UNIVERSITY OF ILLINOIS LIBRARY AT URBANACHAMPAIGN GEOLOGY NOTICE: Return or renew all Library Materials! The Minimum Fee for each Lost Book la $50.00. The person charging this material is responsible for its return to the library from which it was withdrawn on or before the Latest Date stamped below. Theft, mutilation, and underlining of books are reasons for discipli- nary action and may result in dismissal from the University. To renew call Telephone Center, 333-8400 UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN APR 2 8 19<3 4 MAN njfc ?no7 6 L161— O-1096 50.5 FIELDIANA Geology NEW SERIES, NO. 18 The Ear Region in Xenarthrans (= Edentata: Mammalia) ?Part I. Cingulates Bryan Patterson Walter Segall William D. Turnbull November 30, 1989 Publication 1405 PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY Information for Contributors to Fieldiana General: Fieldiana is primarily a journal for Field Museum staff members and research associates, although manuscripts from nonaffiliated authors may be considered as space permits. The Journal carries a page charge of $65 per printed page or fraction thereof. Contributions from staff, research associates, and invited authors will be considered for publication regardless of ability to pay page charges, but the full charge is mandatory for nonaffiliated authors of unsolicited manuscripts. Payment of at least 50% of page charges qualifies a paper for expedited processing, which reduces the publication time. Manuscripts should be submitted to Scientific Editor, Fieldiana, Field Museum of Natural History, Chicago, Illinois 60605-2496, USA. Three complete copies of the text (including title page and abstract) and of the illustrations should be submitted (one original copy plus two review copies which may be machine copies). No manuscripts will be considered for publication or submitted to reviewers before all materials are complete and in the hands of the Scientific Editor. Text: Manuscripts must be typewritten double-spaced on standard-weight, 8V2- by 1 1-inch paper with wide margins on all four sides. For papers longer than 100 manuscript pages, authors are requested to submit a "Table of Contents," a "List of Illustrations," and a "List of Tables." In most cases, the text should be preceded by an "Abstract" and should conclude with "Acknowledgments" (if any) and "Literature Cited." All measurements should be in the metric system. The format and style of headings should follow those of recent issues of Fieldiana. For more detailed style information, see The Chicago Manual of Style (13th ed.), published by The University of Chicago Press, and also recent issues of Fieldiana. In "Literature Cited," authors are encouraged to give journal and book titles in full. Where abbreviations are desirable (e.g., in citation of synonymies), authors consistently should follow Botanico-Periodicum-Huntianum and TL-2 Taxonomic Literature by F. A. Stafleu & R. S. Cowan (1976 et seq.) (botanical papers) or Serial Sources for the Biosis Data Base (1983) published by the BioSciences Information Service. References should be typed in the following form: Croat, T. B. 1978. Flora of Barro Colorado Island. Stanford University Press, Stanford, Calif., 943 pp. Grubb, P. J., J. R. Lloyd, and T. D. Pennington. 1963. A comparison of montane and lowland rain forest in Ecuador. I. The forest structure, physiognomy, and ftoristics. Journal of Ecology, 51: 567-601. Langdon, E. J. M. 1979. Yage among the Siona: Cultural patterns in visions, pp. 63-80. In Browman, D. L., and R. A. Schwarz, eds., Spirits, Shamans, and Stars. Mouton Publishers, The Hague, Netherlands. Murra, J. 1946. The historic tribes of Ecuador, pp. 785-821. In Steward, J. H., ed., Handbook of South American Indians. Vol. 2, The Andean Civilizations. Bulletin 143, Bureau of American Ethnology, Smithsonian Institution, Washington, D.C. Stolze, R. G. 1981. Ferns and fern allies of Guatemala. Part II. Polypodiaceae. Fieldiana: Botany, n.s., 6: 1- 522. Illustrations: Illustrations are referred to in the text as "figures" (not as "plates"). Figures must be accompanied by some indication of scale, normally a reference bar. Statements in figure captions alone, such as "x 0.8," are not acceptable. Captions should be typed double-spaced and consecutively. See recent issues of Fieldiana for details of style. Figures as submitted should, whenever practicable, be P/i by 1 1 inches (22 x 28 cm) and may not exceed 1 1 Vi by 16% inches (30 x 42 cm). Illustration* should be mounted Oft boards ifi the arrangement you wish to obtain in the printed work. This original $ei should be suitable for transmission to the printer as follows: Pen and ink drawings may be originals (preferred) or photostats, shaded drawings should be Originals, but within the size limitation; and photostats should be high-quality, glossy, black and white prints AW illustrations should be marked on the reverse with author's name, figure number**), and "fop," Original illustrations will be returned to the author upon publication unless otherwise specified. Authors who wish to publish figures that require costly special paper or color reproduction most make prior jo range meats with the Scientific Ed Page Pwfo. Fieidkma employs a two-Step correction system, fiacb author will normally receive a copy of the edited manuscript 00 which deletions, additions and changes €tm lk made and queries answered. Only one set of proofs will be sort. AM desired corrections of type must he mads on the single set of page proofs. Changes in mats (as opposed to corrections) atr very expensive. Author -generated changes in page proofs can only be made if the author agrees m advance to pay for them. THIS PUBLICATION IS PRINTED ON ACID-FREE PAPER. UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN GEOLOGY FIELDIANA Geology NEW SERIES, NO. 18 The Ear Region in Xenarthrans (= Edentata: Mammalia) Part I. Cingulates t Bryan Patterson Curator, Fossil Mammals Department of Geology Field Museum of Natural History Chicago, Illinois 60605-2496 Agassiz Professor Museum of Comparative Zoology Harvard University Cambridge, Massachusetts 02138 tWalter Segall Research Associate Department of Zoology Field Museum of Natural History Chicago, Illinois 60605-2496 William D. Turnbull Curator, Fossil Mammals Department of Geology Field Museum of Natural History Chicago, Illinois 60605-2496 Accepted March 22, 1988 Published November 30, 1989 Publication 1405 PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY © 1989 Field Museum of Natural History ISSN 0096-2651 PRINTED IN THE UNITED STATES OF AMERICA <? Table of Contents V^C* KJt^\ Preface v Abstract 1 Introduction 1 Dasypodoids The Utaetus Group t Utaetus buccatus 4 The Priodontes Group The Priodontines Priodontes gigas 5 Cabassous loricatus 8 Cabassous lugubris 8 The Tolypeutines Tolypeutes tricinctus 11 The Dasypus Group The Dasypodines Dasypus novemcinctus 12 The Stegotheriines fStegotherium tesselatum 14 The Euphractus Group The Euphractines Euphractus sexcinctus 16 Chaetophractus (E.) villosus 16 Zaedius pichiy 16 fPaleuphractus argentinus 22 ^Proeuphr actus scalabrinii 23 "fProzaedius exilis 23 The Chlamyphorines Chlamyphorus truncatus 26 The Eutatines fDoellotatus prominens 29 \Doellotatus sp. cf. prominens 29 fEutatus seguini (= E. brevis) 30 fProeutatus oenophorus 30 The Peltephilus Group "f Peltephilus pumilis (= ^Peltephilus ferox) 33 The Chlamytherium Group fPlaina subintermedia 35 Glyptodontoids Glyptodontidae 37 fPropalaeohoplophorus australis ... 38 fStromaphorus comprissideus 38 "fHoplophr actus (= Plohophorus) proximus 38 fEleutherocercus solidus 38 Summary and Conclusions 42 Acknowledgments 43 Literature Cited 43 Appendix 46 List of Illustrations 1 . Eutatus buccatus, cranial fragment; Priodontes gigas, posterior part of skull . 6 2. Cabassous loricatus, basicranium of skull; Cabassous lugubris, posterior part of skull 9 3. Dasypus novemcinctus, posterior portion of skull 13 4. Stegotherium tesselatum, posterior por- tion of skull 15 5. Euphractus sexcinctus, posterior portion of skull 17 6. Euphractus sexcinctus, sectioned skull ... 18 7. Chaetophractus (= Euphractus) villosus, left bulla, part of cranium, and right bulla 19 8. Zaedius pichiy, posterior part of skull; Paleuphractus argentinus, cranium 20 9. Proeuphractus scalabrinii, posterior por- tion of skull 24 10. Prozaedius exilis, posterior portion of skull 25 1 1 . Chlamyphorus truncatus, posterior por- tion of skull 27 12. Doellotatus prominens, three partial skulls 31 13. Proeutatus cf. oenophorus, two partial crania and a right tympanum 32 1 4. Peltephilus (= P. pumilis) ferox, poste- rior part of skull 34 15. Plaina cf. subintermedia, cranium 36 16. Hoplophr actus proximus, cranium 39 17. E leut her ocercus solidus, basicranium and left otic region of skull 40 in Preface When Bryan Patterson left Field Museum in 1955 to take up the Agassiz Professorship at Har- vard, he took with him the only copy of this study, which had resulted from a decade of Wednesday afternoon collaborations.* This he subsequently revised, and in part revised again, but for unknown reasons it somehow never was completed and made ready for publication. I believe Patterson simply never refined it to his complete satisfaction. Aware of the scope of this ambitious project, following the deaths of Patterson in 1 979 and Segall in 1 98 1 , I have attempted to pull together those parts of the scattered manuscript that could be found, and to salvage the 100+ pen-and-ink illustrations that had been rendered expressly for it, mostly by John Conrad Hanson and R. Norris. Hanson was the staff artist for the Department of Geology during most of the decade of the original collaboration, and Norris served briefly as the Zoology staff artist between 1949 and 1950. These descriptions con- stitute the first attempt to provide a comprehen- sive coverage since the work of van der Klaauw (1931), adding much that was new at the time the work was done. The study by Guth ( 1 96 1 ) in many respects parallels this work but is difficult to ob- tain, and in any case covers somewhat different ground and works from a different set of speci- mens. It is an essential companion to this study. Subsequently, still more fossil materials have be- come available, notably pampatheres under study by Edmund, and the Glyptotherium specimens de- scribed by Gillette and Ray (1 98 1). Except for the Field Museum specimen Vassallia (= Plaina), no attempt to incorporate new pampathere data has been made. Glyptotherium, to judge by the Gillette and Ray report (their figs. 7-8, 10c, 12a), fit the more or less stereotypic pattern of all glyptodon- toids, and their illustrations correspond closely to those offered here (figs. 15-16). Patterson had used the illustrations for lectures and student projects over the years, with inevitable wear and other damage resulting. In order to save those most deteriorated, Ron Testa, Field Mu- seum photographer, photographed each on high * Segall, a practicing physician at the time, specializing in eye, ear, nose, and throat, had been trained in the famed Vienna Clinics. He brought to the collaboration anatomical and medical insights not usually available to the paleontologist and comparative anatomist. Wednes- day afternoon was the physician's half-day off, and in this case was regularly devoted to this study by the two senior authors. contrast negatives to eliminate the stains and smudges and to emphasize the original pen-and- ink work. Finally, to keep the results as uniform as possible, we treated all the drawings in this manner. I then montaged them into the figures and decided on the features to be labeled. Credit for the labeling is due Marlene Werner, currently of the Museum's Scientific Services staff. In a few instances Patterson had penciled notations di- rectly on the drawings. In those cases, I could be certain of what was intended. But for the vast majority, I have labeled them according to my interpretation of what the text called for and what I think had been intended, or according to need so as to make each illustration immediately com- prehensible. For this the responsibility is entirely mine. In some cases I have modified the text either to expand upon it, or to provide greater consis- tency, or to conform to present terminology, elim- inate repetition, or correct an occasional lapsus. Wherever this has been done in the text, I have employed the device of setting off the changes in square brackets [ ] so that the reader could distin- guish between the original and my tampering. There can be little doubt that Patterson was the senior and pivotal author of the pair, for he clearly had the broader phylogenetic and taxonomic un- derstanding. Segall's contributions provided func- tional and comparative anatomic understanding, based upon his training in that subject, upon his long experience with human ear study, and upon his inquiring mind. It is a difficult matter for a third party to join an effort of this sort, for one does not know for sure why the initial study was never finished. Were the authors unhappy with the result? Or did they just run out of time? Or would they wish to have another meddle with their effort? Would the third party be doing them an injustice? I have carefully weighed the pros and cons. I believe that both Patterson and Segall would want the work finished. My decision to do so rests heavi- ly on the realization that if I did not complete the work, far too great an investment in their time and effort, insight, and knowledge would be lost. By pulling together the parts of this work that could be found and by my filling the gaps, I hope I do not do the others a disservice. At least the reader should be able to be clear about separating my part of the final work from theirs, and by so doing at least the descriptive aspects and illustrations would survive and be made available to all in a useful form. W.D.T. fc The Ear Region in Xenarthrans (= Edentata: Mammalia) Part I. Cingulates Abstract This report is the first of two planned works detailing the descriptive and comparative anato- my of the xenarthran ear region (basicranium and its immediate surrounds, including middle ear, but not the inner ear). The study was begun by Pat- terson and Segall in 1945 and continued actively well into 1955, and only fitfully by Patterson there- after. This part, "Cingulates," covers a represen- tation of the Cingulata (= Loricata), including da- sypodoids (Utaetus group, Priodontes group, Dasypus group, Euphractus group, Peltephilus group, and Chlamytherium group) and glyptodon- toids. In all, details are given for 10 living and 15 extinct species. The anatomical details largely sup- port most accepted views of cingulate systematic relationships; there are few surprises. It is con- firmed that the pampatheres have a decidedly glyptodont-like basicranium, and it is noted that Bordas had stressed this same point in a 1939b paper. Introduction This work ("Cingulates") constitutes Part I (the first two-fifths) of the original Patterson-Segall manuscript on the ear region of the Xenarthra. The remainder (Pilosa: Anteaters and Sloths) is now planned for publication at a later date as Part II of the whole. The original outline called for an Introduction with a general discussion of the au- ditory region, the main Systematics Section (de- scriptions with discussions), and a Conclusion. No text of any sort has been found for the first and last of these sections, which I know existed at times past. I have tried to fill these gaps as directly and briefly as possible. My introduction attempts to take cognizance of the probable thinking of the two original authors as well as to provide a frame- work for the study. It is cast in as nearly modern systematic terms as is compatible with their text. The main body of their text is but little changed. It includes entirely new descriptions of both fossil and Recent forms (25 taxa). The Conclusions sec- tion, which I have tried to present in the light of our current understanding of cingulate interrela- tionships, is entirely mine, but in it I have at- tempted to stress their implied or stated conclu- sions. Three major goals were certainly in their minds when they began the work: ( 1 ) to increase the level of knowledge beyond that of their predecessors, especially van der Klaauw (193 1) and before him van Kampen ( 1 905); (2) to treat fossil and modern materials together; and (3) to integrate the new knowledge with the old so as to reinterpret xe- narthran relationships. That this was so is evident from the detailed descriptions and discussions they wrote, from the illustrations they had made, and from Patterson's several revision attempts. The approach of both men to systematics was conser- vative, so it is extremely unlikely that they would overemphasize evidence drawn from a single char- acter complex. They did of course consider the petrosal to be a generally conservative structure, and the mastoid to be far more variable, and thus the whole to be a most reliable and informative guide to understanding phylogeny, a concept dat- ing back at least to Flower (1870). With xenarthran systematics as uncertain as it was during the decade of this work (and in some ways as it remains today; see below), it is easily understood that Patterson and Segall would try to PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES bring forward new evidence from this relatively neglected source. I am convinced that Patterson's attempts to sort this out and to fathom the im- plications account in part for the long delays ne- cessitated by his revision attempts. At least one insight (and probably many others) gained from this early collaboration with Segall spilled over into the Patterson and Pascual works ( 1 968, 1 972) in the form of their suggestions about xenarthran relationships, especially that of pampatheres and glyptodonts. Since Segall did not have as extensive an insight into systematics and evolution as Pat- terson did, he was content to have Patterson make whatever changes he thought appropriate. Hence to sort out the thinking in this area, we need only search for the clues that Patterson may have left us. One of these, and I believe this is the key one, comes from the mentioned works of Patterson and Pascual (1968, 1972) wherein the Xenarthra are subdivided in the modern way, into the Cingulata, Vermilingua, and Pilosa.1 They included three families within the cingulates, subdivided as fol- lows: 1 Modern usage usually divides Xenarthra into these three subordinal groups based upon the distinctiveness of each in terms of overall morphology (Engelmann, 1985) and supported by evidence from serum protein (Sarich, 1 98 5) and eye lens proteins (de Jong etal., 1985). As Engelmann points out, however, the older notion of there being a dichotomy rather than a trichotomy may yet prove to be correct, for the many morphologic fea- tures unite the Pilosa (sensu lato), and the modern pro- tein evidence merely indicates remoteness. fPalaeopeltidae: Dasypodidae: fUtaetinae: Priodontinae: Dasypodinae: tStegotheriinae: Euphractinae: Priodontini: Tolypeutini: Euphractini: Chlamyphorini: tEutatinae: fPeltephilinae: fPampatheriinae (= fChlamytheri- inae and = |Chla- mydotheridae): "fPalaeopeltis (= ^Pseudorophodori) "fUtaetus Priodontes Cabassous Tolypeutes Das y pus ^Astegotherium fPseudostegotherium "fStegotheriopsis \Stegotherium Euphractus Zaedius Chaetophr actus fPaleuphractus ^Macroeuphractus ^Proeuph ract us fStenotatus2 fProzaedius Chlamyphorus Burmeisteria \Eutatus fDoellotatus fProeutatus \Meteutatus fPeltephilus fParapeltocoelus \Peltocoelus fAnantiosodon fEpipeltephilus "fChlamytherium "\Holmesina ^Plaina ^Kraglievichia At the time of the study, this was considered to be a stegotheriine. FIELDIANA: GEOLOGY fGlyptodontidae: fPropalaeohoplophor- inae: fSclerocalyptinae (= fHoplophor- inae): fDoedicurinae: fPanochthini: fDoedicurini: fGlyptodontinae: t Vassallia fMachlydotherium ^Glyptatelus "fPropalaeohoplophorus "\Cochlops \Eucinepeltus fAsterostemma fMetopotoxus ^Palaehoplophorns fProtoglyptodon fEosclerophorns t Trachycalyptus fPlohophorus (= "fHoplophractus and fHoplophorus, in part) fParahoplophorus fLomaphorops fStromaphorus "fStromaphoropsis \Eosclerocalyptus fPlohophoroides fLomaphorus fHoplophorus fBrachyostracon = fGlyptotherium [a glyptodont] fPseudoeuryurus "fUrotherium "fNeuryurus (= fEuryurus) fNopachtus fPropanochthus \Panochthus fComaphorus fEleutherocercus fProdoedicurus fXiphuroides fPlaxhaplous fDoedicurus fParaglyptodon "fGlyptotherium fNeothoracophorus \Glyptodon fBoreostracon = ^Glyptotherium The present work covers only the Cingulata. The Pilosa, the Vermilingua, and possible ancestral stock(s) will be treated in a subsequent report. I had hoped that the long-awaited monographic study of the evolution of the Xenarthra, edited by G. G. Montgomery (1985), would shed much light upon the origin and evolution of the order. How- ever, although many details of interrelationships are presented in it, especially those among the ar- madillos, the phylogenetic systematics of the order remains unclear in several respects: ( 1 ) the group's origin is still uncertain, but now it is at least de- monstrably remote; (2) its closest relatives within the Mammalia remain a mystery; (3) its major PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES subdivisions are thus far distinct, without trou- blesome (or enlightening) intergradations, and these subdivisions also appear to represent early events; (4) within the cingulates, the exact relationship of armadillos and glyptodonts remains a debatable topic. For this work the most directly useful chapters in the Montgomery volume (1985) are the first seven. That by Glass reviews the history of eden- tate classification and concludes with a formal pro- posal for use of the ordinal name Xenarthra to replace Edentata. (Bordas, 1 939a,b, had done this, but except for some paleontologists [Hoffstetter, 1958, 1 969], it was not widely accepted and used.) Wetzel in two other useful introductory chapters deals with identification, distribution, and tax- onomy of living taxa exclusively. Sarich's (1985) report on albumin immunological evidence and de Jong's on eye lens protein structures are nec- essarily based upon modern materials. They add significant new dimensions to our understanding of relationships within the order by bringing to bear on the subject such different kinds of evidence from the usual classical morphological kinds. Only two of the seven works deal directly with fossil cingulates, that of Edmund (1985) on the fasci- nating North American pampatheres (giant ar- madillos); and an entirely theoretical one by Engelmann (1985), which is a thoughtful and far- reaching cladistic analysis that seeks to examine the phylogeny of the entire order. Abbreviations used are for the most part stan- dard; the Appendix provides a detailed listing. The systematic section of the original manuscript is written in a style that combines description with discussion and sometimes conclusions. The over- all conclusions will await the final section, but matters dealing with just the cingulates are dis- cussed and conclusions drawn by Turnbull in a Conclusions section of this work. Dasypodoids The Utaetus Group fUtaetus Ameghino, 1902. Figure 1A. A cranial fragment of Utaetus buccatus Ameghi- no, AMNH 28668, from the early Eocene Casa- mayor, provides the only evidence of skull struc- ture in the Xenarthra prior to the Oligocene Deseadan and in the Dasypoda prior to the Mio- cene Santa Cruz. The specimen has been well de- scribed by Simpson (1948, pp. 82-83), but further study of it in light of the extended comparisons made during the preparation of this paper has re- vealed a little information additional to that al- ready published. The squamosal is the only part contributing to the auditory region that is present in the specimen. The glenoid surface is low on the skull, a point emphasized by Simpson, lower than in Dasypus, a form outstanding in this respect among living armadillos, but not as low as in Peltephilus. The articular area is almost flat, not quite transverse but extending a little anterolaterally. Behind the glenoid surface is a rather small process, com- pressed anteroposteriorly and terminating ven- trally in a fairly sharp point, that sends a spur of bone medially in the direction of a posterolaterally extending ridge formed by the alisphenoid and the medial part of the squamosal. Simpson referred to this process when he described the glenoid fos- sae as being "... underhung by the external au- ditory meati posteriorly . . . ," a statement which we are inclined to question (see below). Posterior to this process there is a rectangular area gently convex transversely and decidedly concave anteroposteriorly; a large foramen is situated near the center. The rectangular area narrows ventro- medially and is followed posteriorly by a second, somewhat smaller process. From the gap between the spur on the anterior process and the ridge formed by the squamosal and alisphenoid, a groove runs posteriorly in the endocranial surface of the squamosal. In the matrix that covered this area, small fragments of bone were present between spur and ridge. There can be no doubt that these were remnants of a bridge that connected spur and ridge and formed the ventral rim of a foramen leading to the groove just described. This description has been taken from the left side of the specimen. The right side is less com- plete, but the foramen in the rectangular area and part of the groove medial to it in the endocranial surface of the squamosal can be seen. These two foramina are the clues to the inter- pretation of the structure preserved. The medial one is posterolateral to the foramen ovale and pos- teromedial to the glenoid cavity, occupying essen- tially the same position as a similar opening de- scribed here in Tolypeutes (p. 1 2). From it, a groove in the endocranial surface of the squamosal, cov- ered and converted into a canal by the periotic, leads posteriorly and hence in the direction of the mastoid foramen, with which the corresponding foramen in Tolypeutes is in communication. The FIELDIANA: GEOLOGY identity of the openings in the two forms would seem to be virtually certain. The foramen in the rectangular area leads to a short canal, lateral to the groove just described, that passes directly up- ward through the squamosal to open internally low in the cranial cavity. From this internal opening, a groove, also no doubt covered by the periotic in life, extends upward and backward to cross the low, inconspicuous tentorium. Canal and groove obviously transmitted a vein tributary to the lat- eral sinus system, and it is, we think, evident that this second foramen is the postglenoid. It therefore follows that the process anterior to it is a true postglenoid and that the posterior process is the posttympanic. This is a most important point; Utaetus is the only armadillo, and in fact the only xenarthran, thus far known in which such a process occurs. Its presence adds another significant item to the list of resemblances between Utaetus and the Palaeanodonta3 drawn up by Simpson (1948, p. 88). Postglenoid and posttympanic processes are closer together than in palaeanodonts. The lat- ter process appears to have been somewhat more independent of the mastoid than in other xenar- thrans. Due to the presence of the postglenoid, it is probable that there was no open space above the porus (see below). There is no indication whatever of an epitym- panic sinus in the squamosal and indeed no certain evidence of any appreciable epitympanic recess. It is most unfortunate that no portion of the tympanic is preserved, since the evidence of the squamosal upon the size of this element is some- what equivocal. The size and shape of the area posterior to the postglenoid process are such as almost to suggest the existence of a well-developed bony meatus of palaeanodont or Euphractus type that would have underhung the glenoid to some extent; yet if such had been the case, it is wholly unlikely that the large postglenoid foramen would have occupied the position it does, where it would have been at or very near the porus. We are strong- ly inclined to suspect that the tympanic was sim- ple, little or not at all advanced beyond the ring form, hence of the type that was certainly primi- tive for the Xenarthra and that persisted with little change to Pleistocene and Recent time in many sloths and armadillos, and, apparently, in all glyp- todonts. 3 Epoicotherium among palaeanodonts appears to lack this process (Simpson, 1927), and in the related Xeno- cranium it is feebly developed (Colbert, 1 942). The Priodontes Group The Priodontines Priodontes F. Cuvier, 1825. Figure 1B-C. One young adult skull of Priodontes gigas E. Geoffroy, FMNH 2527 1 , has been available to us. The tympanic is horseshoe-shaped with an irreg- ular, slightly expanded medial border. At the junc- tion of the rather slender anterior crus with the medial border, there is a thin, rather blunt and short styliform process that imparts a right-angled appearance to this portion of the bone. A similar but less prominent process is figured by van Kam- pen(1905, p. 491, fig. 43). The bone is small in comparison with the size of the skull, being actually somewhat smaller than that of the closely related and much smaller Ca- bassous. The crista tympanica and, correspond- ingly, the sulcus tympanicus are well developed. A recessus meatus is present, but there is no in- dication of a cylindrical part. A chordafortsatz (the term applied by Bondy, 1907, to a process origi- nating from the inside of the dorsal part of the posterior crus that runs parallel to and provides a bony support for that part of the n. chorda tympani posterior to the malleus) is present. The anterior crus approaches but is not in contact with the squa- mosal at the posterior extremity of the medial bor- der of the glenoid cavity and also may contact a part of the peculiar, characteristically xenarthran process from the crista facialis of the periotic. As in most xenarthrans, the sulcus malleolaris is very well developed, and the bony anterior end of the malleus is plainly visible externally when in po- sition. The posterior crus rests on the tympanohyal laterally and on the medial extension from this element medially. (The probability that this me- dial extension is the caudal entotympanic is dis- cussed below.) It closely approaches but does not unite with the squamosal. The porus opens anter- oventrolaterally and is notably deeper than wide. The squamosal does not approach the wide inci- sure tympanica in this and most other armadillos (in later members of the Euphractus group the crura meet and the superficies meatus grows down toward them), but arches high above it, leaving a conspicuous open space above the porus, in which the ossicles, when in place, may be seen in a lateral view of the skull. This is characteristic of cingu- lates, although conditions in Utaetus would sug- gest that it may have been primitive for the Xe- narthra as a whole. PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 1 cm foramen ovale glenoid fossa postglenoid process — postglenoid foramen Up post-tympanic process B tympanohyal P°slerior oa,0,id ,oramen i 3 cm inflated medial border of glenoid fossa glenoid fossa postglenoid foramen tympanic mastoid process pars mastoidea paroccipital process occipital condyle foramen lacerum posterior petrosal foramen magnum Fig. 1 . Eutatus buccatus, amnh 28668: A, cranial fragment shown in ventral view. Priodontes gigas, fmnh 25271: B-C, posterior part of skull shown in ventral and left lateral views. The glenoid fossa is elongate anteroposteriorly and deeply concave transversely; it does not over- hang the tympanic. The medial border is formed by the squamosal, which is here considerably in- flated, the inflation involving part of the alisphe- noid as well. [There is no developed postglenoid process.] Van derKlaauw( 1924, 1931, p. 267) has shown that two different elements (or at least centers of ossification) may enter into the composition of the entotympanic, a rostral part that is connected dur- ing early ontogeny with the cartilage of the Eu- stachian tube, and a caudal part that is connected with the tympanohyal. These two parts may, and very often do, grow toward each other and fuse. As a result, in most adult edentates no trace of a former separation remains. Very young specimens occasionally reveal indications of a division, and some of these are described in this paper. In some armadillos the entotympanic does not become os- sified and in others ossification is not extensive. These last, exemplified by Priodontes and to a less- er extent by its relatives, retain traces or at least suggestions of the division into the adult stage. Priodontes possesses what is clearly a rostral en- totympanic. This is a small, irregularly shaped FIELDIANA: GEOLOGY bone, loosely connected to the anterointernal edge of the tympanic and thus usually lost during mac- eration. Van Kampen (1905, pp. 491-492) com- mented that he had encountered no description of it previous to his own, but stated that it had been figured by G. Cuvier (1825 [1834-1836 edition originally cited, but not seen by W.D.T.]) in the Ossemens Fossiles. The figure in question, how- ever, is so poor as regards the auditory region that we are quite unable to determine what is or is not represented in it. Our example of this element does not appear to be as well ossified as the one figured by van Kampen; it is not as broad transversely and, as it does not extend as far posteriorly, there is a pronounced gap between it and the medial extension from the tympanohyal. The surface of the bone is pierced by numerous minute vascular foramina, by a deep, irregular groove for the ca- rotid present in the medial face. There is a semi- circular excavation on the lateral for the Eusta- chian tube which is not completely surrounded by bone. Anteromedially, the rostral entotympanic sends up a spur that comes in close contact with the basisphenoid. The bone does not extend for- ward to enclose a foramen lacerum medium. The question of whether or not a caudal ento- tympanic is present in Priodontes is of some in- terest. Van Kampen (1905, p. 495) and van der Klaauw (1931, pp. 108, 243) describe tympano- hyal in this form, and in Cabassous and Tolypeutes to a lesser degree, as broadening to a leaflike ex- pansion at the tip. The expansion is anteroexter- nal, anteromedial and posteromedial in direction, extending posteriorly to a sutural connection with a ridge running anterointernally from the tip of the paroccipital process and anteromedially to- ward the rostral entotympanic. Anterolaterally the expansion extends beneath the main portion of the tympanohyal, with which its lateral extremity is fused. It is on this expansion, and not primarily upon the tympanohyal proper, that the posterior crus of the tympanic rests. It is possible that this apparent expansion of the tympanohyal is actually the caudal entotympanic. Position relation to the exoccipital and anteromedial expansion toward rostral entotympanic are all features of the caudal entotympanic rather than of the tympanohyal. No suture between tympanohyal and expansion is vis- ible, to be sure, but if present this would have closed during intrauterine life. In the related Tol- ypeutes, Bondy (1907, pp. 349-350) regarded a comparable structure as a process from the peri- otic, a determination with which we are not in agreement (see below). The tympanohyal proper is inclined postero- medially and is fused with the periotic, continuing without interruption into the crista facialis in this form, and, apparently, in all Xenarthra. It is also fused with the pars mastoidea posterior to the sty- lomastoid foramen. This is an age character in certain other armadillos, and possibly here also. Posterointernal to this foramen, as noted by van Kampen, the tympanohyal, together with the mas- toid, exoccipital, and supposed posterior entotym- panic, enclose a second opening comparable in size to the stylomastoid [foramen]. The articular sur- face for the stylohyal is depressed relative to the surrounding structure. The ventral surface of the periotic [petrosal] is unusually irregular. The promontorium is elon- gate, oval in outline and slopes laterally more grad- ually than in other armadillos; it is surrounded anteriorly and medially by a flat, horizontal shelf of bone, which is not in contact with either the basioccipital or basisphenoid. A periotic [petrosal] shelf in this position is unique among Xenarthra. Posteromedial to the fenestra rotunda there is a low blunt ridge that forms the anterior border of the foramen lacerum posterior. This ridge ap- proaches a bluntly angular process from the basi- occipital, which underhangs the medial shelf of the periotic [petrosal]. The recessus epitympanicus is shallow and separated only by a slightly raised border from the glenoid cavity. There is no epi- tympanic sinus in the Priodontes group. The crista facialis is deep [i.e., is more dorsally positioned than the rest of the ventral surface of the petrosal] and has a slight inward curvature. Anterior to the apertura tympanica canalis facialis, the crista gives origin to an anteromedioventrally-directed pro- cess of irregular, although essentially tripartite, shape and delicate structure. The anterior portion of this crest is lanceolate, acutely pointed and slightly concave ventrally; posterior to this are two smaller, more ventrally directed projections with slightly expanded extremities. This process, which reaches its greatest known development in Priodontes, is characteristic of the Xenarthra as a whole. Its significance as a mor- phological character of the group does not seem to have been fully realized heretofore, and its na- ture has been misinterpreted. A structure of such delicacy is particularly liable to damage during maceration of the skulls of Recent forms and to loss prior to burial or during preparation of the skulls of extinct ones. In forms with well-devel- oped bullae, it may be reduced in complexity and is not visible externally. These reasons no doubt PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES [IBRARY I). OF I. URBANA-CHAMPAIGN account for the failure to appreciate the signifi- cance of the structure. Van Kampen (1905, pp. 476, 484; figs. 39—40, O) observed this process in sloths and described its position as medial to the processus anterior to the malleus, but failed to note its connection with the crista facialis, even de- scribing it as loose in Choloepeus. He regarded it as probably homologous with the ossiculum ac- cessorium malleoli. Van der Klaauw ( 1 924, p. 118; 1931, p. 236) believed the process to be a part of the periotic, calling it a processus perioticus su- perior, although he used the abbreviation "oss. ace. mi." in several of his figures of ground sloths (1931, figs. 1, 4, 6). Curiously enough, neither of these authors refers to conditions in the armadil- los, in several of which the process is well devel- oped and its relation clear. As will be evident from the description given above, we agree with van der Klaauw that the process is not an ossiculum accessorium malleoli. We have examined large series of xenarthrans, in- cluding many young specimens, and in none have we found any indication that it is a separate ele- ment; all the evidence— position, union with the crista facialis, and relation to other elements of the auditory region— overwhelmingly indicates that the process is a part of the periotic. Van der Klaauw's term processus perioticus superior is not strictly accurate topographically; the process is ac- tually lateral rather than superior as regards its point of origin. We suggest the name processus cristae facialis in allusion to the demonstrated con- nection between process and crista. The pars mastoidea of the periotic is widely ex- posed on the surface of the skull. It forms the posterior and inferior portions of the massive and very prominent mastoid process, and participates to a slight extent in the root of the paroccipital process. The latter is better developed than in any other living armadillo. A deep, broad excavation runs anteromedially between the two processes. The basicranial portion of the exoccipital is more extensive in priodontines than in other armadillos. In Priodontes a prominent ridge runs anterome- dially from the tip of the paroccipital process to the supposed caudal part of the entotympanic; be- tween this ridge and the condyle there is a trian- gular, concave area. The internal carotid artery runs in the groove on the medial face of the entotympanic described above, and enters the cranial cavity through the anterior part of the wide gap between periotic [pe- trosal] and basisphenoid. A poorly defined groove in the medial part of the basioccipital indicates the presence of a second artery entering the skull at the same point as the internal carotid, as in other armadillos. The postglenoid foramen is situated in a slitlike recess at the posterior extremity of the glenoid cavity. The fairly large subsquamosal openings are variable in number and situated in the posterior root of the zygoma. Posterior to the foramen ovale is a wide gap between squamosal, tympanic, and periotic [petrosal] that probably gives passage to a vein (see below, Tolypeutes). The mastoid foramen is large and opens poste- riorly on the occiput. There is no closed channel in the mastoid ventral to it. The foramen lacerum [posterior] is bordered by the exoccipital, in which it forms a deep, semilunar notch posteriorly, and by the periotic [petrosal] anteriorly. The foramen stylomastoideum primitivum is large and bor- dered by the tympanohyal medially; a slight groove in the mastoid process, bordered posteriorly by a sharp ridge, runs ventrally from it. Cabassous McMurtrie, 1831. Figure 2A-C. One skull of Cabassous lugubris, FMNH 22437 (fig. 2B-C), and three of Cabassous loricatus have been examined, including FMNH 2647 1 (fig. 2A), 47959, and 47960. The tympanic is horseshoe-shaped and is slight- ly larger and considerably better developed than in Priodontes, being notably expanded in the me- dial direction, displaying approximately the same width throughout and forming a very slight ru- diment of a cylindrical meatus laterally. In C. lor- icatus, a medially projecting styliform process is present, which borders the Eustachian tube ante- riorly; in C. lugubris this is barely indicated. The crista tympanica, the sulcus, and the chordafort- satz do not differ from those of Priodontes. The spina tympanica posterior is very well defined, and the posterior crus is somewhat thickened and has an extensive attachment surface for cartilaginous or fibrous connection with the tympanohyal, the spina extending up to contact the squamosal. An- teriorly, the tympanic, as in Priodontes, has only a loose connection of this sort with a ridge on the squamosal, and the crus is free dorsally on the bony skull. In contrast to Priodontes and Toly- peutes, the crus does not closely approach the pro- cessus cristae facialis of the periotic. Medially, and postero- and anteromedially, the tympanic and entotympanic are in close contact in fully adult skulls. The porus, incisura tympanica, and the space above it are essentially as in Priodontes; the same is also true of the glenoid fossa in C. lugubris, but FIELDIANA: GEOLOGY 2cm mastoid tympanic foramen ovale styliform process entotympanic posterior opening carotid canal foramen lacerum posterior eustachian opening foramen lacerum medium forward extension of petrosal foramen ovale fenestra oval is mastoid paroccipital process promontorium anterointernal the promontorium , region mastoid foramen 3 cm Fig. 2. Cabassous loricatus, fmnh 26471: A, right side of basicranium of skull shown in ventral view. Cabassous lugubris, fmnh 22436: B-C, posterior part of skull shown in ventral and left lateral views. PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES in C. loricatus this area is somewhat wider and flatter. The entotympanic is much better developed than in Priodontes and forms, together with the tym- panic, a well-developed bulla. In adult skulls, it is an irregularly shaped but essentially curved bone, pierced by several minute vascular foramina in addition to the large carotid foramen. It is united with the tympanohyal posteroexternally and has sutural connections, which may fuse in old age, with the paroccipital [paroccipital process of ex- occipital], basioccipital, and basisphenoid, ptery- goid, and alisphenoid. The anterior extremity curves dorsolateral^ to the squamosal. The large Eustachian opening is surrounded by the entotym- panic, except laterally where it is bordered by the tympanic. The specimens at our disposal throw some light on the ossification of the entotympanic and on the relations between its rostral and caudal portions. The rostral portion, which forms rather more than half of the bone, completes its ossification first. In C. loricatus, ossification is completed with the growth of a spur or plate of bone that extends back to the exoccipital and thus forms the lateral border of the foramen lacerum posterior. A conspicuous gap is left between this spur and the tympanic. Concurrently with the growth of the spur, the cau- dal entotympanic ossifies medially from the tym- panohyal, passes between the tympanic and pars mastoidea, and finally grows anteriorly to fill the gap between the tympanic and the spur or plate from the rostral portion. In one skull of C. loricatus (FMNH 47958) spicules and thin sheets of bone can be seen in place in the dried cartilage occu- pying this gap. There can, we think, be little doubt that we are dealing here with a caudal entotym- panic; these conditions increase the probability that a comparable, although much less completely ossified, element is also present in Priodontes. In the only specimen of C. lugubris at our disposal, a young adult, the rostral entotympanic does not send a spur to the exoccipital. The tympanohyal is less complex than that of Priodontes; in C. loricatus a rudiment of a vagina processus hyoidei is developed. The ventral surface of the periotic [petrosal] is less complex than in Priodontes. The sulcus and crista facialis and the process from the latter are, in comparison, only moderately developed. A small, irregularly shaped, flat plate of bone, vary- ing in degree of development, extends forward from the promontorium to partially enclose the fora- men lacerum medium laterally. Anterointernal to the promontorium, the periotic [petrosal] is rounded ventrally and passes over the lateral edges of the basioccipital and basisphenoid, to which it is closely appressed. The recessus epitympanicus in C. lugubris is more sharply defined than in C. loricatus and Priodontes. It is not deeper than in these forms but is excavated posteriorly in such a way as to appear as a small fossa, in which the crus breve of the incus lies. The mastoid process is relatively smaller in Cabassous than in Pri- odontes—and. much smaller in C. loricatus than in the larger C. lugubris— and. the surface exposure of the pars mastoidea is in consequence less ex- tensive. The space between mastoid and paroc- cipital process, the latter very small in C. lugubris and almost nonexistent in C. loricatus, is very shal- low in comparison with the deep cleft present in Priodontes. The basicranial portion of the exoc- cipital is not as large, relatively, as in Priodontes. There is, however, a concave area immediately in advance of the condyle, and beyond this the an- terior extremity of the bone curves down to meet the caudal portion of the entotympanic. The posterior opening of the carotid canal is situated at the center of the entotympanic. The foramen lacerum medium is situated between the basisphenoid, pterygoid, and periotic [petrosal], and is completely concealed by the entotympanic. [I find this fits the condition in C. loricatus: f.l.m. is less concealed in C. lugubris.] Other minor open- ings, presumably vascular, lead to this foramen; in both C. lugubris and C. loricatus there is a small opening situated between basisphenoid and en- totympanic near the anterior extremity of the lat- ter, and in C. lugubris there is another in the ven- tral surface of the entotympanic anterointernal to the Eustachian tube. Between entotympanic and basioccipital, ante- rointernal to the foramen lacerum posterior, there is a foramen, again presumably vascular, that is sometimes double and leads to a short canal be- tween basioccipital and periotic [petrosal]; this opens into the cranial cavity posterointernal to the foramen lacerum medium. There is no postglenoid foramen. Possibly in correlation with this loss the subsquamosal foramina are large, as indeed they are in Priodontes, in which the postglenoid fora- men is much reduced. Posteroexternal to the fo- ramen ovale is a groove in the squamosal leading to a gap between the squamosal and periotic that probably transmits a vein. The mastoid foramen is in the usual position; the vein runs ventrally in a groove in the occipital surface that is partially covered by a bridge of bone supplied by the squa- 10 FIELDIANA: GEOLOGY mosal and pars mastoidea of the periotic. The fo- ramen lacerum posterior is large. The foramen stylomastoideum primitivum resembles that of Priodontes. The Tolypeutines Tolypeutes Illiger, 1811 Seven skulls of Tolypeutes tricinctus have been available to us. [Presumably these were FMNH 21407, 28339-28342, 28345, and 54353. Appar- ently Patterson and Segall thought that they re- sembled the priodontines closely enough that it was not necessary to have drawings made.] The tympanic is similar to those of Priodontes and Cabassous, but shows a surprising amount of variation in medial expansion. In degree this var- ies from slight (no greater than in Priodontes) to considerable (very nearly as in Cabassous lorica- tus); in one specimen the anterior portion is much more expanded than the posterior. The skulls at hand are all adult, which suggests that the observed variability is not dependent upon age. The styli- form process varies from practically absent to well developed, and is variable as to form. Crista tym- panica, sulcus, spina posterior, and caudal chordafortsatz4 are as in the preceding forms. The posterior crus rests mainly on the tympanohyal, not on the periotic as maintained by Bondy, but also extends dorsally to form a sutural union with the squamosal. The same author stated that the anterior crus is widely separated from the skull. This is only partly true. The large sulcus malleo- laris together with the very well-developed ante- rior process of the malleus divides the anterior face of the dorsal portion of the crus into two parts. The lateral portion is freely projecting and well separated from adjoining portions of the skull, but the medial is in close contact with the ridge formed by the squamosal and the underlying processus * Bondy (1907, pp. 349-350) stated that the posterior crus of the tympanic in Tolypeutes rests on a process from the periotic (this is actually the tympanohyal), to which he applied the name "chordafortsatz." In the in- troductory part of his paper, however, he defined this term as applying to a process originating from the inside of the dorsal part of the posterior crus of the tympanum and running parallel to and providing a bony support for that part of the n. chorda tympani posterior to the malleus. His use of this term later in the same paper for a process supposedly from the periotic is difficult to un- derstand, particularly since a caudal chordafortsatz as defined by him is present. cristae facialis, to which it is bound by fibrous tissue. The glenoid area is similar to that of Cabassous, but is less concave medially and the ridge on the squamosal that forms the medial border of the area is much less pronounced. The rostral entotympanic is similar in all re- spects to that of C loricatus, including the devel- opment, in some specimens, of a spur of bone extending back to the rudimentary paroccipital process. Due to the variability in the degree of medial expansion shown by the tympanic, the de- gree of union between the two bones and the def- inition of the bony Eustachian tube varies. The caudal entotympanic is very poorly ossified; only a rudimentary projection from the tympanohyal is present in some of the specimens. Tolypeutes is thus to a considerable degree intermediate be- tween Cabassous and Priodontes in the degree of ossification of the entotympanic as a whole. The tympanohyal exhibits no differences from that of Cabassous lugubris. The ventral surface of the periotic [petrosal] is structurally very similar to that of Cabassous but exhibits certain differences in its relations to sur- rounding elements. Anterointernally it abuts against and does not overlie the basisphenoid. The processus cristae facialis is extraordinary in that it underlies and is closely connected to the ridge on the squamosal that forms the medial boundary of the glenoid area, and thus forms a large part of the attachment surface for the anterior crus of the tympanic. The epitympanic recess is precisely as in Cabassous loricatus. Bondy (1907) stated that the recess was bounded laterally by the periotic. This is not entirely correct; the squamosal forms the greater part of the lateral wall, the pars mas- toidea contributing posteriorly to a small degree. [Here Patterson and Segall were probably thinking "petrosal" and Bondy really meant "periotic," i.e., including petrosal and mastoid.] The pars mastoidea of the periotic is similar to that of Cabassous; due to the narrow occiput the squamosal is excluded from the occipital surface. For the greatest part of its length the mastoid pro- cess is in contact with the posterior crus of the tympanic, a point of resemblance to the euphrac- tines. The basicranial portion of the exoccipital is less extensive than in priodontines, but neverthe- less a rudiment is present of the crest that in Pri- odontes runs from the paroccipital process to the presumed caudal entotympanic. The foramina present few differences from those of Cabassous. The canal for the internal carotid is PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 11 better developed, due to the greater degree of os- sification of the rostral entotympanic, and the groove in the basisphenoid for the second artery is converted into a canal as a result of the contact of this bone and the periotic [petrosal]. A small foramen pierces the entotympanic in a dorso- ventral direction, opening into the cranial cavity behind the common opening for the two arteries. A postglenoid foramen is invariably present an- teroexternal to the recessus epitympanicus. Pos- teroexternal to the foramen ovale and immedi- ately anterior to the tympanic there is a groove in the squamosal, as in Cabassous; in Tolypeutes this is converted into a canal by the presence of the underlying processus cristae facialis. The canal leads to the venous system of this region of the skull. The identity of this canal with the compa- rable one in the Casamayoran Utaetus is virtually certain. Priodontes, Cabassous, and Tolypeutes ex- hibit an interesting structural transition from the open gap seen in the first to the closed canal present in the last. Due to the close connection between mastoid process and tympanic, the stylomastoid foramen opens ventrally on the anterior side of the process, anteroexternal to the posterior cms, a point of interest in view of the conditions ob- taining in the euphractines. The Dasypus Group The Dasypodines Dasypus Linnaeus, 1758 Sixty-eight skulls of Dasypus novemcinctus, in- cluding FMNH 14008 (fig. 3A-B), nine of Dasypus septemcinctus, and two of Dasypus kappleri have been examined.5 As noted in the [missing original] introduction, in much of the literature this genus is referred to as Tatu or Tatusia and the name Dasypus is applied to Euphractus and Zaedius. The tympanic is of primitive ring type, the me- dial side enlarging to some extent toward the Eu- stachian tube. The styliform process is either not 5 The suite of materials available to Patterson and Se- gall during the years that this section was under study includes the following FMNH specimens: D. novemcinctus-8858, 11089, 14008, 15857, 15967- 69, 18748-61, 20537, 21131, 21405-6, 28347^19, 30476-87, 30885, 34191, 34351-53, 39304-7, 41585, 41890, 43285-87, 45354-55, 48357, 51392, 51801-2, 51962-64, 52246, 54204, 54245, 55664, 63920. D. septemcinctus— 29331-38, 52354. D. kappleri -30348, 30350. developed or is very poorly defined. There is some variation in the degree of expansion, but this is by no means so extreme as in Tolypeutes. The medial border is irregular in the majority of specimens, the indentations and projections being very marked in some. The posterior crus is attached along much of the length of the tympanohyal and also has a slight sutural connection with the squamosal, but does not come into contact with the mastoid pro- cess. As in Tolypeutes, the anterior crus is in con- tact with both the squamosal and the processus cristae facialis. Other details of structure are sim- ilar to those observed in the forms described above. The glenoid region differs from that of all other armadillos. The articular surface is flat and rela- tively low on the side of the skull, being situated on a level below that of the dorsal extremity of the anterior crus of the tympanic. It is thus farther below the incisura tympanica than in other living forms (only Utaetus and the peltephilids among the extinct exceed it in this respect), and the post- glenoid surface slopes abruptly upward to form a rather acute angle with the base of the mastoid process. The [dorsal crest of the] zygomatic pro- cess of the squamosal is more prominent than in other living forms; it sweeps upward and backward almost to the squamoso-parieto-supraoccipital junction. The entotympanic is very poorly ossified in Das- ypus, and in the great majority of specimens there is no trace of it whatever in the dried skull. Three specimens of D. novemcinctus at our disposal show conclusively, however, that ossification may take place. In one of these (14008) there are traces of a rostral element attached to the tympanic by dried cartilage posterior to the vicinity of the Eustachian tube. In another (18761) there is a suggestion of a caudal ossification near the tympanohyal. The third (43286) has a comparatively well developed ros- tral entotympanic that forms a posterior border for the Eustachian tube, and gradually tapers in a posterior direction; between the posterior portion and the tympanic there is a gap, precisely as in Cabassous and Tolypeutes. The bone has only a cartilaginous connection with the tympanic and does not extend very far medially; thus only a small portion of the floor of the hypotympanic sinus is formed by bone. The tympanohyal is fused proximally with the periotic and is in close contact with the pars mas- toidea distal to the foramen stylomastoideum primitivum; fusion may take place here with ad- vancing age. The articular surface for the stylohyal projects ventrally to varying degrees and is medial 12 FIELDIANA: GEOLOGY PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 13 to the ascending portion of the posterior cms of the tympanic. The promontorium of the periotic [petrosal] is globular in form, very much more rounded than in any of the forms described above. Medial to it there is a moderate shelf. The processus cristae facialis is well developed, decidedly expanded dis- tally, where it is concave and in some specimens actually cup-shaped. The epitympanic recess is small and variable in depth and extends poste- riorly, although not to so marked a degree as in Cabassous and Tolypeutes. There is no epitym- panic sinus. The surface exposure of the pars mastoidea is small in comparison with those of the forms hith- erto described, widening only slightly as it passes to the ventral surface of the skull. It contributes equally with the squamosal to the formation of the short, blunt mastoid process. The basicranial region is narrow in comparison with those of other armadillos and the mastoid processes extend but little laterally beyond the level of the condyles. There is no trace of a paroccipital process. The condyles reach well below the level of the tym- panic, a condition not shown by any other living armadillo except Priodontes. The course of the internal carotid cannot be determined in the dried skull; no trace of a groove for its reception is present in the most complete rostral entotympanic preserved. There is, how- ever, no reason to doubt that it lay medial to this element and entered the cranial cavity through the gap anteromedial to the periotic [petrosal]. The groove for the second artery may be seen in the basisphenoid. The postglenoid foramen is im- mediately behind the glenoid surface and at nearly the same level. Two subsquamosal foramina may be present, one in the zygomatic root above and lateral to the postglenoid, the other caudocranial from it and near the point of origin of the upswept ridge of the zygomatic portion of the squamosal. The former is frequently absent, the latter more constant. The mastoid foramen, or foramina, and enclosed groove ventral to it are in the usual po- sition. The foramen stylomastoideum primitivum is anteromedial to the mastoid process. The Stegotheriines fStegotherium Ameghino, 1887 Two skulls of Stegotherium tesselatum Ameghi- no, PU 15565 (fig. 4A-B) and 15566, both studied previously by Scott (1903, p. 16) and by van der Klaauw (1931, pp. 3 1 2-3 1 3), have been examined. Scott called attention to the many features in which Stegotherium resembled Dasypus and expressed the opinion that the two were closely related. De- tailed study of the auditory region thoroughly sup- ports this view. The tympanic, present on the right side [of the illustration, the left side of the specimen] of 1 5566, is similar to that of Dasypus in structure, degree of expansion, and points of attachment of the crura. The incisura tympanica is likewise steeply arched. As van der Klaauw (1931) has pointed out, the malleus is in position, a feature not clearly shown in Scott's figure (1903, pi. 3, fig. 6), which is in- accurate in certain other respects as well. The glenoid cavity agrees with that of Dasypus in being equally low on the skull— an important point, but here the resemblance ends. The articular surface is long and semicylindrical, much as in Priodontes. Scott ( 1 903-1 904) has given a detailed description of the very peculiar jugal; the descend- ing process of this bone projects straight ventrally well beyond the lateral surface of the glenoid ar- ticulation. Some specimens of D. novemcinctus show a slight reduction of the ventral surface of the zygomatic process of the maxillary and thus exhibit an approach to the extreme degree of re- duction that has occurred in this region in Ste- gotherium. Dasypus kappleri has a much larger jugal than either D. novemcinctus or D. s. septem- cinctus, and the upper border of the zygoma is straighter than in these species; in both characters this large species resembles Stegotherium to a greater extent than do the smaller ones. There is no trace of an entotympanic in either specimen despite the fact that a tympanic is pres- ent in one of them. This is an important resem- blance to conditions in the dried skull of Dasypus and strongly suggests that the entotympanic was largely cartilaginous here also. Van der Klaauw (1931, p. 275) stated that Scott's (1903-1904) ac- count of a part of the petrosal might have referred to an entotympanic. From the specimens and Scott's account itself, however, it seems clear that the part under description was the promontorium. The tympanohyal, pars petrosa, and pars mas- toidea of the periotic, recessus epitympanicus, and mastoid process are all essentially as in Dasypus; only the promontorium differs, being less globular. A paroccipital process is equally lacking. The pro- cessus cristae facialis is broken away in the spec- imens. The very close similarity with Dasypus is also 14 FIELDIANA: GEOLOGY c 5 S d) CO -5? c/) o PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 15 apparent in the foramina, with the exception of the venous system which, save for the mastoid foramen, is reduced in comparison with that of other armadillos. The Euphractus Group The Euphractines Euphractus Wagler, 1 830, and Zaedius Ameghino, 1889 Thirteen skulls of Euphractus sexcinctus Lin- naeus, seven of Euphractus villosus Desmarest [now usually placed in Chaetophractus], and ten of Zae- dius pichiy Desmarest have been available for study.6 [Two of E. sexcinctus, FMNH 28350 (fig. 5A-C) and 34348 (fig. 6A-B), one of Chaetophrac- tus (Euphractus) villosus, FMNH 63865 (fig. 7A- C), and one of Z. pichiy, FMNH 23809 (fig. 8A- B) are illustrated.] Euphractus, Chaetophractus, and Zaedius are so similar in all characters of the au- ditory region that they may most conveniently be described together. As already noted, the name Dasypus is applied to these genera in much of the literature. Some authors place E. villosus in a dis- tinct genus, Chaetophractus Fitzinger (1871). Members of the Euphractus group, and partic- ularly the euphractines, possess well-developed, inflated bullae, better developed than in any other armadillos with the exception of members of the Peltephilus group. In these (q.v.) the bulla is equal- ly as complete but the structural details are quite different. Van Kampen (1905, p. 492) suspected but was unable to prove that the bulla was compound in the euphractines. Van der Klaauw (1924) found a cartilaginous entotympanic in an embryo of E. sexcinctus and accordingly believed that two ele- ments were involved. This is completely con- firmed by a very young individual of C. (E.) vil- losus (FMNH 63865, fig. 7 A) in which ossification of the entotympanic is well advanced but fusion with the tympanic has not yet taken place. Slightly more than half of the bulla is formed by the tym- 6 The suite of materials available to Patterson and Se- gall during the time that this section was under study includes the following FMNH specimens: E. sexcinctus-21403-4, 26427, 28350-54, 34346-49, 54325. C. (E.) vj7/osus-24331-35, 54352, 63865. Z. pichiy- 1 uncat, 2171 1, 23809-11, 25617, 28505- 7, 49942. panic, whose medial border extends in a rather irregular line from the opening for the Eustachian tube to a point just lateral to the external margin of the condyle. Tympanic and entotympanic fuse early in life, leaving no or at most very little in- dication of a previous separation. Externally, the inflated portion of the tympanic is on the whole smoothly convex, the smoothness being interrupted by a wide, shallow groove run- ning anterointernally from the vicinity of the mas- toid process and by a prominence anterior to it that extends medially from the ventral border of the meatus. Both groove and prominence are much more pronounced in Euphractus [and Chaeto- phractus] than in Zaedius. Internally, crista tym- panica and sulcus tympanicus are well developed, the abrupt medial border of the crista forming a sharply defined boundary to the sinus hypotym- panicus. The tympanic extends but little beyond the crista in its posterior and central portions, but, anterointernally [where it may in part be rostral entotympanic], the space between the crista and the opening for the Eustachian tube is extensive, more so in Zaedius than in Euphractus [and Chae- tophractus]. The bone forms a complete ring below and medial to the inwardly sloping superficies meatus. This is obscured in adult specimens, but is particularly well shown in the young C. (E.) villosus mentioned above. Here the two crura may be seen to meet [and fuse] beneath the superficies meatus, forming a rather deep, very thin plate of bone, which continues anteriorly and posteriorly into the crista tympanica. An oblique suture is still present between the crura. This reveals that the anterior crus runs for the entire length of the su- perficies of the squamosal, thus forming the greater part of the bony plate, while the posterior crus does not extend anteriorly beyond the pars mas- toidea of the periotic. A clearly defined recessus meatus is present. The attachments of the crura in euphractines are quite different from those in the forms hitherto described. The anterior crus is not loosely attached by soft parts but on the contrary has a complete and extensive sutural connection to the adjacent part of the squamosal. The posterior crus retains the usual relationship with the tympanohyal, but this is here unimportant in comparison with the much more extensive sutural connections with the mastoid portion of the mastoid process laterally and with the squamosal medially, which go far beyond anything seen in the Priodontes or Dasypus groups. The proximal portions of both crura have enlarged to meet each other in the manner just 16 FIELDIANA: GEOLOGY PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 17 18 FIELDIANA: GEOLOGY A y \ glenoid fossa entotympanic \ foramen lacerum posterior external auditory meatus tympanic mastoid process tympanohyal occipital condyle foramen magnum external auditory meatus epitympanic recess and sinus opening crista tympanica and sulcus tympanicus processus cristae facialis promontorium petrosal Fig. 7. Chaetophractus (= Euphractus) villosus, fmnh 63865: A-B, left bulla and part of cranium shown in ventral and lateral views; C, broken right bulla (shown at larger scale), looking anterodorsolaterally into the bulla through the gap afforded by the missing entotympanic and basicranial elements. PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 19 1 (J <N - o o ,2 c 03 O a E e £> a u o o CO 3 •c o o a CO o M F a <! o O C\J O =\ •C ^) 6, 3 ■w > fc H 2 03 oo ^_» g § £ •c c 20 FIELDIANA: GEOLOGY described. Due to the extensive union between tympanic and squamosal, the sulcus malleolaris becomes obscured early. It is plainly visible in the young Chaetophr actus, and the tip of the processus malleolaris may be seen forming part of the bulla wall in some adults. In the interior of the bulla the tympanic fuses with the adjacent portion of the processus cristae facialis of the periotic [petrosal]. A very well developed external auditory meatus, which ossifies early in postnatal ontogeny, is pres- ent. It is inclined dorsolateral^ and the porus pre- sents upward and outward, the inclination being more pronounced in the smaller Zaedius. The ven- tral and anterior surfaces of the meatus ossify from the anterior crus of the tympanic, which in these forms is very much more extensive than the pos- terior, and come into sutural contact with the mas- toid process, which forms the greater part of the posterior wall. In old individuals the suture be- comes obliterated. Lateral to the stylomastoid fo- ramen, the meatus sends down a short, blunt pro- cess in some individuals. The posterior crus, which is small in comparison with the anterior although equal in relative size to the posterior crura in the Priodontes and Dasypus groups, is largely con- cealed by the mastoid process in lateral view and contributes very little to the posterior wall. The dorsal wall, lateral to the tympanic ring, is formed largely by the squamosal, the anterior crus con- tributing at the margin. In the very young speci- men of Chaetophr actus, the meatus has barely be- gun to ossify and there are traces of a cleft extending medially between the crura. This is an interesting point of resemblance to the eutatines and to the Tertiary euphractines described below. The glenoid surface is high on the skull and slightly convex. Posterior to it is a large, deep re- cess in the squamosal that extends back above and internal to the auditory meatus; the large post- glenoid foramen opens in the caudal extremity of the recess. The structure of this area is a charac- teristic feature of the entire Euphractus group. The entotympanic has sutural contacts with the exoccipital posteriorly— the contact here extend- ing ventrally almost to the level of the condyle, the basioccipital, and to a slight extent the basi- sphenoid medially. It reaches the pterygoid ante- riorly, the two bones forming a short crest above the level of the Eustachian opening. Laterally, it extends above the anterior margin of the tympan- ic, between this element and the alisphenoid, to the squamosoalisphenoid junction, as in Cabas- sous and Tolypeutes. Internally, the entotympanic is smooth, and its posterior extremity forms a sharp, prominent transverse ridge that continues upward into the crista facialis. There is no trace, in this or any other xenarthran, of a septum bullae, a structure that frequently occurs in other mam- mals with a compound bulla. Anteroexternally it fuses extensively in adult specimens with the pro- cessus cristae facialis. The medial wall of the bone is thickened and cancellous. The aperture for the Eustachian tube is completely surrounded by this element and by the tympanic. The entotympanic may be a compound bone, formed from rostral and caudal elements in these genera, but the ma- terial at hand, even the young individual of Chae- tophractus, throws no light on the matter. The paleontological evidence suggests, however, that the part, if any, played by the caudal portion was minor. The tympanohyal is in the usual position. It is clearly visible in very young specimens, in which it may be seen as a thin strip of bone lying medial to the united crura of the tympanic, but later be- comes fused with the surrounding elements. A slight depression for the reception of the stylohyal sometimes remains to mark its position in Eu- phractus, into which, as noted by van Kampen (1 905), the tip of the otherwise fused tympanohyal occasionally protrudes. The ventral surface of the pars petrosa of the periotic shows no outstanding peculiarities. The promontorium is prominent but not as globular as in Dasypus, being quite similar to that of Tol- ypeutes. The processus cristae facialis is a very well developed, gently concave bony plate that is rounded anteriorly and sharply triangular poste- riorly. As noted above, it fuses during postnatal ontogeny with the tympanic and entotympanic. The recessus epitympanicus passes directly into a large epitympanic sinus, which extends anteriorly beneath the posterior part of the recess behind the glenoid articulation and posteriorly into the pars mastoidea. In the young specimen of Chaetophr ac- tus already referred to, it is partially divided by a horizontal septum into two portions, of which the lower, and smaller, appears to correspond in part to the posterior evagination of the recessus epi- tympanicus seen in Cabassous and Tolypeutes. Van Kampen noted the presence of an epitympanic sinus in Euphractus and Zaedius but was uncertain as to whether it lay wholly in the squamosal. Our material shows that it is bounded medially by the periotic. In correlation with the width of the cranium, the pars mastoidea is widely exposed on the oc- cipital surface. It forms the whole of the mastoid PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 21 process, which is wide transversely, compressed anteroposteriorly, and inclined anteroventrally. The relations between this process and the tym- panic have been described above. The process is in contact laterally for most of its length with the outgrowth from the anterior crus of the tympanic; the rounded tip may project for a very short dis- tance in some specimens. The posterior face bears a small protuberance near the center. Between the process and the exoccipital there is a wide, shallow groove [leading from the mastoid foramen]. No trace of a paroccipital process is present. The internal carotid enters a canal in the ante- rior part of the entotympanic, and passes to the cranial cavity via the concealed foramen lacerum medium in all specimens but one of Zaedius, all of C. (E.) villosus, and slightly under half of E. sexcinctus. In the single Zaedius and the remaining E. sexcinctus, there is a groove in the entotym- panic, and the foramen lacerum medium is visible in the dried skull. The artery enters the cranial cavity by a small opening situated directly dorsal to the foramen lacerum medium and posterolater- al to the sella turcica. A second artery enters the skull immediately anterointernal to the foramen lacerum posterior and runs anteriorly in a canal, formed by the entotympanic below, periotic lat- erally and basioccipital medially, that opens in the cranial cavity at the posterolateral extremity of the sella. Several small squamosal foramina, rarely a single large one, occur above the porus in Eu- phractus. In Zaedius there is usually a single fo- ramen above and behind the porus, and occasion- ally one or two smaller ones as well. The mastoid foramen is in the usual position with a channel running ventrally from it, which is only excep- tionally bridged over in part by bone. Above it is a second, smaller venous opening. Both foramina lead into canals that pass anteriorly above the au- ditory region, through the diploe in this very thick part of the cranial wall, to open in the postero- lateral part of the cerebral cavity. Here the dorsal of the two canals (which becomes ventral in po- sition posteriorly) receives a sinus that enters the cerebral cavity through a venous opening in the posterior part of the orbit. Both canals are in com- munication with each other posteriorly and also with the transverse sinus, the canal from the post- glenoid foramen, the subsquamosal foramina, and the numerous small venous openings in the vicin- ity of the squamosoparietal suture. The foramen lacerum posterior, as usual, is oval and situated between exoccipital and entotympanic. The course of the facial nerve in armadillos of the Euphractus group is rather unusual. The nerve follows the usual course within the tympanic cav- ity and then turns laterally to pass through a canal formed by the pars mastoidea posteriorly and the tympanohyal and posterior crus of the tympanic posteriorly. On emerging from this canal at the foramen stylomastoideum definitivum, it runs an- teroventrolaterally along the sulcus in the anterior face of the mastoid process— and hence lies within the external auditory meatus— and finally leaves the skull at a point near the tip of the mastoid process, between this element and the outgrowth of the anterior crus of the tympanic. In young animals, this is in a slit between the two, but as ossification of the crus progresses the point of exit is converted into a foramen, a "foramen stylo- mastoideum tertium," so to speak. fPaleuphractus Kraglievich, 1934 Two skulls of Paleuphr actus argent inus Moreno and Mercerat, FMNH PI 44 12 (fig. 8C) and P 1 4442, from the Andalgalense of Catamarca, have been available for study. Both have suffered to some extent from distortion and breakage. The tympanic, as far as revealed, is very similar to those of Zaedius and Euphratus, differing only in the structure of the anterior crus. In degree of enlargement and lateral projection, this portion of the bone is comparable to that of Zaedius but it is not reflected posteriorly to floor the meatus. The well-defined ventral slit, here V-shaped, between anterior crus and mastoid process is therefore present, as in all extinct euphractine genera now known. The relation of the anterior wall of the tympanic to the deep recess, posterior to the glenoid surface, is precisely as in the living forms. The boundary between tympanic and entotym- panic is visible in the posterior half of the bulla. It corresponds closely with conditions seen in the young specimen of Euphractus [Chaetophractus] described above, except that here again, as in Proeuphr actus, there is a gap posteriorly between the two elements and no contact between exoc- cipital and entotympanic at the level of the con- dyle. The gap is smaller than in that form and is confined entirely to the posterior face of the bulla. This lack of ossification in the immediate vicinity of the tympanohyal, coupled with the absence of any indication of a division in the entotympanic of the young Euphractus [Chaetophractus], strong- ly suggests that the caudal portion of the entotym- 22 FIELDIANA: GEOLOGY panic played a very minor role, if indeed it par- ticipated at all, in the formation of the bone. The ventral surface of the pars petrosa of the periotic is comparable in all respects with that of Euphractus. The same is true of the pars mastoi- dea, save that the mastoid process projects freely in the cranioventral direction, due to the lack of reflection of the anterior crus of the tympanic. All foramina that can be seen in the auditory region differ in no respect from those of the living forms, except that due to the lesser degree of os- sification of the anterior crus, there is no "foramen stylomastoideum tertium" between crus and mas- toid process. tion between entotympanic and exoccipital at the level of the condyle. It is rather remarkable to find these primitive characters persisting in so large and specialized a form. Both available specimens appear to be fully grown, so that the lack of os- sification can hardly be regarded as due to im- maturity. The later, Macroeuphr actus retusus Ameghino, has a fully ossified bulla, to judge from Lydekker's figure. It is probable that an epitympanic sinus is pres- ent in this and other Tertiary euphractines, but sections would be necessary to conclusively dem- onstrate its presence. fProeuphractus Ameghino, 1886 Two specimens of Proeuphractus scalabrinii Moreno and Mercerat, FMNH PI 4360 (fig. 9A- B) and PI 5435, from the Andalgalense of Cata- marca, have been available; both are unfortu- nately incomplete, but nevertheless most of the structure can be determined. Proeuphractus and its close relative Macroeuphractus are similar in general to Euphractus and Zaedius and it will suf- fice here to discuss those points in which they differ. The bulla is less inflated than in the living forms and its anterior border shows an uninterrupted, gently concave sweep when seen in ventral aspect, a feature that is even more pronounced in the giant Macroeuphractus (Lydekker, 1894, pi. 35). The external meatus is as long as in Euphractus and Zaedius but is not closed ventrally by bone. The bony extension of the anterior crus, as in Macro- euphractus, did not turn posteriorly to meet the mastoid process, thus leaving an open V-shaped cleft between the two elements. The bony exten- sion continues ventrally below the porus, forming a groove between the crura. Tympanic and mas- toid are united medially to the point of exit of the facial nerve. The mastoid process is broken in our material; in Macroeuphractus it is very massive. In only one other character is there a major dif- ference between Proeuphractus and the living eu- phractines. As in the Santacruzian Prozaedius (see below), a cleft, reaching almost to the middle of the bulla, separates the entotympanic and tym- panic posteriorly. The position of the cleft is the same as that occupied by the caudal portion of the entotympanic in Cabassous, etc., a fact which sug- gests that this portion of the bone remained unos- sified in Proeuphractus. There is thus no connec- fProzaedius Ameghino, 1 89 1 An excellent skull of Prozaedius exilis Ameghi- no, PU 15579 (fig. 10A-B), from the Santa Cruz Formation of Patagonia, permits the following rather full account of this earliest euphractine. Scott's description of the skull (1903, pp. 70-72), which includes a brief summary of the auditory region, was largely based on this specimen, and van der Klaauw ( 1 93 1 , pp. 3 1 2-3 1 3) has published a brief note upon it. The bulla of this form is, as would be expected, more primitive than in any later euphractine, and resembles the bulla of the Pliocene eutatine Doel- lotatus (see below) in several respects. The tympanic retains some trace of the primi- tive ring form,7 being only moderately expanded medially in the posterior portion. It is considerably prolonged anteromedially, however, in the vicin- ity of the Eustachian tube. The external auditory meatus is barely developed, and above it there is a parallel-sided cleft between the mastoid process and anterior crus, which is U-shaped at the base. The shorter, V-shaped cleft of the Pliocene Paleu- phractus indicates that formation of the tubular meatus in euphractines involved bone growth at the base of the cleft as well as extension of the anterior crus posteriorly. The anterior crus forms a straight, nearly vertical ridge sharply edged lat- erally; there is a steplike indentation at the upper extremity, as in Doellotatus. In contrast to the liv- ing forms, there is no posterior continuation from the anterior portion of the crista tympanica, and 7 It should be noted that what appears at first glance to be a tympanic in Scott's figure (1903, pi. 6, fig. 20) of the closely related but longer skulled Stenotatus pata- gonicus Ameghino is actually the occipital condyle, which, as he states (p. 81), is deeply incised laterally. PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 23 \ <3 1 -« r ^ \ d i S i tt. I / o\ o U- 24 FIELDIANA: GEOLOGY PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 25 the tympanic consequently does not form a com- plete ring. The posterior cms is, as usual in this group, short in comparison with the anterior and curves upward medial to the mastoid process, which conceals its upper portion when the skull is viewed from the lateral side. As far as its attach- ment is concerned, it is nearly identical with Eu- phractus, the only difference being that the sutural connection to the mastoid process is less extensive. Prozaedius thus shows conclusively that, of the two crura, it is the posterior that earlier attains the definitive form and relationships encountered in the later members of the Euphractus group. This is not surprising since the structure here is merely an exaggeration of that seen in incipient form in the Priodontes and Dasypus groups, whereas the modification of the anterior crus has been much more extensive. The superficies meatus is nearly vertical, as in Doellotatus and juvenile Recent eu- phractines. In contrast to these later forms and due to the fact that the tympanic ring is not com- plete, there is a gap between the arched ventral border of the superficies and the incisura tympan- ica, in which the malleus and incus are visible [but not shown in fig. 10, however]. This is a primitive character and an interesting point of resemblance to armadillos of the Priodontes and Dasypus groups. The projecting tip of the anterior process of the malleus may be seen on both sides of the specimen. The fossa in front of the anterior wall of the meatus and behind the glenoid surface which is so characteristic of the entire Euphractus group is less extensive in Prozaedius than in later euphractines, being narrower transversely, due to the less ex- panded zygoma and anterior crus, and not ex- tending so far posterodorsally. The relations between tympanic and entotym- panic are very clear, the bulla not being "broken up into pieces" as van der Klaauw supposed. The rostral portion of the entotympanic, the only part ossified, is less developed than in later forms. An- teriorly, it completely fills the space between the tympanic and neighboring elements, extending laterally on the anterior face of the bulla to the projecting tip of the processus anterior of the mal- leus, as in Cabassous and Tolypeutes, and forming a curved suture posterior to the Eustachian tube with the anterior half of the medial border of the tympanic. Due to a fracture, it is impossible to be certain whether or not a pterygoid-entotympanic crest was present. This anterior portion of the bone is cancellous. Posteriorly, the entotympanic di- verges from the tympanic, forming a fairly wide medial strip that extends back to the exoccipital. A wide gap is thus formed between the two bones (not within the tympanic, as van der Klaauw thought might be the case), in which no trace of a caudal entotympanic is to be seen. In later forms this gap is closed by progressive ossification of the rostral portion. As already stated, there is no clear evidence that a caudal portion participated at all in the formation of the bulla. The form and po- sition of the opening for the Eustachian tube are as in the living forms. In degree of prominence, the promontorium re- sembles those of Euphractus and Zaedius. The recessus epitympanicus is well developed; the presence of the ossicles makes it impossible to determine whether or not a sinus was present. The pars mastoidea is comparable to that of these Re- cent forms in its degree of exposure, but, due to the much less expanded posterior root of the zygo- ma, the anterior half faces laterally rather than posterolaterally. The mastoid process projects more freely than in later forms, and bears a conspicuous buttress on the lateral edge a short distance below the junction with the occipital crest. Its squamosal component extends far ventrally, reaching to the posterior crus of the tympanic. Prozaedius resembles Paleuphractus in the ab- sence of a "foramen stylomastoideum tertium." The other foramina in the vicinity of the auditory region are precisely as in the living forms. Van der Klaauw's (1931, p. 313) "opening in the rostro- medial part of the bulla" is the usual carotid fo- ramen in the entotympanic. The Chlamyphorines Chlamyphorus Harlan, 1825 Specimens of this and of the related, somewhat larger Burmeisteria are rare in collections, and we are fortunate to have been able to examine a skull of Chlamyphorus truncatus Harlan, PU 538 (fig. 11A-B), an adult individual. The extraordinary little animals, so distinctive in external characters and in the relations of the pelvis and tail to the posterior portion of the carapace, are dwarfed rel- atives of the euphractines, from which group they almost certainly arose. We thus agree with Simp- son's remark (1927, p. 292) that Chlamyphorus is "... a peculiar and highly aberrant derivative of an essentially Dasypus [i.e., Euphractus]-like form." So far as the auditory region is concerned, this is not open to question, and evidence from other parts of the body would appear to be in accord. 26 FIELDIANA: GEOLOGY PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 27 The striking features of the skull of Chlamy- phorus, those immediately noticed on comparison with skulls of euphractines— triangular appear- ance, wide, deep, and short cranium, zygomata narrowing anteriorly, frontal protuberances, re- duced orbits, relatively enormous foramen mag- num, and so on— are all consequent upon the size reduction undergone by the chlamyphorines. They are the results of adjustments that have taken place in the head. Compromises have ensued beween overall size reduction and disproportionate re- duction of various parts, some of which (e.g., brain, depth of teeth, sinuses) have undergone relatively less reduction, while others (e.g., eyes) have under- gone more. In the auditory region and vicinity, adjustment to the relatively large size of the brain has resulted in certain differences from the eu- phractines in general appearance. For example, the recess posterior to the glenoid cavity, so conspic- uous in euphractines, is here reduced. The well- inflated bullae are aligned more obliquely, and, as van Kampen has noted, lie in a more ventral plane relative to the basicranium. Posteromedially the bullae have rather extensive contacts with the oc- cipital condyles, which are situated farther forward than in euphractines and greatly expanded in the anteroposterior direction. These distinctions are relatively minor, however, and in no way disguise the fundamental similarity between chlamyphor- ines and euphractines in this area. The bulla closely resembles that of Zaedins in shape, but is smoother and more convex. There is no trace of the prominence extending medially from the meatus, and the groove running ante- rointernally from the vicinity of the mastoid pro- cess is barely discernible; both these features, it will be recalled, were less prominent in the smaller Zaedius than in the larger Euphractus. [It should be noted that Segall has subsequently studied C. truncatus (1 976) and in that work this feature does not appear on the specimen illustrated (Segall, 1976, fig. 1, p. 433; MCZ 3209).] That portion of the tympanic anterointernal to the crista tympan- ica, between this crest and the opening for the Eustachian tube, is relatively more extensive than in Zaedius, in which, in turn, this part is relatively larger than in Euphractus. The attachments of the crura are as in the euphractines, the posterior being partially concealed by the mastoid process and the anterior united suturally with the squamosal. In the structure of the auditory meatus, how- ever, Chlamyphorus is sharply distinct from the Pliocene and later euphractines. The external ca- nal is short, the anterior cms not extending lat- erally and posteriorly to form a long bony meatus. The porus is consequently low on the side of the skull, at a level only slightly above that of the tip of the mastoid process. It is nearly circular in out- line, presents upward and outward, and has a prominent lip. From it a groove, situated mainly in the squamosal, curves upward and forward, ter- minating immediately lateral to the occipital crest. In this groove, and fitting over the lip of the porus, lies the extraordinary tube (van Kampen, 1 905, p. 493), ossified from the cartilaginous meatus,8 which has attracted the attention of all who have studied the chlamyphorine auditory region, from Harlan (1825) and Hyrtl (1854) on [and including Segall (1976, p. 431)]. The tube extends beyond the groove, curves forward over the root of the zygo- ma, and ends above the glenoid articulation. [Se- gall's 1976 dissection (pp. 433, 437) shows this to be a two-part tube, the proximal part occupying the groove, the articulating distal part extending more freely forward nearly to the frontal protu- berance and the eye.] The structure of this area is, of course, decidedly different from that of the Re- cent and Pliocene euphractines, but is derivable from that of the earlier forms. Acquisition of an ossified tube by forms comparable in meatal struc- ture to the Santacruzean Prozaedius would lead to the conditions seen in the chlamyphorines. The position of the glenoid articulation is ex- actly as in the euphractines, but the articular sur- face is rather more convex transversely. As a result of the dwarfing of the animal and the dispropor- tionate [lesser] reduction in brain size, the poste- rior root of the zygoma stands out from the side of the cranium to a greater extent than in euphrac- tines, and, as already noted, the recess behind the articular surface is reduced, although still fully rec- ognizable. No trace of a suture between tympanic and en- totympanic remains, but there can be no doubt that the bulla is compound. The crest formed by the entotympanic and pterygoid is prominent, but, as in Euphractus and Zaedius, it is above the level of the opening for the Eustachian tube, which is here very small. The tympanohyal is completely fused with the surrounding elements and no details of its struc- ture can be seen. A small pit on the occipital sur- face immediately above the bulla and medial to the mastoid process marks the position of the at- tachment of the stylohyal. 8 Van der Klaauw (1931, p. 257) implies that the me- atus in Euphractus and Zaedius is formed in the same way. We do not believe this to be the case. 28 FIELDIANA: GEOLOGY External examination and X-ray photography fail to reveal any trace of an epitympanic sinus. If present in the ancestry of Chlamyphorus, as may have been the case, elimination of this cavity is to be attributed to dwarfing. The pars petrosa of the periotic is not visible in the specimen at our disposal. The sutures between the pars mastoidea and the squamosal and occip- ital elements cannot be seen, but, from the position of the mastoid foramen, it is evident that the pars mastoidea forms the ventrolateral corner of the occipital surface. In comparison with the euphrac- tines, the mastoid process is small and extends laterally to a lesser degree. It conceals the posterior crus of the tympanic in side view and may con- tribute in a minor way to the formation of the groove in which the meatal tube lies. The medial face of the entotympanic is grooved for the reception of the internal carotid but no canal is formed, the artery passing anteriorly to the foramen lacerum medium, which opens at the anterior extremity of the groove. The second ar- tery passes forward in a second, more dorsally situated groove between entotympanic and basi- occipital, runs beneath a bar formed by the en- totympanic and pterygoid, and also enters the cra- nial cavity via the foramen lacerum medium. The postglenoid foramen, as in euphractines, is situ- ated at the posterior extremity of the recess behind the glenoid surface; it is relatively enormous in comparison with the size of the skull. There are no subsquamosal foramina. The mastoid foramen occupies the same position as in euphractines, but there is no groove running ventrally from it. The foramen lacerum posterior resembles those of Eu- phractus and Zaedius. The stylomastoid foramen, again as in these forms, leaves the skull between tympanic and mastoid process. The Eutatines fDoellotatus Bordas, 1932 Three specimens that are apparently referable to Doellotatus prominens (Moreno & Mercerat),9 9 Described as Eutatus prominens. Bordas (1933, p. 584) makes only passing mention of this species in his discussion of the eutatines and leaves the reader in doubt as to the generic assignment. Associated carapacial ma- terial permits at least tentative specific identification of our specimens, and the skulls are very similar to that of Doellotatus inornatus Rovereto, the type species, leaving no doubt as to the generic determination. [Earlier, FMNH PI 4351 (fig. 12D) and PI 4358 (fig. 12C) from the Araucanense and P14526 (fig. 12A-B) from the Corral Quemado, Catamarca, Argentina, have been examined. None of the three is complete and all are either broken or somewhat crushed, but between them most of the structural details can be determined. The tympanic in the later eutatines is an ex- traordinary bone, thick and so dense that it resem- bles the pars petrosa of the periotic in texture. The porus, due to the prominence of the crura, is deep set in the bulla, and has a rather irregular border. The anterior crus is high, reaching to the root of the zygoma, and very robust, expanded in the transverse plane, and with a sharp lateral margin. The upper extremity is indented in steplike fash- ion. There is no indication of a posterior reflection, such as occurs in Recent euphractines, but its pos- terior wall is concave, forming a recess above and slightly anterior to the porus. The posterior crus, as in euphractines, is short and largely covered laterally by the mastoid process, but it also has a prominent lateral margin; the attachments are also almost precisely similar. The margins of both crura continue ventrally beyond the porus, with the re- sult that a conspicuous groove is formed in the lateral face of the bulla, as in Proeuphr actus. The crura meet above the porus, as in Euphractus, ex- cluding the narrow superficies meatus of the squa- mosal from its border. The sutures in this region can be seen clearly in FMNH PI 4526 (fig. 12A- B). As already mentioned above (p. 28), the glenoid area, and in particular the fossa posterior to it, agrees in all essentials of structure with that of the euphractines. The entotympanic, nearly complete on the left side of FMNH PI 4351 (fig. 12D) and on the right of P14526, resembles that of Recent euphractines. It is cancellous in its anterior half, extends laterally above the medial half of the anterior face of the tympanic, where it is concealed externally by the enlargement of the squamosal in this area, and posteriorly to the foramen lacerum posterior. The bulla is thus completely enclosed by bone, in sharp contrast to the Miocene and Pliocene euphractines in which the entotympanic is not completely os- sified posteriorly. The Eustachian tube and styli- form process are comparable to those of Euphrac- tus. Ameghino (1891) had equated E. prominens of Moreno and Mercerat with his species Eutatus distans.] PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 29 The ventrolateral face of the pars petrosa of the periotic is rather flat, and has a rounded anterior extremity. [Here I suspect that Patterson has seen some other specimen(s), for only the anterior tip and ventromedial edge are exposed on the pe- trosals of the specimens listed here. The best ex- posure is seen in FMNH PI 43 51.] The promon- torium is low and inconspicuous in comparison with that of the euphractines. The ventral portion of the medial face is deeply grooved, and lower border of the groove forming a sharp edge. [The cranial side of FMNH PI 4526 possibly may have been further prepared subsequent to the Patterson- Segall study of it. That surface is irregular, and the meatus internus acusticus is prominent and deep.] Unfortunately, nothing can be seen of the crista facialis or recessus epitympanicus in any of the specimens. Whether or not an epitympanic sinus was present is uncertain. Doellotatus resembles Prozaedius in the degree of expansion of the posterior root of the zygoma, and the surface exposure of the pars mastoidea is consequently as much lateral as posterior in po- sition, as in that form. The freely projecting mas- toid process is stouter and less flattened antero- posteriorly than in Prozaedius; its base is closer to the condyle than in any euphractine. It bears a buttress on the lateral edge, exactly as in Prozae- dius, and a second eminence, dorsal to this and above and behind the meatus, to which the squa- mosal contributes. There is no groove or opening in the entotym- panic for the carotid, and there would appear to have been a visible foramen lacerum medium an- terointernal to the bulla, as in certain living eu- phractines. The arterial canal between entotym- panic, periotic and basioccipital was present, occupying the groove in the medial face of the pars petrosa. The postglenoid foramen, as in euphrac- tines, opens at the posterior extremity of the deep fossa posterior to the glenoid surface. The sub- squamosal foramina vary in number and size and are situated above the meatus. The mastoid fo- ramen is ventral in position, opening between the base of the mastoid process and the vestigial par- occipital process. In euphractines, the vein runs down the occiput in a groove, which is here com- pletely bridged over by bone. The narrow foramen lacerum posterior resembles that of the euphrac- tines very closely. As in the Tertiary members of this group, the stylomastoid foramen opens lat- erally between the posterior crus of the tympanic and the mastoid process. tEutatus Gervais, 1867 Examination of a specimen of Eutatus seguini Gervais in the American Museum of Natural His- tory (AM 11231, type of Eutatus brevis Ameghino) permits the statement that this latest and largest eutatine is in nearly every respect comparable to the earlier and much smaller Doellotatus. [This also appears reasonable, to judge by the figure giv- en by Ameghino (1906, p. 380, fig. 236).] fProeutatus Ameghino, 1891 Three skulls of Proeutatus oenophorus Ame- ghino, FMNH PI 3 197 (fig. 13A-B), PI 3 199 (fig. 13C-D), and PI 3203, from the Santa Cruz For- mation of Patagonia, have been examined. Scott (1903, p. 44) has stated that "the tympan- ic, which was loosely attached, is a small, incom- plete ring of bone, with large, irregular external meatus." It was not attached in any of the skulls that he figured. An incomplete tympanic is for- tunately preserved in FMNH PI 3 199, permitting us to confirm and to some extent amplify the brief account just quoted. [Unfortunately, it is now lost, or at least is no longer with the specimen. It is good that Patterson and Segall discussed and il- lustrated it, for that is now all we have to go on.] The specimen is incomplete anteriorly, the prox- imal end of the anterior crus being broken off. In contrast to Doellotatus, the tympanic is of prim- itive type, comparable in general to those of Das- ypus and of the Priodontes group. The posterior portion is more expanded than the anterior, and the two parts meet in a very sharp angle, almost precisely as in Priodontes. Again as in this form, there is no groove in the region of the Eustachian tube. No trace is apparent of the thickening that is so striking a feature of Doellotatus. The porus, as Scott states, is irregular; the outline corresponds closely to that of Priodontes. The spina tympanica posterior is conspicuous, the bony meatus short, and the crista, sulcus, and recessus well developed. The posterior crus is attached mainly to the tym- panohyal, with only the spina extending up to the squamosal as in the Priodontes and Dasypus groups; more of it is visible in lateral view than in later eutatines, due to the relatively smaller and shorter mastoid process. The anterior crus is evidently broken off immediately distal to the sulcus mal- leolaris. The small part of the ascending portion that is preserved is very slender, and it would seem 30 FIELDIANA: GEOLOGY 2 cm anterior crus of tympanic meatal porus squamosal mastoid process tympanic stylomastoid foramen mastoid foramen foramen lacerum posterior foramen lacerum medium opening for eustachian tube mastoid process -IL occipital condyle pars mastoidea petrosal entotympanic 2 cm Fig. 1 2. Doellotatus prominens: A-D, three partial skulls shown in right lateral and anterior views (fmnh PI 4526), right lateral view (fmnh P14358), and ventral view (fmnh P14351). likely that the missing portion was similarly con- structed. In all probability the whole resembled the anterior crus in the Priodontes and Dasypus groups, was bound to the squamosal by soft parts only, and did not extend dorsally much above the level of the posterior crus. The structure of the malleus confirms this inference. The adjacent sur- face of the squamosal, anterointernal to the su- perficies meatus, forms a conspicuous, thick ridge, second in prominence only to Priodontes among the forms studied. The recess in the squamosal posterior to the glenoid area, characteristic of all members of the Euphractus group, is present in Proeutatus also. PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 31 B basisphenoid pit basioccipital glenoid postglenoid process 3 cm occipital condyle mastoid foramen occipital crest pars petrosa foramen magnum foramen rotundum foramen ovale alisphenoid anterior crus of tympanic crista tympanica sulcus tympanicus< recessus meatus glenoid fossa posterior crus of tympanic 1 cm paroccipital process groove below mastoid foramen occipital condyle pars petrd^a fenestra oval is : fen. vestibuli) external auditory meatus tympanohyal 1 cm Fig. 13. Proeutatus: A-C, two partial crania shown in right lateral and ventral views (P. cf. oenophorus, fmnh PI 3 197) and ventral view (P. oenophorus, fmnh PI 3 199); D, right tympanum shown in internal (left) and external views (fmnh PI 3 199). In this form, however, the distance between the base of the mastoid process and the rim of the recess is much less than in other members of the group, the arched superficies meatus actually being anterior to the rim rather than posterior to it as in Prozaedius. Proeutatus thus indicates that ac- quisition of the postglenoid recess preceded, and may have been the structural prerequisite for, en- largement of the anterior crus in members of the Euphractus group, but at the same time shows that enlargement of the crus was accompanied by fur- ther enlargement of that portion of the squamosal forming the rim of the process. No trace of an entotympanic is present in our material nor in that figured by Scott. The tympanohyal is very well preserved in FMNH PI 3 1 99. At the base it is indistinguishably fused with the crista facialis and curves laterally, 32 FIELDIANA: GEOLOGY providing a base of attachment for the posterior cms of the tympanic. The posterior surface is grooved for the passage of the facial nerve. The pars petrosa of the periotic runs to a blunt point anterointernally, and has a sharp ventro- medial border that is only slightly above the level of the basicranium. Above this border, as in Doel- lotatus, there is a groove in the medial face of the bone. The promontorium agrees with that ofDoel- lotatus in being notably less prominent than in euphractines. The crista facialis is very sharply defined and bears the usual xenarthran process, which is unfortunately broken away anteriorly in the available specimens. The recessus epitympan- icus is rather shallow toward the lateral side but is sharply bounded medially due to the promi- nence of the crista facialis. There is no epitympanic sinus. The pars mastoidea is less extensive than in Doellotatus, and is more widely exposed on the occipital than on the lateral surface of the skull. The mastoid process is very notably smaller and shorter than in any other member of the Euphrac- tus group, and is further distinguished by being widened in the anteroposterior rather than in the transverse direction. Due to the absence (or lack of ossification of the entotympanic) nothing can be said concerning the carotid foramen. To judge from the space between the anterior extremity of the pars petrosa and the alisphenoid, a foramen lacerum medium was present. The groove in the medial surface of the pars petrosa, which is converted into a canal by the basioccipital, extends toward a foramen that pierces the alisphenoid anterior to the tip of the petrosal. This in turn leads to a canal that opens in the posterior part of the sella. The postglenoid foramen, as in other members of the Euphractus group, is situated in the posterior portion of the postglenoid recess. There are several small venous foramina above the recess. The mastoid foramen is in the usual position; the groove running ven- trally from it does not appear to have been roofed over by bone. The foramen lacerum posterius has the usual slitlike form. The stylomastoid foramen is situated between the tympanic and mastoid pro- cess, and is lower in position than in other mem- bers of the group. It will be evident from the figure that the fora- men ovale and the opening of the transverse canal are more lateral in position than is usual in eden- tates, the transverse canal opening occupying the usual position of the foramen ovale. Although out- side of the auditory region, a note on the transverse canal may be interpolated here; among armadillos, the canal is present in the Dasypus and Euphractus groups and absent in the others. Glyptodonts, ant- eaters, and sloths do not have it, and there has been no record of its occurrence in palaeanodonts. The Peltephilus Group fPeltephilus Ameghino, 1887 Two skulls of Peltephilus pumilus (= Peltephilus ferox) Ameghino (PU 1 539 1 , fig. 1 4A-B; AM 9524) from the Santa Cruz Formation of Patagonia have been examined. The peculiarities of structure ex- hibited by these remarkable armadillos extend also to the auditory region, the group being highly dis- tinctive in the structure of this area. The bulla is complete and well inflated. The tympanic forms a long tubular meatus as in Recent euphractines, but the resemblance between the two ends here. In Peltephilus the meatus extends not only dorsally but anteroexternally to a marked de- gree as well, a unique feature. Both crura, and not the anterior alone, contribute to the formation of the ventral half of the tube. How much of the dorsal half may be supplied by them and how much by the superficies meatus of the squamosal cannot be determined. Scott (1903, p. 90) states that the dorsal surface is formed by the zygomatic process of the squamosal and the anterior by the postglenoid process. This is quite uncertain and at least partly erroneous. There is no postglenoid pro- cess, and it is virtually certain that a strip from the anterior cms of the tympanic lines the anterior part of the meatus, and [it is] possible that it may extend to the dorsal surface as well. Whatever its composition, the dorsal half is less arched than in other armidillos. In consequence, the poms is oval with the long axis in the anteroposterior direction. The anteroexternal inclination of the meatus is so pronounced that its anterior portion actually lies above the external portion of the wide glenoid cavity; the tympanic and squamosal are indistin- guishably fused in this area. Again as a result of the direction of the meatus, the posterior cms is well separated from the mastoid process. The ven- tral surface of the meatus bears a rounded, irreg- ular ridge that curves posteriorly to the vicinity of the tympanohyal, and joins the bulla proper in an irregular manner. Anteriorly, the bulla extends al- most to the glenoid cavity, the two being separated only by a rather shallow but well defined fissura Glaseri. The posterior end of the bulla is in very close contact with and in places fused with the PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 33 C 2 t r«-> O m CO "" «, t 34 FIELDIANA: GEOLOGY adjoining portion of the pars mastoidea of the peri- otic. The glenoid cavity, it should be noted, is lower on the skull than in any other armadillo. The boundary between entotympanic and tym- panic cannot be determined. There is no reason to doubt, however, that the relations of the two bones were essentially as in Euphractus and allies. The opening for the Eustachian tube is in the same position as in these forms, but is larger. As Scott has noted, Peltephilus is the only Santa Cruz armadillo that has a completely ossified bul- la. Members of this group are in fact the first xe- narthrans known to have acquired this condition. The Myrmecophagidae are a close second, but there is some doubt (see below) as to whether the tym- panic is completely fused with the surrounding elements in the Santa Cruz Promyrmephagus. The tympanohyal cannot be distinguished; it is, however, almost surely fused with surrounding elements in the shallow depression medial to the mastoid process. Nothing can be seen of the pars petrosa in our material, except for a narrow strip anterior to the foramen lacerum posterius that is not covered by the bulla. The pars mastoidea is very well devel- oped, and is broadly exposed both on the occiput and on the lateral side of the skull. The mastoid process is massive and rather short, situated lat- erally on the skull and sharply set off from the posterior cms of the tympanic by a wide and deep groove. The extent to which the squamosal par- ticipates in the process cannot be seen in the spec- imens. The lateral extension of the exoccipital be- yond the condyle is well separated from the mastoid process. The carotid foramen and canal occupy the same position as in the Euphractus group. The presence of the second arterial canal is uncertain; if present it was small and ran forward between the dorsal surface of the entotympanic and the ventral sur- face of the periotic. There is a small foramen lac- erum medium at the anterointernal corner of the bulla. In other edentates the foramen ovale is sit- uated well in advance of the bulla, and is hence outside the region under consideration. In Pelteph- ilus, however, as Scott noted, it adjoins the bulla, being above and slightly in front of the opening for the Eustachian tube. There is no postglenoid foramen, but a number of vascular foramina occur in the dorsal surface of the squamosal. The mas- toid foramen is in the usual position, and a very well defined groove runs ventrally from it. Some minor foramina, presumably vascular, are present around the posterior margin of the bulla. The fo- ramen lacerum posterior is larger than in members of the Euphractus group. The stylomastoid fora- men is slitlike and opens externally at the ventro- medial extremity of the cleft between the meatus and the mastoid process. The Chlamytherium Group tPlaina Castellanos, 1937'° A magnificent skull of this form, referred ten- tatively to Plaina subintermedia Rovereto, has been available and is currently under study by A. G. Edmund. The specimen, FMNH PI 4424 (fig. 1 5A- B), comes from the Corral Quemado horizon, Ca- tamarca, Argentina. The tympanic and entotympanic (if this element was ossified) are missing in this specimen and in other skulls of chlamytheres thus far described (Castellanos, 1937; Bordas, 1939b; the bulla in- dicated in fig. 1 of this paper is the ventral surface of the periotic"). There can be no doubt that the tympanic was of the primitive, horseshoe-shaped type seen in such forms as Priodontes and Dasypus, and was, likewise, loosely attached by soft parts to the squamosal anteriorly and resting almost en- tirely on the tympanohyal posteriorly. The meatus faced laterally and was farther posterior to the glenoid cavity than in other armadillos, a point of resemblance to the glyptodonts. The glenoid surface differs from that of other armadillos and strongly resembles that of the glyp- todonts. The articular surface is at right angles to the long axis of the skull, narrow anteroposteriorly, wide transversely and gently convex in both these directions. A very large, U-shaped notch separates the glenoid articulation from the mastoid process; this is labeled as the glenoid cavity in Bordas's Figure 1 . Between the glenoid surface and the au- ditory region proper there is a conspicuous, ven- trally facing fossa within which the postglenoid foramen is situated. The fossa is separated from the auditory region by the thick ridge formed by 10 [A. G. Edmund, who is reviewing the Pampatheres, tells me that Plaina will have to be abandoned, since its genus type is a synonym. Vassallia is the appropriate replacement generic name, the species being V. maxima Castellanos, 1946.] " Bordas appears to have been influenced in his ac- count of the auditory region of Chlamytherium by the works of Huxley (1865) and Burmeister (1870-1874) on glyptodonts. Van Kam pen's correction of the errors made by these authors was apparently not available to him. PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 35 OQ CO CO E o in c CO 00 , ■ CD CD o 00 "O o CD O jo c o E CO o O c 0 E o CO 00 CO CD O 2 n CO CO o "55 Q. CO CD 'o W CO GO o a E CD CD O O Q. B Q. C z> O O en "•£ 05 c '2 CO oo CO a E CO O ro c CD E CO "o o o "i§ CO "oo o CL a5 CD E CO a CO CL •"- tf'f £>^>. o 36 FIELDIANA: GEOLOGY the squamosal, which is so characteristic a feature of the Xenarthra. In the somewhat earlier, late Miocene Kraglievichia, this ridge would appear to have been more prominent than in Plaina, to judge from Castellanos's figure (1937, fig. 2). The tympanohyal is present, although incom- plete ventrally, on the right side of the skull. Here it projects medially, as a continuation of the crista facialis, as in other armadillos. The promontorium is well rounded, and slopes evenly upward and forward to the rather sharp anterointernal tip [of the petrosal], which is, how- ever, less pointed than in glyptodonts. There is a wide gap anteriorly between periotic [petrosal] and neighboring elements. Behind this promontorium and set off from it by a groove, the ventral surface of the pars petrosa forms a rather flat shelf. The fenestra rotundum ("fosa para el atrohial" in Bor- das's figure) is situated in the posterior part of the promontorium just beneath the junction of pro- montorium and shelf. The recessus epitympanicus is comparable in degree of definition to those of other armadillos. There is no indication of an epi- tympanic sinus. The crista facialis has evidently been damaged somewhat, especially in its poste- rior portion, during the preparation of the skull from the very hard matrix that enveloped it. On the left side, part at least of the processus cristae facialis remains; the extremity is thick and blunt, much as in the glyptodonts. The pars mastoidea has a large occipital exposure and forms the ven- trolateral corner of the occiput. The mastoid pro- cess (referred to as the postglenoid process by Bor- das), nearly all of which is formed by the pars mastoidea, is a massive structure directed antero- ventrally and terminating in a blunt tip. On the lateral side, near the junction of process and oc- cipital crest, there is a prominent protuberance. As in Priodontes, and [as in that form] presumably correlated with large size, the paroccipital process is well developed— for an armadillo.12 Situated immediately lateral to the condyle, it consists of a ventrally projecting knob of bone, which, again as in Priodontes, descends slightly lower on the side of the skull than the mastoid process. The two processes are separated by a deep and rather nar- row cleft, considerably narrower than the corre- sponding one in Priodontes. The internal carotid evidently entered the cra- 12 Large euphractines show no such development, but in these forms formation of a paroccipital process would appear to have been prevented by union of bulla and pars mastoidea immediately lateral to the occipital con- dyle. nial cavity via the gap anterior to the periotic. Bordas identifies the lateral portion of this gap as the foramen ovale, but this is not the case; the foramen ovale opens farther forward, well within the alisphenoid, as in all edentates. The second arterial canal appears to have run forward, as usu- al, between the periotic and the basicranial ele- ments. The postglenoid foramen is small and is situated within the fossa posterior to the glenoid articulation, as noted above; in glyptodonts the same foramen is similarly situated. The small subsquamosal foramen opens in a groove between the root of the mastoid process and the ridge that forms the upper border of the zygomatic process of the squamosal, a position precisely similar to that occupied by this foramen in glyptodonts. In Chlamytherium the subsquamosal is evidently larger than in Plaina. Bordas identified this as the porus, which is certainly not correct (the parts la- beled as external auditory meatus and crista meati in his fig. 1 are evidently portions of the mastoid process). The mastoid foramen is in the usual po- sition. The foramen lacerum posterior, as noted by Bordas, is relatively small. The facial nerve, after leaving the foramen stylomastoideum prim- itivum, ran ventrally in a groove in the medial surface of the mastoid process. The general appearance of the auditory region and of the surrounding parts of the chlamythere skull, taken as a whole, is highly distinctive. The very broad, ventrally convex, tuberculate basis cranii with the periotics set well above it; the bluntly terminating processus cristae facialis; the close ap- proximation of the mastoid process and exoccip- ital; the wide, transversely aligned glenoid surface and the zygomata that are so notably wider than the cranium; the fossa posterior to the glenoid sur- face, into which the postglenoid foramen opens; the relatively short basicranial region; the stout pterygoid processes— all these combine, as ob- served by Bordas, to form a whole that is decidedly glyptodont-like. Glyptodontoids Glyptodontidae Available material of this group is limited in extent.13 Fortunately, this lack is not particularly 13 This was true when Patterson and Segall did the study; while still somewhat limited today, such materials are more abundant now than when that statement was PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 37 serious, since the glyptodont auditory region is stereotyped in comparison with the diversity en- countered among armadillos and sloths. In this section, therefore, individual descriptions are not given and a general account only is presented, based on the following skulls (those in Field Museum's collections bear "P" prefix): Propalaeohoplopho- rus australis Ameghino, PI 3205, from the early Miocene Santa Cruz Formation of Patagonia; Stromaphorus compressideus14 Castellanos (fide Hoffstetter, 1958) from the Andalgala Formation, Catamarca, Argentina; Hoplophractus proximus Moreno and Mercerat, PI 4501 (fig. 16A-B); and Eleutherocercus solidus14 Rovereto, P14475 (fig. 17), both from the Corral Quemado formation, Catamarca, Argentina. [These last two skulls are nearly complete, and are deserving of additional study. For example, that of Hoplophractus pre- serves the right premaxillary, which is small and endentulous but has a pair of anteriorly directed pits, one dorsal to its tip, the other ventral and adjacent the midline. To me these suggest attach- ment surfaces for tendons of a proboscis. Gillette and Ray (1981) did not find a preserved premax- illary in any of their Glyptotherium material, but they conclusively inferred its presence on the basis of thickness and form of the adjacent bones.] No trace of a tympanic or an entotympanic has yet been found in any glyptodont. Description of parts of the tympanic by earlier authors, Huxley (1865) and Burmeister (1870-1874), are now known to be erroneous and to have been based on misidentifications of other parts of the auditory region (see below). The tympanic very evidently was of the primitive horseshoe-shaped type, loose- ly attached to the tympanohyal posteriorly and to the squamosal ridge anteriorly; this ridge is more rounded and much more massive than in arma- dillos. Ridge and tympanohyal are well separated, and the incisura tympanica must therefore have been relatively wide. It is clear that the meatus written. There is as yet undescribed new material from South America, and two significant published works con- siderably improve the situation for glyptodont cranial study: that of Vinacci Thul (1945) [which Patterson and Segall were unaware of], whose description of the skull appears to be the first serious attempt in that direction since Burmeister (1 870-1 874) and Owen ( 1 845), and the more recent thorough and detailed monograph of the North American glyptodonts by Gillette and Ray (1981). 14 [The species compressideus is a Moreno and Mer- cerat species that Ameghino (1891) denied belonged to the genus Neuryurus, where M. & M. had placed it. The species solidas was named by Ameghino (1891) and questionably assigned to the genus Protoglyptodon.] was situated far behind the glenoid articulation, notably farther posteriorly than in all armadillos except the chlamytheres. The glenoid articulation is highly characteristic of the group. It is very wide transversely and nar- row anteroposteriorly, and stands out at a right angle to the skull. Posterior to and above the me- dial portion there is a large fossa [mandibular fos- sa] in which the postglenoid foramen (or foramina) is situated. ' 5 [Gillette & Ray (1981) show the man- dibular fossa in their illustrations of Glyptothe- rium taxonicum and Glyptotherium arizonae (their figs. 7-8, lOb-c), but they make no mention of a postglenoid foramen. In both Field Museum spec- imens here described, that foramen opens into the fossa.] Glyptodonts differ from all armadillos, again with the exception of the chlamytheres, in the structure of this area. The tympanohyal is best shown in our material on the left side of the skull of//, proximus, and even in this specimen the bone may not be com- plete ventrally. [Since this was written there has been further damage, so it is fortunate that the illustration, fig. 16B, was made at an earlier date.] As preserved, it is a relatively stout bar extending posterointernally from the crista facialis and pars mastoidea to the posterior portion of the ventral surface of the pars petrosa, where it comes in con- tact with a small spur from this surface. The ventral surface of the periotic [petrosal] is extensive. The promontorium is not at all sharply defined, no more so than in the eutatines, which are outstanding among armadillos in this respect. Anterior to it is a laterally compressed antero- ventromedially extending portion that terminates more sharply than in any armadillo. Medial and dorsal to this compressed portion is a shelf, vary- ing in width in the different forms, that slopes down to a large quadrangular, somewhat irregu- larly surfaced area behind the promontorium. This area, which is much larger than in any armadillo, continues posterolaterally into the pars mastoidea. On the left side of//, proximus, above the medial extremity of the tympanohyal, the quadrangular area sends down a small, posteriorly curving pro- cess, with which the tip of the tympanohyal comes into contact. [This can now only be seen on the right side due to subsequent damage.] No such process is to be seen on any of the other specimens available to us, but a small structure of this sort 15 Several of Lydekker's plates (1894, pis. 8, 16, 22, 31-32) show the condyle of the jaw resting in this fossa and not in articulation with the glenoid surface. 38 FIELDIANA: GEOLOGY PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 39 ■g "o c _CD Ui CD CO b CO o CO CO o CO CD o CD CO CD CD a O T3 c O £ CO CO CO Q. CO Cl E E CO Cl CD CO Q. c CD E CO o p o H — • CO CO E CO C 3 CO O Q) C) l o <U X 0 yE O as o c CD CD E to Q 10 CL hj 40 FIELDIANA: GEOLOGY would be very liable to damage during prepara- tion. That it was characteristic of at least some of the later glyptodonts would appear to be demon- strated by Burmeister's figure of Glyptodon ( 1 870- 1874, pi. 23, fig. 5), in which the tympanohyal (p.t.) is shown fused to a small process from the periotic. Development of the epitympanic recess varies considerably within the family. In Eleuthe- rocercus it is very small, no larger relatively than in Plaina, whereas in Hoplophractus, PI 4501, it is considerably larger, pear-shaped in outline, and very sharply defined. In accordance with van Kampen's (1905) findings, there is no indication of an epitympanic sinus.'6 The crista facialis has a more robust ventral margin than in armadillos. The processus cristae facialis is poorly preserved in our material; an anterior projection, represent- ing part of it, at least, may be seen on the left side of//, proximus, and all specimens show a decided thickening of the crista lateral to and above the apertura tympanici canalis facialis. The pars mas- toidea, although large, plays a relatively small part in the formation of the occipital surface, the great- er part of its exposure being lateral where it forms the major portion of the mastoid process. The process is massive, comparatively long antero- posteriorly, directed laterally as well as antero- ventrally, and short, not extending below the level of the condyle. That portion of the exoccipital lat- eral to the condyle is more extensive and rugose in most glyptodonts (but not in all; Stromaphorus, for example, has only a very slight lateral portion) than it is in most armadillos, although no paroc- cipital process in the strict sense is developed. No details can be given concerning the courses of the arteries going to the circle of Willis. The postglenoid foramen (or foramina) is invariably present and opens at or near the base of the fossa posterior to the medial portion of the glenoid ar- ticulation. The subsquamosal foramen (or foram- ina), as in chlamytheres, is situated in the groove between the root of the mastoid process and the overhanging dorsal rim of the zygomatic process of the squamosal. The mastoid foramen usually opens ventrally near the mastoid-exoccipital su- ture. In some specimens, e.g., E. solidus, it is more dorsal in position, and a short, deep, narrow groove 16 The cavity in the squamosal described by Burmeis- ter (1870-1874, pp. 15, 248, 253-254, 267, pi. 28, fig. 2r) is, we suspect, an excavation for the reception of a venous sinus. His passing mention of a connection be- tween this cavity and the cavum tympani is, as van Kampen suspected, certainly erroneous. is present in the posterior portion of the pars mas- toidea. The foramen lacerum posterior is relative- ly small and slitlike. After leaving the foramen stylomastoideum primitivum, the facial nerve ran ventrolaterally in a groove in the anteromedial surface of the mastoid process. The earlier writers on glyptodonts, Owen ( 1 845), Huxley (1865, pp. 55-56), and Burmeister (1870- 1 874), were all to varying degrees mistaken in their descriptions of the auditory region. Most of their errors have been pointed out and corrected by van Kampen, with whose interpretations we are in hearty agreement, but for the sake of completeness a brief report of them may be included here. Ow- en's account (1845, pp. 384-385) is a straightfor- ward description of the structure seen in Glypto- don, which goes astray only in the identification of certain of the foramina. Thus, he mistook the mastoid foramen for a perforated depression for the digastric muscle, believed that the jugular vein went through the foramen lacerum medium (his foramen lacerum), and stated that the "wide ap- ertures" (postglenoid foramina) in the "deep ex- cavation behind the glenoid articulation . . . seem to comunicate with the tympanic cavity." Huxley (1865, pp. 55-56), who studied the same specimen described by Owen, correctly recognized that the jugular left the skull by the foramen lacerum pos- terius and that the postglenoid foramina did not, as Owen had supposed, communicate with the tympanic cavity but with the cranial cavity. Un- fortunately, however, he conceived the idea that Glyptodon had a tympanic of Euphractus type and this led him into one error after another. He be- lieved the mastoid process (Owen's "petromastoid below the digastric depression") to be a part of the tympanic, and this in turn led him to suppose that the paroccipital portion of the exoccipital included the pars mastoidea. He mistook the mastoid fo- ramen (Owen's "perforated depression for the di- gastric") for the external auditory meatus, com- menting, in some surprise, on so small a passage in so large an animal. The characteristic rugosities for muscle attachment on the ventral surface of the basicranium were stated to "mark the place where the thick inner walls of the bullae impinge upon the basioccipital." The basis for this last statement is hard to imagine, since in no arma- dillo, not even in the dwarfed Chlamyphorus, do the bullae encroach medially upon the basicrani- um. All these errors might have been avoided had Huxley selected an armadillo with an unspecial- ized auditory region, such as Dasypus or Pri- odontes, as his standard of comparison. Burmeis- PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 41 ter (1870-1874, pp. 12-13, 255) correctly recognized that the tympanic was missing in the specimens studied by him. He supposed the tym- panic to have been large, reaching to the exoccip- ital posteriorly (p. 255), whereas the position of the tympanohyal clearly shows that a gap existed between the two bones, as in Priodontes and Da- sypus. As a result, he mistook the posterior portion of the sulcus facialis for the auditory meatus (pp. 18, 255; pis. 12, fig. 28, figs. 5-6i). Like Huxley, Burmeister was misled by comparing the glypto- donts with Euphractus. [Had Patterson and Segall been able to have the Gillette & Ray (1981) mono- graph before them, doubtless they would have commented on the basic similarities of the basi- cranium in South and North American forms. The few differences I see are minor: the presence of a postglenoid foramen in the South American forms, its apparent absence in the North American ones, and the degree of development of the paroccipital process (PoP). Gillette & Ray label one, while in this paper we comment that none is developed in the strict sense in Hoplophr actus. This is really a difference in emphasis, for the details of the region as shown in their fig. 8 and in our fig. 1 6 are quite alike. In the FMNH specimen of Eleut her ocercus, however, the paroccipital process is definitely a well-formed feature, being low, massive, and with a rugose surface.] Summary and Conclusions When Patterson left Chicago for Harvard, there was a Conclusions section to the original work, but it has not survived among any of the drafts or revisions that he had made. It is clear to me that he wanted to revise it further, but I do not know either the original content or anything about plans for its revision. I shall therefore here try briefly to set forth the points that seem to be most important as regards this portion of the whole study. Quoted material is taken directly from original text (de- scriptive sections). 1. For Eutatus, the postglenoid foramen is identified from its position and from its connec- tion into the cranial cavity. Building upon this evidence, the posttympanic process is identified and the concluding point is made that "its presence adds another significant item to the list of resem- blances between Eutatus and the Palaeanodonta drawn up by Simpson (1948, p. 88)." It is spec- ulated, based upon location of a number of struc- tures, that the tympanic was a simple [type?], "lit- tle or not at all advanced beyond the ring form, hence of the type that was certainly primitive for the Xenarthra and that persisted with little change to Pleistocene and Recent time in many sloths and armadillos, and, apparently in all glyptodonts." 2. For Priodontes it is expressly noted that the crista facialis gives origin to a delicate tripartite process that is characteristic of the Xenarthra as a whole, a feature not fully realized heretofore (p. 8) and for which the name processus cristae facialis is suggested. 3. For Cabassous new information on ossifi- cation of the entotympanic is presented in which rostral and caudal portions are described. 4. For Tolypeutes the degree of ossification of the entotympanic is intermediate between those of Cabassous and Priodontes. In most other re- spects Tolypeutes was found to be similar to Ca- bassous. As regards the degree of enclosure of the venous canal between squamosal and processus cristae facialis, Priodontes, Cabassous, and Toly- peutes "exhibit an interesting structural transition from open gap seen in the first to closed canal in the last." 5. For Dasypus the unique condition of the glenoid region among all other armadillos is stressed, and variation in degree of ossification of the entotympanic is described. 6. For Stegotherium the many features in which Stegotherium resembles Dasypus as noted by Scott (1903-1904) had suggested to him that the two were closely related. This work reinforces that con- clusion. 7. For the Euphractus group, six genera are il- lustrated showing the basically similar morphol- ogies for all. For the modern genera the suites of specimens included young and adults which en- abled developmental data to be presented. 8. For the eutatines the description of the ex- traordinarily thick and dense tympanic of the later forms (Doellotatus) is a contribution significant in its contrast with the primitive condition seen in Proeutatus. 9. For the Peltephilus group, in Patterson and Segall's words, "the peculiarities of structure ex- hibited by these remarkable armadillos extend to the auditory region, the group being highly dis- tinctive in the structure of this area." The bulla is complete, well inflated, with the tympanic forming a long tubular meatus as in Recent euphractines. The glenoid cavity is lower on the skull than in any other armadillo. 42 FIELDIANA: GEOLOGY 10. For the Chlamytherium group the glypto- dont-like form of the auditory region and its im- mediate surrounding cranial area is stressed. 11. For the glyptodonts it is concluded that "the tympanic very evidently was of the primitive horseshoe-shaped type, loosely attached to the tympanohyal posteriorly and to the squamosal ridge anteriorly. ..." Further, the glenoid articu- lation is characteristic for the group, being very wide transversely and narrow anteroposteriorly. This section, comprising about half of the orig- inal, is the first and most straightforward of the two currently planned publications. It makes but a few suggestions for changes in the systematics of the cingulates (cf. Hoffstetter, 1958; Engelmann, 1 985), and is notable for its depiction of the extent of diversity of morphology of the otic regions among the armadillos. In contrast, and equally striking, is the uniformity of that area in glypto- donts, a point stressed in the text. The few sys- tematic changes that had been suggested by earlier workers, usually upon other grounds, are con- firmed by the ear region structures. For the sloths (to be reported in the subsequent companion vol- ume) there is far more uncertainty in this regard, and so the ear region study offers more and greater insights. However, discussions of those consider- ations, as well as any broad conclusions relating to the entire study, will await the publication of that second section of the work. It is unfortunate that the extensive data set rep- resented by this effort by Patterson and Segall was not published 25 to 30 years ago, for it would have provided a far broader base for others— such as Guth (1961), whose important study most closely parallels this one, Maclntyre (1972), Szalay (1972), Hunt (1974), Novacek (1977), Archibald (1977), Hunt and Korth (1980), Wible (1980, 1983), Gil- lette and Ray (1981), Engelmann ( 1 985), and Wyss et al. (1987)— to have worked from. All of them worked with a broad comparative perspective as regards their several concerns with therian otic region development and evolution. Equally, it is a pity that Patterson and Segall were not aware of the Guth study, which probably was being pre- pared at about the same time as their early draft, or the Vinacci Thul study of Glyptodon (1945), which appeared about the time that they began this work, for both are directly pertinent. At this point, we can only move on, for the past is past. If the xenarthran stock is truly one of the basic early adaptive lines of the eutherians, as it appears now to be (McKenna, 1975; Hoffstetter, 1958; Wyss et al., 1987), then the better based our un- derstanding of that group is, the more meaningful should be all broad extrapolations from it to the metatheria and to the various other stem euthe- rians. At least now, at this late date, this docu- mentation for cingulate ear region morphology is finally available. Acknowledgments I am indebted to Ron Testa, Field Museum pho- tographer, Marlene Werner of the Museum's Sci- entific Support Services, and Elaine Zeiger, sec- retary in the Department of Geology, for their help in pulling the entire work together. W.D.T. Literature Cited Ameghino, F. 1886. Contribuciones al conocimiento de los mamiferos fosiles de los terrenos terciarios an- tiguos del Parana (Memoria Cuarta). Bol. Acad. Nac. Ciencias Cordoba (Argentina), IX: 5-228. . 1887. Enumeracion sistematica de los especies de mamiferos fosiles coleccionados por Carlos Ameghino en los terrenos eocenos de la Patagonia austral y depositados en el Museo de la Plata. Bol. Mus. de La Plata, 1: 1-26. . 1889. Contribucion al conocimiento de los mamiferos fosiles de la Republica Argentina. Actas de la Academia Nacional de Ciencias exactas de la Re- publica Argentina en Cordoba, 6: 1-1028. Atlas with 98 plates. 1891. Mamiferos y aves fosiles Argentinas, y nuevos restos de mamiferos fosiles. Revista Argentina de Historia Natural, 1: 240-259, 289-328. . 1902. Notices preliminaires sur des ongules nouveaux des terrains cretaces de Patagonie. Bol. Acad. Nac. Ciencias, Cordoba (Argentina), 16: 350-426. . 1906. Les formations sedimentables du cretace superieur et du tertiaire de Patagonie. Anal. Mus. Nac. Buenos Aires, 16(ser. Ill, tome VIII): 1-568. Archibald, J. D. 1977. Ectotympanic bone and in- ternal carotid circulation of eutherians in reference to anthropoid origins. J. Human Evol., 6: 609-622. Bondy, G. 1 907. Beitrage zur vergleichenden Anato- mie des Gehororgans der Sauger (Tympanicum, Mem- brana chrapnelli und chordaverlauf). Anatomische Hefte, Heft 106, 35(2): 293-408, Taf. X-XIII. Bordas, A. F. 1932. Proposition de un nuevo genero para "Eutatus inornatus" (Rov.). Physis, 11: 1 67-1 68. . 1933. Notas sobre los Eutatinae: Nueva subfamilia extinguida de Dasypodidae. Anal. Mus. Nac. Buenos Aires, 37: 583-614. PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 43 . 1939a. Diagnosis sobre algunos mamiferos de las capas con Colpodon del Valle del Rio Chubut (Republica Argentina). Physis, 14: 413-433. 1939b. Craneometria y region auditiva de "Chlamytherium typum" Ameghino. Physis, 14: 447- 460. Burmeister, H. 1870-1874. Monografia de los glyp- todontes en el Museo Publico de Buenos Aires. Anal. Mus. Pub. Buenos Aires, 2: 1-412. (Issued in undated parts from 1870 to 1874.) Castellanos, A. 1937. Anotaciones sobre la linea filo- genetica de los clamiterios. Publicaciones de Facultad de Ciencias, Universidad Nacional del Litoral, Santa Fe, Argentina, Serie Tecnico-Cien tinea, 8: 1-35. . 1946. Una nueva especie de clamiterio Vas- sallia maxima n. sp. Pub. Inst. Fisiografia y Geologia, Univ. Nac. Litoral, Rosario, Argentina, 6(26): 1-47. Colbert, E. H. 1942. An edentate from the Oligocene of Wyoming. Notulae Naturae, Acad. Nat. Sci., Phila., 109: 1-16. Cuvier, F. 1825. Des dentes des mammiferes, consi- derees comme caracteres zoologiques. Strasbourg, F. G. Levrault, 258 pp. Cuvier, G. 1825. Recherches sur les ossemens fossiles, ou Ton retablit les caracteres de plusieurs animaux dont les revolutions du globe ont detruit le especes, 3rd ed. G. Dufour et F. d'Ocagne, Paris. [4th ed., 1934-1936, not seen.] . 1831. The Animal Kingdom ..., Translated from the French by H. M'Murtrie. ... 4 vols. G. and C. and H. Carvill, New York. Desmarest, A. G. 1804. Nouveau Dictionnaire d'His- toire Naturelle. Paris, 24: 28. Edmund, G. 1985. The fossil giant armadillos of North America (Pampatheriinae, Xenarthra = Edentata), pp. 83-93. In Montogomery, G. G., ed., The Evolution and Ecology of Armadillos, Sloths, and Vermilinguas. Smithsonian Institution Press, Washington, D.C. Engelmann, G. F. 1985. The Phylogeny of the Xe- narthra, pp. 51-64. In Montgomery, G. G., ed., The Evolution and Ecology of Armadillos, Sloths, and Ver- milinguas. Smithsonian Institution Press, Washing- ton, D.C. Fitzinger, L. J. 1871. Die Naturliche Familie der Gur- telthiere (Dasypodes). Osterreichs. Akad. Wiss., Math.- nat., 64(1): 268, 329-390. Flower, W. H. 1870. An Introduction to the Osteology of the Mammalia. Macmillan and Co., xi + 344 pp., 126 figs. . 1882. On the mutual affinities of the animals composing the order Edentata. Proc. Zool. Soc. Lon- don: 358-367. Geoffroy Saint-Hilaire, E. 1803. Catalogue des Mammiferes du Museum National d'Histoire Natu- relle. Paris, 272 pp. Gervais, F. L. P. 1855. Mammiferes. Animaux nou- veaux, ou rares, recueillis pendant l'expedition dans les parties centrales de 1'Amerique du Sud. P. Ber- trand, Paris, 1 16 pp., 20 color pis. Gervais, P. M. 1867. Comptes Rendu Academie des Sciences, 65: 280. Gillette, D. D., and C. E. Ray. 1981. Glyptodonts of North America. Smithsonian Contributions to Pa- leobiology, 40: i-vi, 1-255, figs. 1-96, tables 1-70. Glass, B. P. 1985. History of classification and no- menclature in Xenarthra (Edentata), pp. 1-3. In Mont- gomery, G. G., ed., The Evolution and Ecology of Armadillos, Sloths, and Vermilinguas. Smithsonian Institution Press, Washington, D.C. Guth, C. 1961. La region temporale des Edentates. Le Puy, Haute Loire, xviii + 192 pp. Harlan, R. 1825. Description of a new genus of mam- miferous quadrupeds, of the order Edentata. Ann. Ly- ceum Nat. Hist. New York, 1: 235. Hoffstetter, R. 1958. Xenarthra, pp. 535-636. In Piveteau, J., Traite de Paleontologie, 6(2): 1-962. . 1969. Remarques sur la phylogenie et la clas- sification des Edentes Xenarthres (Mammiferes) ac- tuels et fosiles. Bull. Mus. Nat. d'Hist. Nat., 1: 92- 103. Hunt, R. 1974. The auditory bulla in Carnivora: An anatomical basis for reappraisal of carnivore evolu- tion. J. Morph., 143(1): 21-76. Hunt, R. M., Jr., and W. W. Korth. 1980. The au- ditory region of Dermoptera: Morphology and func- tion relative to other living mammals. J. Morph., 164(2): 167-211. Huxley, T. H. 1865. On the osteology of the genus Glyptodon. Phil. Trans. Roy. Soc, 155(1): 31-70. Hyrtl, J. 1854. Beitrage zur vergleichenden Angio- logie. V. Das arterielle Gefasssystem der Edentaten. Osterreichs. Akad. Wiss., Math.-nat. 7. . 1855. Chlamydophori truncati cum Dasypode gymnuro comparatum examen anatomicum. Oster- reichs. Akad. Wiss., Wien, Math.-nat. Klasse, 9: 1-66. Illiger, J. K. W. 1811. Prodromus systematis mam- malium et avium additis terminis zoographicis utri- usque classis, p. 111. C. Salfeld, Berlin, xviii + 201 pp. Jong, W. W. de, A. Zweers, K. A. Joysey, J. T. Gleaves, and D. Boulter. 1985. Protein sequence analysis applied to xenarthran and pholidote phylogeny, pp. 65-76. In Montgomery, G. G., ed., The Evolution and Ecology of Armadillos, Sloths, and Vermilinguas. Smithsonian Institution Press, Washington, D.C. Kampen, P. N. van 1905. Die Tympanalgegend des Saughtierschadels. Morph. Jahrb., 34: 321-722, figs. 1-96. Klaauw, C. J. van der 1924. 'Bau und Entwickelung der Gehorknochelschen. Literature 1899-1923.' Er- gebnisse der Anatomie und Entwicklungsgeschichte, 25: 565-622. . 1931. The auditory bulla in some fossil mam- mals. Bull. Amer. Mus. Nat. Hist., 62: 1-352. Kraglievich, L. 1934. La antiquedad pliocena de las faunas de Monte Hermoso y Chapadmalal, deducidas de su comparacion con las que le precedieron y su- cedieron. El Siglo Illustrado, Montevideo, No. 938, 136 pp. Linnaeus, C. 1758. Systema naturae per regna tria naturae, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis, p. 51. 10th ed., vol. 1., Laurentii Salvii, Stockholm, 824 pp. 44 FIELDIANA: GEOLOGY Lydekker, R. 1 894. Contributions to a knowledge of the fossil vertebrates of Argentina. Part 2. The extinct edentates of Argentina. Anales del Museo de La Plata, Paleontologia Argentina, 3(2): 1-118, pis. 1-61. MacIntyre, G. T. 1972. The trisulcate petrosal pattern of mammals, pp. 275-303. In Dobzhansky, T., et al., eds., Evolutionary Biology, vol. 6. Appleton-Century- Crofts, New York. McKenna, M. C. 1975. Toward a phylogenetic clas- sification of the Mammalia, pp. 21-46. In Luckett, W. P., and F. S. Szalay, eds., Phylogeny of the Primates. Plenum Press, New York, 483 pp. McMurtrie, H. 1831. See Cuvier, G., 1831. Montgomery, G. G., ed. 1985. The Evolution and Ecology of Armadillos, Sloths, and Vermilinguas. Smithsonian Institution Press, Washington, D.C., 45 1 pp. Moreno, F. P., and A. Mercerat. 1891. Catalogo de los pajaros fosiles de la Republica Argentina. Pa- leontologia Argentina, 1: 1-71, pis. I-XXI. Nov acek, M.J. 1977. Aspects of the problem of vari- ation, origin and evolution of the eutherian auditory bulla. Mamm. Rev., 7(3&4): 131-150. Owen, R. 1845. Descriptive and illustrated catalogue of the fossil organic remains of Mammalia and Aves contained in the Museum of the Royal College of Sur- geons, England. London. Patterson, B., and R. Pascual. 1968. V. The fossil mammal fauna of South America, pp. 409-451. In Evolution of Mammals on Southern Continents. Quarterly Review in Biology, vol. 43. . 1972. The fossil mammal fauna of South America, pp. 247-309. In Keast, A., F. C. Eck, and B. Glass, eds., Evolution, Mammals, and Southern Continents. State University of New York Press, Al- bany, 543 pp. Rovereto, C. 1914. Los estratos araucanos y sus fos- iles. Anal. Mus. Nac. Hist. Nat. Buenos Aires, 25: 1- 247. Sarich, V. M. 1985. Xenarthran systematics: Albumin immunological evidence, pp. 77-81. In Montgomery, G. G., ed., The Evolution and Ecology of Armadillos, Sloths, and Vermilinguas. Smithsonian Institution Press, Washington, D.C. Scott, W. B. 1903-1904. Mammalia of the Santa Cruz beds. Part 1. Edentata. Reports of the Princeton Uni- versity Expedition to Patagonia, 5(1): 1-226, pis. 1- 35 (1903), 227-364, pis. 36-63 (1904). Segaix, W. 1976. Further observations on the ear in fossorial mammals with special considerations of Chlamyphorus truncatus (Harlan). Acta Anatomica, 94(3): 431-444. Simpson, G. G. 1927. A North American Oligocene edentate. Ann. Carnegie Mus., 17: 283-298 & pi. 24. . 1948. The beginning of the Age of Mammals in South America. Part I. Bull. Amer. Mus. Nat. Hist., 91: 1-232. Szalay, F. 1972. Cranial morphology of the Early Ter- tiary Phenacolemur and its bearing on primate phy- logeny. Amer. J. Phys. Anthrop., 36(1): 59-76. VrNACCi Thul, Enriqueta, L. 1945. Osteografia ce- falica de Glyptodon reticulatusOv/. Physis, 20(55): 24- 30. Wagler, J. G. 1830. Naturliches system der amphi- bien, mit vorangehender classification der saugethiere und vogel. J. G. Cotta, Munich, 354 pp. Wetzel, R. M. 1985a. The identification and distri- bution of Recent Xenarthra (= Edentata), pp. 5-21. In Montogmery, G. G., ed., The Evolution and Ecol- ogy of Armadillos, Sloths, and Vermilinguas. Smith- sonian Institution Press, Washington, D.C. . 1985b. Taxonomy and distribution of arma- dillos. Dasypodidae, pp. 23-46. In Montgomery, G. G., ed., The Evolution and Ecology of Armadillos, Sloths, and Vermilinguas. Smithsonian Institution Press, Washington, D.C. Wible, J. 1980. The primitive course of the mam- malian internal carotid artery. Amer. J. Phys. An- throp., 52: 292-293. . 1983. The internal carotid artery in early eu- therians, pp. 281-293. In Kielan-Jaworowska and H. Osmolska, eds., Second Symposium on Mesozoic Ter- restrial Ecosystems. Acta Paleontologica Polonica, 28: 281-293. WYSS, A. R., M. J. NOVACEK, AND M. C. McKjENNA. 1987. Amino acid sequences versus morphological data and the interordinal relationships of mammals. Molecular Biol, and Evol., 4(2): 99-1 16. PATTERSON ET AL.: EAR REGION IN XENARTHRANS. I. CINGULATES 45 Appendix The following abbreviations are used to desig- nate institutions, usually in relation to specimens: AMNH = American Museum of Natural His- tory, New York. FMNH = Field Museum of Natural History, Chicago (prefix P indicates a pa- leontological specimen; numbers lacking a prefix indicate speci- mens in the Recent mammal col- lections). MCZ = Museum of Comparative Zoology, Harvard University, Cambridge, Mass. PU = Princeton University, Princeton, N.J. can. a. car. int. = canal for internal carotid artery can. a. ? = arterial canal ? can. int. pet. sin. = canal internal the petrosal sinus eust. = opening for Eustachian tube fen. r. or fen. rotunda = fenestra rotunda (= fen. cochleae) f.l.m. = foramen lacerum medium (= entocarot- id foramen) f.l.p. = foramen lacerum posterior m.a.e. = meatus auditorius externus PoP = paroccipital process post, cms of tympanic = posterior crus of tym- panic PN = posterior nasal aperture p.t. = processus tympanohyal. The following abbreviations are used in the text, figures, or figure captions: The dagger symbol (f) is used to indicate that the two senior authors are deceased; in systematic references, the dagger designates an extinct taxon. ant. crus of tympanic = anterior crus of tym- panic 46 FIELDIANA: GEOLOGY Field Museum of Natural History Roosevelt Road at Lake Shore Drive Chicago, Illinois 60605-2496 Telephone: (312) 922-9410