes, A. C. S. 1998. A primeira ocorrcncia do icnogenero Athrophycus Hall, 1852 na Formacao Pumas (Bacia do Parana). Anais da Academia Brasileira de Ciencias, 70(1): 1 5 1 . [In Portuguese] Moreira, M. I. C., and Borghi, L. 1999. Facies scdimentares c sistemas deposicionais das Formacocs Alto Garcas e Vila Maria na regiao de Chapada dos Guimaracs (MT), borda noroeste da Bacia do Parana. Revista Brasileira de Geociénias 29(3):419-428. [In Portuguese] Mukherjee, K. K., Sen, P., and Santra, D. K. 1987. Occurrence of ichnofossils in Gulcheru Quartzite of Lower Cuddapah sequence in Andhra Pradesh. Geol. Surv. India Spec. Publ. no. 11, pp. 118-122. Nathorst, A. G. 1881. Om spar av négra cvertebrerade djur m.m. och deras paleontologiska betysdelse. (Memoire sur quclques traces d'animaux sans vertebres etc. et de leur portee paleontologique.) Kongliga Svenska Vetenskaps-Akademiens Handlingar 18:1-59. (not seen, trusting Rindsberg and Martin 2003) Neto de Carvalho, C., Femandes, A. C. S., and Borghi, L. 2003. Diferenciacao das icnospecies e variantes de Arthrophycus e sua utilizacao problematica cm 218 icnoestratigrafia: o resultado dc homoplasias comportarnentais entre anelidcos e artropodes? Revista Espanola de Paleontologia 18(2):221-228. [In Portuguese] Niediwiedzki, G., Szrek, P., Narkiewicz, K., Narkiewicz, M., and Ahlbcrg, P. E. 2010. Tetrapod trackways from the early Middle Devonian period of Poland. Nature 463 :43-48. Nogueira, A. C. R, Truckenbrodt, W., and Soares, E. A. A. 1999. O icnogénero Arthrophycus de depositos sublitoraneos da Formacao Nhamunda (Siluriano inferior) da Bacia do Amazonas regiao dc Presidente Figueiredo. Revista Brasileira de Geociencias 29(2):]35-140. [In Portuguese] Ogg, G. 2008. International stratigraphic chart. International Commission on Stratigraphy. Orr, P. J. 1994. Trace fossil tiering within event beds and preservation of frozen profiles: an example fi'om the lower Carboniferous of Menorca. Palaiois 9:202-210. Osgood, R.G. 1970. Trace fossils of the Cincinnati Area. Palaeontolographica Americana 6(41):281-438. Osgood, R. G. 1975. The history of invertebrate ichnology. In Frey, R. W. (ed.), The study of trace fossils: a synthesis of principles, problems, and procedures in ichnology. Springer-Verlag, New York 3-12. Osgood, R. G. 1975b. 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 87-108. Paczesna, J. 1996. The Vendian and Cambrian ichnocoenoses fiom the Polish part of the east-European platform. Prace Panstwowego Instytutu Geologicznego CLl 1, 77 pp. Peeples, C. B. III, Isaacson, P. E. and Rember, W. C. 1997. Paleogeography of low- diversity faunas from the Swan Peak Sandstone (Middle Ordovician), southeastern Idaho. Abstracts with Programs, Geological Society of America, 29(6):98. Pemberton, S. G., and Risk, M. J. 1982. Middle Silurian trace fossils in the Hamilton, Ontario region: their identification, abundance, and significance. Northeastern Geology 4(2):98-104. Perez, G. and Salazar, A. 1978. Estratigrafia y facies del Grupo Guadalupe. Geologia Colombiana No. 10, pp. 7-85. [In Spanish] 219 Pettijohn, S. J ., and Potter, P. E. 1964. Atlas and glossary of primary sedimentary structures. Springer-Verlag, Berlin, 370 pp. Pfltiger, F. 1999. Matground structures and redox facies. Palaios 14:25-39. Pickerill, R. 1994. Nomenclature and taxonomy of invertebrate trace fossils. In Donovan, S., The palcobiology of trace fossils. The Johns Hopkins University Press, pp. 3-42. Pickerill, R. K., Fillion, D., and Brenchley, P. J. 1991. A note on the occurrence of Arthrophycus in the Bell Island Group of eastern Newfoundland. Atlantic Geology 27:73-77. Pickerill, R. K., Romano, M., and Melendez, B. 1984. Arenig trace fossils from the Salamanca area, western Spain. Geological Journal 19:249-269. Plotnick, R., and Wagner, P. 2006. Round up the usual suspects: common genera in the fossil record and the nature of wastebasket taxa. Paleobiology 32(1):126- 146. Poiré, D. G., Spalletti, L. A., and Del Valle, A. 2003. The Cambrian-Ordovician siliciclastic platform of the Balcarce Formation (Tandila System, Argentina): Facies, trace fossils, palaeocnvironments and sequence stratigraphy. Geologica Acta, 1(1):41-60. Prouty, W. F ., and Swartz, C. K. 1923. Vermes. In Maryland Geological Survey: Silurian. The Johns Hopkins Press, Baltimore 8:402-405. Rindsberg, A. 1998. Workshop on ichnotaxonomy, Bomholrn, Denmark. WIT Group. Accessed Apr 7, 2009. http://www.envs.emory.edu/faculty/MARTIN/ichnology/IN99-WITREP~1 .htrn Rindsberg, A. K., and Martin, A. J. 2003. Arthrophycus in the Silurian of Alabama (USA) and the problem of compound trace fossils. Palaeogeography, Palaeoclimatology, Palaeoecology 192:187-219. Romano, M. 1991. Lower to Middle Ordovician trace fossils fi'om the Central Iberian Zone of Portugal and Spain. In: Barnes, C. R., Williams, S. H. (eds.), Advances in Ordovician Geology. Geological Survey of Canada, Paper 90-9, pp. 191-204. Roniewicz, P. and Pienkowski, G. 1977. Trace fossils of the Podhale Flysch Basin. In Crimes, T. P., and Harper, J. C. (eds.), Trace fossils 2. Geological Journal Special Issue 9:273-288. 220 Sarle, C. J. 1905. The burrow origin of Arthrophycus and Daedalus (V exillum). Science 22:335. Sarle, C. J. 1906. Arthrophycus and Daedalus of burrow origin. Proceedings of the Rochester Academy of Science 4:203 -210. Schiller, W. 1930. Die tektonische Natur von Arthrophycus- und Spirophyton- aehnlichen Gebilden im Altpalaeozoikum dcr Provinz Buenos Aires (Argentinien). Geologische Rundschau 21(3)]45-151. [In German] Schimper, W. P. 1869. Traite' de Paleontologie Ve’gétale on La Flore du Monde Primitif dans ses Rapports avec les Formations Géologiques et La Flore du Monde Actuel, Vol. I. J. B. Baillierc et F ils, Paris. [In French] Schimper, W. P, and Schenk, A. 1879-1890. Palaeophytologie. In: Zittel, K. A. (ed.), Handbuch der Palaeontologie. Munchen, Druck und Verlag von R. Oldenbourg, 958 pp. [In German] Schuchert, C. 1916. Silurian formations of southeastern New York, New Jersey, and Pennsylvania. Geological Society of America Bulletin 27:531-554. Scotese, C. R. 2002. http://www.scotese.com, (PALEOMAP website). Seilacher, A. 1953. Studien zur Palichnologie. I. Ueber die Methoden der Palichnologie. Neues Jahrbuch fiir Geologic und Palaeontologie 96(3):421- 452. Seilacher, A. 1955. Spuren und Facies irn Unterkarnbrium. In Schindenwolf, Seilacher, A. (eds.), Beitrage zur Kenntnis des Karnbriums in der Salt Range (Pakistan). Akad. Wiss. Lit. Mainz, Abh. math.-naturw. KL, pp. 373-399. Seilacher, A. 1964a. Biogcnic sedimentary structures. In Imbrie, J. and Newell, N. (eds.), Approaches to paleoecology. John Wiley and Sons, Inc., New York, pp. 296-3 16. Seilacher, A. 1964b. Sedimentological classification and nomenclature of trace fossils. Sedirnentology 3:253-256. Seilacher, A. 1970. Cruziana stratigraphy of “non-fossiliferous” Palaeozoic sandstones. In Crimes, T. P., and Harper, J. C. (eds.), Trace fossils 3. Geological Journal Special Issue, pp. 447-476. Seilacher, A. 1997. Fossil art: an exhibition of the Geologisches Institut Tuebingen University Germany. Royal Tyrrel Museum of Palaeontology, Drumheller (Alberta), 64 pp. 221 Seilacher, A. 2000. Ordovician and Silurian Arthrophycid ichnostratigraphy. In Sola, M. A., and Worsley, D. (eds.), Geological exploration in Murzuq Basin. Elsevier, Amsterdam, pp. 237-258. Seilacher, A. 2007a. Trace Fossil Analysis. Springer, Berlin, 226 pp. Seilacher, A. 2007b. Principles of Ichnostratigraphy. In Bromley, R. G., Buatois, L. A., Mangano, G., Genisc, J. F ., and Melchor, R. N. (eds.) Sediment-Organism Interactions: A Multifaceted Ichnology. SEPM Special Publication No. 88, pp. 53-56. Seilacher, A. and Alidou, S. 1988. Ordovician and Silurian trace fossils from northern Benin (W-Africa). Neues Jahrbuch fuer Geologic und Palaeontologie. 1988(7):43 1-439. Seilacher, A., Ltlning, S., Martin, M. A., Klitzsch, E., Khoja, A., and Craig, J. 2002. Ichnostratigraphic correlation of Lower Palaeozoic clastics in the Kufi'a Basin (SE Libya). Lethaia 35(3):257-262. Seilacher, A., Cingolani, C., and Varela, R. 2003. Ichnostratigraphic correlation of early Palaeozoic sandstones in North Africa and central Argentina. In Salem, M. J ., and Oun, K. M. (eds.), The geology of northwest Libya; sedimentary basins of Libya, second symposium 2:275-292. Selley, R. C. 1970. Ichnology of Palaeozoic sandstones in the southern desert of Jordan: a study of trace fossils in their scdimentologic context. In: Crimes, T. P., Harper, J. C. (eds.), Trace fossils. Geological Journal 3 (special issue), pp. 477-488. Selley, R. C. 1972. Diagnosis of marine and non-marine environments from the Cambro-Ordovician sandstones of Jordan. Journal of the Geological Society, 128(2):]35-150. Shimer, H. W., and Shrock, R. R. 1944. Index fossils of North America. John Wiley & Sons, Inc., New York, 837 pp. Silva, S. dc O. 1951. Siluriano no rio Tapajos. Revista da Escola de Minas 16(5):9- 11. [In Portuguese] Simpson, S. 1975. Classification 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 39-54. Siveter, D. J ., Briggs, D. E. G., Siveter, D. J., Sutton, M. D., and Fortey, R. A. 2008. Horns, eggs, and legs: exceptionally preserved new arthropods from the Herefordshire (Silurian) lagerstatte. In RAbano, I., Gozalo, R., and Garcia- 222 Bellido, D. (eds.) Advances in trilobite research. Cuademos de Museo Geominero, no. 9. Instituto Geologico y Minero dc Espana, Madrid, pp. 371- 374. Sneath, P. H. A., and Sokal, R R 1973. Numerical taxonomy: the principles and practice of numerical classification. W. H. Freeman and Company, San Francisco, 573 pp. Sokal, R. 1966. Numerical taxonomy. Scientific American 215(Dec.):106-116. Stanley, T. M., and Feldmann, R. M. 1998. Significance of nearshore trace-fossil assemblages of the Cambro-Ordovician Deadwood Formation and Aladdin Sandstone, South Dakota. Annals of Carnegie Museum 67: 1-51. Taylor, R. C. 1834. A description of a fossil vegetable of the family Fucoides, in the Transition Rocks of North America, and some considerations of geology connected with it. London’s Magazine of Natural History 7:27-32. Taylor, R C. 1835. On the geological position of certain beds which contain numerous fossil marine plants of the Family F ucoidcs, discovered near Lewistown, Mifflin County, Pennsylvania. Transactions Geological Society of Pennsylvania 1 :5-15. Terrell, F. M. 1972. Lateral facies and paleoecology of Permian Elephant Canyon Formation, Grand County, Utah. Brigham Young University Geology Studies 19(2):3-44. Trewin, N. H. 1994 (1995). A draft system for the identification and description of arthropod trackways. Palaeontology 37:811-823. Turner, B. R. and Benton, M. J. 1983. Paleozoic trace fossils form the Kufra Basin, Libya. Journal of Paleontology 57:447-460. Turner, B. R., Makhlouf, I. M., and Armstrong, H. A. 2005. Late Ordovician (Ashgillian) glacial deposits in southern Jordan. Sedimentary Geology 181:73- 91. Uchman, A. 1998. Taxonomy and ethology of flysch trace fossils: revision of the Marian Ksiazkiewicz collection and studies of complementary material. Annales Societatis Geologorum Poloniae, 68:105—218. Uchman, A. 1999. Ichnology of the Rhcnodanubian Flysch (Lower Cretaceous- Eocene) in Austria and Germany. Beringeria 25:67—173. 223 Uchman, A., and Demircan, H. 1999. Trace fossils of Miocene deep-sea fan fi’inge deposits from the Cingoz Formation, southern Turkey. Annales Societatis Geologorum Poloniae, 69(3-4):125-135. Wagner, J. R. 1978. Facies comparison of Silurian clastics in Tennessee and Pennsylvania; contributions toward a regional synthesis of depositional environments. Abstracts with Programs, Geological Society of America 10(4):201. Webby, ED. 1977. Trace-fossil assemblages in early Paleozoic quartz-rich clastics of western New South Wales. Journal of Paleontology 51, no. 2, Supplement Part III, p. 30. Wolfart, R. 1961. Strategraphie und Fauna des aelteren Palaeozoikums (Silur, Devon) in Paraguay. Geologisches Jahrbuch 78:29-102. [In German] Wolfart, R. 1981. Lower Palaeozoic rocks of the Middle East. In Holland, C. H. (ed.), Lower Palaeozoic of the Middle East, eastern and southern Africa, and Antarctica: with essays on Lower Palaeozoic trace fossils of Africa and Lower Palaeozoic palaeoclimatology. Wiley, London, pp. 5-130. Yang, S. 1994. Cambrian-Ordovician trace fossils fi'om Kuruktag Mt., northeastern part of Tarim Basin, Xinjiang. Hsien tai ti chih = Geoscience 8(4):371-379. [In Chinese] Yang, R, Wang, Z., and Zhu, S. 1996. The late Triassic Zhuwo Formation in Maladun, Songpan, Sichuan: trace fossils and their depositional environments. Yanxiang Gudili = Sedimentary Facies and Palaeogeography 16:34-41. [In Chinese] Young, R S. 1955. Arthrophycus alleghaniensis as a guide fossil in northern Virginia. Journal of Paleontology 29:550-551. Zhang, X., and Wang, D. 1996. A restudy on Silurian-Devonian ichnofossils from northwestern Hunan. Acta palaeontologica Sinica 35(4):475-489. [In Chinese] Zimmermann, U., and Spalletti, L. A. 2009. Provenance of the Lower Paleozoic Balcarce Formation (Tandilia System, Buenos Aires Province, Argentina): implications for paleo geo graphic reconstructions of SW Gondwana. Sedimentary Geology 219:7-23. 224