Upper Cambrian Chitons (Mollusca, Polyplacophora) from Missouri, USA
John Pojeta, Jr., M. J. Vendrasco, and Guy Darrough
Bulletins of American Paleontology
Number 379, September 2010 BULLETINS OF AMERICAN PALEONTOLOGY Established 1895
Paula M. Mikkelsen Warren D. Allmon Editor-in-Chief Director
Editorial Board
Jason S. Anderson, University of Calgary Kenneth Angielczyk, Field Museum of Natural History Carlton Brett, University of Cincinnati Ann F. Budd, University of Iowa Peter Dodson, University of Pennsylvania J. Thomas Dutro Jr., United States Geological Survey Daniel Fisher, University of Michigan Dana H. Geary, University of Wisconsin-Madison Peter J. Harries, University of South Florida John Pojeta, United States Geological Survey Carrie E. Schweitzer, Kent State University Geerat J. Vermeij, University of California at Davis Emily H. Vokes, Tulane University (Emerita) William Zinsmeister, Purdue University
Bulletins of American Paleontology is published semiannually by Paleontological Re- search Institution. For a list of titles and to order back issues online, go to http://museum- oftheearth.org. Numbers 1-94 (1895-1941) of Bulletins of American Paleontology are available from Periodicals Service Company, Germantown, New York, http://www.pe- riodicals.com. Subscriptions to Bulletins of American Paleontology are available in hard-copy and electronic formats to individuals or institutions; non-US addressees pay an additional fee for postage and handling. For current rates, additional information or to place an order, see http://museumoftheearth.org or contact: Publications Offi ce Paleontological Research Institution 1259 Trumansburg Road Ithaca, New York 14850-1313 Tel. (607) 273-6623, ext. 20 Fax (607) 273-6620 [email protected]
Th is paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Upper Cambrian Chitons (Mollusca, Polyplacophora) from Missouri, USA
John Pojeta, Jr., M. J. Vendrasco, and Guy Darrough
Bulletins of American Paleontology
Number 379, September 2010 ISSN 007-5779 ISBN 978-0-87710-488-9 Library of Congress Catalog Card Number 2010935745 © 2010, Paleontological Research Institution, 1259 Trumansburg Road, Ithaca, New York 14850, U. S. A. UPPER CAMBRIAN CHITONS MOLLUSCA, POLYPLACOPHORA FROM MISSOURI, USA
John Pojeta, Jr. U. S. Geological Survey and Department of Paleobiology, Smithsonian Institution, P. O. Box 37012, NHB MRC 121, Washington, DC 20013-7012, U.S.A.; email [email protected]
M. J. Vendrasco Department of Biological Sciences (MH-282), California State University, Fullerton P. O. Box 6850, Fullerton, California 92834, U.S.A.; email [email protected], [email protected]
and Guy Darrough 2075 Rosedale Court, Arnold, Missouri 63012, U.S.A.; email [email protected], [email protected]
ABSTRACT Numerous new specimens reveal a greater presence of chitons in Upper Cambrian rocks than previously suspected. Evidence is presented showing that the chiton esthete sensory system is present in all chiton species in this study at the very beginning of the known polyplacophoran fossil record. Th e stratigraphic occurrences and paleobiogeography of Late Cambrian chitons are documented. Th e 14 previously-named families of Cambrian and Ordovician chitons are reviewed and analyzed. Aulochitonidae n. fam. is defi ned, based on Aulochiton n. gen.; A. sannerae n. sp. is also defi ned. Th e long misunderstood family Preacanthochitonidae and its type genus Preacanthochiton Bergenhayn, 1960, are placed in synonymy with Mattheviidae and Chelodes Davidson & King, 1874, respectively; Eochelodes Marek, 1962, also is placed in synonymy with Chelodes, and Elongata Stinchcomb & Darrough, 1995, is placed in synonymy with Hemithecella Ulrich & Bridge, 1941. At the species level, H. elongata Stinchcomb & Darrough, 1995, and Elongata perplexa Stinchcomb & Darrough, 1995, are placed in synonymy with H. eminensis Stinchcomb & Darrough, 1995. Th e Ordovician species H. abrupta Stinchcomb & Darrough, 1995, is transferred to the genus Chelodes as C. abrupta (Stinchcomb & Darrough, 1995). Th e Ordovician species Preacanthochiton baueri Hoare & Pojeta, 2006, is transferred to the genus Helminthochiton as H. ? baueri (Hoare & Pojeta, 2006). Th e Ordovician species H. marginatus Hoare & Pojeta, 2006, is transferred to the genus Litochiton as L. marginatus (Hoare & Pojeta, 2006). Matthevia walcotti Runnegar, Pojeta, Taylor, & Collins, 1979, is treated as a synonym of Hemithecella expansa Ulrich & Bridge, 1941. In addition, other multivalved Cambrian mollusks are discussed; within this group, Dycheiidae n. fam. is defi ned, as well as Paradycheia dorisae n. gen. and n. sp. Cladistic analysis indicates a close relationship among the genera here assigned to the Mattheviidae, and between Echinochiton Pojeta, Eernisse, Hoare, & Henderson, 2003, and mattheviids. Th e results suggest treating these taxa as stem-lineage chitons, and do not support the hypothesis that they are aplacophorans.
INTRODUCTION ing Gasconade Dolomite. Th e Gunter Sandstone Member is Th is study deals with species belonging to Chelodes Davidson essentially equal to the Van Buren Formation of various au- & King, 1874 (= Preacanthochiton Bergenhayn, 1960; thors (Th ompson, 1991). Eochelodes Marek, 1962), Hemithecella Ulrich & Bridge 1941 All of our specimens are small, disarticulated, and pre- (= Elongata Stinchcomb & Darrough, 1995), Aulochiton n. served in white chert; most are molds, but some have the shell gen., and Paradycheia n. gen. All of the new material is from replaced by chert. Th e chert occurs as masses formed by si- the Upper Cambrian Eminence Dolomite of east-central licifi cation of coarse bioclastic molluscan grainstones that ac- Missouri, USA. (Text-fi g. 1). In this area, we use the stra- cumulated between domal stromatolites. Th e fauna collected tigraphy that was presented by Pojeta et al. (2005), in which with these multivalved mollusks is sparse (Pl. 1, Figs 1-7) and the Eminence Dolomite is Late Cambrian in age, as is the consists of (a) internal molds of indeterminate gastropods, (b) Gunter Sandstone Member the lowest member of the overly- the orthid brachiopod Finkelnburgia missouriensis Ulrich & 2 Bulletins of American Paleontology, No.
MATERIAL AND METHODS In the fi eld, striking cherts with geology hammers results in sharp-edged fragments that "fl y" through the air and can pierce skin and/or damage eyes. Th us, care must be taken when collecting specimens preserved as chert. Appropriate clothing and eye protection are necessary. Part and counterpart of the same specimen are infre- quently collected, largely because either the internal or exter- nal portion shatters when the chert is struck with a hammer. Laboratory preparation is largely limited to using rock saws to reduce the amount of adhering rock and nipping the edges of the rock matrix with large steel wire cutters; these proce- dures expose more of the specimens and make them easier to photograph. For this study, we had 205 disarticulated specimens of multivalved mollusks. One hundred twenty-seven of the spec- imens were collected by Darrough over a period of 20 years; he also prepared these specimens. Darrough's collections were supplemented by 82 type, fi gured, or studied specimens used by Bergenhayn (1960) and Stinchcomb & Darrough (1995) in their monographs. Th e specimens studied by Bergenhayn (1960) were collected by E. O. Ulrich and his fi eld parties in the early part of the 20th century. Pojeta spent a week with Darrough in the fi eld and in Darrough's laboratory.
COLLECTING LOCALITIES Darrough's collections came from three localities in Washington County, east-central Missouri, U.S.A (Text-fi g. 1). All collections are from the Eminence Dolomite. Th e lo- cality numbers are Darrough fi eld numbers: Locality 1: A road cut north of Potosi on State Highway Route 21, 3 km (1.9 mi) from the junction with State Highway Route 8; Sec. 1, T. 37 N., R. 2 E. Locality 2: A railroad cut north of Potosi where State Highway Route 21 crosses railroad tracks, 5.4 km (3.5 mi) from the junction with State Highway Route 8; Sec. 25, T. 38 N, R. 2 E. Locality 3: A road cut near Palmer and Berryman. Follow State Highway Route 8 west for 27 km (17 mi) from its junc- tion with State Highway Route 21 in Potosi. Turn south (left) Text-fi g. 1.—Outline map of the state of Missouri, USA, showing on Pigeons Roost Road (Machell Hollow) for 4.8 km (3 mi) the counties. Enlargement of Washington County shows the loca- and turn west (right) onto service road to exposures on both tions of our localities 1-3. Some of Ulrich's Missouri localities are in sides of the road; Sec. 33, T. 37 N., R. 1 W. the general area of east-central Missouri, and some of them are near Localities 1 and 2 can be plotted on the U.S. Geological Potosi and Palmer/Berryman. However, the information provided Survey Potosi 1:24,000 scale Topographic Map. Locality 3 can in the Ulrich locality register is not suffi cient to plot the exact places be plotted on the U.S. Geological Survey Berryman 1:24,000 from which his specimens came. scale Topographic Map.
Cooper (1936, 1938), identifi ed by J. T. Dutro, Jr. (in litt., Ulrich Collecting Localities July 2007), (c) indeterminate trilobite fragments, examined Collections made by E. O. Ulrich and his fi eld parties between by John F. Taylor (in litt., July 2007), (d) and a poorly pre- 1905 and 1924, including some from east-central Missouri, served probable rostroconch. POJETA ET AL.: UPPER CAMBRIAN CHITONS 3 are listed in the Ulrich locality register at the U. S. National 1923)." Our localities 1 and 2 are near Potosi, Missouri. Museum of Natural History (USNM). Th e information is Ulrich locality 453n: "Ozarkian-Gasconade. Top of hill 2 here quoted from the register. Th is information is variable in miles south of Centerville. On Centerville-Ellington Road, clarity, but overall lacks specifi c locality details, and it was not Missouri. (Coll.[ectors] E. O. Ulrich and party, June 1923." possible to relocate Ulrich's exact localities. Collections from Ulrich locality 453y: "Ozarkian-Van Buren[.] Top of Hill, these localities were studied by Bergenhayn (1960). Ulrich 1 mile E. Berryman, Mo. North side of state road. Top beds used the stratigraphic term "Ozarkian" for rocks that are now of Eminence. (Coll. E.O. Ulrich and party, June 1923)." Our regarded as belonging to the Upper Cambrian and part of the locality 3 is near Berryman. Lower Ordovician. Ulrich locality 455v: "Ozarkian-Near Top Eminence[.] Ulrich locality 101-v: "Ozarkian-Upper Gasconade. 1¼ Fossiliferous chert from near top of Eminence formation. mi.[les] north of Decaturvill[e], Mo. [Missouri], Hahatonka Near top of south slope of hill 4¾ to 5 mi[les] south of Potosi, Spgs. [Springs]. E. O. Ulrich's N. B. [notebook], p. 56. Colls. Mo.[Missouri]. Has many trilobites. (Colls. E. O. Ulrich, C. E. O. Ulrich and H. E. Dickhaut, Sept. 20, 1905." L. Dake, and Martin, June 30, 1924)." Our localities 1 and 2 Ulrich locality 217: Information about this locality is are near Potosi. unclear. Th e locality register reads: "Cambrian-Mendota Ulrich locality 479a: "Lower Ozarkian[.] South side of hill Dolomite Member of St. Lawrence. [Th e word Cambrian has north of Mrs. Josie Hall's house. Whitehall, N.[ew] Y.[ork]. been crossed out in pencil and Middle Ozarkian was written (Coll. C. D. Walcott)." in its place in what is probably Ulrich's handwriting. Also the words "Member of St. Lawrence" have been crossed out OVERVIEW OF CAMBRIAN CHITONS in pencil.] Silurian? (Lower) [these words have been crossed Among shelly fossils, multivalved mollusks such as chitons are out in pencil]. L.C.W, 1883) (C.C., 1884) (J.J.? 1889) [prob- poorly known; this is especially the case in rocks of Cambrian ably initials of collectors between 1883 and 1889] Lower and Ordovician age. In addition, some Cambrian taxa as- Magnesium, Whitfi eld. Eikey quarry, six miles E. From signed to the Polyplacophora have been questioned. As re- Baraboo, Sauk County, Wisconsin, in town of Greenfi eld? cently as 2006, Sirenko (2006: 30) expressed doubt about the Sec. 25, T. 12, R. 7 E., northern portion on south fl ank of presence of chitons in Cambrian rocks, when he wrote: "Th e Northern quartzite range (See Geol. Wis., Vol. II, pp. 594, fi rst chitons probably appeared at the end of the Cambrian. I 595). Th is stuff seems to be combined with Ulrich's collec- will not discuss here the insuffi ciently substantiated or doubt- tion at the same locality." A handwritten label with the col- ful placement of Cambrian (Preacanthochiton in Bergenhayn, lection reads: "217=Mendota Dol.[omite] Eikey's quarry, 6 1960, ... and Matthevia in Runnegar et al., 1979 ...) remains mi[les] E. Of Baraboo, Wisc[onsin]." Runnegar et al. (1979: of animals in the class Polyplacophora." 1391) noted that the Mendota Dolomite is now the Black A few recent studies have begun to show that there is more Earth Member of the St. Lawrence Formation of early Late species diversity and morphological disparity of chitons and Cambrian age. other multivalved mollusks in Cambrian and Ordovician Ulrich locality 261n1: "Ozarkian-Eminence. Fossils in re- rocks than had previously been thought (Stinchcomb & sidual chert in dolomite in depressions on surface of porphyry Darrough, 1995; Pojeta et al., 2003; Vendrasco & Runnegar, hill on which Jerktail mine is situated. Evidently rather low 2004; Hoare & Pojeta, 2006; Pojeta & Derby, 2007; Pojeta & in the Eminence formation. Mo. [Missouri]. Ulrich and Lee Dufoe, 2008; and herein). Aug. 4, 1912; Nb. 11[notebook 11], p. 76. Jerktail mine Loc. To date, almost all Late Cambrian multivalved mollusks are [locality] N.E. ¼ sec. 5, T. 29, N. R. 3 W on the head of a known only from North America; among these are the oldest small tributary of Th ompson Creek, 6 mi. [miles] due N. E. known undoubted polyplacophorans. One species, Chelodes Eminence, Missouri." whitehousei Runnegar, Pojeta, Taylor, & Collins, 1979 (Hoare Ulrich locality 399c: "Trilobite beds in Lower Eminence, & Pojeta, 2006), from Datsonian age rocks of the Ninmaroo Little Indian Creek 6 to 7 miles south of St. Clair, Mo. Formation in Queensland, Australia, is the only known non- [Missouri], 2½ miles above mouth creek. Collectors Ulrich North American Late Cambrian chiton (Shergold, 1996). To and St. Clair, 1915." Th e word "south" has been underlined date, no Cambrian articulated specimens have been found. and someone entered the following: "Th is is, no doubt, a mis- Yu Wen (1984, 1987, 1993) reported and described pu- take. I am quite sure it should be north." tative small polyplacophorans from the Lower Cambrian Ulrich locality 453L: "Ozarkian-Eminence[.] Small tri- Meishucun Stage of China. Yi & Bengtson (1989) were criti- lobites and gastropods, Eminence chert or formation Potosi cal of Yu Wen's report. Th ey assigned some of Yu Wen's taxa quad.[rangle], Mo.[Missouri]. N.E. ¼ N.E. ¼ sec. 17, 37, 2 to the paracarinachitids, which is an Early Cambrian sclero- E. About 3 miles E. of Potosi, Mo. (Coll. E. O. Ulrich, May, tome group "having elongated sclerites, transversely divided 4 Bulletins of American Paleontology, No. into distinct segments with spines along the crest" (Yi & Orthriochiton Vendrasco & Runnegar, 2004, might be- Bengtson, 1989: 47). Some other of Yu Wen's taxa were as- long to Group A. However, the apices of the specimens of signed by Yi & Bengtson (1989: 102) to their Early Cambrian the known species are too poorly preserved to determine the Ocruranus-Eohalobia group, which consists of smooth shells extent of the apical area. with subapical fi elds or reentrants and that superfi cially re- Group B: Chiton taxa in this group lack lacunae and have semble brachiopods or univalved mollusks. a small apical area. In Late Cambrian time, Group B is repre- Runnegar (1996: 82) summarized thoughts about Yu sented by the single polyplacophoran Aulochiton n. gen. Wen's putative Early Cambrian polyplacophroans, and came Group C: Taxa in this group have been newly discovered to the conclusion that "the earliest known chitons are ... spe- (Pojeta & Derby, 2007). Th e group has two shallow median cies of Matthevia from the latest Cambrian of the United States lacunae in the valves, the intermediate valves gape anterior- ..." However, Yu Wen (2001) strongly reiterated his support ly, posteriorly and ventrally, and lack apical areas. Group C for the occurrence of small polyplacophroans in the Lower contains the two genera Dycheia Pojeta & Derby, 2007, and Cambrian rocks of China. Vendrasco et al. (2008b) supported Paradycheia n. gen. Group C is placed with question in the Yu Wen's interpretation of the Ocruranus-Eohalobia group as Polyplacophora. possibly being the earliest known polyplacophorans. Until Bergenhayn (1960) assigned one Upper Cambrian valve the interpretation of the affi nities of the Ocruranus-Eohalobia from New York to Priscochiton? Dall, 1882. Th is specimen is group is resolved, the Late Cambrian taxa treated in this study too poorly preserved to determine its systematic position (Pl. are the oldest known undoubted chitons. 1, Figs 8-9). Yates et al. (1992) regarded Triplicatella disdoma Conway Of the eight genera of Upper Cambrian multivalved mol- Morris, 1990, as the earliest known chiton; the species oc- lusks recognized here, four also are known from Ordovician curs in the Lower Cambrian Parara and Ajax Limestones of rocks: Matthevia (Pojeta et al., 2005), Hemithecella (Stinch- South Australia. Skovsted et al. (2004) showed that species of comb & Darrough, 1995), Orthriochiton (Hoare & Pojeta, Triplicatella are best interpreted as opercula of hyolithiform 2006), and Chelodes, which ranges into Lower Devonian animals similar to orthothecid hyoliths. To the best of our rocks (Cherns, 1998a; Hoare & Farrell, 2004). knowledge, no putative chitons have been reported from In the area of our study in east-central Missouri, and from Middle Cambrian rocks. Ordovician rocks, Stinchcomb & Darrough (1995) named In Upper Cambrian rocks, there are three presumed prim- the genera Conodia and Robustum. Conodia has not been re- itive groups of multivalved mollusks: studied, but Stinchcomb & Darrough's fi g. 7.1 of C. acumi- Group A: Chitons having one or two deep lacunae ("tun- nata Stinchcomb & Darrough, 1995, looks much like a speci- nels" of authors) and large apical areas on the intermedi- men of Chelodes. Robustum has been restudied by Vendrasco ate valves. In Upper Cambrian rocks, Group A contains & Runnegar (2004); they regarded R. nodum Stinchcomb & four known genera: Matthevia Walcott, 1885; Hemithecella Darrough, 1995, as a morphological link between mattheviids Ulrich & Bridge, 1941 (= Elongata Stinchcomb & Darrough, and septemchitonids. Vendrasco & Runnegar (2004) noted 1995); Chelodes Davidson & King, 1874 (= Preacanthochiton that Robustum has subcylindrical intermediate valves with a Bergenhayn, 1960; Eochelodes Marek, 1962); and [short] posterior lacuna and an upturned anterior lip that they Eukteanochiton Vendrasco & Runnegar, 2004. regarded as a remnant of an anterior lacuna. Although none of these genera is known from Cambrian As shown below, reports of Preacanthochiton from articulated specimens, for a variety of reasons most workers Ordovician rocks (Stinchcomb & Darrough, 1995; Hoare & regard them as eight-valved polyplacophorans (Runnegar Pojeta, 2006) are incorrect. et al., 1979; English, 2002; Vendrasco & Runnegar, 2004; Cherns et al., 2004; Hoare & Pojeta, 2006). An exception to PALEOBIOGEOGRAPHY AND STRATIGRAPHY OF UPPER CAMBRIAN this consensus is Stinchcomb & Darrough (1995) who did POLYPLACOPHORA not accept Hemithecella as a chiton, but regarded it as a multi- Th e oldest-known undoubted chitons occur in Upper valved mollusk with up to 16 valves. Th ey regarded their pro- Cambrian rocks. Chelodes whitehousei (see Hoare & Pojeta, posed genus Elongata (treated as a junior subjective synonym 2006) is known from Upper Cambrian rocks of the Ninmaroo of Hemithecella here) as related to Hemithecella. As noted Formation (Datsonian) of Queensland, Australia (Shergold, below, because of inadequate original illustrations, the genus 1996). All other known Upper Cambrian occurrences are in Preacanthochiton Bergenhayn, 1960, has been misunderstood; North America (Text-fi g. 2). In point of fact, more is known it is here included in Group A as a junior subjective synonym about both Cambrian and Ordovician species of chitons from of Chelodes. Van Belle (1975a) synonymized Eochelodes Marek, North America than from all the other parts of the world. 1962, with Chelodes; we agree with Van Belle. Th e six known genera of Late Cambrian polyplacophorans are POJETA ET AL.: UPPER CAMBRIAN CHITONS 5 morphologically diverse, suggesting a rapid early evolution in et al., 1979; Stinchcomb & Darrough, 1995; English, 2002; Laurentia (Vendrasco & Runnegar, 2004). Vendrasco & Runnegar, 2004; and herein). Matthevia is the Th e fossil records of molluscan lineages such as monopla- most widespread, occurring in all of the states mentioned cophorans, pelecypods (bivalves), rostroconchs, and gas-tro- except Missouri, Oklahoma, and Wisconsin. In addition, pods extend back to the Early Cambrian. In addition, some Late Cambrian multivalved mollusks such as Dycheia and Early Cambrian Problematica, such as halkieriids, have been Paradycheia n. gen., which might be related to chitons, are regarded as closely related to chitons (Vinther & Nielson, known from Missouri and Oklahoma (Pojeta & Derby, 2007, 2005). Caron et al. (2006) discussed and redefi ned the con- and herein). cept of the large, Middle Cambrian, Burgess Shale species Ordovician chitons are also widespread in North America, Odontogriphus omalus Conway Morris, 1976. Odontogriphus occurring in Alabama, Iowa, Kentucky, Minnesota, Missouri, omalus looks much like a shell-less chiton, as does the Ediacaran New Mexico, Oklahoma, Ontario, Virginia, and Wisconsin fossil Kimberella Glaessner & Wade, 1966. Kimberella was in- (Billings, 1865; Smith & Hoare, 1987; Stinchcomb & Dar- terpreted as a mollusk or mollusk-like animal by Fedonkin rough, 1995; Pojeta et al., 2003, 2005; Cherns et al., 2004; & Waggoner (1997) and Fedonkin et al. (2007). Butterfi eld Hoare & Pojeta, 2006; Pojeta & DuFoe, 2008). However, (2008: 543) described what he regarded as Early Cambrian outside of North America, Ordovician chitons are more "pock-marked cuticular fragments interpreted as having sup- widespread than those known from the Cambrian, and oc- ported multiple biomineralized sclerites/spines in the manner cur in Th e Czech Republic, Scotland, Sweden, ailand,Th and of a modern chiton girdle." Th us, it is possible that polypla- perhaps Australia (Bergenhayn, 1955; Stait & Burrett, 1984; cophorans have an older, but as yet largely unknown, deeper Smith & Hoare, 1987; Cherns et al., 2004). Th e diversifi ca- geological history. Current information suggests that the fi rst tion of chitons that began in Late Cambrian time continued diversifi cation of chitons occurred along the coastlines of throughout the Early and Middle Ordovician (Cherns et al., Laurentia during Late Cambrian and Ordovician time, and 2004; Hoare & Pojeta, 2006). this could be where the Polyplacophora evolved. Late Cambrian chitons commonly occur with stromato- In North America, Late Cambrian chitons are known from lites and oncolites (Yochelson, 1966; Runnegar et al., 1979; Arizona, California, Maryland, Missouri, Nevada, New York, Stinchcomb & Darrough, 1995; Cherns et al., 2004; and Oklahoma, Texas, Utah, Wisconsin, and perhaps Alberta herein). Th ese stromatolites tend to be unbroken and un- (Yochelson et al., 1965; Yochelson & Taylor, 1974; Runnegar abraded (Kouchinsky, 2001), suggesting that the earliest chi-
Text-fi g. 2.—Paleobiogeographic distribution of chitons in Late Cambrian time. Continental reconstruction after the Late Cambrian map (514 Ma), PALEOMAP project, Christopher R. Scotese, 2001. 6 Bulletins of American Paleontology, No. tons lived on or around the stromatolite heads and fed on defi nition of the base of the Ordovician (Cooper et al., 2001), the microbes that formed those structures (Taylor & Halley, concluded that in east-central Missouri, the Cambrian- 1974; Runnegar et al., 1979). Th e chitons are disarticulated Ordovician boundary lies somewhere in the lower Gasconade and occur as molluscan grainstones between domal stroma- Dolomite, above the Eminence Dolomite. tolites, suggesting that the sediments were deposited in inter- Th e distribution of fossil chitons in four Cambrian-Lower tidal to shallow-subtidal marine environments, like those in Ordovician stratigraphic sections across North America is which extant chitons are most common and abundant. shown in Text-fi g. 3. Th ese sections preserve the oldest known Th is stromatolitic ecosystem, which is heavily dominated chitons from the Rasettia magna trilobite Subzone, and reveal by cyanobacterial and algal structures, was widespread in Late that chitons appear in the fossil record with a relatively high Cambrian and Early Ordovician time in Laurentia. During diversity, which is best known in Missouri. the Phanerozoic, the time from Middle Cambrian through Early Ordovician was one when reefs were dominated by SYSTEMATIC PALEONTOLOGY microbialites. Rowland & Shapiro (2002) discounted three hypotheses advanced to explain this event: extinction of reef- INTRODUCTION building metazoans, lack of photosynthesis, and lack of graz- At almost all levels of classifi cation, the systematics of ing animals. In their view, this left four possible explanations, Cambrian and Ordovician chitons is poorly understood. all of which probably contributed to microbialite domination Many names have been proposed, but the concepts behind during this time interval: nutrient defi ciency due to low terri- the names are often vague. Th is is particularly the case with a genous runoff , high levels of atmospheric CO2, global warm- number of the 14 family names proposed before 2008; here ing, and a drop in the Mg/Ca ratio (Rowland & Shapiro, we review, discuss, and, where possible, provide concepts for 2002). Also, during Late Cambrian time, Laurentia under- the families (Appendix I). As an example, the Late Cambrian went widespread transgression (Seslavinsky & Maidanskaya, family Preacanthochitonidae Bergenhayn, 1960, has been 2001). It is not clear why these conditions might have favored widely cited in the literature. However, Bergenhayn misin- chitons. By Middle Ordovician time, chitons are found in terpreted and misunderstood the holotype of the type species additional environments, where they form minor elements Preacanthochiton cooperi Bergenhayn, 1960, of the type genus in shell beds deposited in nearshore subtidal, lagoonal, and Preacanthochiton Bergenhayn, 1960. Th us, subsequent au- open-marine facies (Hoare & Pojeta, 2006: 24). thors have not had a conceptual base for understanding these Th e specimens used in this study are all derived from taxa. In fact, Sirenko (2006) doubted that there were any Late the Upper Cambrian Eminence Dolomite in Missouri. Th is Cambrian chiton fossils. formation has yielded numerous mollusks, including gastro- In this study, we had a large number of specimens and pods, monoplacophorans (Stinchcomb, 1975, 1980, 1986; have provided useful morphological defi nitions for taxa at all Yochelson & Stinchcomb, 1987), and multivalved mollusks levels. Other major contributions of this study include (a) a (Stinchcomb & Darrough, 1995), most of which are likely large increase in the knowledge and revision of the systemat- chitons (Vendrasco & Runnegar, 2004; and herein). ics of poorly understood early multivalved mollusks, (b) the Bridge (1930) noted two distinct faunal zones within the highly likely interpretation that esthete sense organs penetrat- Eminence Dolomite: a lower zone characterized by the tri- ing the shell of chitons (Moseley, 1885; Baxter & Jones, 1981, lobites Calvinella Walcott, 1914, and Triarthropsis Ulrich, 1984; Fernandez et al., 2007; Vendrasco et al., 2008a) were 1930, and the gastropod Matherella Walcott, 1912, and present in Late Cambrian time, at the beginning of the known an upper zone characterized by the trilobites Plethopeltis fossil record of the class Polyplacophora; and (c) criteria for Raymond, 1913, and Entomaspis Ulrich, 1930. According to determining the anterior and posterior ends (and thus right Stinchcomb (1986: 607), "Th ese two zones are generally rec- and left sides) in unfamiliar disarticulated valves of Cambrian ognized in most fossiliferous outcrop areas;" however, earlier, multivalved mollusks. Stinchcomb (1965: 4) had questioned these as distinct zones. Specimens of the fossils used in this study are reposited Bridge (1930) also noted that biostratigraphy is diffi cult for at the Department of Paleobiology, United States National the Eminence Dolomite because most specimens were collect- Museum of Natural History (USNM), Smithsonian ed as "fl oat" material and very few were in situ; this point was Institution, Washington, DC, USA. echoed by Stitt (1977: 28-29). Stitt also questioned the iden- tifi cation of Plethopeltis. Stinchcomb (1986) considered the Phylum MOLLUSCA Cuvier, 1797 Cambrian-Ordovician boundary to lie within the Eminence Class POLYPLACOPHORA de Blainville, 1816 Dolomite. Pojeta et al. (2005) summarized conodont and Diagnosis.—Mollusks having a head, elongated body, and trilobite biostratigraphic data, and based on the most recent a dorsal shell consisting of a longitudinal series of eight articu- POJETA ET AL.: UPPER CAMBRIAN CHITONS 7
Text-fi g. 3.—Stratigraphic occurrences of chitons in four Upper Cambrian-Lower Ordovician stratigraphic sections across North America. Utah stratigraphy derived from Vendrasco & Runnegar (2004); Oklahoma stratigraphy derived from Stitt (1977); Missouri stratigraphy derived from Pojeta et al. (2005); New York stratigraphy derived from Taylor & Halley (1974). lated valves; valves embedded in and surrounded by a thick In extant polyplacophorans, below the organic shell layer, muscular mantle border (girdle) that is covered by pellicle and it is widely agreed that the outermost calcareous shell layer be in which occur a variety of calcareous spicules, scales, spines, called the tegmentum, a term coined by Middendorf (1848). and/or hair-like structures. Th e tegmentum contains large numbers of both horizon- Occurrence.—Upper Cambrian to Holocene. tal and vertical canals for the passage of sense organs known as Chiton shell microstructure.—Ordinal-level taxa of chi- esthetes, or aesthetes (Greek, for "one who perceives," Neilson tons are defi ned on the basis of the shell layers in the valves. et al., 1951: 42). However, the nomenclature of the shell layers in extant chi- In extant chitons, for the calcareous shell layer below the tons is not fully agreed upon. In general, there is agreement tegmentum, Middendorf (1848) used the term articulamen- that, as in other mollusks, a thin outer organic layer is pres- tum. Knorre (1925) proposed the term hypostracum for the ent. Haas (1981: 403) noted that the organic layer is very ventralmost layer of the valve and restricted the term articu- thin and incompletely polymerized and that it is not a true lamentum to the shell layer between the tegmentum and hy- periostracum; he termed it a properiostracum. Baxter & Jones postracum. Th is naming scheme recognizes three calcareous (1981: 65; 1984: 550) used the term periostracum for this shell layers, the dorsalmost tegmentum, below which is the ar- outer organic layer. Eernisse & Reynolds (1994: 63) noted "... ticulamentum, and the ventralmost layer is the hypostracum. Specialization of epithelial cellular surfaces proximal to the Baxter & Jones (1981: 67; 1984: 550) used Knorre's (1925) chiton shell plate [valve] indicates the presence of the 'pro- shell layer naming scheme; they noted that all of the calcare- periostracal groove,' a more primitive [than conchiferans], ous shell layers are composed of aragonite and that they are but homologous, type of shell-secreting system that produces easily distinguished. Hoare et al. (1972: 676) used the terms an organic, weakly polymerized pellicle or 'properiostracum' tegmentum and hypostracum for the outermost and inner- whose protein component might have some limited propen- most shell layers of Pennsylvanian age chitons from Ohio. sity for later action by tanning agents." Haas (1981: 403) recognized two calcareous shell layers 8 Bulletins of American Paleontology, No. in extant chiton valves, a dorsal tegmentum and a ventral hy- Ulrich & Bridge, 1941 (= Elongata Stinchcomb & Darrough, postracum. He regarded the hypostracum as being modifi ed 1995); Robustum Stinchcomb & Darrough, 1995; Spicuchelodes into the articulamentum where it projects from underneath Cherns, 1998a. the tegmentum as the sutural laminae and insertion plates and Discussion.—We use the term lacunae for the deep holes in into the myostracum at muscle attachments sites. Haas (1981: the intermediate and tail valves of mattheviids rather than call- 403-405) noted the occurrence of diff erent crystal structures ing them tunnels (Stinchcomb & Darrough, 1995; Vendrasco for the calcareous shell layers, as did Carter & Hall (1990). & Runnegar, 2004). Th e term tunnel implies something open Eernisse & Reynolds (1994) described and used Haas' nam- at each end, whereas the lacunae end blindly in the inside of ing scheme for shell layers in their review of chiton biology, the valves (Runnegar et al., 1979: pl. 1, fi g. 28; English, 2002: and we adopt this scheme here. fi g. 2.11) and are only open to the outside at one end. Matthevia is the only genus that clearly has two lacunae. Order PALEOLORICATA Bergenhayn, 1955 Eukteanochiton has one lacuna, above which there is a median Diagnosis.—Polyplacophorans having valves that lack the groove that might be a relict second lacuna or the result of sutural laminae and insertion plates that are characteristic of abrasion of a second lacuna; Robustum has a posterior lacuna neoloricates. Th us, it is presumed that paleoloricates have and can have a relict anterior lacuna (Vendrasco & Runnegar, valves composed of tegmentum and hypostracum but lack the 2004). Th e other genera have a single lacuna. Most workers articulamentum subdivision of the hypostracum. regard the lacunae as having possessed structures that attached Occurrence.—Upper Cambrian to Lower Devonian. the valves to the soft body. Whatever their function, the pres- Discussion.—Paleoloricata is a convenient morphological ence of one or two lacunae is not regarded as a suffi cient mor- grouping. However, preservation of lower Paleozoic chitons phological criterion for separating families, especially because often makes it diffi cult to determine whether or not the su- some genera can have relict lacunae. Mattheviidae is the old- tural laminae and insertion plates are present. In addition, est name available for this grouping and it is used here. Vendrasco & Runnegar (2004: 681) noted that paleoloricates Valve orientation.—Th e standard convention for descrip- are largely "defi ned by a primitive character state—lack of the tions of disarticulated mattheviid intermediate valves treats articulamentum. ... So the Paleoloricata is probably paraphyl- the side with the apical area as ventral (Cherns, 1998a: pl. 1, etic and this name should be used more as a descriptive term fi g. 1b; Vendrasco & Runnegar, 2004: fi g. 8.1). Th is makes than as a taxonomic name that implies monophyly." However, the jugal side of the valves dorsal. Th e open end of the valve in the present state of knowledge of lower Paleozoic chitons, having the openings of the lacunae is designated anterior and we see Paleoloricata as a useful morphological grouping. the closed apex becomes posterior (Text-fi gs 4.1-4.4). Runnegar et al. (1979: pl. 2, fi gs 1-10) followed the stan- Family MATTHEVIIDAE Walcott, 1885 dard convention when describing the intermediate valves of Chelodes whitehousei; however, when describing intermediate Mattheviidae Walcott, 1885: 17. and tail valves of Matthevia variabilis Walcott, 1885, they re- Chelodidae Bergenhayn, 1943: 298. garded the open end of the valve having the openings of the Preacanthochitonidae Bergenhayn, 1960: 169. lacunae as ventral, the jugal side of the valve as anterior, the Eochelodidae Marek, 1962: 373. part of the valve having the apical area as posterior (Runnegar Hemithecellidae Stinchcomb & Darrough, 1995: 58. et al., 1979: pl. 1, fi gs 9-13; Text-fi gs 4.5-4.8), and the apex Robustidae Stinchcomb & Darrough, 1995: 63. of the valve as dorsal. Th is diff erence in valve orientation nomenclature is explained by the way the two species were Type genus.—Matthevia Walcott, 1885, by monotypy. reconstructed. For C. whitehousei, they showed the valves as Diagnosis.—Paleoloricates having elongate, thick interme- overlapping (Runnegar et al., 1979: pl. 2, fi g. 29) whereas for diate and tail valves and an apical area that covers at least the M. variabilis, they showed a reconstruction having the inter- posterior one-third of the median length of the valves, one or mediate and tail valves erect (1979: 1375). two elongate conical lacunae (= "tunnels" of authors) present. Because Late Cambrian mattheviids are known only from Jugal angle can be narrow and raised or low and fl at. Where disarticulated valves, it is not clear how the elongate pointed known, the head valve varies from fl at and thin to thick and intermediate and tail valves were positioned on the body. Th e arched. Large circumsomal spines lacking. positioning of the valves determines what parts of the valves Occurrence.—Upper Cambrian to Lower Devonian. to call dorsal, ventral, anterior, and posterior. As reconstruct- Genera included—Matthevia Walcott, 1885; Calceochiton ed by Runnegar et al. (1979: 1375), Matthevia variabilis has Flower, 1968; Chelodes Davidson & King, 1874 (= Preacantho- almost erect intermediate and tail valves that slope slightly in chiton Bergenhayn, 1960, and Eochelodes Marek, 1962); the posterior direction of the body, and which do not overlap. Eukteanochiton Vendrasco & Runnegar, 2004; Hemithecella POJETA ET AL.: UPPER CAMBRIAN CHITONS 9
Th is reconstruction places the apices in the dorsalmost part Genus CHELODES Davidson & King, 1874 of each valve, makes the lacunate end ventralmost, and places the apical area in a posterior position. English's (2002: 296) Chelodes Davidson & King, 1874: 167. reconstruction is similar to that of Runnegar et al.'s in depict- Preacanthochiton Bergenhayn, 1960: 169. ing a subvertical orientation of the valves. Eochelodes Marek, 1962: 373. Th is type of reconstruction is supported by a largely artic- ulated specimen of the Ordovician species Echinochiton dufoei Type species.—Chelodes bergmani Davidson & King, 1874, Pojeta et al., 2003. Echinochiton dufoei is known from several by monotypy; Middle Silurian (Wenlockian) of Gotland, partially and one completely articulated specimen (Pojeta & Sweden. DuFoe, 2008) that are preserved parallel to bedding. However, Diagnosis.—Here we use Cherns' (1998a: 548) diagnosis one specimen preserves four articulated valves so that they can of Chelodes: "Wedge-shaped to cordate, arched intermediate be seen in lateral view (Text-fi g. 4.9) and the specimen pre- sclerites [valves] with posterior apex; becoming elongate, mas- serves the fi lling of the body space below the valves (Pojeta et sive. Ventral apical area fl attened, up to more than half the al., 2003: 652; Pojeta & DuFoe, 2008). Th is specimen has length [and having a broadly triangular single lacuna], anteri- the apices of the valves dorsalmost; the lacunate end is ven- or rim elevated slightly above smooth ventral surface. Growth tralmost; the apical area is posterior; the valves do not overlap; lines across dorsal surface and ventral apical area sometimes and the specimen looks much like the Runnegar et al., 1979, with ridged and granulate dorsal ornament. Shell fi elds lack- and English, 2002, reconstructions of Matthevia variabilis. ing to well-defi ned." Th is specimen suggests that the valves were attached to the Occurrence.—Late Cambrian to Early Devonian, world- body in the anatomically most effi cient position for clamping wide. the valves onto the body. Comparisons.—In mattheviids having a single lacuna, In describing the valves of Matthevia wahwahensis Calceochiton Flower, 1968, has intermediate valves that dif- Vendrasco & Runnegar (2004: 680) used the conventional fer from those of Chelodes in their posterior taper and apical description and treated the apex as posterior, the lacunate end angle (Hoare & Pojeta, 2006: fi g. 5). Hemithecella has a long as anterior, the part of the valve having the apical area as ven- narrow lacuna above the apical area of intermediate valves tral, and the jugal side of the valve as dorsal. Th eir reconstruc- (Pl. 4, Figs 6-9; Pl. 5, Figs 1, 3-4; Pl. 11, Figs 1-2), whereas tion of M. wahwahensis (Vendrasco & Runnegar, 2004: fi gs Chelodes has a broad triangular lacuna (Pl. 2, Figs 4-5, 12- 4, 9.31-32) shows the intermediate and tail valves as recum- 13; Pl. 3, Figs 1-3). Spicuchelodes diff ers from Chelodes in its bent and overlapping and not erect. In this reconstruction, sharply triangular apical area on intermediate valves (Cherns, the apices of the valves remain posterior, the lacunate end be- 1998a: pl. 7; Hoare & Pojeta, 2006: fi g. 3). Here, Elongata comes anteroventral, the jugal side becomes anterodorsal, and is regarded as a junior subjective synonym of Hemithecella, the apical area becomes posteroventral; this reconstruction is and Preacanthochiton and Eochelodes are junior subjective syn- somewhat at odds with the descriptive nomenclature applied onyms of Chelodes; we agree with Van Belle (1975a) who re- to the valves by Vendrasco & Runnegar (2004: 680). garded Eochelodes Marek, 1962, as a synonym of Chelodes. Th e reconstruction of Eukteanochiton milleri Vendrasco & Runnegar, 2004 (p. 678) is more like the Runnegar et al. PREACANTHOCHITON Bergenhayn, (1979) reconstruction of Matthevia variabilis than it is like the Type species.—Preacanthochiton cooperi Bergenhayn, 1960, reconstruction of M. wahwahensis. Vendrasco & Runnegar by original designation. (2004: 676) illustrated various other reconstructions of mat- Discussion.—Bergenhayn's defi nition of Preacanthochiton theviids by other authors; except for Yochelson (1966), these (1960: 169) was "Sculpture consisting of small rounded pus- other reconstructions are variations on the possibilities noted tules. Circumference of body valves heart-shaped or rectan- above. gular with rounded lateral margins. Covering total or par- Because there are no known Late Cambrian articulated tial." None of the fi gured specimens of the species assigned mattheviids, and because at least two reconstructions are to Preacanthochiton by Bergenhayn show "pustules" (Pl. 2, possible, the decision of how to describe the valves is an ar- Figs1-3; Pl. 3, Figs 10-11; Pl. 4, Figs 1-5). bitrary convention. Both Cherns (1998a) and Vendrasco & Bergenhayn's illustrations showing pustules (1960: text- Runnegar (2004) used the descriptive convention of having fi gs 1.1, 1.3-1.4, 1.6; Text-fi g. 5 here) are artistic creations that the apex posterior and the apical area ventral in position. For place the seemingly pustulose surface of what is here the new the sake of consistency, this convention is also followed here genus Aulochiton on what Bergenhayn called Preacanthochiton. for mattheviids (Text-fi gs 4.1-4.4). Bergenhayn was the fi rst person to study Eminence Dolomite multivalved mollusks and it not surprising that he confused taxa. 10 Bulletins of American Paleontology, No. POJETA ET AL.: UPPER CAMBRIAN CHITONS 11
In addition to the lack of pustules, Bergenhayn's illustra- 1-3) is a relatively large, poorly preserved internal mold of tion of the holotype (USNM 137370) of Preacanthochiton a tail valve (USNM 137373) and it is the only specimen of cooperi (1960: text-fi g. 1.2) is incorrect; the holotype is the the taxon fi gured by Bergenhayn (1960: text-fi gs 1.7-1.8). It only specimen of the species that he fi gured. For reasons that shows a median jugal ridge and perhaps a remnant of the mu- cannot now be understood, someone brushed white enam- cro; the lateral sides diverge slightly in the anterior direction. el paint around the holotype, which obscured some of the Th e broken anterior and posterior ends, and the white enamel specimen; the paint cannot be removed without damaging the paint on the posterior end, make it diffi cult to determine the specimen. Th e holotype is illustrated here (Pl. 2, Figs 1-3) and dorsal profi le. Chiton tail valves occasionally have a median it diff ers signifi cantly from Bergenhayn's illustration (1960: jugal ridge. In addition to the holotype, Bergenhayn (1960: text-fi g. 1.2; Text-fi g. 5.2, here). Th e holotype and paratypes 173) assigned four other specimens to P. productus, three of (Pl. 2, Figs 4-6; Pl. 3, Fig. 6) of P. cooperi fi t the concept of which are still available for study. Two of these three paratypes the genus Chelodes, and the generic name Preacanthochiton is are unidentifi able fragments (USNM 537325a, 537325b). here regarded as a subjective junior synonym of Chelodes. Th e Th e third paratype (USNM 537325) is a tail valve that is type species of Preacanthochiton is redescribed here as Chelodes incomplete posteriorly (Pl. 12, Fig. 7), the lateral margins cooperi (Bergenhayn, 1960). diverging slightly in an anterior direction, but it lacks a mid- In addition to the type species, Bergenhayn (1960) named jugal ridge. Preacanthochiton depressus, P. productus, and P. n. sp. aff . P. Hoare & Pojeta (2006) reassigned Preacanthochiton pro- cooperi (Text-fi gs 5.4-5.10). Th e holotype of the Ordovician ductus to the genus Listrochiton Hoare & Pojeta, 2006. Th e taxon P. depressus (Pl. 3, Figs 10-11) consists of part and coun- illustrations presented here of the holotype of P. productus terpart internal and external molds (USNM 137372) that are show its poor preservation; the tail valve of the type species poorly preserved and triangular in shape; the part is the only of Listrochiton, L. weiri Hoare & Pojeta, 2006, also lacks the specimen of the taxon fi gured by Bergenhayn (1960: text-fi gs jugal ridge. Th us, the concept implied by the name P. produc- 1.4-1.6). Someone brushed white enamel paint around both tus is unclear but it does not fi t the defi nition of Listrochiton. the part and counterpart. In our opinion, the concept implied by the name P. productus Bergenhayn (1960: 173) assigned a second specimen to is uncertain, and therefore, as of now we recommend that the Preacanthochiton depressus, which he did not fi gure. Th is para- name be restricted to the type suite. type (USNM 325306) is shown here (Pl. 9, Fig. 6). Th e speci- Th e specimen (USNM 137374) of Preacanthochiton n. sp. men is very poorly preserved and quite diff erent in shape than aff . P. cooperi fi gured by Bergenhayn is a small internal mold the holotype. In our opinion, the holotype and paratype are of a tail valve (Pl. 4, Figs 4-5). Th e specimen shows the mucro, taxonomically indeterminate valves, the concept implied by it lacks a median jugal ridge, it is rounded posteriorly, but the the name P. depressus is uncertain, and therefore, as of now we white enamel paint brushed around the specimen obscures recommend that the name be restricted to the type suite. the rest of the dorsal profi le. Th is specimen is here treated as a Th e holotype of Preacanthochiton productus (Pl. 4, Figs tail valve of Aulochiton sannerae n. gen and n. sp.
Text-fi g. 4 (at left).—4.1-4.8. Valve orientation in mattheviids. A specimen of Chelodes sp.(USNM 537352a) illustrating two possibilities of valve orientation in mattheviids. Th e left column is the conventional orientation having the valve in a horizontal position and treating the api- cal area as ventral and the lacunate end as anterior; 4.1, left-lateral view; 4.2, anterior view showing apical area above which is the lacuna; 4.3, dorsal jugal area; 4.4, ventral view showing the apical area and opening to the lacuna. Th e right column treats the valve as erect and places the apical area in a posterior position and the lacunate end in a ventral position; 4.5, left-lateral view; 4.6, ventral view showing apical area above which is the lacuna; 4.7, anterior jugal view; 4.8, posterior apical area. Th e two orientations determine how the length and height of the valves are measured, and the interpretation of the position of the valves in life; however, they do not aff ect width measurements or the perceived right and left sides of the specimen. A, anterior; AP, apical area; D, dorsal; H, height; L, length; P, posterior; V, ventral; W, width. Chelodes sp. (USNM 537352a) is from the Fort Sill Limestone, 284 m (930 ft) above the base of the Chandler Creek Section, in the Rasettia magna Trilobite subzone (Upper Cambrian, Sunwaptan) of the Wichita Mountains, Oklahoma, USA (Stitt, 1977). 4.9. Specimen showing the right side of four valves and the sedimentary fi lling of the body space below the valves of Echinochiton dufoei Pojeta et al., 2003 (USNM 517481), an Ordovician taxon closely allied to mattheviids. Th e valves are erect with the apical area posterior and the lacunate end ventral. Th is supports the right column orientation given in Text-fi gs 4.5-4.8. Nevertheless, for consistency with previous publications herein we use the conventional orientation when describing mattheviids. Th e downward-facing straight-tailed barbed arrow points to a valve having two sedimentary fi llings of the lacunae. The downward-facing, straight-tailed, unbarbed arrow points to the bottom of the sedimentary fi lling of the body space, and possible impressions of spicules. Th e broken wavy arrow at the far left of the image points to a fifth valve that is only partially preserved. Th e solid triangular tailess arrow points to the anteriormost preserved valve. Th e specimen is from Forreston Member, Grand Detour Formation (Middle Ordovician) near Beloit, Wisconsin, USA. Scale bars = 5 mm. 12 Bulletins of American Paleontology, No.
Text-fi g. 5.—Reproductions of Bergenhayn's (1960) fi gures of species of Preacanthochiton. Published with the permission of Th e Paleontological Society. 5.1-5.3. P. cooperi, reconstruction, holotype, drawing of holotype with surface of Aulochiton sannerae superimposed on it (Bergenhayn, 1960: fi gs 1.1-1.3). See Pl. 2, Figs 1-3, for photographs of the holotype, USNM 137370. Herein, this species is assigned to Chelodes as C. coo- peri. Th e holotype is from the Eminence Dolomite at Ulrich locality 453L. 5.4-5.6. P. depressus, reconstruction, holotype, drawing of holotype with surface of A. sannerae superimposed on it (Bergenhayn, 1960: fi gs 1.4-1.6). See Pl. 3, Figs 9-10, for photographs of the holotype, USNM 137372. Herein, it is recommended that the name P. depressus be restricted to the type suite. Th e holotype is from the Gasconade Dolomite (Ordovician) at Ulrich locality 101-v. 5.7-5.8. P. productus, fi gure of holotype and drawing of holotype (Bergenhayn, 1960: fi gs 1.7-1.8). See Pl. 4, Figs 1-3, for photographs of the holotype, USNM 137373. Herein, it is recommended that the name P. productus be restricted to the type suite. Th e holotype is from the Eminence Dolomite at Ulrich locality 261n1. 5.9-5.10. P. n. sp. aff . P. cooperi, fi gure of a specimen and a drawing of the specimen (Bergenhayn, 1960: fi gs 1.9-1.10). See Pl. 4, Figs 4-5, for photographs of the specimen, USNM 137374. Herein, this specimen is regarded as a tail valve of A. sannerae. Th e specimen is from the Eminence Dolomite at Ulrich locality 399c.
Chelodes cooperi (Bergenhayn, 1960) valves are also fi gured. Pl. 2, Figs 1-13; Pl. 3, Figs 1-9 Occurrence.—Upper Cambrian Eminence Dolomite of the state of Missouri, USA. Preacanthochiton cooperi Bergenhayn, 1960: 169, text-fi gs 1.1-1.3; Comparisons—Chelodes cooperi and C. whitehousei Runnegar et al., 1979: 1391, pl. 2, fi g. 64, not pl. 2, fi gs 62-63, Runnegar et al., 1979, are the two known Late Cambrian 65-66. species of Chelodes. Chelodes whitehousei occurs in Datsonian [non] Preacanthochiton Stinchcomb & Darrough, 1995: fi gs 6.2- age rocks of Queensland, Australia. Th e intermediate valves of 6.6. C. cooperi can be distinguished from C. whitehousei by their [non] Preacanthochiton Hoare & Pojeta, 2006: 9. much smaller size, lack of a prominent anterior sinus, and the presence of a broad to well-defi ned medial jugal ridge separat- Diagnosis.—Small Chelodes; intermediate valves fl ared and ing the expanded anterolateral areas. laterally expanded anteriorly; apical area and single lacuna Discusssion.—Th e Early Ordovician specimens of Chelodes broad anteriorly and tapering to point posteriorly; anterior and Preacanthochiton fi gured by Stinchcomb & Darrough edge of apical area scalloped, commonly having median de- (1995: fi gs 6.1-6.6) are not mattheviids and they are prob- pression and lateral projections; jugal ridge present, variable ably not polyplacophorans. None of these fi gures shows the in its strength from marked to subdued. Head and tail valves presence of an apical area and some of them show multiple unknown. paired muscle scars similar to tryblidiid monoplacpohorans Measurements.—See Table 1. (Runnegar, 1987: 289). Vendrasco & Runnegar (2004: Specimens.—Th e holotype is the intermediate valve shown 675) argued that some of these specimens (Stinchcomb & on Pl. 2, Figs 1-3 (USNM 137370); 11 other intermediate Darrough, 1995: fi gs 6.1-6.2, 6.4) are most likely monopla- POJETA ET AL.: UPPER CAMBRIAN CHITONS 13
Table 1.—Measurements (in mm) of Chelodes cooperi. All specimens Th e specimen that Stinchcomb & Darrough illustrated in are internal molds of intermediate valves. H, maximum dorsal- their fi g. 6.3 was not given a published USNM number by ventral height; L, anterior-posterior length; W, maximum right-left those authors. In the USNM collections, it now has number dimension. 489542. Th e label with the specimen in part reads: "Preacan- thochiton sp., Stinchcomb and Darrough, 1995, F[ig]. 6.3, Museum No. Pl./Fig. L H W not received from author, not cited by number." USNM 137370 2/1 4.6 2.9 4.1 Th e specimen that Stinchcomb & Darrough illustrated in USNM 537303 2/4 5.0 3.0 4.9 their fi g. 6.5 was not given a published USNM number by USNM 537308 2/12 7.7 4.3 7.1 those authors. In the USNM collections, it now has the num- ber 489543. Th e label with the specimen in part reads: "Prea- USNM 537309 3/1 9.7 5.8 8.0 canthochiton sp., Stinchcomb and Darrough, 1995, F[ig]. 6.5, not received from author, not cited by number." cophorans. Runnegar et al. (1979), followed Bergenhayn's (1960) de- Although most of Stinchcomb & Darrough's fi gured scription of Preacanthochiton cooperi; thus, they fi gured speci- specimens were assigned USNM numbers, some of their mens that are here assigned to both Chelodes and Aulochiton specimens having USNM numbers were never reposited in n. gen. Using Bergenhayn's illustration of the holotype of P. the museum by those authors; other specimens were not as- cooperi, Hoare & Pojeta (2006) named P. baueri. Th is species signed USNM numbers and were not reposited in the USNM is Middle Ordovician (Turinian) in age, it has a very small collections. Th us, various specimens used by Stinchcomb & apical area limited to the immediate border of the posterior Darrough were not available to us and we could not restudy margins and therefore it is not a mattheviid. Here we reas- them. For Chelodes sp., Stinchcomb & Darrough (1995: 57, sign P. baueri to Helminthochiton with question as H.? baueri fi g. 6.1) did not cite a published USNM number for the (Hoare & Pojeta, 2006). specimen; at the museum, a USNM label and number were One specimen of Chelodes cooperi shows the probable pres- created for this specimen; in part, the USNM label for this ence of esthete canals (Pl. 3, Figs 8-9). See the discussions of specimen reads "USNM 489544; Chelodes sp., Stinchcomb- esthete canals in the sections on Hemithecella and Aulochiton Darrough, 1995, F[ig]. 6.1, not received from author." below. Stinchcomb & Darrough (1995: 57, fi gs 6.2-6.6) illustrat- ed fi ve specimens as Preacanthochiton sp. Th ree of these speci- Genus HEMITHECELLA Ulrich & Bridge, 1941 mens were assigned published USNM numbers (409457, fi g. 6.2; 409457B, fi g. 6.4; 409458, fi g. 6.6; fi gs 6.3 and 6.5 were Hemithecella Ulrich & Bridge, 1941: 19. not assigned published USNM numbers). Only one of the fi ve Elongata Stinchcomb & Darrough, 1995: 62. specimens was sent by the authors to the USNM. Th e infor- mation on the USNM labels referring to the Stinchcomb & Type species.—Hemithecella expansa Ulrich & Bridge, Darrough paper indicates that the Stinchcomb and Darrough 1941, by monotypy and original designation (Pl. 4, Fig. 9; Pl. number USNM 409457 was sent to the museum and was re- 5, Figs 1-2; Pl. 6, Fig. 4). numbered 409457A; the label reads in part: "Preacanthochiton Description—Mattheviids having a single, elongated, nar- sp., Stinchcomb and Darrough, 1995, F[ig.] 6.2." Th is speci- row lacuna above the apical area of intermediate valves; valves men is illustrated here (Pl. 13, Fig. 3). It lacks any indication triangular to markedly elongate in shape; in lateral view, con- of an apical area. Th e specimen is poorly preserved and has vex ventrally, fl at to concave dorsally, with thin apical areas been weathered and eroded. It does show multiple marginal that are extended laterally to midlength or beyond; valves marking on one side that might be remnants of muscle scars; tapering to a pointed posterior apex, broadened anteriorly the corroded apex is displaced toward one end. Th ese features and sometimes separated into vague, raised central area and are most like tryblidiid monoplacophorans (Runnegar, 1987: laterally extended fi elds; anterior edge raised and sometimes 289) and we would classify this specimen with that group. slightly refl ected posteriorly. Ornament of growth lines that Th e label of USNM 409457B in part reads: "Preacan- can be raised. thochiton sp., Stinchcomb-Darrough, 1995, F[ig]. 6.4, not re- Head valve subquadrangular, arched dorsally, with lateral ceived from author." Stinchcomb & Darrough used the num- margins convex, posterior margin shallowly concave, ventral ber USNM 409458 for both fi gs 6.6 and 6.10. Th e specimen surface with transverse ridge near midlength separating shal- shown in fi g. 6.6 is now numbered USNM 489541; in part, lower anterior area from deeper posterior area. Ornament of the label reads: "Incorrectly cited as 409458...Preacanthochi- growth lines. Tail valve poorly known. ton Stinchcomb-Darrough, 1995, F[ig]. 6.6, not received Occurrence.—Upper Cambrian to Lower Ordovician, in from author." the states of Missouri, Virginia, and Wisconsin, USA. 14 Bulletins of American Paleontology, No.
Discussion.—Th e head valve description is based on two Figs 1-3; Pl. 12, Figs 1-6; Pl. 13, Figs 1-2 specimens of Hemithecella expansa Ulrich & Bridge, 1941, from Lower Ordovician rocks in Virginia described by Hoare Hemithecella eminensis Stinchcomb & Darrough, 1995: 61, fi gs 7.7- & Pojeta (2006); they also described a possible tail valve. 7.9, 7.12, 8.15-8.16. In internal molds of intermediate valves of Hemithecella Hemithecella elongata Stinchcomb & Darrough, 1995: 62, fi gs 7.10- expansa from Missouri, the apical area is a thin plate ventral 7.11. Elongata perplexa Stinchcomb & Darrough, 1995: 62, fi gs 7.14- to the lacuna; commonly an external mold of the ventral side 7.15. of the apical area is present and sometimes this external mold shows growth lines (Pl. 4, Fig. 9; Pl. 5, Figs 1, 3-4; Pl. 6, Fig. Diagnosis.—Narrow Hemithecella, intermediate valves 4). In silicifi ed specimens of H. expansa from Virginia, the markedly elongated, slowly expanding anteriorly, apical angle apical area of intermediate valves is also a thin plate showing less than 25º; apical area recessed dorsal to ventral edge of ventral growth lines (Hoare & Pojeta, 2006: fi gs 6.5, 6.13). valve. Head and tail valves unknown. In molds of Hemithecella eminensis Stinchcomb & Measurements.—See Table 2. Darrough, 1995, from Missouri, there is also evidence of a Specimens.—Holotypes and paratypes of Hemithecella em- thin plate in the apical region of intermediate valves, and its inensis, H. elongata, and Elongata perplexa, and 20 new speci- ventral impression showing growth lines is preserved on sev- mens. eral specimens (Pl. 4, Figs 6-8; Pl. 7, Figs 3-4; Pl. 10, Fig. 2). Th e specimens of Hemithecella eminensis are fi gured here Th is plate is either a recessed apical area or an internal plate with the posterior end facing up. Th is allows for ease of com- separating lucanae. Ventral to the thin plate is a mold of an ad- parison with the monograph by Stinchcomb & Darrough ditional area of shell, which sometimes preserves growth lines (1995), who also fi gured specimens of Hemithecella with the (Pl. 4, Fig. 6; Pl. 6, Fig. 6; Pl. 7, Fig. 3; Pl. 8, Fig. 2). Because posterior end facing up. none of the specimens available for this study preserves a clear Occurrence.—Upper Cambrian Eminence Dolomite of view of the ventral side of the valves of H. eminensis, we can- Missouri, USA. not determine if the growth lines on this additional shell area Comparisons.—Th e type species of Hemithecella, H. ex- below the thin plate continue around the valves on the ven- pansa, diff ers from H. eminensis in having a thin, unrecessed tral side, or if the area of additional shell formed a curtain- apical area and an apical angle of 25º or more (Pl. 4, Fig. 9; like shell fl ange around a dorsally recessed apical area. If the Pl. 5, Figs 1-4; Pl. 6, Figs 3-4; Pl. 9, Figs 3-5). Th e Ordovician growth lines continued around the ventral side of the valve, species H. abrupta Stinchcomb & Darrough, 1995 (text-fi gs the thin plate would be separating two lacunae, similar to the 6.16-6.17), is much like Chelodes cooperi, having anteriorly condition in Matthevia (Runnegar et al., 1979; Vendrasco expanded, triangular-shaped intermediate valves and an api- & Runnegar, 2004) and the apical area would be below the cal area that is broad and that lacks the elongated lacuna of lower lacuna. If the shell below the thin plate were a curtain- H. expansa and H. eminensis. Here H. abrupta is placed in like fl ange, the thin plate would be a dorsally recessed apical Chelodes as C. abrupta (Stinchcomb & Darrough, 1995). area. A latex replica of one specimen (Pl. 8, Fig. 3) suggests Hemithecella quinquelites Stinchcomb & Darrough, 1995, is the presence of a curtain-like fl ange of shell around a recessed subcircular in shape, lacks the elongate shape of other species apical area. Although no articulated specimens of Cambrian multi- valved mollusks are known, Stinchcomb & Darrough (1995: Table 2.—Measurements (in mm) of Hemithecella eminensis. Not fi g. 6.9; Pl. 9, Figs 3-5 here) showed an Ordovician specimen all specimens were complete enough to be measured, nor could all of Hemithecella expansa having two attached, thick interme- measurements be made on all specimens. USNM 489533 preserves diate valves preserved parallel to bedding. As preserved, the the shell; all other measured specimens are internal molds. All speci- upper valve completely overlaps the lower valve. We could not mens are intermediate valves. H, maximum dorsal-ventral height; L, determine with certainty whether the attached valves repre- anterior-posterior length; W, maximum right-left dimension. sent cone-in-cone preservation caused by currents, or if they represent articulation. Th e cone-in-cone preservation seems Museum No. Pl./Fig. L H W the more likely explanation because of the way the valves USNM 409456 6/2 ca. 26.1 4.1 ca. 7.0 overlap. USNM 537314 7/3 ca. 25.6 5.5 6.6 Hemithecella eminensis Stinchcomb & Darrough, 1995 USNM 409447-B 8/7 26.6 ca. 3.0 4.0 Pl. 4, Figs 6-8; Pl. 5, Figs 5-7; Pl. 6, Figs 1-2, 5-6 ; Pl. 7, Figs USNM 489533 9/2 ca. 36.0 5.9 11.3 1-7; Pl. 8, Figs 1-7; Pl. 9, Figs 1-2; Pl. 10, Figs 1-7; Pl. 11, USNM 537322 11/1 ca. 31.0 4.1 12.5 POJETA ET AL.: UPPER CAMBRIAN CHITONS 15 of Hemithecella, and has a very short posterior lacuna; its sys- regarded as a paratype of H. elongata. In addition, the USNM tematic placement is unclear to us. labels use the number 409448A for the holotype of H. elon- Matthevia walcotti Runnegar et al., 1979 (Pl. 14, Figs 1-2), gata (Pl. 6, Fig. 1) and 409448B for a previously unfi gured from the Upper Cambrian rocks of Wisconsin, has a single paratype (Pl. 5, Fig. 7) of the taxon. lacuna, an apical angle of more than 25º, and probably had Th e holotype of Hemithecella elongata (USNM 409448A a thin, unrecessed apical area; it is here placed in synonymy = 409448 of Stinchcomb & Darrough, 1995), is elongate, with Hemithecella expansa. incomplete both anteriorly and posteriorly, and has extensive Discussion.—Th e species name Hemithecella eminen- secondary deposits on its surface (Pl. 6, Fig. 1); it has an apical sis is used here because its description begins on p. 61 of angle of approximately 22º. Stinchcomb & Darrough's (1995) monograph, and thus has Two specimens labeled USNM 409450A are part of the page priority over H. elongata and Elongata perplexa, which type suite of Hemithecella elongata as designated by Stinchcomb are described on p. 62. Comparing the descriptions and type & Darrough (1995: 62); these paratypes were not fi gured by specimens of these three names, we fi nd no signifi cant and those authors. Th e previously unfi gured paratypes are shown consistent morphological diff erences between them and place herein (Pl. 5, Figs 3-4; Pl.6, Fig. 3); they have expanded, pad- H. elongata and E. perplexa in subjective synonymy with H. dle-like shapes, thin apical areas, and apical angles of 30º or eminensis. more. Th ese specimens are herein reassigned to H. expansa. USNM 409439B was designated as the holotype of Th e holotype of Elongata perplexa (USNM 449447A) con- Hemithecella eminensis by Stinchcomb & Darrough (1995: sists of part and counterpart, internal and external molds, of 62), however, in the explanation of their fi g. 8.16 on p. 61, an elongated intermediate valve (Pl. 8, Figs 1-4) that is incom- USNM 409439B is called a paratype. Glued to the under- plete anteriorly, has an almost complete fi lling of the lacuna side of USNM 409439B is a slip of paper labeled "Holotype." to the posterior end, and has an apical angle of approximately Th us, the explanation in their fi g. 8.16 is treated as a lapsus 15º. In the explanation of their illustrations of E. perplexa, calami by us, and we regard USNM 409439B as the holotype Stinchcomb & Darrough (1995: 58, text-fi g. 7) wrote: "14, of H. eminensis. Th e holotype is shown here on Pl. 4, Figs 6-8. 15, Elongata perplexa n. gen. and sp. Eminence Formation, Th e specimen is incomplete anteriorly, slightly bent clockwise locality H-2. 14, holotype with high contrast showing shal- posteriorly, and the fi lling of the lacuna is broken and incom- low and elongate posterior tunnel, partially fi lled with druze, plete posteriorly. Growth lines of the ventral side of the apical USNM 409447, X2 [on page 62, they gave the holotype area are present below the lacuna fi lling as well as on the area number as USNM 409447A, which corresponds with the of shell below the apical area. Because of the bending of the USNM label number]; 15, paratype, USNM 409447B, X1." posterior end, the apical angle is diffi cult to measure; it is ap- However, both fi gures are of the holotype at diff erent magni- proximately 19º. fi cations. A latex cast (Pl. 8, Fig. 3) of the holotype (USNM Th e type suite of Hemithecella elongata, reposited at the 449447A) shows that the edges of the presumed apical area USNM has the numbers 409448A, 409448B, 409450A are recessed above the surrounding fl ange-like ventral edge of (two specimens), and 409456; all specimens in this type the valve. Stinchcomb & Darrough (1995: 62) cited two pre- suite are fi gured here. Th e paratype USNM 409449 cited by viously unfi gured paratypes of E. perplexa (USNM 409447B, Stinchcomb & Darrough (1995: 62) was not received at the 409447C); these specimens are shown on Pl. 8, Figs 5-9. USNM from the authors. Paratype USNM 409447B is an elongate internal mold Stinchcomb & Darrough (1995: 58, fi gs. 7.10-7.11) having an apical angle of approximately 9º. Dorsally, the fi gured two specimens that they identifi ed as Hemithecella specimen shows right and left linear features (Pl. 8, Figs 5-7) elongata. In the fi gure explanation, they noted: "10, 11, that anteriorly encompass a raised area (which would be a Hemithecella elongata n. sp. 10, internal mold of valve show- depressed area in the shell). Th e posterior end of this raised ing elongated posterior tunnel, part of which is coated with area is probably where the lacuna began, but the fi lling of the secondary deposits, holotype, USNM 409448, locality 1, X1; lacuna is not preserved. Th e linear features are composed of a 11, chert slab with two specimens, largest specimen is the ho- series of knobs (which would be depressions in the shell) (Pl. lotype, ... locality 3, USNM 409456, X1.2." [boldface type 8, Figs 5, 7). Th ese linear features could be point attachments not in original]. Th us, they designated two specimens as the of the epidermis of the mantle to the shell. Such attachments holotype of H. elongata. In the text (1995: 62), they noted to the shell occur in the mantle epithelium over the general that the holotype of H. elongata is USNM 409448. USNM mantle surface of various pelecypods (bivalves) such as the 409448 has line priority in the fi gure explanation as the holo- Devonian species Modiomorpha concentrica (Conrad, 1838) type and the designation of two holotypes is herein treated as (Pojeta et al., 1986: pl. 11) and the Holocene species Mytilus a lapsus calami. Th e specimen USNM 409456 (Pl. 6, Fig. 2) is californianus Conrad, 1837 (Pojeta et al., 1986: pl. 12, fi g. 7). 16 Bulletins of American Paleontology, No.
In Modiomorpha concentrica, the point attachments can be ar- Figs 3-4, show an epoxy cast of the ventral side of horizon- ranged in rows. Although these attachments can leave scars on tal esthete canals of the extant chiton Dendrochiton fl ectens the shell, Smith (1983: 67) noted that in unionids: "Mantle (Carpenter, 1864); as in H. eminensis, these canals vary in di- attachment involves modifi cation of mantle epithelial cells and ameter. Plate 14, Fig. 4, is an enlargement of Pl. 14, Fig. 3, in associated connective fi bers within the mantle. Th e connec- the center of which are the vertical esthete canals proceeding tive fi bers within the mantle do not appear to have contractile dorsally. Fernandez et al., (2007) and Vendrasco et al. (2008a) properties, but along with the attachment cells probably pro- provided extensive discussions and illustrations of esthetes in vide support for the mantle." Eernisse & Reynolds (1994: 64) extant chitons and the technique for making epoxy casts of wrote that in chitons: "Th e mantle epidermis [of the girdle] is esthete canal systems. continuous with a thin epithelium in close contact with the In Placiphorella velata, as many as three horizontal canals valve undersurface and extending through pores to the dorsal occur in the space of 100 μm. In a study of 12 species of esthete organs." Th ese linear features in the internal mold of extant chitons, Fernandez et al. (2007) found that the diam- paratype USNM 409447B could explain similar linear fea- eter of horizontal esthete canals ranged from 9-50 μm. Th us, tures in the thin external shell of the paratype USNM 489533 the horizontal tubes of Hemithecella eminensis are larger than (Pl. 9, Figs 1-2) of Hemithecella eminensis. the horizontal canals in extant chitons that have been studied. Elongata perplexa paratype USNM 409447C (Pl. 8, Figs Th is is understandable because H. eminensis occurs at the be- 8-9) most probably is not a Hemithecella. Th e specimen has ginning of the known fossil record of chitons. a circular cross section. Stinchcomb & Darrough (1995: 62) No known specimens of Hemithecella eminensis preserve noted: "Valves of Elongata were at fi rst thought to be occipital evidence of vertical esthete canals. However, an Ordovician spines of a saukiid trilobite such as Calvinella." Th is seems to specimen of H. expansa (Pl. 15, Figs 5-8) shows numerous be a more likely explanation for USNM 409447C. pustule-like markings on its exterior surface. Th ese pustule- like markings are interpreted as being the upper ends of EVIDENCE OF ESTHETE CANALS IN HEMITHECELLA EMINENSIS vertical esthete canals and not as valve ornament. Hoare et We have six specimens of Hemithecella eminensis that show al. (1972) noted the presence of pores in the tegmentum of right-left, or horizontal, fi llings of tubes immediately inside Pennsylvanian chitons from Ohio and that these pores are the dorsum of intermediate valves (Pl. 5, Figs 5-6; Pl. 7, Figs expressed externally as small granules. Cherns (2004: 445), 5-7; Pl. 11, Fig. 3; Pl. 12, Figs 1-6; Pl. 13, Fig. 1). One of the working with Silurian chitons from Sweden, wrote: "Th e typi- specimens preserves part and counterpart (Pl. 11, Figs 1-3; Pl. cally granular dorsal ornament is comparable with Recent chi- 12, Figs 1-2) showing that the tubes are just to the inside of tons, and is hence possibly linked functionally with sensory the dorsum above the body of the valve and in the dorsalmost aesthetes." Hoare & Karasawa (2008) noted that in species of part of the lacuna. Th e average diameter of the tube llingsfi in Permian chitons from Japan, the apparent external pustules the largest specimen (Pl. 11, Fig. 3; Pl. 12, Figs 1-2) is 280 are the ends of esthete canals. Care should be taken when μm. In the smallest specimen (Pl. 7, Figs 5-6), the average describing the external surfaces of fossil chiton valves, not to diameter of the tubes is 110 μm. Th e average diameter of the confuse the upper ends of esthete canals with shell surface tubes in fi ve specimens is 150 μm; the sixth specimen (Pl. 7, ornament. Fig. 7) is not well enough preserved to measure the diameter of the tubes. DICHOTOMOUS KEY TO THE GENERA OF CAMBRIAN AND Th e tubes in Hemithecella eminensis are interpreted as ORDOVICIAN MATTHEVIIDAE horizontal esthete canals. In extant chitons, there are nu- merous esthete canals that penetrate the tegmentum of the (1a) Two lacunae present in intermediate valves ...... 2 valve and serve sensory and perhaps secretory functions. (1b) One lacuna present in intermediate valves ...... 4 Some esthetes have been demonstrated to be photoreceptors (Moseley, 1885). Plate 13, Fig. 4, shows a section through (2a) Intermediate valves elongate ...... 3 the tegmentum-articulamentum junction in the extant chiton (2b) Intermediate valves subcylindrical ...... Robustum Placiphorella velata Carpenter, 1879; the lowest part of the tegmentum shows a dense series of layers of large horizon- (3a) Two fully developed internal lacunae ...... Matthevia tal esthete canals in cross section. Th e tubes of H. emeninsis (3b) Upper lacuna poorly developed as an external groove . also occur in multiple horizontal layers (Pl. 5, Fig. 5; Pl. 13, ...... Eukteanochiton Fig. 1). Th e upper part of the tegmentum in P. velata (Pl. 13, Fig. 4) shows the numerous small holes of vertical esthete (4a) Intermediate valves wide, taper rapidly to a poste- canals that arise from the horizontal esthete canals. Plate 14, rior apex ...... 5 POJETA ET AL.: UPPER CAMBRIAN CHITONS 17
(4b) Intermediate valves narrow, taper slowly to a posterior orly; ornament of growth lines. apex ...... Calceochiton Occurrence.—Upper Cambrian Eminence Dolomite of the state of Missouri, USA. (5a) Intermediate valve lacuna broadly triangular ...... 6 Etymology.—Greek, aulon, channel, in allusion to the ca- (5b) Intermediate valve lacuna narrow and linear ...... nals that penetrate the outer shell layer; chiton, tunic (Brown, ...... Hemithecella 1956). Discussion.—Th e seemingly pustulose surfaces of all valves (6a) Anterior end of intermediate valves broadly expanded is not ornament; each of the apparent pustules represents the laterally ...... Chelodes outer ends of vertical canals for the esthetes in the tegmen- (6b) Anterior end of intermediate valves not broadly expand- tum. ed ...... Spicuchelodes Th e anterior-posterior axis of intermediate valves was es- tablished by the position of the boss as posterior; in extant Family AULOCHITONIDAE n. fam. chitons, the apex of intermediate valves is posterior in posi- Type genus.—Aulochiton n. gen., is here designated as the tion. In Aulochiton, there is a median ridge that extends from type genus of the new family Aulochitonidae. the boss in an anterior direction toward the open end of the Diagnosis.—Small paleoloricates; intermediate valves snout. Th e head valves also have a boss from which a median elongately rounded in shape and mucronate, having a midline ridge passes in the anterior direction. Th e mucro of the tail posterior boss from which a jugal ridge extends anteriorly; api- valves is posterior in position, as it is in many extant chitons. cal area small; head valve almost circular in shape, mucronate with a central boss from which a jugal ridge extends anteri- Aulochiton sannerae n. sp. orly; tail valves elongated with posterior mucro from which a Pl. 4, Figs 4-5; Pl. 15, Figs 1-4; Pl. 16, Figs 6-11; Pl. 17, Figs poorly defi ned jugal ridge can extend anteriorly. 1-7; Pl. 18, Figs 1-11; Pl. 19, Figs 1-13; Pl. 20, Figs 1-19; Pl. Occurrence.—Known only from the Upper Cambrian 21, Figs 1-3 Eminence Dolomite of the state of Missouri, USA. Discussion.—See Appendix 1 for reviews and analyses of [non] Preacanthochiton cooperi Bergenhayn, 1960: 169, text-fi gs 1.1- the 14 previously named families of Cambrian and Ordovician 1.3. chitons. Preacanthochiton aff . cooperi Bergenhayn, 1960: 173, text-fi gs 1.9- 1.10. Preacanthochiton cooperi Runnegar et al., 1979: 1391, pl. 2, fi gs 62- Genus AULOCHITON n. gen. 63, 65-66; [non] pl. 2, fi g. 64. Type species.—Aulochiton sannerae n. sp. Diagnosis.—As for the family. Description.—As for the genus. Description.—Aulochitonids having circular head valve in Measurements.—See Table 3. dorsal profi le, mucronate with central boss from which a mid- Specimens.—Two head valve paratypes (Pl. 17, Figs 4-5; jugal ridge extends anteriorly, lacking anterior snout; lateral, Pl. 18, Figs 8-11); 10 tail valve paratypes (Pl. 4, Figs 4-5; Pl. anterior, and posterior profi les forming low isosceles triangles, 17, Figs 6-7; Pl. 20; Figs 5-19); 15 fi gured intermediate valve ornament of growth lines. paratypes (Pl. 15, Figs 1-4; Pl. 16, Figs 6-11; Pl. 17, Figs 1-3; Intermediate valves elongately rounded, extended anteri- Pl. 18, Figs 1-7; Pl. 19, Figs 1-13; Pl. 20, Figs 1-4; Pl. 21, Figs orly into a snout-like projection that is raised and open ante- 1-3 ; and 21 unfi gured intermediate valve paratypes (USNM riorly, mucronate with a midline posterior boss from which a suite number 537375). midjugal ridge extends anteriorly; in dorsal profi le, lateral and Th e holotype (USNM 537336; Pl. 18, Figs 1-4) of posterior margins strongly arcuate; anterior and posterior pro- Aulochiton sannerae is an intermediate valve preserving the fi les forming low isosceles triangles; lateral profi les forming part and counterpart. Th is specimen clearly shows that the low scalene triangles; ornament of growth lines; apical area part with the esthete-penetrated tegmentum and the coun- highly reduced. terpart having the smooth internal mold belong to the same Tail valve anteriorly-posteriorly elongate, with posterior valve. Th e holotype is the only known intermediate valve that central mucro from which externally there is a vague midju- preserves both part and counterpart. gal ridge; in dorsal profi le, anterior margin rounded, raised, Occurrence.—Upper Cambrian Eminence Dolomite, state and open, posterior margin rounded; lateral margins straight of Missouri, USA. to slightly rounded; anterior and posterior profi les forming Etymology.—Th e species is named for JoAnn Sanner, low isosceles triangles; lateral profi le of internal molds straight Department of Paleobiology, National Museum of Natural dorsally, rounded anteriorly and ventrally, concave posteri- History, in recognition of her exceptional help in creating 18 Bulletins of American Paleontology, No.
Table 3.—Measurements (in mm) of Aulochiton sannerae. All EVIDENCE OF ESTHETE CANALS IN AULOCHITON SANNERAE measurements taken from valves that preserve the exterior of the All of the specimens preserving the exterior of Aulochiton san- shell, not from internal molds. Not all specimens were complete nerae have the shell replaced by silica and the thickness of the enough to be measured, nor could all measurements be made on all shell can be measured. Th e average thickness of the lower left specimens. H, maximum dorsal-ventral height; L, anterior-posterior side of the intermediate valve seen on Pl. 21, Fig. 3, is 283 length; W, maximum right-left dimension. μm. Th e average thickness of the lower right side of the inter- mediate valve seen on Pl. 21, Fig. 2, is 187 μm. Museum No. Pl./Fig. L H W Th e inner surface of the shell appears to be full of holes Head Valve: (Pl. 15, Fig. 1; Pl. 16, Figs 6, 8, 10; Pl. 17, Figs 2-4, 6; Pl. USNM 537339 18/9 5.5 3.3 5.1 18, Figs 3, 8; Pl. 21, Figs 2-3). However, these are not holes; rather, they are the inner openings of canals that have length Tail Valve: and penetrate the outer shell layer; this penetration can be USNM 537335 17/6 7.5 3.0 3.5 seen in cross section (Pl. 21, Figs 2-3) and in latex replicas (Pl. Intermediate Valves: 16, Figs 7, 9, 11; Pl. 17, Figs 5, 7; Pl. 18, Figs 4, 9-11). Th us, USNM 537331 16/10 7.5 3.1 6.0 it is reasonable to consider the preserved shell to be the teg- mentum shell layer and the canals to be the spaces for vertical USNM 537332 17/2 5.9 2.8 ca. 5.0 esthetes. As measured on the intermediate valve seen on Pl. 16, Figs 10-11, the canals nearest to the posterocentral boss (the fi rst formed part of the valve) have an average diameter electronic montages of fi gures and plates for this manuscript of 62 μm; away from this area, toward the shell margin, the and in several other papers for JP. canals increase in average diameter to 122 μm. Discussion.—Aulochiton sannerae is the species in which None of the known specimens of Aulochiton sannerae the seemingly pustulose exterior that Bergenhayn (1960: shows the presence of horizontal esthete canals, which would text-fi gs 1.1, 1.3) incorrectly placed on his illustrations of have been at the tegmentum-hypostracum junction; only the Preacanthochiton cooperi (= Chelodes cooperi here). Based on tegmentum has been silicifi ed. All of the internal molds are what was known at the time, Runnegar et al. (1979) accepted smooth and show no evidence of esthetes (Pl. 15, Figs 3-4; Bergenhayn's illustrations. Pl. 18, Figs 6-7; Pl. 19, Figs 1-11; Pl. 20, Figs 1-18). Th us, Th e small apical area (Pl. 21, Fig. 1) diff erentiates Aulo- it seems probable that the internal molds were formed below chiton sannerae from mattheviids. Th e rounded head valve of the hypostracum, which has been diagenetically dissolved. A. sannerae is somewhat similar in shape to the head valve Perhaps the tegmentum was silicifi ed because of the great sur- of the Middle Ordovician species Alastega martini Hoare & face area created by the vertical esthete canals for replacement Pojeta, 2006; however, the head valve of A. martini has a of the carbonate by silica. Th e dense hypostracum was dis- small anterior sinus and straighter lateral margins. As in the solved rather than being replaced. Th is interpretation is sup- Late Cambrian species Matthevia wahwahensis Vendrasco & ported by the specimen shown on Pl. 20, Fig. 11, in which Runnegar, 2004, the head valve of A. sannerae shows holope- there is a space between the silicifi ed outer shell layer and the ripheral growth, but it diff ers in not being fl at and in having internal mold of the valve. the anterior midjugal ridge. In addition to evidence for the presence of esthetes in Th e intermediate valves of Aulochiton sannerae have an an- Aulochiton sannerae and Hemithecella eminensis, one interme- terior snout (Pl. 18, Figs 6-7), are expanded laterally, and have diate valve of Chelodes cooperi (Pl. 3, Figs 7-8) shows the fi ll- arcuate lateral and posterior margins in dorsal profi le (Pl. 19, ings of esthete canals. Th us, all of the species of undoubted Figs 7-8). Th is diff ers from other Cambrian and Ordovician polyplacophorans in this study show evidence for the pres- chitons in which the lateral margins are straight. In Chelodes ence of esthetes at the beginning of the known fossil record cooperi, the intermediate valves are expanded anterolaterally, of chitons. but they tend to be straight-sided (Pl. 3, Figs 1, 6), and C. cooperi has a large apical area (Pl. 3, Fig. 1). Class ?POLYPLACOPHORA Th e slightly-rounded to straight-sided tail valves of Family DYCHEIIDAE n. fam. Aulochiton sannerae are similar to the tail valves of many other Diagnosis.—Small multivalved mollusks having two lacu- Cambrian and Ordovician chitons. One tail valve (Pl. 20, Fig. nae on the ventral side of the dorsum of the valves, one lacuna 9) shows prominent ventral folds on the left side. Th ese folds placed anteriorly and the other placed posteriorly; apical areas along the ventral margin are where the valve would join the lacking; intermediate valves gaping anteriorly, posteriorly, and mantle girdle; the folds might have increased the surface area ventrally; tail valve gaping ventrally and anteriorly. for better attachment of the valve to the girdle. POJETA ET AL.: UPPER CAMBRIAN CHITONS 19
Type genus.—Dycheia Pojeta & Derby, 2007, is here desig- the lacunae of the intermediate valves located medially on the nated the type genus Dycheiidae n. fam. valve far from the anterior and posterior edges of the valves Occurrence.—Dycheia occurs in the Upper Cambrian Fort (Pl. 21, Fig. 10; Pl. 22, Figs 4-5, 7, 9). In Dycheia, the lacunae Sill Limestone in the state of Oklahoma, USA. Paradycheia n. of the intermediate valves are located marginally and almost gen. occurs in the Upper Cambrian Eminence Dolomite in touch the anterior and posterior edges (Pl. 16, Fig. 2). Th e the state of Missouri, USA. Both taxa are known only from tail valves of Dycheia have large deep lacunae (Pl. 21, Fig. 6), disarticulated valves. whereas in Paradycheia the lacunae are small and shallow (Pl. Discussion.—Dycheiids have features occurring in various 23, Figs 5-7). mollusks including accretionary growth lines, multiple valves, and probable muscle-scar growth tracks (Pl. 21, Figs 4, 6; Pl. MORPHOLOGY AND ORIENTATION 23, Figs 1-2; Pojeta & Derby, 2007: fi gs 3.A, F, 4.D). Paradycheia dorisae is known from two types of disarticulated Th e family Dycheiidae is proposed to unite multivalved valves. Except for one specimen, the valves are preserved as mollusks of similar morphologies, but whose systematic po- chertifi ed internal molds known from three localities. We sition in the Mollusca is poorly understood. Some of the regard P. dorisae as a multivalved mollusk closely allied to morphological features of dycheiids can be understood by Dycheia; however, at the present time, it is not possible to comparing them to mattheviid paleoloricates; however, the determine how many valves covered the body. absence of an apical area and the gaping of the intermediate Th e reasons for regarding both types of valves as belonging valves anteriorly, posteriorly, and ventrally, make placement of to the species Paradycheia dorisae are: (1) the presence of ante- dycheiids in the Polyplacophora uncertain. rior and posterior lacunae in both types; (2) the occurrence of Dycheiids have the two tapering lacunae that also occur the two types of valves in the same collection at locality 3; and in both the intermediate and tail valves in Late Cambrian and (3) the symmetry of the valves. Early Ordovician species of Matthevia (Runnegar et al., 1979; English, 2002; Pojeta et al., 2005; Vendrasco & Runnegar, Type 1 Valve 2004). Like paleoloricate chitons, dycheiids show no sign of Th is is the most abundant valve; of the 15 available Type 1 an articulamentum. Th e valves of dycheiids were most prob- valves, three are from locality 1, six from locality 2, and six ably dorsal to the body and the lacunae are on the ventral side from locality 3. of the dorsum. Th e tissue that fi lled the lacunae could not In lateral and dorsal profi les, Type 1 valves are rectangular function in any other position. in shape (Pl. 21, Fig. 10; Pl. 22, Figs 4, 7, 9). Except for the tips of the lacunae, the valves have a rounded cross section; Genus PARADYCHEIA n. gen. they are open ventrally and lack an apical area (Pl. 21, Fig. Type species.—Paradycheia dorisae n. sp. is here designated 9; Pl. 23, Fig. 3). None of the known specimens allows di- the type species of Paradycheia n. gen. rect observation of the shell edges at either end of the Type 1 Diagnosis.—Dycheiids having intermediate valves rounded valves; however, lateral views indicate the presence of shallow, in center cross section; anterior and posterior lacunae placed inverted U-shaped anterior and posterior openings (Pl. 22, medially, far from the anterior and posterior shell margins; tail Figs 5, 7-8, 10). valve with very small, medially placed lacunae. Exterior features are poorly known. One poorly preserved Description.—Multivalved mollusks having two types of external mold shows a few comarginal growth lines (Pl. 23, valves, both types with two lacunae placed on the ventral side Figs 1-2); another mold shows probable growth lines on one of the dorsum; lacunae separated by a ridge of shell; apical area side (Pl. 22, Fig. 11). absent. Intermediate valves elongated rectangular in lateral As in the paleoloricate Matthevia, Type 1 valves of Para- and dorsal profi les with prominent lacunae; shell thickened dycheia dorisae have two lacunae on the underside of the medially between lacunae, somewhat auriculate anterior and closed dorsum (Pl. 22, Figs 3, 9); the lacunae are expressed posterior to lacunae; cross section between lacunae broadly as projections in the internal molds (Pl. 22, Figs 1-10). Th e rounded. Tail valve elongated rectangular in lateral and dorsal projections surround a space made by a shell thickening that profi les with mid-dorsal ridge and triangular in cross section; separated the lacunae (Pl. 22, Fig. 3). In some specimens, the lacunae very small. two lacunae almost touch each other dorsally (Pl. 22, Fig. 4), Etymology.—Para, Greek, meaning near or beside; Dycheia, but the shell thickening between them is still present. Slight, a genus of Dycheiidae, Greek, dyo, two holes (Brown, 1956). oblique shell thickenings set off the anterior and posterior Occurrence.—Upper Cambrian Eminence Dolomite in parts of the shell from the central part and create auricle-like the east-central area of the state of Missouri, USA. anterodorsal and and posterodorsal portions of the valve (Pl. Discussion.—Paradycheia diff ers from Dycheia in having 22, Figs 3-5, 7-9). 20 Bulletins of American Paleontology, No.
Because the lacunae are near each end of the elongated shallow inverted Us (Pl. 21, Figs 7-8), the posterior opening valve, they are regarded as being anterior and posterior in po- is smaller than the anterior one. Th e cross-sectional shape of sition. It is likely that the open anterior and posterior end of the Type 2 valve is triangular (Pl. 21, Figs 7-8) rather, than the valves were at right angles to the soft parts and that the broadly rounded as in Type 1 valves (Pl. 21, Fig. 9), and the broad lateral sides of the valves paralleled the elongate soft Type 2 valve has a mid-dorsal ridge. Th e right side of the Type parts; thus, the lacunae are regarded as having been placed in 2 valve shows prominent ventral folds (Pl. 23, Fig. 7); as in anterior and posterior positions on the dorsum. Th e lacunae a tail valve of Aulochiton sannerae (Pl. 20, Fig. 9), the folds probably housed tissues or muscles that attached the valves to might have increased the surface area for better attachment of the body. the valve to the soft parts. Type 1 valves are bilaterally symmetrical through the Type 1 valves are regarded as intermediate valves and Type long axis (length) of the valve (Pl. 21, Fig. 10; Pl. 22, Figs 3, 2 valves are regarded as terminal valves. Based on the likely 9; Pl. 23, Fig. 3). Th us, it is diffi cult to distinguish anterior interpretation of the anterior and posterior ends of Type 2 from posterior and to determine the right and left sides of valves, they are probably tail valves. Tail valves of the paleolo- the valves. In several of the better preserved specimens, it is ricate Matthevia wahwahensis Vendrasco & Runnegar (2004: noticeable that one of the two lacunae is slightly larger than 683) are narrower than intermediate valves and have a very the other hole (Pl. 22, Figs 2- 4, 9). Th us, the valves are asym- small anterior lacuna as in Paradycheia dorisae. metrical through the narrow (width) dimension of the valve. In species of the paleoloricate Matthevia, the larger lacuna is SYSTEMATIC PLACEMENT OF DYCHEIIDAE posterior to the smaller hole (Runnegar et al., 1979; Pojeta As noted above, and by Pojeta & Derby (2007), some mor- et al., 2003; Vendrasco & Runnegar, 2004). Using this com- phological features of dycheiids can be interpreted by com- parison, and for the convenience of description, we regard the parisons with paleoloricate polyplacophorans. Typically, larger and more robust lacuna of Paradycheia dorisae as being Cambrian and Ordovician chitons with small apical areas posterior. Having the anterior and posterior ends postulated, have one large ventral opening to the intermediate valves as it is now possible to recognize the right and left sides. in Aulochiton (Pl. 16, Figs 6, 8, 10; Pl. 21, Fig. 2) and various taxa presented by Hoare & Pojeta (2006); the anterior margin Type 2 Valve can be raised slightly, producing a small anterior gape (Pl. 18, Th e Type 2 valve is known from only one internal mold from Fig. 6). As discussed above, the position of the single ventral locality 3 (Pl. 21, Figs 7-8; Pl. 23, Figs 5-7), where it occurs opening to the intermediate valves of taxa with large apical with six Type 1 valves. areas has various interpretations, but it is a single opening; Th e Type 2 valve is elongated rectangular in dorsal profi le this opening can lead to one or two lacunae. Dycheiids have (Pl. 23, Fig. 5); it is bilaterally symmetrical through the longer two lacunae in the intermediate valves. However, the valves dimension of the valve. Projecting dorsally from the midline are bent so that there is a continuous shell gape anteriorly, are two small lacunae placed closer to one end of the valve ventrally, and posteriorly (Pl. 16, Figs. 2, 4-5); the lacunae than to the other end; thus the valve is asymmetrical through straddle the mid-dorsal line of the valves (Pl. 16, Fig. 2; Pl. 22, the narrow dimension. By comparison with Type 1 valves and Figs 2-3, 9; Pl. 23, Fig. 3); and an apical area is lacking. Matthevia, the larger lacuna is regarded as posterior. Th us, the Th e bent intermediate valves of dycheiids have a solid shell lacunae are off set closer to the anterior end than to the pos- dorsum, anterior, posterior, and ventral gapes, and soft-part terior end. In dorsal profi le, the anterior margin is essentially impressions across the dorsal midline; these characteristics ap- straight; whereas, the posterior margin is broadly rounded (Pl. proximate morphological features seen in the pseudobivalved 23, Fig. 5). Th e Type 2 valve lacks the slight oblique shell primitive Cambrian and Ordovician ribeiriid rostroconch thickenings seen in Type 1 valves, and the Type 2 valve is not mollusks (Pojeta & Runnegar, 1976: pls. 5-7; Pojeta et al., auriculate. 1977: pl. 1). Ribeiriid rostroconchs have unique morphologi- In lateral profi le (Pl. 23, Figs 6-7), the posterior lacuna cal features not seen in dycheiids, and rostroconchs are not fi lling projects well above the dorsal margin, whereas the an- multivalved. Nonetheless, dycheiids show features that are not terior lacuna fi lling does not. Th e seemingly bifi d structure of seen in undoubted Cambrian and Ordovician chitons; hence the posterior lacuna (Pl. 23, Figs 5-6) is the result of a small we treat them as polyplacophorans with question and place piece of chert fl aking away when the specimen was collected. them in a separate family. Th e fi nding of articulated speci- In lateral profi le, posterior to the posterior lacuna, the dorsal mens could resolve the question of the systematic placement margin slopes posteriorly; anteriorly it is straight (Pl. 23, Figs of dycheiids. 6-7). Th e anterior and posterior ends of the valve are open as POJETA ET AL.: UPPER CAMBRIAN CHITONS 21
Paradycheia dorisae n. sp. Diff erences between the species are: (1) the interme- Pl. 21, Figs 7-11; Pl. 22, Figs 1-11; Pl. 23, Figs 1-7 diate valves of Dycheia shergoldi are triangular in cross sec- tion, whereas Paradycheia dorisae has a rounded or inverted Matthevia variabilis Stinchcomb & Darrough, 1995: p. 60, fi g. 8.1. U-shaped cross section; (2) the lacunae of the intermediate [non] Matthevia variabilis Walcott, 1885: 18, fi gs 1-6; Yochelson, valves of P. dorisae are placed medially on the valves and are far 1966: 3, pl. 1, fi gs 1-45; Runnegar et al., 1979: 1390, pl. 1, fi gs removed from the anterior and posterior edges, whereas those 19-29; English, 2002: 293, fi g. 2. in D. shergoldi are marginal in position and almost touch the anterior and posterior edges; and (3) the tail valves of D. sher- Diagnosis.—As for the genus. goldi have deep and large lacunae, whereas those of P. dorisae Etymology.—Th e species is named for Doris Darrough, are shallow and small. the wife of Guy Darrough. Figured types.—Eleven specimens are fi gured. Th e holo- CLADISTIC ANALYSIS type (USNM 534821) is an intermediate valve (Pl. 21, Figs Th e results of the cladistic analysis (Text-fi g. 6) suggest a 9-10; Pl. 22, Fig. 5); it is from locality 3. Ten paratypes are close relationship among the taxa here considered to be mat- fi gured (USNM 534822-534831). theviids, and also suggest a close link between Echinochiton Unfi gured types.—USNM suite number 534832, has fi ve and mattheviids. Th e results also indicate that Robustum, intermediate valve paratypes from locality 1, three from local- Aulochiton, Dycheia, Paradycheia, and Septemchiton are stem ity 2, and one from locality 3. lineage chitons. Th e cladistic analysis was done using PAUP* Measurements.—See Table 4. ver. 4.0b10 (Swoff ord, 2000), and all characters were treat- Discussion.—Although the methods of preservation are ed as unordered. Details about the characters and states are quite diff erent (chertifi ed internal molds versus silicifi ed repli- provided in Appendix 2 and Table 5. Our results diff er from cas), Paradycheia dorisae is morphologically similar to Dycheia those of Sigwart & Sutton (2007), because in their analysis shergoldi Pojeta & Derby, 2007 (Pl. 16, Figs 1-5; Pl. 21, Figs Robustum, Matthevia, and Septemchiton fall out in the total 4-6): (1) both have two types of valves possessing two lacunae group Aplacophora. We prefer the hypothesis that those taxa, that are located anteriorly and posteriorly on the valve im- and similar forms, are most likely members of the total group mediately ventral to the dorsum; (2) both lack apical areas on Polyplacophora, a conclusion reached by others (Runnegar the intermediate and tail valves; (3) both have tail valves that et al., 1979; Smith & Hoare, 1987; Vendrasco & Runnegar, slope posteriorly to small inverted U-shaped openings, and in 2004). dorsal profi le, the tail valves have straight anterior margins; Even though there is uncertainty about the nature of (4) both have intermediate valves with subequally-sized open- Septemchiton and the family Septemchitonidae (Appendix 1), ings at the posterior and anterior ends and they gape ventrally; we included it in the analysis because a fully-articulated topo- and (5) in both, the lacunae of the tail valves are displaced type of the type species S. vermiformis is known (Rolfe, 1981: anteriorly. text-fi gs 1-2, as S. grayiae), and it has a unique form that sug- gested to Hoare (2000) and Sigwart & Sutton (2007) that it is not a chiton. We included Septemchiton to maximize the Table 4.—Measurements (in mm) of Paradycheia dorisae. Not number of multivalved mollusks used in our analysis and to all specimens are complete enough to measure, nor could all assess the proposed hypothesis (Sigwart & Sutton, 2007) that measurements be made on all specimens. All measured specimens Septemchiton is an aplacophoran. are internal molds. H, dorsal-ventral height; L, anterior-posterior length; W, right-left dimension at the center of the valve; -, not In our analysis, Acaenoplax (Sutton et al., 2004) falls out in available. the basal stem-chiton lineage, between the two modern apla- cophoran taxa Chaetoderma and Epimenia and other chitons. However, we consider Acaenoplax to be a problematic taxon, Museum No. Pl./Fig. L H W the closest extant relatives of which might be the aplacoph- Intermediate valves oran s. Th e exclusion of Acaenoplax from the modern apla- USNM 534824 22/4 5.2 4.1 - cophoran taxa in this analysis resulted in large part from the USNM 534821 22/5 ca. 10.0 4.8 - weighting of the state of no valves, done to prevent inapplica- ble character states from being applied to taxa in the analysis USNM 534825 22/7 11.2 6.2 - without valves (see Appendix 2: discussion of character 8). USNM 534823 22/9 - 3.4 4.3 Th e Cambrian chitons, except Aulochiton, form a clade, Tail valve with strong support for the close relationship of matthevi- USNM 534829 23/5 14.7 4.7 5.1 ids (Hemithecella, Calceochiton, Eukteanochiton, Matthevia, 22 Bulletins of American Paleontology, No.
a
c b
d e g
f
h
i
Text-fi g. 6—Majority-rule consensus tree of 58 most parsimonious trees (tree length = 39 steps). Th e brachiopods Lingula and Novocrania are outgroup taxa; all character states are unordered and each character is unweighted. Numbers on branches are 50% majority consensus scores. Th e data matrix is provided in Table 5 and descriptions of characters and character states as well as other details of the analysis are provided in Appendix 2. Nodes on the cladogram that are supported by characters in this analysis are indicated by letters representing the following: (a) vermiform or dorsoventrally fl attened body; (b) presence of spicules and 7-8 repeated structures; (c) no shell; (d) eight valves in a central row; (e) unopposed, overlapping valves; (f) prominent shell length posterior or ventral to the apex; (g) elongate-subcylindrical, elongate-triangular, or elongate-rounded valves; (h) intermediate valves subconical, elongate, and thick; and (i ) articulamentum layer present.
Chelodes, and Echinochiton). Robustum falls out with Dycheia areas, and are thus diff erent from the valves of other known and Paradycheia in this analysis (in 71% of the 58 most Cambrian chitons. Th erefore, dycheiids are assigned to the parsimonious trees), which form a sister group to matthevi- Polyplacophora with question (see discussion of this family in ids. However, Robustum shares some key derived characters the Systematic Paleontology section). with matteviids, and we have classifi ed it within that fam- Th e results of the cladistic analysis also indicate a robust ily. In this analysis, Dycheia and Paradycheia fall out in the separation between paleoloricates (those taxa lacking the ar- Polyplacophora although the morphology of their valves dif- ticulamentum shell layer) and neoloricates (those taxa hav- fers from that of typical chitons. For example, dycheiids have ing the projecting articulamentum shell layer). However, bent valves with anterior and posterior gapes and lack apical the placement of Polysacos (a multiplacophoran) relative to POJETA ET AL.: UPPER CAMBRIAN CHITONS 23
Table 5.—Data matrix for cladistic analysis, the result of which is shown in Text-fi g. 6. See Appendix 2 for character and character state descrip- tions. OG, outgroup.
123456789101112131415
Lingula (OG) 200001221000030 Novocrania (OG) 200001221100030 Acaenoplax 001101003?0?012 Chaetoderma 00?100034221250 Echinochiton 121111?13?0?101 Epimenia 001100034221250 Glaphurochiton 111101113111022 Halkieria 12?001201?0?030 Lepidochitona 111101113111022 Neopilina 101001240001030 Polysacos 121?11?1211?021 Robustum 0???01?1??0?111 Matthevia 1???01?1?10?101 Eukteanochiton 1???01?1??0?101 Calceochiton 1???01?1??0?101 Hemithecella 1???01?1?10?101 Chelodes 1???01?1?10?101 Septemchiton 0?1?01?13?0??12 Dycheia 0???01????0?111 Paradycheia 0???01????0?111 Aulochiton 1???01?1?10?012 Ribeiria ?0000124000?040 the other neoloricates is not well supported by characters in Th e characters that support each of the nodes in the cla- this analysis. Th e characters that unite Glaphurochiton and dogram are listed in the caption for Text-fi g. 6. Th ose nodes Lepidochitona, compared with Polysacos, are spicule distribu- without characters are not well supported by characters tion just in the dorsolateral margins, and anterior to posterior in this analysis. Th e weakness of the hypothezied clade of distinct strips of left-lateral, central, and right-lateral skeletal Glaphurochiton and Lepidochitona to the exclusion of Polysacos zones, such as valves, spicules, spines, etc. However, these is described above. Likewise, the relationships among the characters are not preserved in most fossil chitons (and so are non-amphineuran mollusks and the distinction between coded as "?" in those cases); these could be characters that Septemchiton and other chitons are not well supported in this evolved earlier in chiton history (appearing higher on the cla- analysis. dogram in Text-fi g. 6), and, if so, it would leave the relation- Th e conclusions that are well-supported by characters in ships of the most derived part of the tree unsupported by key this analysis, and that we emphasize here are: (1) the close characters. relationship of the taxa in the Mattheviidae: (2) Echinochiton 24 Bulletins of American Paleontology, No. as a member of the Mattheviidae (it diff ers in the presence Baltimore, Maryland, 882 pp. of large hollow circumsomal spines and scutes), (3) the in- Butterfi eld, N. J. 2008. An Early Cambrian radula. Journal of clusion of Septemchiton, mattheviids, and Echinochiton in the Paleontology, 82: 543-554. Polyplacophora, and (4) support for the historical separation Caron, J.-B., A. Scheltema, C. Schander, & D. Rudkin. 2006. A soft-bodied mollusc with radula from the Middle Cambrian of paleoloricates from neoloricates. Burgess Shale. Nature, 442: 159-163. Carpenter, P. P. 1864. Supplementary report on the present state of ACKNOWLEDGMENTS our knowledge with regard to the Mollusca of the west coast of We thank J. T. Dutro, Jr., U.S. Geological Survey (USGS), North America. British Association for the Advancement of Science, for identifying the brachiopod species associated with the Report for 1863: 517-686. multivalved mollusks and for reviewing the manuscript; Scott Carpenter, P. P. 1879. Placiphorella velata. In: Report on the limpets Whittaker, National Museum of Natural History (USNM) and chitons of the Alaskan and Arctic regions, with descriptions for his help with digital light photography of small specimens, of genera and subspecies believed to be new, by W. H. Dall. Proceedings of the United States National Museum, 1: 298. and measuring very small morphological features; and JoAnn Carter, J. G. 1979. Comparative shell microstructure of the Mollusca, Sanner (USNM) for creating the electronic montages of the Brachiopoda, and Bryozoa. Pages 439-446, 456, in: Scanning plates and Text-fi gs 4-5. John F. Taylor, Indiana University Electron Microscopy, O. Johari (dir.), Chicago Press Corporation, of Pennsylvania, examined the trilobite fragments. David Chicago. Bohaska (USNM) Robert Purdy (USNM), and R. D. Hoare Carter, J. G., & R. M. Hall. 1990. Polyplacophora, Scaphopoda, (Bowling Green State University, Ohio) reviewed drafts of the Archaeogastropoda, and Paragastropoda (Mollusca). Pages manuscript. Steven Jabo (USNM) was very helpful with the 29-52, in: Skeletal Biomineralization: Patterns, Processes, and making of latex casts of a specimen. Christine Z. Fernandez Evolutionary Trends, volume 2, J.G. Carter (ed.), Van Nostrand (Los Angeles, California) reviewed the manuscript, provided Reinhold, New York. helpful discussions, and aided in the creation of Text-fi gs 2-3. Cherns, L. 1998a. Chelodes and closely related Polyplacophora (Mollusca) from the Silurian of Gotland, Sweden. Palaeontology, Th e late James H. Stitt (University of Missouri) kindly donated 41: 545-573. the specimen of Chelodes sp. used in Text-fi g. 4. M. J. Grygier Cherns, L. 1998b. Silurian polyplacophoran molluscs from Gotland, (Lake Biwa Museum, Japan, and ICZN Commissioner) was Sweden. Palaeontology, 41: 939-974. very helpful with written information about a potential no- Cherns, L. 2004. Early Palaeozoic diversifi cation of chitons menclatural problem. (Polyplacophora, Mollusca) based on new data from the Silurian of Gotland, Sweden. Lethaia, 37: 445-456 LITERATURE CITED Cherns, L., D. M. Rohr, & J. Fryda. 2004. Polyplacophoran and Baxter, J. M., & A. M. Jones. 1981. Valve structure and growth symmetrical univalve mollusks. Pages 179-195, in: Th e Great in the chiton Lepidochitona cinereus (Polyplacophora: Ordovician Biodiversifi cation Event, B. D. Webby & M. L. Ischnochitonidae). Journal of the Marine Biological Association of Drosser (eds.), Columbia University Press, New York. the United Kingdom, 61: 65-78. Cohen, B. L., L. E. Holmer, & C. Lüter. 2003. Th e brachiopod fold: Baxter, J. M., & A. M. Jones. 1984. Th e valve morphol- a neglected body plan hypothesis. Palaeontology, 46: 59-65. ogy of Callochiton achatinus (Mollusca: Polyplacophora: Conrad, T. A. 1837. Descriptions of new marine shells, from Upper Ischnochitonidae). Journal of Zoology, 202: 549-560. California. Collected by Th omas Nuttall, Esq. Journal of the Bergenhayn, J. R. M. 1943. Preliminary notes on fossil polypla- Academy of Natural Sciences of Philadelphia, 7: 227-268. cophoras from Sweden. Geologiska Föreningens, Förhandlingar, Conrad, T. A. 1838. Report on the Palaeontological Department 65: 297-303. of the Survey, State of New York, No. 200. Pages 107-119, Bergenhayn, J. R. M. 1955. Die fossilen Schwedischen loricaten in: Assembly, Communication from the Govenor, Relative to the nebst einer vorläufi gen revision des systems der ganzen klasse Geological Survey of the State (New York Geological Survey, Annual Loricata. Lunds Universitets Årsskrift, n.f., 51: 3-43. Report 2.) Bergenhayn, J. R. M. 1960. Cambrian and Ordovician loricates Conway Morris, S. 1976. A new Cambrian lophophorate from the from North America. Journal of Paleontology, 34: 168-178. Burgess Shale of British Columbia. Palaeontology, 19: 199-122. Billings, E. 1865. Palaeozoic fossils 1861-1865. Geological Survey of Conway Morris, S. 1990. Triplicatella. Pages 231-232, in: Early Canada, 1: 345-426. Cambrian Fossils from South Australia, S. Bengtson, S. Conway Bischoff , G. C. O. 1981. Cobcrephora n. g., representative of a new Morris, B. J. Cooper, P. A. Jell, & B. N. Runnegar (eds.), polyplacophoran order Phosphatoloricata with calciumphos- Association of Australasian Palaeontologists, Memoir 9. phatic shells. Senckenbergiana Lethaea, 61: 173-215. Cooper, R. A., G. S. Nowlan, & S. H. Williams. 2001. Global Bridge, J. B. 1930. Geology of the Eminence and Cardarevera quad- stratotype section and point for base of the Ordovician System. rangles. Missouri Bureau of Geology and Mines, ser. 2, 24: 228 Episodes, 24: 19-28. pp. Cowper Reed, F. R. 1907. Crustacea, etc. from Girvan. Geological Brown, R. W. 1956. Composition of Scientifi c Words. Reese Press, Magazine, new ser., decade 5, 4: 108-115. Cuvier, G. 1797. Tableau Élémentaire de l'Histoire Naturelle des POJETA ET AL.: UPPER CAMBRIAN CHITONS 25
Animaux. Imprimeur du Corps législatif et du l'Institut nation- (Mollusca). Journal of Paleontology, 76: 570-573. al, place de Carrousel, no. 662, Paris, 710 pp. Hoare, R. D., & J. R. Farrell. 2004. Lower Devonian Polyplacophora Dall, W. H. 1882. On the genera of chitons. Proceedings of the United from New South Wales, Australia. Palaeontology, 47: 1495- States National Museum, 4: 279-291. 1506. Davidson, T., & W. King. 1874. On the Trimerellidae, a Palaeozoic Hoare, R. D., & H. Karasawa. 2008. Permian Polyplacophora family of the palliobranchs or Brachiopoda. Quarterly Journal of (Mollusca) from the Akaska Limestone, Gifu, Japan. Paleonto- the Geological Society, 30: 124-173. logical Research, 12: 321-327. De Blainville, H. M. D. 1816. Prodrome d'une nouvelle distribution Hoare, R. D., & J. Pojeta, Jr. 2006. Ordovician Polyplacophora Systématique du rène animal. Bulletin des Sciences, Philomathique (Mollusca) from North America. Paleontological Society Memoir de Paris, pp. 105-124. 64: 27 pp. Dunn, C. W., A. Hejol, D. Q. Matus, K. Pang, W. E. Browne, S. A. Hoare, R. D., M. T. Sturgeon, & T. B. Hoare. 1972. Middle Smith, E. Seaver, G. W. Rouse, M. Obst, G. D. Edgecomb, M. Pennsylvanian (Allegheny Group) Polyplacophora from Ohio. V. Sorensen, S. H. D. Haddock, A. Schmidt-Rhaesa, A. Okusu, Journal of Paleontology, 46: 675-680. R. M. Kristensen, W. C. Wheeler, M. Q. Martindale, & G. International Commission on Zoological Nomenclature. 1999. Giribet. 2008. Broad phylogenomic sampling improves resolu- International Code of Zoological Nomenclature, 4th ed. Inter- tion of the animal tree of life. Nature, 452: 745-749. national Trust for Zoological Nomenclature, London, 306 pp. Dzik, J. 1986. Turrilepadida and other Machaeridia. Pages 116-134, Knorre, H. Von. 1925. Die Schale und die Rückensinnesorgane von in: Problematic Fossil Taxa, A. Hoff man & M. H. Nitecki (eds.), Trachydermon (Chiton) cinereus L. und die ceylonischen Chitonen Oxford University Press, New York. der Sammlung Plate. Jenaische Zeitschrift für Naturwissenschaft Dzik, J. 1994a. Machaeridians, chitons, and conchiferan molluscs herausgegeben von der medizinisch-naturwissenschaftlicken Gessell- of the Mójcza Limestone. In: Ordovician Carbonate Platform schaft zu Jena, 61: 469-632. Ecosystem of the Holy Cross Mountains, J. Dzik, E. Olempska, & Kouchinsky, A. 2001. Mollusks, hyoliths, stenothecoids, and coelo- A. Pisera (eds.). Palaeontologia Polonica, 53: 213-252. scleritophorans. Pages 326-349, in: Th e Ecology of the Cambrian Dzik, J. 1994b. Evolution of 'small shelly fossils' assemblages of the Radiation, A. Y. Zhuravlev & R. Riding (eds), Columbia Univer- early Paleozoic. Acta Palaeontologica Polonica, 39: 247-313. sity Press, New York. Eernisse, D., & P. D. Reynolds. 1994. Polyplacophora. Pages 55- Marek, L. 1962. New polyplacophorid family Eochelodidae n. 110, in: Microscopic Anatomy of Invertebrates, Volume 5, Mollusca fam. in the Ordovician of Bohemia. Věstník Ústředního Ústavu I, F. W. Harrison & A. J. Kohn (eds), Wiley-Liss, New York. Geologického [Bulletin of the Geological Survey, Prague], 37: 373- English, A. M. 2002. Th e enigmatic mollusk Matthevia variabilis 375. (Walcott, 1885): new insights regarding the morphology and Middendorf, A. T. von. 1848 Beschreibung und Anatomie Ganz systematic relationships of a probable Late Cambrian polypla- Neuer, oder für Russland Neuer Chitonen. Beiträge zu einer cophoran. Northeastern Geology and Environmental Sciences, 24: Malacozoologica Rossica (aus den Mémoires Sciences Naturelles Tome 293-297. VI de l'Academie Impériale des Sciences Besonders Abgedruckt). Fedonkin, M. A., A. Simonetta, & A. Y. Ivantsov. 2007. New data Gedruckt bei der Kaiserlichen Academie der Wissenschaften, St. on Kimberella, the Vendian mollusc-like organism (White Sea Petersburg, 151 pp. (Title page dated 1847, but penciled nota- region, Russia): paleoecological and evolutionary implications. tion in W. H. Dall's copy reads "issued Feb. 1848.") Pages 157-179, in: Th e Rise and Fall of the Ediacaran Biota, P. Moseley, H. N. 1885. On the presence of eyes in the shells of cer- Vickers-Rich & P. Komarower (eds), Geological Society of tain Chitonidae. Quarterly Journal of Microscopical Science, 25: London, Special Publication 286. 37-60. Fedonkin, M. A., & B. M. Waggoner. 1997. Th e Late Precambrian Neilson, W. A., T. A. Knott, & P. W. Carhart. 1951. Webster's New fossil Kimberella is a mollusc-like bilaterian organism. Nature, International Dictionary of the English Language, 2nd ed. G. and 388: 868-871. C. Merriam Company, Springfi eld, Massachusetts, 3,214 pp. Fernandez, C. Z., M. J. Vendrasco, & B. Runnegar. 2007. Aesthete Nielsen, C. 1991. Th e development of the brachiopod Crania canal morphology in twelve species of chiton (Polyplacophora). (Neocrania) anomala (O. F. Müller) and its phylogenetic signifi - Th e Veliger, 49: 51-69. cance. Acta Zoologica, 72: 7-28. Flower, R. H. 1968. Some El Paso guide fossils. New Mexico Bureau Pilsbry, H. A. 1892-1893. Monograph of the Polyplacophora. of Mines and Mineral Resources, Memoir 22: 1-19. Manual of Conchology, 14: 350 pp. Glaessner, M. F., & M. Wade. 1966. Th e Late Precambrian fossils Pojeta, J., Jr., & J. R. Derby. 2007. Dycheia shergoldi, a new genus and from Ediacara, South Australia. Palaeontology, 9: 599-628. species of Late Cambrian multivalved mollusc from Oklahoma, Gray, J. E. 1847. On the genera of the family Chitonidae. Proceedings U.S.A. Pages 489-497, in: Papers in Honour of John H. Shergold of the Zoological Society of London, 15(174): 70. 1938-2006, J. R. Laurie & J. R. Paterson (eds), Association of Haas, W. 1981. Evolution of calcareous hardparts in primitive mol- Australasian Palaeontologists Memoir 34. luscs. Malacologia, 21: 403-418. Pojeta, J., Jr., & J. DuFoe. 2008. New information about Echinochiton Hoare, R. D. 2000. Considerations on Paleozoic Polyplacophora dufoei, the Ordovician spiny chiton. American Malacological including the description of Plasiochiton curiosus n. gen. and sp. Bulletin, 25: 25-34. American Malacological Bulletin, 15: 131-138. Pojeta, J., Jr., D. J. Eernisse, R. D. Hoare, & M. D. Henderson. Hoare, R. D. 2002. New genera of Paleozoic Polyplacophora 2003. Echinochiton dufoei: a new spiny Ordovician chiton. 26 Bulletins of American Paleontology, No.
Journal of Paleontology, 77: 646-654. the Oklahoma Academy of Science, 45: 94-98. Pojeta, J., Jr., J. Gilbert-Tomlinson, & J. H. Shergold. 1977. Sanders, R. B. 1966. Th e incertae sedis group tegmates revealed as Cambrian and Ordovician rostroconch mollusks from Northern plates of a chiton. Journal of Paleontology, 40: 1240-1241. Australia. Australian Bureau of Mineral Resources, Geology, and Scotese, C. R. 2001. Atlas of Earth History, Volume 1, Paleogeography. Geophysics Bulletin 171: 54 pp. PALEOMAP Project, Arlington, Texas, 52 pp. Pojeta, J., Jr., & B. Runnegar. 1976. Th e paleontology of rostro- Seslavinsky, K. B., & I. D. Maidanskaya. 2001. Global facies dis- conch mollusks and the early history of the phylum Mollusca. tributions from Late Vendian to Mid-Ordovician. Pages 47-68, United States Geological Survey Professional Paper 968: 88 pp. in: Th e Ecology of the Cambrian Radiation, A. Y. Zhuravlev &R. Pojeta, J., Jr., J. H. Taylor, & G. Darrough. 2005. Matthevia Riding (eds), Columbia University Press, New York. (Polyplacophora) invades the Ordovician: the fi rst reported Shergold, J. H. 1996. Cambrian (Chart 1). Pages 63-76, in: An post-Cambrian occurrence. Journal of Paleontology, 79: 1021- Australian Phanerozoic Timescale, G. C. Young & J. R. Laurie 1027. (eds), Oxford University Press, Melbourne. Pojeta, J., Jr., R. Zhang, & Y. Yang. 1986. Systematic paleontology Sigwart, J. D. 2007. Th e Irish fossil Polyplacophora. Irish Journal of of the Devonian pelecypods of Guangxi and Michigan. United Earth Sciences, 25: 27-38. States Geological Survey Professional Paper 1394G: 57-108. Sigwart, J. D., & M. D. Sutton. 2007. Deep molluscan phylogeny: Raymond, P. E. 1913. A revision of the species which have been synthesis of palaeontological and neontological data. Proceedings referred to the genus Bathyurus. Victoria Memorial Museum of the Royal Society of London, 274B: 2413-2419. Bulletin, 1: 51-69. Sirenko, B. 2006. New outlook on the system of chitons (Mollusca: Rhoads, D. C. 1962. Microfossils of problematic affi nity from the Polyplacophora). Venus, 65: 27-49. Maquoketa Formation of eastern Iowa and western Illinois. Skovsted, C. B., J. S. Peel, & C. J. Atkins. 2004. Th e problematic fos- Journal of Paleontology, 36: 1334-1340. sil Triplicatella from the Early Cambrian of Greenland, Canada, Robson, G. C. 1913. Helminthochiton aequivoca, n. sp., Lower and Siberia. Canadian Journal of Earth Sciences, 41: 1273-1283. Ordovician, Bohemia. Geological Magazine, decade 10, 10: 302- Smith, A. G. 1960. Amphineura. Pages I41-I76, in: Treatise on 304. Invertebrate Paleontology, R. C. Moore (ed.), Part I, Mollusca Rolfe, W. D. I. 1967. Septemchiton Bergenhayn = Solenocaris. Page 1, by J. B. Knight et al., Geological Society of America and 423, in: Th e Fossil Record, W. B. Harland, C. H. Holland, M. R. University of Kansas Press, Lawrence. House, N. F. Hughes, A. B. Reynolds, M. J. S. Rudwick, G. E. Smith, A. G. 1964. Systematic descriptions. In: Chitons From the Satterthwaite, L. B. H. Tarlo, & E. C. Willey (eds), Geological Kindblade Formation, by A. G. Smith & D. F. Toomey, Oklahoma Society of London. Geological Survey Circular 66: 20-41. Rolfe, W. D. I. 1981. Septemchiton – a misnomer. Journal of Paleon- Smith, A. G., & R. D. Hoare. 1987. Paleozoic Polyplacophora: a tology, 55: 675-678. checklist and bibliography. Occasional Papers of the California Rowland, S. M., & R. S. Shapiro. 2002. Reef patterns and envi- Academy of Sciences, 146: 71 pp. ronmental infl uences in the Cambrian and earliest Ordovician. Smith, D. G. 1983. On the so-called mantle muscle scars on shells of SEPM Special Publication 72: 95-128. the Margaritiferidae (Mollusca, Pelecypoda), with observations Runnegar, B. 1987. Class Monoplacophora. Pages 297-304, in: Fossil on mantle-shell attachment in the Unionidae and Trigonioida. Invertebrates, R. S. Boardman, A. H. Cheetham, & A. J. Rowell Zoologica Scripta, 12: 67-71. (eds), Blackwell Scientifi c Publications, Palo Alto, California. Stait, T., & C. Burrett. 1984. Early Ordovician polyplacophoran Runnegar, B. 1996. Early evolution of the Mollusca: the fossil re- Chelodes whitehousei from Tarutao Island, southern Th ailand, cord. Pages 77-87, in: Origin and Evolutionary Radiation of the Alcheringa, 8: 112. Mollusca, J. D. Taylor (ed.), Oxford University Press, U. K. Stinchcomb, B. L. 1965. New Mollusca of the Potosi and Eminence Runnegar, B., J. Pojeta, Jr., M. E. Taylor, & D. Collins. 1979. New Formations (Cambrian). Master's Th esis, Washington University, species of Cambrian and Ordovician chitons Matthevia and Saint Louis, Missouri, 53 pp. Chelodes from Wisconsin and Queensland: evidence for the early Stinchcomb, B. L. 1975. Paleoecology of two new species of Late history of polyplacophoran mollusks. Journal of Paleontology, 53: Cambrian Hypseloconus (Monoplacophora) from Missouri. 1374-1394. Journal of Paleontology, 49: 416-421. Runnegar, B., & M. J. Vendrasco. 1997. Early Paleozoic stem group Stinchcomb, B. L. 1980. New information on Late Cambrian chitons from Utah and Missouri. No Problematica! Program and Monoplacophora Hypseloconus and Shelbyoceras (Mollusca). Abstracts, Combined Annual Meeting, American Malacological Journal of Paleontology, 54: 45-49. Union and Western Society of Malacologists, Santa Barbara, Stinchcomb, B. L. 1986. New Monoplacophora (Mollusca) from California, p. 52. Late Cambrian and Early Ordovician of Missouri. Journal of Salter, J. W. 1846. Addenda. Pages 71-72, in: A Synopsis of the Paleontology, 60: 606-626. Silurian Fossils of Ireland, F. M'Coy (ed.), University Press, Stinchcomb, B. L., & G. Darrough. 1995. Some molluscan Dublin, Ireland. Problematica from the Upper Cambrian-Lower Ordovician of Salter, J. W. 1847. Description of a fossil chiton from the Silurian the Ozark uplift. Journal of Paleontology, 69: 52-65. rocks, with remarks on the fossil species of the genus. Quarterly Stitt, J. H. 1977. Late Cambrian and earliest Ordovician trilobites, Journal of the Geological Society of London, 3: 48-52. Wichita Mountains Area, Oklahoma. Oklahoma Geological Sanders, R. B. 1965. A new species of Septemchiton. Proceedings of Survey Bulletin 124: 79 pp. POJETA ET AL.: UPPER CAMBRIAN CHITONS 27
Sutton, M. D., D. E. G. Briggs, D. J. Siveter, & D. J. Siveter. cephalinae. Smithsonian Miscellaneous Collections, 57: 345-512. 2004. Computer reconstruction and analysis of the vermiform Weedon, M. J., & P. D. Taylor. 1995. Calcitic nacreous ultrastruc- mollusc Acaeoplax hayae from the Herefordshire Lagerstätte tures in bryozoans: implications for comparative biomineraliza- (Silurian, England), and implications for molluscan phylogeny. tion of lophophorates and mollusks. Biological Bulletin, 188: Palaeontology, 47: 293-318. 281-292. Swoff ord, D. L. 2000. PAUP*. Phylogenetic Analysis Using Parsimony Williams, A. 1997a. Shell structure. Pages 267-320, in: Treatise (*and Other Methods), Version 4. Sinauer Associates, Sunderland, on Invertebrate Paleontology, R. L. Kaesler (ed.), Part H revised, Massachusetts. Brachiopoda 1, by A. Williams et al., Geological Society of Taylor, M. E., & R. B. Halley. 1974. Systematics, environment, and America and University of Kansas, Lawrence. biogeography of some Late Cambrian and Early Ordovician Williams, A. 1997b. Brachiopoda: introduction and integumentary trilobites from New York State. United States Geological Survey system. Pages 237-296, in: Microscopic Anatomy of Invertebrates, Professional Paper 834: 38 pp. Volume 13: Lophophorates, Enteroprocta, and Cyclophora, F. W. Th ompson, T. L. 1991. Paleozoic succession in Missouri, Part 2, Harrison & R. M. Woollacott (eds), Wiley-Liss, New York. Ordovician System. Missouri Division of Geology and Land Williams, A., M. A. James, C. C. Emig, S. Mackay, & M. C. Survey Report of Investigations, 70: 262 pp. Rhodes. 1997. Anatomy. Pages 7-188, in: Treatise on Invertebrate Ulrich, E. O. 1930. Entomaspis and Triarthropsis. In: Geology of the Paleontology, R. L. Kaesler (ed.), Part H revised, Brachiopoda 1, by Eminence and Cardareva Quadrangles, by J. Bridge. Missouri A. Williams et al., Geological Society of America and University Bureau of Geology and Mines, 24: 212-215. of Kansas, Lawrence. Ulrich, E. O., & J. Bridge. 1941. Hemithecella expansa n. gen. and Woodward, H. 1885. On a new species of Helminthochiton from the sp. In: Geology of the Appalachian Valley in Virginia, by C. Butts. Upper Bala (Silurian) of Girvan, Ayrshire. Geological Magazine, Virginia Geological Survey Bulletin 52(2): 19. new series, decade 3, 2: 352-358. Ulrich, E. O., & G. A. Cooper. 1936. New genera and species of Yates, A. M., K. L. Gowlett-Holmes, & B. J. McHenry. 1992. Ozarkian and Canadian brachiopods. Journal of Paleontology, 10: Triplicatella disdoma Conway Morris, 1990, reinterpreted as the 616-631. earliest known polyplacophoran. Journal of the Malacoplogical Ulrich, E. O., & G. A. Cooper. 1938. Ozarkian and Canadian Society of Australia, 13: 71. Brachiopoda. Geological Society of America Special Paper 13: 323 Yi, Q., & S. Bengtson. 1989. Palaeontology and biostratigrphy of pp. the Early Cambrian Meishucunian Stage in Yunnan Province, Van Belle, R. A. 1975a. Sur la classifi cation des Polyplacophora, Part south China. Fossils and Strata, 24: 156 pp. I. Informations de la Societe Belge de Malacologie, serie 4, 5: 121- Yochelson, E. L. 1966. Mattheva, a proposed new class of mollusks. 131. United States Geological Survey Professional Paper 523B: 11 pp. Van Belle, R. A. 1975b. Sur la classifi cation des Polyplacophora, Part Yochelson, E. L., J. F. McAllister, & A. Reso. 1965. Stratigraphic II. Informations de la Societe Belge de Malacologie, serie 4, 6: 135- distribution of the Late Cambrian mollusk Matthevia Walcott, 145. 1885. United States Geological Survey Professional Paper 525B: Vendrasco, M. J. 1999. Early Evolution of the Polyplacophora (Chitons). 73-78. Unpublished Ph.D. dissertation, University of California, Los Yochelson, E. L., & B. L. Stinchcomb. 1987. Recognition of Angeles, 232 pp. Macluritella (Gastropoda) from the Upper Cambrian of Missouri Vendrasco, M. J., & B. Runnegar. 2004. Late Cambrian and Early and Nevada. Journal of Paleontology, 61: 56-61. Ordovician stem group chitons (Mollusca: Polyplacophora) Yochelson, E. L., & M. E. Taylor. 1974. Late Cambrian Matthevia from Utah and Missouri. Journal of Paleontology, 78: 675-689. (Mollusca, Matthevida) in North America. Geological Society of Vendrasco, M. J., C. Z. Fernandez, D. J. Eernisse, & B. Runnegar. America, Northeastern Section, Abstracts with Program, 6(7): 88. 2008a. Aesthete canal morphology in the Mopalidae (Polyplaco- Young, & J. Young [no initial or fi rst name was provided for the phora). American Malacological Bulletin, 25: 51-70. fi rst author]. 1868. Solenocaris solenoides. In: On New Forms of Vendrasco, M. J., G. Li, S. M. Porter, & C. Z. Fernandez. 2008b. Crustacea, from Silurian Rocks at Girvan, by J. Young, Proceeding A reassessment of the enigmatic Ocruranus-Eohalobia group of the Natural History Society of Glascow, 1: 173. of small skeletal fossils From the Early Cambrian of eastern Yu, W. 1984. Early Cambrian molluscan faunas of Meishucun Stage Yunnan, China. PaleoBios, 28: 7-8. with special reference to Precambrian-Cambrian boundary. Vendrasco, M. J., S. M. Porter, A. Kouchinsky, G. Li, & C. Z. Pages 21-33, in: Academia Sinica, Developments in Geoscience,. Fernandez. 2010. New data on molluscs and their microstruc- Contribution to the 27th International Geological Congress, Moscow. ture from the Middle Cambrian Gowers Formation, Australia, Science Press, Beijing. Palaeontology, 54: 97-136. Yu, W. 1987. New molluscan materials of the Tethys. Pages 51-59, Vinther, J., & C. Nielsen. 2005. Th e Early Cambrian Halkieria is a in: Proceedings International Symposium on Shallow Tethys, K. G. mollusc. Zoologica Scripta, 34: 81-89. McKenzie (ed.), A. A. Ballkema, Rotterdam, 2. Walcott, C. D. 1885. Note on some Paleozoic pteropods. American Yu, W. 1993. [On the diff erent affi nities between Yangtzechiton Yu Journal of Science, 30: 17-21. and Paracarinachites Qian et Jiang.] Acta Micropalaeontologica Walcott, C. D. 1912. Th e Cambrian Faunas of China, Vol. 3. Carnegie Sinica, 10: 307-316. In Chinese, with English summary. Institution of Washington, Washington, DC, 54: 375. Yu, W. 2001. Th e earliest Cambrian polyplacophorans from China. Walcott, C. D. 1914. Dikelocephalus and other genera of the Dikelo- Records of the Western Australian Museum, 20: 167-185. 28 Bulletins of American Paleontology, No.
APPENDIX 1 (1966: 1240) corrected his nomenclatural error and syn- REVIEW OF PREVIOUSLY USED FAMILY onymized Rhoads' genera with S. iowensis: "I believe that NAMES FOR CAMBRIAN AND Rhoads' three incertae sedis genera, Triangulata, Bursata, and Subcylindrica, represent the anterior plate, intermediate plate, ORDOVICIAN CHITONS and a weathered intermediate plate of the chiton Septemchiton iowensis... Th e structure termed 'columella' by Rhoads was Previous to this study, at least 14 family names have been used found to be a phosphatized scaphopod-like form fortuitously for Cambrian and Ordovician chitons. As noted above, (1) emplaced under the arch of the shell. Th e species name is de- Preacanthochitonidae Bergenhayn, 1960, (2) Hemithecellidae rived from Rhoads' Bursata iowensis, and the correct citation Stinchcomb & Darrough, 1995, (3) Robustidae, and (4) should be Septemchiton iowensis (Rhoads) rather than as ap- Eochelodidae Marek, 1962, are regarded as junior subjective peared earlier (Sanders, 1965)." synonyms of (5) Matheviidae Walcott, 1885. Mattheviids Rhoads' (1962) illustrations of the tegmates are better ex- have large apical areas and lacunae both of which are lacking ecuted than Sanders' (1965) illustrations of Septemchiton io- in aulochitonids. wensis. Th e presumed anterior plate (head valve = Triangulata (6) Echinochitonidae Pojeta et al., 2003, includes taxa hav- simplex Rhoads, 1962: pl. 178, fi gs 8, 11; Sanders, 1965: fi gs ing a large lacuna and apical area, but diff ers from mattheviids 2, 5) is rectangular in dorsal profi le with slightly concave par- in the possession of large circumsomal spines. Th e head and allel sides; one end has a sinus and the other end is straight. tail valves each have four spines and each intermediate valve Th ere is a dorsal ridge, and Rhoads noted that both the cross has a pair of spines (Pojeta & DuFoe, 2008). To date, echi- section and sagittal section are triangular. Sanders also men- nochitonids are known only from Middle Ordovician rocks tioned the presence of punctae in this valve. Th is morphol- in the state of Wisconsin, USA. Aulochitonids lack the large ogy diff ers from head valves of known lower Paleozoic chitons circumsomal spines, as do all of the other families discussed (Runnegar et al., 1979; Rolfe, 1981; Vendrasco & Runnegar, here. 2004; Hoare & Pojeta, 2006; Sigwart, 2007) in which head (7) Septemchitonidae Bergenhayn, 1955, was based on valves usually have rounded anterior margins and are gently the mistaken impression that Septemchiton Bergenhayn, convex to fl at. Rolfe (1981) regarded what Sanders called the 1955, had only seven valves. Rolfe (1981) restudied the ar- anterior plate of S. iowensis as more like valve 2 of S. vermi- ticulated specimens in the type series and topotypes of S. ver- fromis and noted that the head valve of S. iowensis probably miformis Bergenhayn, 1955, and demonstrated the presence remains yet to be found. of a small eighth (head) valve. As noted by Rolfe (1981: 675), Sanders (1966) described punctae in the presumed inter- Bergenhayn's illustrations are poorly executed and he did not mediate and posterior plates (valves) of Septemchiton iowensis, understand the preservation of the specimens. Septemchiton which are prominent in fi gure 3 of Sanders (1965); if the so- vermiformis occurs in Upper Ordovician rocks of Scotland. called punctae can be shown to be openings to esthete canals, From the published literature, it is not possible to determine as assumed by Sanders, then S. iowensis is most likely a chi- the extent of the apical area in S. vermiformis, or if lacunae ton. were present. Based on Rolfe's (1981) illustrations of a topo- Th e dorsal surface of the intermediate valves of Bursata io- type, the intermediate valves end posterodorsally in a pointed wensis Rhoads (1962: pl. 178, fi gs 13-14) has an upper raised horizontal apex. In lateral profi le, valves 4-7 are elongate rect- triangular area and, posteriorly, a lower fl ange-like area (bursa angular in shape, having a straight ventral margin that paral- of Rhoads, 1962) to each side; this is unlike other known lels the dorsal margin; the posterior margin is erect, and the Cambrian and Ordovician chitons. Th e lateral views of the in- anterior margin of valve 4 slopes forward. In the topotype termediate valves of B. bellevuensis Rhoads (1962: pl. 178, fi gs fi gured by Rolfe (1981), valves 1-3 have been moved diage- 9-10) show growth increments and a profi le very similar to netically and they are slightly out of alignment with respect to valve 4 of Septemchiton vermiformis (see Rolfe, 1981: text-fi g. valves 4-7, thus, their shape in relation to valves 4-7 is unclear. 1). Subcylindrica laddi Rhoads (1962: pl. 178, fi gs 6-7) has a In comparison, the intermediate valves of aulochitonids are large tuberculate knob at one end, which Rhoads termed the strongly mucronate and triangular in lateral profi le. hood; Sanders (1965, 1966) made no mention of this knob, Sanders (1965, 1966) discussed Septemchiton iowensis but he synonymized S. laddi with S. iowensis. If this knob rep- (Rhoads, 1962). Th e species is based on disarticulated valves resents a mucro, it is very diff erent from what Sanders (1965: from Upper Ordovician rocks in Iowa. Rhoads (1962) de- fi gs 9-10) termed the posterior plate (valve). More knowledge scribed three genera, Bursata, Subcylindrica, Triangulata, and of the type species of Septemchiton is needed before the ge- six species of what he called tegmates, which he regarded as neric name, and thus the family Septemchitonidae, can be Problematica. Sanders (1965) restudied the tegmates and as- conceptualized and understood. signed them to Septemchiton iowensis Sanders, 1965. Sanders POJETA ET AL.: UPPER CAMBRIAN CHITONS 29
Dzik (1986) reclassifi ed Helminthochiton aequivoca Dzik did not designate a type genus for the Solenocarididae, Robson, 1913, as a species of Septemchiton. Dzik (1986: 130) the name satisfi es the provisions of the International Code reconstructed S. aequivoca in: "Longitudinal section and dor- of Zoological Nomenclature (1999) Articles 13.1, 13.2, and sal view. Based on several articulated but incomplete speci- 16.2, for family names proposed before January 1, 2000 (M. mens from the Šárka Formation [Middle Ordovician], Osek, J. Grygier, ICZN, Commissioner, written communication, Bohemia (housed at the Narodní Museum, Prague)." He also May 2008). As for the spelling of the family name, based on noted that "Septemchiton had roof shaped valves with fl at areas the type genus Solenocaris, Grygier noted: "As for the stem, and narrow posterior duplicature [apical area]." Dzik did not Dzik's family name is correctly formed. Th e genitive of Latin fi gure any of the Czech specimens; he did fi gure a very small 'caris' (shrimp) is 'caridis', so the stem for the family-name intermediate valve as S. aff . aequivoca that has a small apical formation is carid-, thus giving rise to ... carididae." area; this specimen is from the Ordovician Mójcza Limestone Rolfe (1967) proposed synonymizing Septemchiton with of Poland. It is not clear why Dzik placed H. aequivoca in Solenocaris; however, in 1981, Rolfe reversed this proposal Septemchiton because, as noted above, based on the type spe- and considered them to be separate genera. cies, the apical area of Septemchiton is unknown and Rolfe Th e type species of Solenocaris is S. solenoides Young & (1981) made no mention of the apical area. Young, 1868, from the Upper Ordovician rocks of Scotland, Dzik (1994a: 221) diagnosed the Septemchitonidae as was originally described as a crustacean. Young & Young "Valves V-shaped in cross section and with smooth inner sur- (1868: 171) fi gured a single valve (the holotype) and noted face, the tail valve being of similar length as the intermedi- the presence of an eye spot. Th ey provided an outline draw- ate ones." Rolfe (1981: fi g. 1) showed that the dorsal surfaces ing of a horizontal section through the middle of the valve; of valves 1-4 in the type species Septemchiton vermiformis are this drawing shows a swollen posterior end, constricted from a not V-shaped, and the tail valve is longer than all the other narrower anterior end, and does not look like sections through valves. Dzik made Helminthochiton aequivoca the type species chiton valves (Runnegar et al., 1979). of the genus Sarkachiton Dzik, 1994a, and placed it in the Woodward (1885) named Helminthochiton grayiae from Septemchitonidae. In reference to his 1986 paper, Dzik noted Upper Ordovician rocks of Scotland. He compared H. grayiae (1994a: 222): "Among several specimens of the type species to Solenocaris and noted (1885: 357), "After careful study of housed at the Narodni Muzeum [sic], Prague, at least two are Messrs. Young's [sic] Solenocaris, and comparison with the articulated, but not more than four valves are in the original Chiton-like series of plates above noticed [H. grayiae], we arrangement. Th e head valve has not been found and its shape are led to conclude that there is general agreement between given in the reconstruction of the whole armor (Dzik, 1986: them..." Woodward went on to explain how to tell Solenocaris fi g. 14C) is quite conjectural." As in 1986, Dzik did not fi gure from Helminthochiton. In the explanation of his plate 9, fi g- the type specimens of H. aequivoca; Robson's (1913) original ure 11, Woodward listed "Helminthochiton (Solenocaris) so- fi gures are crude line drawings. lenoides, Young and Young." Woodward's fi gure is a lateral In addition to the type species of Sarkachiton, Dzik view and looks much like the lateral view shown by Young (1994a) named S. keilcensis, based on the specimen that he & Young (1868: 171); however, if Young & Young's fi gure is fi gured as Septemchiton aff . aequivoca in 1986. In the diagno- a left-lateral view, Woodward's fi gure is a right-lateral view. sis of the species, he mentioned that the length of the interme- Both fi gures show prominent comarginal ornament and a diate valves is similar to the height, but that the probable tail prominent anterodorsal to posteroventral radial rib. valve is approximately two times as long as high; however, he Woodward studied new specimens that he assigned to fi gured only the intermediate valve showing the small apical Solenocaris solenoides, not Young & Young's holotype, and area and the V-shaped cross section. Dzik (1994b: 280) reiter- Woodward did not fi nd an eye spot. Smith & Hoare (1987: ated his concept of septemchitonids having "angularly bent, 50) wrote that the location of the original specimen of S. sole- roof-shaped valves." noides Young & Young is unknown. Until the type species of Septemchiton is better understood, Cowper Reed (1907) fi gured a specimen of Solenocaris the Septemchitonidae remains a name looking for a concept. solenoides having a greatly elongated tail valve attached to Hoare (2000: 134) proposed the elimination of the family four more anterior valves; the tail valve termination is acute. Septemchitonidae Bergenhayn (1955). Based on this specimen, Cowper Reed (1907: 112) tenta- (8) Th e Solenocarididae Dzik (1994a: 222) was defi ned as: tively suggested of S. solenoides, "In the face of this new evi- "Valves U-shaped in cross section with pseudoporus (aesthe- dence it may be well to hesitate to refer this fossil to the genus tae?) wall, strongly elongated, especially the acute tail valve." Helminthochiton; and with the suspicion that it may ultimate- Dzik placed Bursata Rhoads, 1962, Carnicoleus Dzik, 1986, ly have to fi nd its place among the Phyllocarida [Crustacea] and Solenocaris Young & Young, 1868, in the family. Although we may return to the old generic name Solenocaris..." 30 Bulletins of American Paleontology, No.
Rolfe (1981) provided a camera lucida drawing of the crebatus Hoare & Pojeta, 2006, is from Lower Ordovician specimen of Solenocaris solenoides fi gured by Cowper Reed rocks in the state of Minnesota, USA. Helminthochiton mar- (1907), and Rolfe summarized information subsequent to ginatus Hoare & Pojeta, 2006, from the Middle Ordovician Cowper Reed about the higher systematic placement of S. so- rocks of the state of Wisconsin, USA, is here transferred to lenoides. Rolfe regarded the species as a chiton. Litochiton as L. marginatus (Hoare & Pojeta, 2006), because Dzik (1994a, b) included the genus Bursata Rhoads, 1962, in dorsal profi le it has rectangular intermediate valves. Th e in the Solenocarididae and named the species B. santacruensis intermediate valves of aulochitonids are elongately rounded from the Upper Ordovician of Poland. Dzik made no mention and not rectangular. of Sanders' (1965, 1966) reclassifi cation of Rhoads' (1962) (11) Helminthochitonidae Van Belle, 1975b (originally tegmate genera, although Dzik did cite Sanders' (1966) pa- proposed as a subfamily of Lepidopleauridae Pilsbry, 1892; per in his bibliography. In B. santacruensis, Dzik mentioned subsequently used as a family name by Hoare, 2002). Th e the presence of pseudopores and reconstructed the species type genus Helminthochiton Salter, 1846, is based on the type with rectangular lateral intermediate valve profi les similar to species H. griffi thi Salter, 1846, from the Lower Silurian of Septemchiton vermiformis except that they have fringed poste- Ireland; in the description of the genus, Salter (1846: 71) not- rior edges. Also, the tail valve of B. santacruensis diff ers from S. ed: "Elongate, plates [valves] as long or longer than wide..." vermiformis in being sharply pointed and elongated triangular In 1847 (p. 51), in the description of Helminthochiton, Salter in lateral profi le; ventrally, the valves are open and lack apical wrote: "Elongate; plates [valves] as long as wide, subquad- areas. Th e intermediate valves of B. santacruensis are inverted rate..." Th us, it is not clear what Salter had in mind as to valve U-shaped and lack the "bursae" of B. iowensis. shape. His 1846 illustrations (pl. 5, fi gs 5a-e) are better and Dzik's main taxobasis for separating Septemchitonidae more extensive than the diagrammatic 1847 fi gure. from Solenocarididae is the shape of the valves in cross sec- Th e name Helminthochiton has been used for species rang- tion, an inverted V-shape in Septemchitonidae and an in- ing in age from Early Ordovician to Pennsylvanian (Smith, verted U-shape in Solenocarididae. A secondary taxobasis 1960; Sigwart, 2007) and for species that both lack and pos- is the shape of the tail valves. However, for the holotype of sess sutural laminae formed by the articulamentum layer of Solenocaris solenoides, Young & Young (1868: pl. 1, fi g. 7b) the shell. Smith & Hoare (1987) limited Helminthochiton showed a horizontal section through the middle of its height, to lower Paleozoic species and assigned upper Paleozoic spe- with a marked constriction separating the anterior and poste- cies to the genus Gryphochiton Gray, 1847. Hoare (oral com- rior parts of the valve; this diff ers from any known Cambrian munication, April 2008) noted that Helminthochiton is used or Ordovician chiton. for pre-Mississippian species that lack sutural laminae, and Th e Solenocardidae shows the problems of basing a fam- Gryphochiton is used for Mississippian and Pennsylvanian ily name on a poorly understood genus, in which the type species that have sutural laminae. Sirenko (2006) placed specimen of the type species of the genus cannot be located. Helminthochiton in the Paleoloricata. It would be better to base the new family name on a well- Sigwart (2007) located and restudied the holotype of known genus of which the type specimens of the type species Helminthochiton griffi thi; the specimen preserves a terminal are known to be in a museum collection. valve, three nearly complete intermediate valves, and a fourth Th e families Alastegiidae Hoare & Pojeta, 2006, and incomplete intermediate valve. Salter (1846: 71) was con- Litochitonidae Hoare & Pojeta, 2006, both have intermedi- fused about the terminal valve. In his description of the genus ate valves with an anterior sinus, and apical areas reduced to Helminthochiton, he called it an anal plate; however, in the the posterior and posterolateral margins. description of H. griffi thi, he called it a cephalic plate. Sigwart (9) Th e Alastegiidae includes genera with triangular dorsal (2007) clearly demonstrated that the terminal valve is a head profi les in the intermediate valves. In addition to the two gen- valve. Th e posterior end of all valves of the holotype are in- era placed in this family by Hoare & Pojeta, 2006, we include complete and this makes it diffi cult to measure valve length. Listrochiton Hoare & Pojeta, 2006, in the Alastegiidae because When the posterior margin of valve 4 is restored to what of its strongly triangular intermediate plates. To date, alastegi- Sigwart (2007: 29) thought would be a straight margin, the ids are known only from Middle Ordovician rocks in the state length measurement of the valve is subequal with the width of Kentucky, USA. In dorsal profi le, the intermediate valves of measurement, and the valves are almost square in dorsal pro- aulochitonids are elongately rounded and not triangular. fi le. Sigwart noted that the holotype is a part and counterpart (10) Th e Litochitonidae includes taxa having intermediate of the dorsal surface and not the ventral surface as assumed by valves that in dorsal profi le are rectangular, having the length Salter (1846) and subsequent authors; thus, it is not possible to greater than the width. Th is monotypic family contains two know whether sutural laminae were present or if an apical area species of the genus Litochiton Hoare & Pojeta, 2006. Litochiton was present. Th e concept behind the name Helminthochiton POJETA ET AL.: UPPER CAMBRIAN CHITONS 31 has yet to be properly defi ned and awaits the fi nding of addi- these plates show structural diff erences from polyplacoph- tional specimens of the type species H. griffi thi. In the mean- orans such as plate growth patterns, microstructure, and sup- time, the suggestions of Smith and Hoare mentioned above posed insertion plates." We agree with Hoare, that it is doubt- provide a temporary way to proceed for species with interme- ful that cobcrephorids are polyplacophorans. diate valves that have subequal width and length and that lack In summary, in our opinion, the family names sutural laminae. Two species from Middle Ordovician rocks Septemchitonidae, Helminthochitonidae, Solenocarididae, in the states of Wisconsin and Kentucky, USA, are placed in and Gotlandochitonidae are not well conceptualized; in large Helminthochiton: Helminthochiton? baueri (Hoare & Pojeta, part, this is because the type genera and the type species of 2006) and H. blacki Hoare & Pojeta, 2006. those genera are poorly understood. Th e Llandeilochitonidae (12) Gotlandochitonidae Bergenhayn, 1955: Tradition- is no longer regarded as being a member of the Polyplacophora. ally, this name has been used for chitons whose intermediate Th e Cobcrephoridae seems to be a doubtful member of the valves are wider than they are long and that have highly reduced Polyplacophora. Th e Preacanthochitonidae, Eochelodidae, apical areas (Bergenhayn, 1955; Smith, 1960, 1964). Cherns Robustidae, and Hemithecellidae are herein regarded as junior (1998b) provided an emended description of Gotlandochiton synonyms of Mattheviidae. Th e family names Mattheviidae, and restudied the type species G. interplicatus Bergenhayn, Echinochitonidae, Alastegiidae, and Litochitonidae are well 1955. Th e holotype of G. interplicatus is from Middle Silurian conceptualized and based on type genera whose type species rocks in Sweden; it is incomplete in the right posterolateral are well known; to this list we add the Aulochitonidae. area and has a width of 12.3 mm and length of 11.3 mm; thus it is subquadrate. Bergenhayn regarded the apical area as be- APPENDIX 2 ing narrow, however, Cherns (1998b: 944) noted that the api- DETAILS OF THE CLADISTIC ANALYSIS cal area is apparently wide. Th us, the concepts implied by the names Gotlandochiton and Gotlandochitonidae are not clear. REASONING FOR TAXA INCLUDED AND OUTGROUP CHOICE In Ordovician rocks, Smith (1964) named fi ve species All known Cambrian genera (Matthevia, Robustum, Chelodes, that he assigned to the Gotlandochitonidae from the Lower Calceochiton, Hemithecella, Aulochiton, Eukteanochiton, Ordovician Kindblade Formation of Oklahoma. All of the Dycheia, and Paradycheia) were included in the analysis to species have highly reduced apical areas and, in four of them, determine their relationships to each other and to test the the width is greater than the length of intermediate valves; hypothesis that they are chitons, or near chitons. In addi- however, in Paleochiton kindbladensis Smith, 1964, the length tion, all known Paleozoic articulated chitons were included of the valves is greater than the width. Hoare (2000) noted (Echinochiton, Glaphurochiton, Polysacos, and Septemchiton), the occurrence of Gotlandochiton Bergenhayn, 1955, in because these taxa have more preserved characters and show Ordovician rocks in the state of Minnesota, USA. more of the features of their skeletons. Th e extant chiton (13) Llandeilochitonidae Bergenhayn, 1955, contains the Lepidochitona was included so that a clear member of the single genus Llandeilochiton Bergenhayn, 1955, having the chiton crown group is represented in the analysis. In addi- single species L. ashbyi Bergenhayn, 1955, from the Middle tion, the well-preserved problematic fossils Halkieria and Ordovician of Scotland [not Australia as listed by Smith, 1960: Acaenoplax were included because they are regarded as having 72]. Th e holotype, and only known specimen, of L. ashbyi is a similarities to chitons. Th e other mollusks (Ribeira, represent- disarticulated valve. Smith (1960) placed Llandeilochiton and ing rostroconchs; Neopilina, representing monoplacophorans; the Llandeilochitonidae in the Insertae Sedis category. Smith Epimenia and Chaetoderma, representing the two groups of & Hoare (1987: 10) rejected Llandeilochiton as a polypla- aplacophorans) were included to determine if some of the cophoran, a position with which we agree. Late Cambrian multivalves share more similarities with other (14) Cobcrephoridae Bischoff , 1981. Bischoff (1981: 188) mollusks than with chitons. placed this family in the new order Phosphatoloricata, which Th e outgroup in the analysis is the Brachiopoda, represent- he defi ned as: "Primitive polyplacophorans with valves of pos- ed by Novocrania and Lingula. Th e brachiopods were chosen tulated periostracum and two layers, tegmentum and hypos- as the outgroup because of similarities in shell morphology tracum, both composed of numerous alternating organic and with mollusks, and because brachiopods are one of the lopho- apatitic [phosphatic] lamellae." trochozoans that is estimated to be closely related to mollusks. No known undoubted polyplacophorans have phosphat- A recent phylogenetic analysis of the Metazoa (Dunn et al., ic shells. In discussing the Cobcrephoridae, Hoare (2000: 2008) indicated that brachiopods are within the clade that is 133) wrote: "Bischoff (1981) included Ordovician-Silurian the sister group to the Mollusca, suggesting that brachiopods phosphatic multiplated organisms from Australia in the are a good choice for an outgroup to the Mollusca. Polyplacophora. Besides being of a diff erent composition 32 Bulletins of American Paleontology, No.
Th e other taxa in this sister clade to the Mollusca are the 3. SEVEN TO EIGHT REPEATED STRUCTURES: 0 = Annelida, Echiura, Sipincula, Phoronida, Nemertea, and no; 1 = yes. Ectoprocta; the tree by Dunn et al. (2008) indicates that Comments.–Th is character refers to structures that are re- these taxa are not more closely related to mollusks than are peated along the anteroposterior axis, including valves, pedal the brachiopods. In addition, these other lophotrochozoans muscles, and scutes, but not spicules or halkieriid sclerites. lack prominent shells and so are not useful as outgroups in an analysis of fossils that must rely heavily on shell features. 4. SPICULES: 0 = absent; 1 = present. In addition, the shells of brachiopods and mollusks share Comments.–Spicules are solid needles of calcium carbon- a number of similarities that suggest homology. For example, ate (in particular, aragonite) that originate in single cells and brachiopods, like mollusks, have a thin shell-secreting man- become extracellular with growth. tle and an organic outermost shell layer, the periostracum. Although shell microstructures in mollusks tend to be more 5. MARGINAL SPINES: 0 = absent; 1 = present. complex and better organized in three dimensions than they Comments.–These are large, hollow skeletal elements that are in brachiopods, all shell microstructures seen in mollusks have growth lines resulting from accretionary growth and also occur in brachiopods (Carter, 1979). Weedon & Taylor which surround the valves. (1995) described striking similarities between the calcitic seminacre in brachiopods and the aragonitic nacre sensu stricto 6. VALVE OR VALVES: 0 = absent; 1 = present. in mollusks. Vendrasco et al. (2010) described cases of calcitic Comments.–Valves are shelly accretions formed by a man- seminacre shell microstructure in early mollusks. Th e shells of tle. Th e homology between the molluscan mantle and shell brachiopods (Williams, 1997b) and chitons (see above) are with those of brachiopods is discussed in a previous section infi ltrated with a canal system that houses organic tissue, the of Appendix 2. Th e mantle of brachiopods, like that of mol- caeca of brachiopods and the esthetes of chitons. lusks, consists of folded epithelial sheets that secrete the pe- riostracum and control biominrealization (Williams, 1997b). DESCRIPTIONS OF CHARACTERS AND STATES USED Th e shells of brachiopods are similar to those of mollusks in IN THE CLADISTIC ANALYSIS overall form and in shell microstructure (Carter, 1979). 1. BODY SHAPE: 0 = vermiform body; 1 = dorsoventrally fl attened body having ventral locomotory sole; 2 = U-shaped 7. SPICULE DISTRIBUTION: 0 = over entire body; 1 = just body enclosed between valves. in dorsolateral margins; 2 = no spicules. Comments.–Th is character refers to the overall body plan. Comments.–See defi nition of spicules under character 4 Character states 0 and 1 are inferred in fossils based on the above. nature of the shell valves; those taxa having anteroposteriorly elongated valves with high lateral sides, such as Septemchiton 8. VALVE DISPOSITION: 0 = unopposed and separated; 1 vermiformis, are regarded as having had a vermiform body. = unopposed and overlapping or nearly overlapping; 2 = op- Th ose taxa having valves that are erect or dorsoventrally re- posed; 3 = no valves; 4 = one valve. cumbent are regarded as having had a ventrally fl attened body Comments.–Th is character refers to how the valves are ori- such as Echinochiton dufoei (Text-fi g. 4.9) and Aulochiton san- ented relative to each other. Opposed valves refers to valves nerae (Pl. 18). Character state 2 is assigned to Novocrania and touching each other along a commissure as in brachiopods. Lingula based on the brachiopod fold hypothesis of Nielson Character states 3 and 4 are provided to avoid coding inappli- (1991), which was named and discussed by Cohen et al. cable character states as "?". Th is adds weighting of presence/ (2003). absence of valves, which strengthens the link between Epimena and Chaetoderma and leads to the exclusion of Acaenoplax 2. LEFT-RIGHT SKELETAL ZONES: 0 = no left-right from the aplacophoran clade. Dycheia and Paradycheia are zones; 1 = left-lateral, central, and right-lateral zones; 2 = left, coded as "?" because it is unclear whether or not there was left-central, right-central, and right zones. separation between the shell valves, but the valves might have Comments.–Th is character refers to the number of discrete been very close to each other. longitudinal, anterior-to-posterior shell bands having diff er- ent skeletal structures than that occuring in adjacent lateral 9. NUMBER OF VALVES IN CENTRAL ROW: 0 = one; 1 bands. Taxa scored "?" are done so because of uncertainty = two; 2 = seven; 3 = eight; 4 = no valves. about whether or not they had spicules, or any other, lateral Comments.–Th is character refers to the number of shell skeletal elements fl anking the shell valves. valves. In those taxa with multiple left-right zones of skeletal POJETA ET AL.: UPPER CAMBRIAN CHITONS 33 plates, such as Polysacos and Echinochiton, this character refers 12. RADULA: 0 = absent; 1 = present. to the number of valves in the central row. 13. LACUNAE: 0 = absent; 1 = present; 2 = no valves. 10. PRESENCE OF A BRANCHING LARGE CANAL Comments.–Lacunae are blind cavities within the interior SYSTEM IN THE VALVES THAT TERMINATES AT, OR of valves that open to the exterior at one end of the valve. See VERY NEAR, THE SHELL SURFACE: 0 = absent; 1 = pres- text for more details. ent; 2 = no valves. Comments.– Th is character refers to the canal system 14. INTERMEDIATE VALVE SHAPE: 0 = subconical, elon- within the valves. Th e many branched caeca or punctae that gate, and thick; 1 = elongate, subcylindrical, elongately trian- penetrate the calcareous shell of the cranioid brachiopod gular, or elongately rounded with high sides on the valves; 2 Novocrania (Williams, 1997a: 301-302) bear morphological = subrectangular and wide in dorsal profi le, high arch; 3 = resemblance to branched chiton esthetes. However, the caeca lozenge-shaped or subcircular in dorsal profi le; 4 = elongate do not penetrate the periostracum, and they accommodate and tall; 5 = no valves. storage out growths of the mantle and are not sensory as are Comments.–Th is character refers to the shape of the in- chiton esthetes. In Novocrania, the lumen of a caecum con- termediate valves of those taxa with three or more valves. For tains membrane-bounded droplets of glycosaminoglycans those with one or two valves, it refers to the outline of one or (mucopolysaccharides), glycoproteins, and lipids (Williams et both valves. al., 1997: 34-35, 42-43). Although the functions of esthetes and caeca are diff erent, their gross branching morphologies 15. SHELL LENGTH POSTERIOR OR VENTRAL TO are similar and we code them in the same way. Th e pores of APEX OF VALVE: 0 = no intermediate valves; 1 = long (ca. the organophosphatic shells of brachiopods, such as Lingula, ½ or more of the length of the valve anterior to the apex); 2 are much fi ner than the caeca of Novocrania. In Lingula, the = short (less than ½ the length of the valve anterior to the pores typically terminate well below the shell surface (Willams, apex). 1997a), thus, Lingula is coded as absent for this character, Comments.– Th is character refers to the length of shell ma- Matthevia is coded as having a canal system because thin terial that is either underneath the apex as the apical area, or sections of the valves of M. variabilis show a vesicular struc- that is directly posterior to the apex in taxa that lack an api- ture that is interpreted as representing oblique sections of large cal area, in comparison with the valve length anterior to the canals (Runnegar & Vendrasco, 1997; Vendrasco, 1999). apex. Character state 1 is applied to taxa with a large apical area, such as mattheviids, or to taxa in which the apex of the 11. ARTICULAMENTS: 0 = absent; 1 = present; 2 = no valve is located relatively far from the posterior margin of the valves. valve, such as Dycheia and Paradycheia. Character state 2 is Comments.–Articulaments are projections of the articula- applied to taxa with a small apical area or an apex close to the mentum shell layer, either as insertion plates (lateral projec- posterior margin. tions) and/or sutural laminae (anterior projections). 34 Bulletins of American Paleontology, No.
Plate Figure Page
Figs 1-7 are examples of the associated non-chiton fauna occurring with the Eminence Dolomite chiton specimens studied in this paper. Scale bars = 5 mm.
1-2. Internal molds of ventral and dorsal valves of the brachiopod Finkelnburgia missouriensis Ulrich & Cooper, 1936, locality 1, USNM 534833 and 537298...... 1
3-4. Trilobite glabella and thorax, locality 1, USNM 537299 and 537300...... 1
5. Indeterminate gastropods, locality 1, USNM 537301...... 1
6-7. Probable rostroconch, external mold and latex cast, locality 1, USNM 537302...... 1
8-9. Valve and latex cast identifi ed as Priscochiton? sp. indet. by Bergenhayn (1960), Ulrich locality 479a, USNM 137377...... 4 POJETA ET AL.: UPPER CAMBRIAN CHITONS 35 36 Bulletins of American Paleontology, No. POJETA ET AL.: UPPER CAMBRIAN CHITONS 37
Plate Figure Page
1-13. Chelodes cooperi (Bergenhayn, 1960); internal molds of intermediate valves. Arrows (in Figs 1-5, 10) point to anterior margin of apical area. Scale bars = 3 mm...... 12 1-3. Holotype, USNM 137370, right-lateral and dorsal views and latex cast of dorsal view, Ulrich locality 459L. Previous to our study, white enamel paint was applied around the specimen. 4-5. Paratype, USNM 537303, dorsal view and latex cast of dorsal view, Ulrich locality 453y. 6. Paratype, USNM 537304, anterior view showing sharp dorsal crest, Ulrich locality 453y. 7. USNM 537305, dorsal view, locality 1. 8. USNM 537306, dorsal view, locality 1. 9-10. Paratype, USNM 537307, anterior and posterior views, Ulrich locality 453y. See Pl. 3, Fig. 6, for dorsal view. 11-13. USNM 537308, left-lateral and dorsal views, and latex cast of dorsal view, locality 1.
38 Bulletins of American Paleontology, No.
Plate Figure Page
Scale bars = 3 mm.
1-9. Chelodes cooperi (Bergenhayn, 1960); internal molds of intermediate valves. Arrows (Figs 1-3, 6) point to anterior margin of apical area...... 12 1-3. USNM 537309, dorsal and right-lateral views and latex replica of dorsal view, locality 1. 4-5. USNM 537309a, posterior and dorsal views, Ulrich locality 455v. 6. Paratype USNM 537307, dorsal view, Ulrich locality 453y. See Pl. 2, Figs 9-10, for anterior and posterior views. 7. USNM 259006, dorsal view of topotype, previously fi gured by Runnegar et al. (1979: pl. 2, fi g. 64), Ulrich locality 455v. 8-9. USNM 537311, dorsal view and latex cast showing esthete canal fi lling on the right side of view of Fig. 9 (arrow), locality 1.
10-11. Preacanthochiton depressus Bergenhayn, 1960. Holotype, USNM 137372, dorsal view of intermediate valve part and counterpart, Ulrich locality 101-v, upper Gas- conade Dolomite (Ordovician). Previous to our study, white enamel paint was applied around both the part and counterpart. See Pl. 9, Fig. 6...... 11
POJETA ET AL.: UPPER CAMBRIAN CHITONS 39 40 Bulletins of American Paleontology, No. POJETA ET AL.: UPPER CAMBRIAN CHITONS 41
Plate Figure Page
Scale bars = 3 mm (Figs 4-5), 5 mm (all others).
1-3. Preacanthochiton productus Bergenhayn, 1960. Holotype, USNM 137373, right-lateral, pos- terior, and dorsal views of internal mold of tail valve, Ulrich locality 261n. Previous to our study, white enamel paint was applied at the posterior end of the specimen. See Pl. 12, Fig. 7. . . . 11
4-5. Aulochiton sannerae n. gen et n. sp. Tail valve internal mold, USNM 137374. Figured as Preacanthochiton n. sp. aff . P. cooperi by Bergenhayn, 1960, Ulrich locality 399c. Previous to our study, white enamel paint was applied around the specimen...... 17
6-8. Hemithecella eminensis Stinchcomb & Darrough, 1995. Holotype, USNM 409439B, internal mold of intermediate valve...... 14 6. Oblique left-lateral view of posterior end showing growth lines made by ventral surface of apical area (arrow) and shell thickening below apical area. 7-8. Dorsal view and enlargement of dorsal view. Stinchcomb & Darrough (1995) listed this specimen as coming from the Eminence Formation [Dolomite], Washington County, Missouri.
9. Hemithecella expansa Ulrich & Bridge, 1941. Holotype, USNM 96211, dorsal view of internal mold of intermediate valve. See Pl. 5, Fig. 1 for lateral view. Th e museum label lists this specimen as com- ing from "Gasconade [Dolomite; Lower Ordovician], near Decaturville, Mo.[Missouri], ½ mile N.W, of pegmatite dike, Camden Co.[unty], Mo." ...... 14 42 Bulletins of American Paleontology, No.
Plate Figure Page
Scale bars = 5 mm.
1-4. Hemithecella expansa Ulrich & Bridge, 1941...... 14 1. Holotype, USNM 96211, right-lateral view showing impression of ventral side of apical area (arrow) without shell thickening below it. See Pl. 4, Fig. 9 for dorsal view. 2. Paratype, USNM 97289, dorsal view of shelled intermediate valve. Th e museum label lists the following information: "Chepultepec [Dolomite; Lower Ordovician], 6 mi.[les] N.E. Bristol, Virginia." 3-4. Dorsal and left-lateral views of intermediate valve internal mold showing impression of ventral side of apical area (arrow) without shell thickening below it. Stinchcomb & Darrough (1995: 62) listed this specimen as part of unfi gured paratype suite of H. elongata, USNM 409450A; herein this number is limited to this specimen. Th e specimen is from the Eminence Dolomite, Washington County, Missouri.
5-7. Hemithecella eminensis Stinchcomb & Darrough, 1995...... 14 5-6. External mold of intermediate valve showing fi llings of horizontal esthete canals and sigmoidal growth lines, and latex cast of same, USNM 409449B. Figured by Stinchcomb & Darrough (1995: fi g. 8.15), but not listed as a paratype of H. eminensis in the fi gure explanation or the text. Th e specimen is regarded as a paratype herein and on the USNM label. Th e specimen is from the Eminence Dolomite, Washington County, Missouri. 7. Previously unfi gured paratype of H. elongata, USNM 409448B. Lateral view of internal mold of intermediate valve. Th e specimen is from the Eminence Dolomite, Washington County Missouri. Th e specimen was not cited by Stinchcomb & Darrough (1995), however, glued to the specimen is their label reading: "Hemithecella elongata unfi gured USNM 409448B." POJETA ET AL.: UPPER CAMBRIAN CHITONS 43 44 Bulletins of American Paleontology, No. POJETA ET AL.: UPPER CAMBRIAN CHITONS 45
Plate Figure Page
Scale bars = 5 mm.
1-2, 5-6. Hemithecella eminensis Stinchcomb & Darrough, 1995 ...... 14 1. Holotype of H. elongata Stinchcomb & Darrough, 1995, USNM 409448A, dorsal view of intermediate valve internal mold. Th is specimen is from the Eminence Dolomite, Washington County, Missouri. 2. Paratype of H. elongata Stinchcomb & Darrough, 1995, USNM 409456, dorsal view of intermediate valve internal mold. Th is specimen is from the Eminence Dolomite, Washington County, Missouri. 5. Previously unfi gured paratype of H. eminensis, right-lateral view of internal mold of intermediate valve. Stinchcomb & Darrough (1995: 62) listed this specimen as part of paratype suite USNM 409450; herein, this number is limited to this specimen. Th is specimen is from the Eminence Dolomite, Washington County, Missouri. 6. USNM 537312, left-lateral view of internal mold of intermediate valve showing the presence of a shell thickening (arrow) below the apical area, locality 2.
3-4. Hemithecella expansa Ulrich & Bridge, 1941...... 14 3. Dorsal view of internal mold of intermediate valve. Stinchcomb & Darrough (1995: 62) listed this specimen as part of unfi gured paratype suite of H. elongata, USNM 409450A; herein, this specimen is renumbered USNM 537312a. Th is specimen is from the Eminence Dolomite, Washington County, Missouri. 4. Latex cast of the apical end of the holotype of H. expansa showing the growth lines on the ventral side of the apical area and the lack of shell below the apical area, USNM 96211. See Pl. 5, Fig. 1. 46 Bulletins of American Paleontology, No.
Plate Figure Page
1-7. Hemithecella eminensis Stinchcomb & Darrough, 1995. Scale bars = 1 mm (Fig. 6), 5 mm (all others) ...... 14 1. Right-lateral view of intermediate valve internal mold, USNM 537313, locality 1. See Pl. 10, Fig. 3. 2. Anterior cross-section of intermediate valve internal mold of a previously unfi gured paratype. Stinchcomb & Darrough (1995: 62) listed this specimen as H. eminensis, part of paratype suite USNM 409450; herein, this specimen is renumbered USNM 537313a. Th e specimen is from the Eminence Dolomite, Washington County, Missouri. 3-4. Oblique right-lateral and dorsal views of internal mold of an intermediate valve showing growth lines made by the ventral side of the apical area and on the shell below the apical area, USNM 537314, locality 2. 5-6. External impressions of three small intermediate valves, one of which shows fi llings of the horizontal esthete canals, and enlargement of specimen showing the fi llings. USNM 537315, locality 1. 7. External impression of intermediate valve showing fi llings of horizontal esthete canals on the right side, USNM 537316, locality 1. POJETA ET AL.: UPPER CAMBRIAN CHITONS 47 48 Bulletins of American Paleontology, No. POJETA ET AL.: UPPER CAMBRIAN CHITONS 49
Plate Figure Page
Scale bars = 3 mm (Figs 8-9), 5 mm (all others).
1-7. Hemithecella eminensis Stinchcomb & Darrough, 1995...... 14 1-4. Holotype of Elongata perplexa Stinchcomb & Darrough, 1995, part and counterpart, incomplete internal and external molds of intermediate valve, USNM 409447A. Th e specimen is from the Eminence Dolomite, Washington County, Missouri. 1-3. Internal mold part, dorsal view showing the linear apical area edges, oblique left-lateral view showing shell below the apical area (arrow), and latex cast suggesting that the apical area is recessed and that the shell below it is a fl ange-like structure (arrow) surrounding the apical area. 4. External mold counterpart. 5-7. Previously unfi gured paratype of Elongata perplexa, internal mold of intermediate valve, USNM 409447B. Left-lateral, dorsal, and right-lateral views, showing linear pitted features. Th e specimen is from the Eminence Dolomite, Washington County, Missouri.
8-9. Occipital spine of trilobite? Previously unfi gured paratype of Elongata perplexa, top and side views, USNM 409447C. Th e specimen is from the Eminence dolomite, Washington County, Missouri...... 16
50 Bulletins of American Paleontology, No.
Plate Figure Page
Scale bars = 5 mm.
1-2. Hemithecella eminensis Stinchcomb & Darrough, 1995. Paratype USNM 489533, a specimen pre- serving the shell. Right-lateral and dorsal views showing linear features (see Pl. 8, Figs 5-7). Stinch- comb & Darrough (1995: 58) gave the museum number as USNM 409453, but this duplicated a number already in use for another specimen, and the USNM staff assigned the number 489533 to this specimen. Th e specimen is from the Eminence Dolomite, Washington County, Missouri. . . 14
3-5. Hemithecella expansa Ulrich & Bridge, 1941. Left-lateral, dorsal, and anterior views of two at- tached valves. Th e upper valve is broken, but the right side (arrow in Fig. 4) shows that it completely overlaps the lower valve. USNM 489527. Previously fi gured by Stinchcomb & Dar- rough (1995: 56). Th ey cited the USNM number as 409459, which was already in use for another specimen; USNM staff renumbered the specimen as 489527. Th e specimen is from the Gasconade Dolomite (Lower Ordovician), near the Franklin/Crawford county border, Missouri. . . 14
6. Preacanthochiton depressus Bergenhayn, 1960. Paratype, USNM 325306; the specimen is from the Gasconade Dolomite (Lower Ordovician) at Ulrich locality 453n. See Pl. 3, Figs 10-11. . . 11 POJETA ET AL.: UPPER CAMBRIAN CHITONS 51 52 Bulletins of American Paleontology, No. POJETA ET AL.: UPPER CAMBRIAN CHITONS 53
Plate Figure Page
1-7. Hemithecella eminensis Stinchcomb and Darrough, 1995. Scale bars = 3 mm (Fig. 6), 5 mm (all others)...... 14 1. Previously unfi gured paratype of H. elongata Stinchcomb & Darrough, 1995. USNM 409448B; specimen not listed by Stinchcomb & Darrough (1995: 62), but glued to the bottom of the specimen is a label with the notation "Hemithecella elongata unfi gured." Th e specimen shows the exterior of an intermediate valve, and is from the Eminence Dolomite, Washington County, Missouri. 2. Previously unfi gured specimen from paratype suite USNM 409450; the specimen is here renumbered USNM 537317. Th e specimen is an internal mold of an intermediate valve showing a very long broken lacuna and growth lines of the ventral side of the apical area. Th e specimen is from the Eminence Dolomite, Washington County, Missouri. 3. Dorsal view of the specimen illustrated on Pl. 7, Fig. 1, showing the drusy quartz that sometimes obscures specimens. USNM 537313, locality 1. 4. Dorsal exterior view of intermediate valve, USNM 537318, locality 1. 5. Dorsal view of internal mold of intermediate valve, USNM 537319, locality 1. 6. Dorsal exterior mold of intermediate valve, USNM 537320, locality 1. 7. Dorsal exterior mold of intermediate valve, showing growth lines, USNM 537321, locality 1. 54 Bulletins of American Paleontology, No.
Plate Figure Page
1-3. Hemithecella eminensis Stinchcomb & Darrough, 1995. Part and counter- part of the same specimen, internal and external molds of intermediate valve, USNM 537322, locality 1. See Pl. 12, Figs 1-2. Scale bars = 5 mm...... 14 1-2. Internal mold part, oblique left-lateral and dorsal views showing the long incomplete lacuna. 3. Stereoscopic pair of external mold of counterpart showing the fi llings of the horizontal esthete canals; fi gure of specimen incomplete anteriorly. POJETA ET AL.: UPPER CAMBRIAN CHITONS 55 56 Bulletins of American Paleontology, No. POJETA ET AL.: UPPER CAMBRIAN CHITONS 57
Plate Figure Page
Scale bars = 3 mm (Fig. 6), 5 mm (all others).
1-6. Hemithecella eminensis Stinchcomb & Darrough, 1995...... 14 1-2. Oblique view of counterpart of USNM 537322, looking at the fi llings of the esthete canals from posterior to anterior, and view of entire counterpart. See Pl. 11, Figs 1-3. 3-5. Oblique view of external mold looking at the fi llings of the horizontal esthete canals from posterior to anterior; view showing the several layers of the fi llings of the horizontal esthete canals, and latex cast of fi gure 4. USNM 537323, locality 1. See Pl. 13, Fig. 1. 6. Small external mold of intermediate valve showing fi llings of horizontal esthete canals, USNM 537324, locality 1.
7. Preacanthochiton productus Bergenhayn, 1960. Paratype, tail valve, internal mold, USNM 537325, Ulrich locality 261n1. See Pl. 4, Figs 1-3...... 11 58 Bulletins of American Paleontology, No.
Plate Figure Page
Scale bars = 5 mm (Figs 1-3) or as indicated on the fi gure (Fig. 4).
1-2. Hemithecella eminensis Stinchcomb & Darrough, 1995...... 14 1. Stereoscopic pair of external mold of intermediate valve, showing multiple layers of fi lled horizontal esthete canals, USNM 537323, locality 1. See Pl. 12, Figs 3-5. 2. Internal mold of intermediate valve. Previously unfi gured paratype from paratype suite USNM 409450; this specimen is here renumbered USNM 537326. Th e specimen is from the Eminence Dolomite, Washington County, Missouri.
3. Tryblidiid monoplacophoran. Dorsal view of poorly preserved specimen classifi ed as Preacanthochiton sp. by Stinchcomb & Darrough (1995: fi g. 6.2). Th e specimen is from the Gasconade Dolomite (Lower Ordovician), Franklin County, Missouri. USNM 409457A. Stinchcomb & Darrough (1995) cited the specimen as USNM 409457; this number was already in use for another specimen and USNM staff renumbered the specimen...... 13
4. Placiphorella velata Carpenter, 1879. SEM photograph of cross section of an intermediate valve. Holocene, Santa Barbara Museum of Natural History number 83161; collected from the intertidal zone at Pacifi c Grove, Monterey Peninsula, California...... 16 POJETA ET AL.: UPPER CAMBRIAN CHITONS 59 60 Bulletins of American Paleontology, No. POJETA ET AL.: UPPER CAMBRIAN CHITONS 61
Plate Figure Page
Scale bars = 200 μm (Fig. 4), 1 mm (Fig. 3), 10 mm (Figs 1-2).
1-2. Hemithecella expansa Ulrich and Bridge, 1941. Internal mold of intermedi- ate valve, lateral and dorsal views of a paratype of Matthevia walcotti Runnegar et al., 1979, USNM 258990, Ulrich locality 217...... 14
3-4. Dendrochiton fl ectens (Carpenter, 1864). Epoxy casts of esthete canals. Two images at diff er- ent magnifi cations of the ventral sides of horizontal canals; the arrow in Fig. 4 shows the ver- tical canals proceeding dorsally from the horizontal canals. Holocene, Santa Barbara Museum of Natural History number 369492; collected by George Hanselman in 1973 from the underside of rocks during a -0.76 m tide at Cactus Island, Washington State...... 16 62 Bulletins of American Paleontology, No.
Plate Figure Page
Scale bars = 5 mm.
1-4. Aulochiton sannerae n. gen. and n. sp., intermediate valves...... 17 1-2. Interior of incomplete chertifi ed tegmentum and latex cast of same, showing the internal openings of the vertical esthete canals, paratype USNM 537327, locality 1. 3-4. Dorsal and right-lateral views of internal mold, paratype USNM 537328, Ulrich locality 455v.
5-8. Hemithecella expansa Ulrich & Bridge, 1941. Th is specimen is from the Gasconade Dolomite (Lower Ordovician), Franklin County, Missouri, and was previously fi g- ured by Stinchcomb & Darrough (1995: fi g. 6.15). Th ey gave the specimen num- ber as USNM 409462; this number had already been used for another specimen and USNM staff assigned the new number 489530...... 14 5-6. External mold of intermediate valve and stereoscopic pair of lower portion of same showing growth lines and numerous surface pustule-like markings. 7-8. Latex casts of Figs 5-6.
POJETA ET AL.: UPPER CAMBRIAN CHITONS 63 64 Bulletins of American Paleontology, No. POJETA ET AL.: UPPER CAMBRIAN CHITONS 65
Plate Figure Page
Scale bars = 2 mm (Figs 1-5), 3 mm (Figs 6-7), 5 mm (all others).
1-5. Dycheia shergoldi Pojeta & Derby, 2007, intermediate valves. Silicifi ed replicas, views of dorsal, ven- tral, right-lateral, USNM 534347, and anterior and posterior cross-sections, USNM 534347 and 534350. Fort Sill Limestone (Upper Cambrian), Murray County, Oklahoma. See Pl. 21, Figs 4-6. . . 21
6-11. Aulochiton sannerae n. gen. and n. sp., intermediate valves...... 17 6, 8, 10. Interiors of chertifi ed tegmentum showing the inner openings of the vertical esthete canals. Paratypes, USNM 537329, 537330, and 537331, all from locality 1. See Pl. 21, Fig. 3. 7, 9, 11. Stereoscopic pairs of latex casts of specimens in Figs 6, 8, and 10, showing the lengths of the esthete canals. 66 Bulletins of American Paleontology, No.
Plate Figure Page
1-7. Aulochiton sannerae n. gen. and n. sp. Scale bars = 3 mm (Figs 1-2), 5 mm (all others). . . . 17 1-2. Latex cast and specimen showing the interior of the chertifi ed tegmentum of an intermediate valve, showing vertical esthete canal openings and growth lines. Paratype USNM 537332, locality 1. See Pl. 21, Fig. 2. 3. Stereoscopic pair showing the interior of the chertifi ed tegmentum of an intermediate valve. Paratype USNM 537333, locality 1. 4-5. Chertifi ed tegmentum of circular head valve and stereoscopic view of latex cast of same showing vertical esthete canals and growth lines; anterior part is facing down. Paratype USNM 537334, locality 1. 6-7. Chertifi ed tegmentum of elongate tail valve and stereoscopic view of latex cast of same showing vertical esthete canals. Paratype USNM 537335, locality 1. POJETA ET AL.: UPPER CAMBRIAN CHITONS 67 68 Bulletins of American Paleontology, No. POJETA ET AL.: UPPER CAMBRIAN CHITONS 69
Plate Figure Page
1-11. Aulochiton sannerae n. gen. and n. sp. Scale bars = 5 mm...... 17 1-4. Part and counterpart of an intermediate valve. Holotype, USNM 537336, Ulrich locality 455v. 1-2. Part, dorsal and right-lateral views of the smooth internal mold. 3-4. Counterpart, interior of chertifi ed tegmentum, and latex cast of same showing vertical esthete canals. Th e apparent hole at the apex of Fig. 4 was caused by the lack of latex penetrating to the apex of the valve. 5. Right-lateral view of internal mold of intermediate valve. Paratype USNM 537337, locality 1. 6-7. Internal mold of intermediate valve, anterior view showing anterior opening of snout and dorsal view. Paratype 537338, locality 1. 8-11. Stereoscopic pair of interior of chertifi ed tegmentum of circular head valve, latex cast of same in stereoscopic dorsal views, and anterior and right-lateral views of latex cast. Th e apparent hole at the top of the latex cast was caused by the lack of penetration of the latex into the apex of the valve. Paratype USNM 537339, locality 1. 70 Bulletins of American Paleontology, No.
Plate Figure Page
1-13. Aulochiton sannerae n. gen. and n. sp., intermediate valves. Scale bars = 3 mm (Figs 1, 3, 6-7), 5 mm (all others)...... 17 1-3. Posterior, anterior, and right-lateral views of internal mold. Paratype USNM 537340, locality 1. 4-5. Anterior view showing raised jugal ridge snout and dorsal view of internal mold. Paratype USNM 537341, locality 1. 6-7. Left-lateral and dorsal views of internal mold. Paratype USNM 537342, locality 1. 8-10. Dorsal, anterior and posterior views of internal mold. Paratype USNM 537343, locality 1. 11. Dorsal view of internal mold preserving the entire jugal ridge snout. Paratype USNM 537344, locality 1. See Pl. 20, Fig. 1. 12-13. Small slab showing fi ve valve exteriors and latex cast of same. Paratypes USNM 537345- 537349, locality 1. See Pl. 21, Fig. 1. POJETA ET AL.: UPPER CAMBRIAN CHITONS 71 72 Bulletins of American Paleontology, No. POJETA ET AL.: UPPER CAMBRIAN CHITONS 73
Plate Figure Page
1-19. Aulochiton sannerae n. gen. and n. sp. Scale bars = 3 mm (Figs 2-4, 16, 19), 5 mm (all others). . . . 17 1. Left-lateral view of specimen shown on Pl. 19, Fig. 11. Paratype USNM 537344. 2-4. Dorsal, anterior, and right-lateral views of a strongly mucronate intermediate(?) valve internal mold. USNM 537350, locality 1. 5-16. Tail valve internal molds. 5-10. Dorsal, anterior, posterior, right-lateral, left-lateral, and right-oblique anterior views. Figs 6 and 10 show the raised anterior edge; Fig. 9 shows the prominent ventral folds. Th e eccentric mucro is probably a diagenetic eff ect. Paratype USNM 537351, locality 1. 11. Dorsal view of a specimen showing the smooth internal mold separated by a space from the partially preserved outer shell layer. Paratype USNM 537352, locality 1. 12-14. Posterior, dorsal, and right-lateral views. Paratype USNM 537353, locality 2. 15. Dorsal view. Paratype USNM 537354, locality 2. 16. Dorsal view of small valve. Paratype USNM 537355, locality 1. 17-19. Part and counterpart internal and external molds. Posterior and dorsal views of internal mold and external mold showing growth lines. Paratype USNM 537356, locality 1. 74 Bulletins of American Paleontology, No.
Plate Figure Page
Scale bars = 2 mm (Figs 4-6), 3 mm (Figs 7-11), 5 mm (Figs 1-3).
1-3. Aulochiton sannerae n. gen. and n. sp...... 17 1. Latex cast of lower right-hand specimen shown on Pl. 19, Fig. 12. Arrow points to a part of the small apical area. Paratype USNM 537345. 2. Enlargement of specimen shown on Pl. 17, Fig. 2. Arrow points to a section of shell where the dark esthete canals can be seen traversing the broken chertifi ed edge of the tegmentum. Paratype USNM 537332. 3. Enlargement of specimen shown on Pl. 16, Fig. 6. Arrow points to a section of shell where the dark esthete canals can be seen traversing the broken chertifi ed edge of the tegmentum. Paratype USNM 537329.
4-6. Dycheia shergoldi Pojeta & Derby, 2007. Left-lateral, posterior, and ventral views of si- licifi ed tail valve. Paratype USNM 534354. Fort Sill Limestone (Upper Cambrian), Murray County, Oklahoma. See Pl. 16, Figs 1-5...... 21
7-11. Paradycheia dorisae n. gen. and n. sp...... 21 7-8. Anterior and posterior views of tail valve internal mold, showing cross-sectional shape. Paratype, USNM 534829, locality 3. See Pl. 23, Figs 5-7. 9-10. Posterior view showing cross-sectional shape, and oblique dorsal view of internal mold of intermediate valve. Holotype, USNM 534821, locality 3. See Pl. 22, Fig. 5. 11. Lateral view showing mantle-like markings on the left lacuna of internal mold of intermediate valve. Paratype USNM 534830, locality 2. POJETA ET AL.: UPPER CAMBRIAN CHITONS 75 76 Bulletins of American Paleontology, No. POJETA ET AL.: UPPER CAMBRIAN CHITONS 77
Plate Figure Page
1-11. Paradycheia dorisae n. gen. and n. sp., internal molds of intermediate valves. Scale bars = 5 mm...... 21 1-2. Right-lateral and dorsal views; arrows point to larger posterior lacuna. Paratype, USNM 534822, locality 1. 3, 9. Latex cast of dorsum of the specimen shown in Fig. 9; arrows point to larger posterior lacuna. Paratype USNM 534823, locality 3. Fig. 9 is a latex cast of the dorsum of the specimen shown in Fig. 9. 4. Left-lateral view; arrow points to the larger posterior lacuna. Paratype USNM 534824, locality 3. 5. Left-lateral view of holotype. USNM 534821, locality 3. See Pl. 21, Figs 9-10. 6. Left-lateral (?) view. Paratype USNM 534831, locality 2. 7-8. Right-lateral view and latex cast of same. Paratype USNM 534825, locality 3. 10. Left-lateral view. Paratype USNM 534826, locality 2. 11. Left-lateral view; arrow points to probable growth line. Paratype USNM 534827, locality 1. 78 Bulletins of American Paleontology, No.
Plate Figure Page
1-7. Paradycheia dorisae n. gen. and n. sp. Scale bars = 5 mm...... 21 1-4. Part and counterpart of an intermediate valve. Paratype USNM 534828, locality 3. 1-2. Part, external mold, and latex cast of same; arrows point to growth lines. 3-4. Counterpart, incomplete internal mold, and dorsal and lateral views showing cross- sectional shape between lacunae. 5-7. Internal mold of tail valve. Paratype USNM 534829, locality 3. 5. Dorsal view; horizontal arrow points to larger posterior lacuna and shows the small missing chip of chert that causes the seemingly bifi d shape of the lacuna. Oblique arrow points to the smaller anterior lacuna. 6-7. Left-lateral and right-lateral views; arrows point to the position of the smaller anterior lacuna. Right-lateral view shows the prominent lateral folds. See Pl. 21, Figs 7-8. POJETA ET AL.: UPPER CAMBRIAN CHITONS 79 80 Bulletins of American Paleontology, No.
INDEX
Abstract ...... 1 expansa ...... 13-15 Acaenoplax ...... 21 quinquelites ...... 14 Acknowledgments...... 24 Hemithecellidae ...... 8, 28, 31 Alastega ...... 18 hypostracum ...... 7 Alastegiidae ...... 30 Introduction ...... 1 Appendix 1 ...... 28 key to genera of Mattheviidae ...... 16 Appendix 2 ...... 31 Kimberella ...... 5 articulamentum ...... 7 lacunae ...... 4, 8 Aulochiton ...... 17, 21-22 Listrochiton ...... 11, 30 sannerae ...... 17 Literature Cited ...... 24 Aulochitonidae ...... 17 Litochiton ...... 30-31 Bursata ...... 28-30 crebatus ...... 30 Calceochiton ...... 8, 17 marginatus ...... 30-31 Calvinella ...... 6, 16 Litochitonidae ...... 30-31 Chaetoderma ...... 21 Llandeilochiton ...... 31 Chelodes ...... 8-10, 17 Llandeilochitonidae ...... 31 abrupta...... 14 Materials and Methods ...... 2 cooperi ...... 11-12, 18 Matherella ...... 6 whitehousei ...... 4 Matthevia ...... 16, 19, 21-22 Chelodidae ...... 8 wahwahensis ...... 18 chiton shell microstructure ...... 7 walcotti ...... 15 cladistic analysis ...... 21 Mattheviidae ...... 8, 21-22, 28, 31 Cobcrephoridae ...... 31 Modiomorpha concentrica ...... 15-16 collecting localities ...... 2 Mytilus californianus ...... 15 Dendrochiton ...... 16 Ninmaroo Formation ...... 3-4 fl ectens ...... 16 Novocrania ...... 22-23 Dycheia ...... 19-22 Ocruranus ...... 4 shergoldi ...... 21 Odontogriphus omalus ...... 5 Echinochiton ...... 21-22 Orthriochiton ...... 4 Echinochitonidae ...... 28, 31 Overview of Cambrian Chitons ...... 3 Elongata ...... 8, 13, 16 paleobiogeography ...... 4 perplexa ...... 14-15 Paleochiton kindbladensis ...... 31 Entomaspis ...... 6 Paleoloricata ...... 8 Eochelodes ...... 8-9 Paradycheia ...... 19-22 Eochelodidae ...... 8, 28, 31 dorisae ...... 19-21 Eohalobia ...... 4 pellicle ...... 7 Epimenia ...... 21 Phosphatoloricata ...... 31 Esthetes ...... 7, 16, 18 Placiphorella velata ...... 16 Eukteanochiton ...... 8, 16 Plethopeltis ...... 6 Finkelnburgia ...... 1 Preacanthochiton ...... 8-12 Gotlandochiton ...... 31 baueri ...... 13 Gotlandochitonidae ...... 31 cooperi (aff .) ...... 11 Gryphochiton ...... 30 depressus ...... 11 Helminthochiton ...... 29-31 productus ...... 11 aequivoca ...... 29 Preacanthochitonidae ...... 8, 28 baueri ...... 13, 31 Priscochiton ...... 4 blacki ...... 31 properiostracum ...... 7 griffi thi ...... 30-31 Robustum ...... 8, 16, 21-22 marginatus ...... 30 Robustidae ...... 8, 28 Helminthochitonidae ...... 30 Sarkachiton ...... 29 Hemithecella ...... 8, 13, 17 Septemchiton ...... 21-23, 28 abrupta ...... 14 iowensis ...... 28 elongata ...... 14-15 vermiformis ...... 28 eminensis ...... 14 Septemchitonidae ...... 28 POJETA ET AL.: UPPER CAMBRIAN CHITONS 81
Solenocaris ...... 29 tegmentum ...... 7 Solenocarididae ...... 29 text fi gures ...... 2, 5, 7, 10, 12, 22 Spicuchelodes ...... 8, 17 Triarthropsis ...... 6 stratigraphy ...... 4 Triplicatella ...... 4 Systematic Paleontology ...... 6 tunnels ...... 4, 8 tables ...... 13-14, 18, 21, 23 valve orientation ...... 8 82 Bulletins of American Paleontology, No. PREPARATION OF MANUSCRIPTS
Bulletins of American Paleontology, the oldest continuously published, peer-reviewed paleontological journal in the Americas, seeks signifi cant, larger monographs (> 50 printed pages, minimum 100 manuscript pages) in paleontological subjects or in neontological sub- jects that are strongly relevant to paleontological problems. Most contributions focus on systematics, placed in biostratigraphic, biogeographic, paleoenvironmental, paleoecologi- cal, and/or evolutionary contexts. Contributions have historically focused on fossil inverte- brates, but papers on any taxon of any age are welcome. Emphasis is placed on manuscripts for which high-quality photographic illustrations and the large quarto format are desirable. Both single- and multi-authored (contributed proceedings) volumes are invited.
Submissions are welcome from authors of any institutional or organizational affi liation. Publication costs of the Bulletins are heavily subsidized by PRI, but authors are asked to pay illustration charges ranging from $35 per text-fi gure to $120 per full-page plate.
Manuscripts must be submitted as electronic copy (preferably pdf) in English, prepared in one of the more widely used word-processing programs. Each manuscript should include Abstract, Introduction, Materials and Methods, Results (fl exible format), Discussion, Ac- knowledgments, Literature Cited, Appendices, Index (subjects only), Tables, Figure and Plate Captions, Text-fi gures and Plates. Th e abstract should be informative and include all new taxa, combinations, and taxonomic decisions (e.g., selection of lectotypes); an addition- al abstract in a second language can be included when appropriate in view of the content or authors. All measurements should be metric, and authories and dates of all species-level taxa must be provided when fi rst mentioned in the text (although citations need not be included in the Literature Cited). Repositories of types and voucher specimens should be indicated, and nomenclature must be in accordance with the relevant International Code of Nomencla- ture. All abbreviations and acronyms used in the paper must be explained. Particular care should be paid to formatting the Literature Cited (see full “Instructions” for examples). Comprehensiveness of the Index is at the discretion of the author(s). Illustrations should be prepared as high-resolution, black-and-white digital images, prearranged on plates as appro- priate; color reproduction is available but at a substantially higher cost, which must be borne by the author. Low-resolution images are acceptable for initial submission. After review, fi nal submission should include the original (word-processor) text fi le(s), plus separate table fi les and a high-resolution image fi le for each text-fi gure and plate.
Th e full “Instructions for Authors” should be consulted during manuscript preparation, and are available from the Editor or online at http://www.museumoftheearth.org/publications/ bookstore.php?page=Info_Authors. 1259 Trumansburg Road Ithaca, New York 14850 U. S. A.
ISSN 007-5779 US $40.00