S-Nft-L ubrahy°°'^ 1973 THE CRANIAL MYOLQ®^^^ AND OSTEOLOGY OF DICOTYLES TAJACU, THE COLLARED PECCARY, AND ITS BEARING ON CLASSIFICATION MICHAEL O. WOODBURNE MEMOIRS OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOLUME 7 July 5, 1968 ’■' ’.••'•■'/few ■ - ” • . I ' < r THE CRANIAL MYOLOGY AND OSTEOLOGY OF DICOTYLES T A J ACU, THE COLLARED PECCARY, AND ITS BEARING ON CLASSIFICATION THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU, THE COLLARED PECCARY, AND ITS BEARING ON CLASSIEICATION MICHAEL O. WOODBURNE MEMOIRS OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOLUME 7 July 5, 1968 MEMOIRS OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES The MEMOIRS of the Southern California Academy of Sciences is a series begun in 1938 and published on an irregular basis thereafter. It is intended that each article will continue to be of a monographic nature, and each will constitute a full volume in itself. Contributions to the MEMOIRS may be in any of the fields of science, by any member of the Academy. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Articles must not duplicate, in any substantial way, material that is published elsewhere. Manuscripts must conform to MEMOIRS style of volume 4 or later, and may be examined for scientific content by mem- bers of the Publications Committee other than the Editor or be sent to other competent critics for review. Because of the irregular nature of the series, authors are urged to contact the Editor be- fore submitting their manuscript. Donald J. Reish Editor Communications concerning the purchase of the Memoirs should be directed to the Treasurer, in care of the Los Angeles County Museum of Natural History, Exposition Park, Los Angeles, California 90007. Communications concerning manuscripts should be directed to Donald J. Reish, Dept, of Biology, California State College at Long Beach, Long Beach, California. 90804. Printed by Anderson, Ritchie & Simon, Los Angeles, California CONTENTS Abstract 1 Introduction 1 Acknowledgments 2 Definitions 2 The Species of Dicotyles 3 Myological Description 4 Cutaneous Muscles 4 Auricular Muscles 4 Muscles of the Eyelids 5 Muscles of the Face 5 Muscles of the Neck and Adjacent Cranium 7 Ventral Cervical and Hyoid Musculature 1 2 Mandibular Muscles 1 2 Variation in Muscle Scars 14 Cranium 1 7 Mandible 1 7 Cranial Osteology of the Living Peccaries 17 Collared Peccary, Dicotyles tajacu 1 7 Cranium 1 7 Mandible 20 White Lipped Peccary, Tayassu pecari 20 Cranium 2 1 Mandible 22 Variation 22 Dental Characteristics of the Living Peccaries 23 Dicotyles tajacu 23 Upper Dentition 23 Lower Dentition 25 Tayassu pecari 27 Upper Dentition 27 Lower Dentition 27 Summary of the Salient Differences Between Dicotyles and Tayassu 28 The Ancestry of Dicotyles and Tayassu 29 Conclusions 33 Literature Qted 34 THE CRANIAL MYOLOGY AND OSTEOLOGY OF DICOTYLES TAJ AC U, THE COLLARED PECCARY, AND ITS BEARING ON CLASSIFICATION MICHAEL O. WOODBURNEi Abstract: The cranial myology of the collared peccary is described with emphasis on the facility with which muscle attachments are determinable from remnant scars on the cleaned cranium. Individual variation with respect to the major muscle scars is noted over a range of 135 crania. The cranial and dental osteology of the collared peccary is compared with that of the white lipped peccary. Individual variation within each sample is recorded in the text and in tabular form. Features of the snout and zygomatic arch are found to be of value in determining the paleontological associates of each living species. The white lipped peccary is a member of a broad lineage including Mylohyus (late Plei- stocene of North America) and Platygonus (Brasiliochoerits) stenocephalus (Lund, 1880) (late Pleistocene of Brazil). The line of descent common to these three peccaries can probably be traced back through Prosthennops niobraren- sis Colbert, 1935, to Dyseohyus fricki Stock, 1937. Removal of “ Platygonus” (B.) stenocephalus from the genus Platygonus is advocated. The collared peccary is considered to be broadly related to Platygonus. Closest resemblance is shown to P. compressus Le Conte, 1848, but P. cumber- landensis Gidley, 1920, and P. (Parachoerus) carlesi Rusconi, 1931, from the late Pleistocene and early Recent of South America are other members of this group. The two broad groups thus described seem to have been separate entities since early Pliocene time. The morphological distinctions between the two living peccaries have historical significance and it is concluded that two sep- arate genera are represented. The collared peccary is Dicotyles tajacu (Lin- naeus, 1758); the white lipped peccary is Tayctssu pecari Fischer, 1814. INTRODUCTION The generic allocation of the living peccaries has fluctuated widely ever since the collared peccary was named by Linnaeus in 1758. The white lipped species was named by Fischer in 1814 and subse- quent workers have almost invariably differed, not only in their opinion as to whether the two species should be placed in the same genus or in separate genera, but also as to the names which should apply in either alternative. Singularly or collectively, the modern peccary species have been variously allo- cated to Tayassu, Dicotyles, Pecari, Notophorous or Tagassu (see Table 14). A recent examination of the cranial myology of the collared peccary, the results of which are present- ed here, has occasioned a review of the problem. Although it was not possible to undertake a dissec- tion of the rarer white lipped peccary, comparison of the cranial osteology of the two species suggests that features of their masseteric and facial musculature are quite different. These differences are of such magnitude as to suggest that the past evolution of the two species occurred along discrete lineages. An investigation of the fossil record suggests that these lineages have probably been distinct and autonomous since at least early Pliocene time. An investigation of the cranial osteology and dental morphology of the two species is presented, with emphasis on the individual variation observed within each species. Sixty-five cranial, mandibular and dental measurements, taken from a large sample, are statistically summarized for each species. It is intended that the range of variation thus documented for the Recent animals will be used as a uniformi- tarian “yardstick” when judging the content of fossil species in future work. The two living peccaries can be differentiated by at least 28 cranial and dental characters. Both Recent forms possess a number of features which are more primitive than those seen in species of such Pleisto- cene genera as Mylohyus and Platygonus. Prosthen- nops and Mylohyus are generally considered to be more closely related to each other than either is to Platygonus and, although all the fossil evidence is not yet in, the white lipped peccary seems basically related to the Prosthennops — Mylohyus “lineage.” The collared peccary is more closely aligned with Platygonus. The status of these three fossil genera is currently being reviewed and it seems likely that the 'Department of Geological Sciences, University of California, Riverside, California 92502 / -> MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 familiar concepts of Platygonus and Prosthennops, at least, will become altered to some extent. For the purposes of this report, however, these names will be used largely in the conventional manner. It is my opinion that the morphological distinc- tions which can be observed in the crania of the two living peccaries are such as to warrant their being placed in separate genera. Paleontological evidence, while incipient at this stage, seems to corroborate such a procedure, and suggests that the modern peccaries are primitive members, with some addi- tional specializations, of two distinct, relatively long- lived lineages. As will be developed below, the white lipped peccary should be known as Tayassu pecari, the collared peccary as Dicotyles tajacii. In the interests of clarity and convenience, these taxa will be utilized throughout the text, the justification for such usage being presented cumulatively. Acknowledgments This report stems from a project involving a revi- sion of the early Tertiary Tayassuidae which was being jointly undertaken by the author and the late R. A. Stirton, of the Museum of Paleontology, University of California, Berkeley. 1 am indebted to Dr. Stirton for encouraging my participation in the latter project and for his suggestion that a dissec- tion of the cranium of the collared peccary be made. Appreciation is also extended to Dr. T. E. Downs, Director of Earth Sciences, Los Angeles County Museum of Natural History, for allowing us to study and illustrate specimens under his care. Sincere thanks are also given to Dr. R. H. Van Gelder, Department of Mammalogy, and Dr. M. C. Mc- Kenna, Department of Vertebrate Paleontology, American Museum of Natural History, for permis- sion to examine specimens under their jurisdiction. In particular, permission to examine peccary mate- rial in the Frick Collection of the American Museum of Natural History and the support of Dr. McKenna in this and other endeavors was invaluable. In addi- tion, generous financial assistance from the Frick Laboratory enabled me to spend the summer of 1966 studying fossil peccaries in its collections and during that time some of the observations used in this report were made. Some of the costs of publica- tion were defrayed by an Intramural Grant from the Academic Senate, University of California, River- side. This is gratefully acknowledged. All illustrations were prepared by me. Abbrevia- tions which apply to specimens discussed in the text are; A.M.N.H: (M.), Department of Mam- malogy, American Museum of Natural History. Definitions Length from the anterior tip of F to the rear of the condyles: This dimension (Tables 2,8) consists of two measurements; that from the tip of F to the rear of M^ and from M^ to the point at which a plane perpendicular to that of the palatal surface intersects the rear of the condyles. In spite of the dual nature of the total dimension, mechanical error is probably less than 3 per cent. Length of the diastema from C* to P^: This is the distance from the rear edge of the alveolus of C* to the anterior edge of that of P^, as measured parallel to the palatal plane (Tables 2,8). Width between the alveoli of P^: The dimension (Tables 2 and 8) is taken between the lingual edges of the anterior root of P^. Width between the alveoli of M^: This is trans- verse distance between the lingual edges of the alveoli of the anterolingual root of M^. In some instances the dorsomedial slant of this root coupled with the very thin bone of the adjacent palate causes the alveolus in question to extend farther lingually than usual. Variability thus introduced into this measurement could be as much as 15 percent (Tables 2,8). Least width of the rostrum behind the canines: The measurement, made from the ventral aspect of the cranium, is the least transverse dimension of the rostrum measured at the level of the dorsal lip of the buccinator fossa in Dicotyles tajacu (Table 2) and at the point of greatest constriction of the lateral expan- sion of the palate in the diastemal area of Tayassu pecari (Table 8). The point to be stressed is that in D. tajacu the measurement is taken above the level of the palatal surface, not at the diastemal crests. Height from the condyles to the nuchal crest: The measurement is taken in the plane perpendicular to that of the palatal surface. Ventrally the calipers rest on the ventral edge of the condyles; dorsally they meet the sagittal plane of the cranium either at or as much as 10 mm. anterior to the inion (Tables 2,8). Breadth across the zygomatic arches: This dimen- sion is also taken from the ventral aspect of the cranium. It represents the greatest transverse dia- meter of the cranium at the zygomatic arches, usual- ly at or slightly anterior to the anterior edge of the glenoid fossa (Tables 2,8). Breadth between the postorbital processes of the frontals: This parameter is measured from the dorsal aspect with the calipers resting on the lateral surface of the postorbital processes (Tables 2,8). Length and width of the upper incisors: This measurement represents the greatest length of the tooth, parallel to the plane of the palatal surface, between its anterolingual and posterolingual edges. 1968 3 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU The width is taken in the same plane as the length, but perpendicular to it. Because of difficulties en- countered in making these measurements and be- cause of variations caused by tooth wear, only the simplest statistical summary was made (Tables 3,9). Length and width of the lower incisors: For Ij and I2 the length was taken in an anteromedial direction across the ventrolabial base of the enamel. The width was taken posterodorsally, through the body of the tooth, in a plane perpendicular to that used for the length. I3 is oriented longitudinally so that its dimensions can be measured in the usual way. Limited statistical summaries were made of the lower incisor dimensions for the same reasons as given above and because I3 is highly variable (Tables 3,9). Length and width of the canines: The length of both upper and lower canines can be made along the cross sectional axis of the tooth at the base of the enamel. The width was measured perpendicular to this. Even though the greatest width of the upper canine occurs at the anterior portion of the tooth, it seems to be essentially unaffected by tooth wear. The width of the lower canine is also unaffected by wear, although the length of both teeth is decreased as wear progresses (Tables 4,10). Length from P to M^: This measurement was taken directly, from the anterior tip of P to the rear of M^. Additional variability was probably intro- duced into the results (Tables 5,11) because of wear between adjacent members of the cheek tooth series and because of occlusal wear on P. Length and width of the upper cheek teeth: These results (Tables 5,1 1) are obtained in the usual way. The length is the greatest longitudinal dimension of the tooth, taken somewhat above the base of the enamel. Fdr the upper premolars, the direction of measurement is aligned slightly anteromedially; for the molars it is roughly parallel to the long axis of the cranium. The width is measured perpendicular to the length, across the widest part of the tooth. Length from L to the condyles: This dimension (Tables 6,12) is the sum of two measurements: L to M3 length, and that from the rear of M3 to the rear of the condyles. Mechanical error introduced by this method is probably less than 5 per cent. Length of the diastema between Ci and P2: The results are given in Tables 6 and 12. The measure- ment was taken from the rear of the alveolus of Ci to the anterior alveolus for P2. Depth from the tip of the coronoid process to the angle of the mandible: The measurement may be taken directly with the Jaws of the calipers oriented essentially vertically. The results are presented in Tables 6 and 1 2. Depth below P2: With the calipers oriented as in the previous measurement, the depth (Tables 6, 1 2) is taken from the edge of the anterior alveolus of P2 to the ventral edge of the ramus directly below. A certain amount of additional variation results from the fact that the alveolus of P2 occasionally extends down toward the diastemal region for as much as 5 mm. Depth below the rear of M] : The orientation ot the calipers is the same as above. Possible mechanical error of about 5 per cent may be introduced because of the necessity of estimating the point at which the ventral edge of the mandible would strike the lower jaw of the calipers. In this orientation the lower point of the calipers can project only slightly medial to the lateral surface of the mandible (Tables 6,12). Width between the alveoli of P2: This is the trans- verse distance between the inner edges of the anterior alveoli of P2 (Tables 6,12). Width between the alveoli of M3: The measure- ment (Tables 6, 1 2) is determined in a manner analo- gous to that described above. Width between the condyles: The points of the calipers are placed between the inner tips of the condyles (Tables 6,12). Length, P2 to M3: This is the greatest longitudinal dimension of the lower cheek tooth series. The results are shown in Tables 7 and 1 3. Length and width of the lower cheek teeth: The length is the greatest anteroposterior diameter of the tooth; the width is the greatest transverse dia- meter (Tables 7, 1 3). The Species of Dicotyles During the early 1900’s when many collections of Recent mammals were being established, a number of peccary species were proposed (see Miller, 1924, pp. 48 1 -483 for a list). Most of these belonged to the angiilatus group within the collared peccaries; the white lipped peccary has rarely been known by more than one species name. More recently, however, the number of specific names for the collared peccary has been reduced to one, commonly Pecari tcijacu or Tayassii tcijacu (compare Miller and Kellogg, 1955, pp. 792-795 with Hall and Kelson, 1959, pp. 994- 999). My own observations on specimens in the collections of the Department of Mammalogy of the American Museum of Natural History would tend to corroborate this practice. While some speci- mens from Mexico and the southwestern United States labelled Tayassii angiilatus (Cope, 1 889) show a tendency toward increased hypsodonty of the premolar dentition in general and in particular of the anterior moiety of the teeth relative to the poste- rior moiety, or heel, other specimens of this “species” are not specialized to such an extent. Also, a few 4 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 specimens of Dicotyles tajacii from Colombia are specialized as described above, but most are not. In summary, during the course of this study 1 observed no character or suite of characters by which the former “species” of Dicotyles could be reliably validated. As far as can be determined from the available cranial and dental material, the osteologi- cal and dental variation in collared peccary occurs gradually from one portion of its geographic range to another, and, at least for the purposes of this study is considered as being contained within a single species. MYOLOGICAL DESCRIPTION In the following section the cranial myology of a single adult specimen of Dicotyles tajacii is des- cribed. It was not possible to perform a series of myological dissections and, as a result, an evaluation of the influence of individual and population varia- tion on the development of the various muscles cannot be made. The terminology and format follow- ed below is based on, but slightly modified from that in Sisson and Grossman (1953) as this reference affords a convenient standard. The names applied in the following description are based on the topo- graphic position of the muscle in question and the position of its origin or insertion. Tracing the ner- vous and vascular associates of a given muscle was not undertaken. Homology of a particular muscle in the peccary to one found in the horse, ox, pig, or dog cannot be conclusively demonstrated. Cutaneous Muscles * Cutaneous (Fig. 1). This is a flat sheet of muscle extending from the superficial aspect of the neck onto the posterolateral part of the face. From the nuchal ligament, the muscle extends anterolaterally behind and over the posterior auricular muscles and connects to deep fascia at the ventral and posterior base of the ear. Anteriorly, the muscle attaches to fascia adherent to the posterior surface of the zygo- matic arch. The muscle leaves the arch ventral to the eye, attaching to fascia at the rear of M . zygomaticus, then continues anteroventrally below M. buccinator and fascia lying on the lateral aspect of the mandible. After wrapping around the ventral edge of the man- dible, the left half of M. cutaneus meets the sheet from the opposite side at the median raphe found along the ventral longitudinal axis of the throat. Auricular Muscles * Frohto-scutularis (Figs. 1, 2). This is a small, flat, strap-like muscle found on the dorsolateral surface of the head just posterior to the postorbital process and anterior to the ear. Origin: Posterior superficial surface of the postor- bital process of the frontal and adjacent portion of the parietal crest. Insertion: Anterior end of the scutiform cartilage. Relations: Anterior to M. interscutularis and the scutiform cartilage, superficial to M. temporalis. * interscutularis (Figs. 1, 2). A flat muscle which passes anterolaterally between the parietal crest and scutiform cartilage. Origin: Posterior parietal crest, nuchal crest and anterior portion of the ligamentum nuchae. Insertion: On medial and posteromedial border of the scutiform cartilage. * Denotes muscles which attach either to the cranium or the mandible. Table 1. Geographic distribution of specimens of Dicotyles tajacn examined sex Geographic area numher male female unknown juvenile Arizona 1 _ 1 5 1 Texas 4 1 — 3 — Mexico 25 1 1 7 6 1 Guatemala 1 — — 1 — Honduras 3 1 2 — — Nicaragua 5 2 2 1 — Costa Rica 2 — 1 1 — Panama 1 — — 1 — Colombia 40 1 — 29 10 Equador 1 — 1 — — Peru 6 2 3 1 — Venezuela 4 2 2 — — • British Guiana 7 — — 7 — .Surinam 1 — — 1 — T rinidad 1 — — 1 — Brazil 27 8 7 12 — TOTAL 135 28 26 69 12 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OF DICOTYLES TAJACU 5 Relations: Posterior to the preceding muscle, anterior to M. cervico-scutularis; superficial to M. temporalis. Cervico-scutularis (Figs. 1, 2). This muscle lies pos- terior to the preceding, between the nuchal ligament and scutiform cartilage. A slip from the anterolateral portion connects to the anterodorsal corner of the concha. Origin: From the ligamentum nuchae behind M. interscutularis. Insertion: On the posterior surface of the scuti- form cartilage and on the anterodorsal corner of the concha of the ear, posterodorsal to the insertion of M. scutulo-auricularis superficialis. Relations: Posterior to M. interscutularis, super- ficial to M. cervico-auricularis superficialis and parieto-auricularis; medial to concha of ear. * Zygomatico-auricularis (Figs. 1, 2). The presence of this muscle has not been completely demonstra- ted. It may be represented by a small cylindrical mass which arises from the posterior portion of the zygomatic arch immediately anteroventral to the base of the ear and connects deeply at the anterior base of the conchal cartilage. Scutulo-auricularis superficialis (Figs. 1, 2). This is a flat triangular muscle lying between the postero- lateral surface of the scutiform cartilage and the anterior base of the concha of the ear. Origin: From the lateral surface of the posterior part of the scutiform cartilage. Insertion: On the anterior base of the conchal cartilage anterior to the insertion of M. cervico- scutularis. Relations: Posterolateral to the scutiform carti- lage; posterior to M. fronto-scutularis and antero- lateral to M. cervico-scutularis. Parieto-auricularis (Fig. 2). A thin flat muscle which tapers transversely from the nuchal ligament to the posterior portion of the concha. Origin: From the ligamentum nuchae anterior to and under the cover of M. cervico-scutularis; poste- rior and above that of M. cervico-auricularis super- ficialis. Insertion: On lower surface of convex portion of concha, posterior and ventral to that of M. cervico- auricularis superficialis. Relations: Deep to M. cervico-scutularis; partly superficial and posterior to M. cervico-auricularis superficialis. Cervico-auricularis superficialis (Fig. 2). A thin flat muscle, posterior and deep to M. interscutularis, which tapers transversely to attach to the posterior base of the concha. Origin: On the ligamentum nuchae deep and posterior to the origin of M. interscutularis, anterior to that of M. parieto-auricularis. Insertion: On the posterodorsal portion of the convex concha anterior and dorsal to that of M. parieto-auricularis. Relations: Anterior to M. parieto-auricularis, posterior and deep to M. interscutularis, deep to M. cervico-scutularis. * Parotido-auricularis (Figs. 3, 4, 5)^ A large strap-shaped muscle directed anteroven- trally between the base of the ear and the parotid gland. Origin: Fleshy interdigitation with the parotid gland. Insertion: A thin sheet along the postzygomatic crest medial to the external auditory meatus, extend- ing up the postzygomatic crest to just below the posterolateral corner of the nuchal crest (Fig. 1 1). Relations: Superficial to Mm. cleido-mastoideus, cleido-occipitalis ventralis, serratus ventralis and parotid gland; posterior to M. masseter lateralis, deep to M. cutaneus. Muscles of the Eyelids * Orbicularis oculi (Fig. 1). This is the only muscle of the eyelid which has been located in this specimen. It is a flat sphincter which encircles the eyelids. Muscles of the Face * Levator nasolabialis (Fig. 1). A thin flat muscle on the face just under the skin; it is oriented obliquely anteriorly from the rear of the snout to the postero- lateral surface of the buttress covering the root of the upper canine. Origin: On the maxillary and nasal bones dorsal to the anterior end of the facial crest of the maxillary (Fig. 9). Insertion: On the lips around the canine buttress and forward to the lateral surface of the snout. Relations: Deep to the skin of the snout; super- ficial to Mm. dilator naris lateralis, depressor rostri and levator labii superioris proprius. * Levator labii superioris proprius (Fig. 1). A super- ficial facial muscle which leads anteriorly from the orbital region to the dorsolateral portion of the snout. The anterior portion of the muscle is tendinous. Origin: A shallow preorbital depression anterior to the fronto-lacrimal notch of the orbital rim and below the supraorbital canal (Fig. 9). Insertion: Dorsolateral part of the snout. Relations: Passes beneath the preceding muscle; lies dorsal to tendon of M. dilator naris lateralis. * Dilator naris lateralis (Fig. 1). A slender muscle which passes anteriorly from the facial crest to the lateral portion of the snout. Origin: In the anterior part of an ovoid pit on the ^ The name for this muscle is misspelled Parotoidoauri- cularis in Figs. 3,4, and 5. 6 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 temporalis depressor lobii inferioris zygomaticus mosseter lateralis superficialis scutulo- auricularis superficialis fronto-scutularis orbicularis ocu levator labii superioris proprius levator nasolabialis interscutularis dilator naris lateralis depressor rostri cervico- scutularis cutaneus zygomatico- auricularis Figure 1. Dicotyles tajacii; superficial cranial musculature as seen in left lateral view. About one-half natural size. interscutularis scutulo-auricularis superficialis cervico-scutularis temporalis fronto-scutularis parieto- auricularis cervico- auricularis superficialis cutaneus zygomatico-auricularis Figure 2. Dicotyles tajacir, dorsal aspect of temporal region showing auricular musculature. About three-fourths natural size. 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU 7 undersurface of the facial crest above M* and ante- rior to a sulcus for the attachment of M. masseter lateralis profundus (Figs. 9,10). Insertion: On lateral part of the snout. Relations: The muscle splits into two bodies, each of which become tendinous anteriorly. The bodies of the muscle lie dorsal to M. depressor rostri and extend anteriorly beneath M. levator nasolabialis. * Depressor rostri (Fig. 1 ). This muscle arises slightly posterior to the preceding, and parallels it to reach the snout. The muscle has also been designated as M. depressor labii superioris by various workers. Origin: The posterior part of the same depression occupied by the origin of the preceding muscle (Fig. 10). Insertion: Ventrolateral edge of snout. Relations: Ventral to M. dilator naris lateralis, dorsal to M. buccinator; passes beneath M. levator nasolabialis. * Buccinator (Fig. 1). This muscle forms the lateral wall of the mouth posterior to the canines. Origin: On and below a strong crest which passes dorsal and parallel to the alveolar border from the rear of the upper canine to a point above M^; then posterodorsally and laterally along the maxillary portion of the base of the zygomatic arch (Fig. 9); in the sulcus at the anterior surface of the coronoid process of the mandible posterior to M3 and along the lower alveolar border anteriorly to the diastema between the premolar and canine (Fig. 12). Insertion: Angle of mouth, blending with M. orbicularis oris. Relations: Ventral to body of M. depressor rostri, deep to M. masseter medialis posteriorly and M. zygomaticus; dorsal to and posteriorly interfingering with M. depressor labii inferioris. Zygomaticus (Fig. 1 ). A short, flat superficial muscle between the anterior edge of M. masseter lateralis superficialis and M. buccinator. Origin: In fascia of anterior surface of M. mas- seter lateralis superficialis and M. cutaneous. Insertion: Fascia of the midlateral portion of M. buccinator near the angle of the mouth. Relations: Ventral to M. depressor rostri, deep to M. levator nasolabialis; anterior to Mm. cutaneus and masseter lateralis superficialis. * Depressor labii inferioris (Figs. 1, 5). A long thin largely tendinous muscle which extends antero- ventrally from the anteromedial surface of the coro- noid process of the mandible to the ventrolateral lips. Origin: In common with M. buccinator along the posterior alveolar border of the mandible and medial surface of the coronoid process (Figs. 1 2, 1 3). Insertion: In the tissue of the lips at the ventro- lateral tip of the mouth anterior to the canines. Relations: Ventral to M. buccinator. Muscles of the Neck and Adjacent Cranium Brachiocephalicus. The muscle occupies the dorsal and lateral aspect of the neck and occipital area deep to M. cutaneus; it is divided into two major parts in the peccary. M. cleido-occipitalis, dorsal of the two divisions, is also divisible into two parts on the basis of their origins; the Mm. cleido-occipitalis dorsalis and ventralis. The insertions tor these mus- cles, and for Mm. rhomboideus and splenius, were not determined during this dissection as the animal’s head had been separated from its body. Cleido-occipitalis dorsalis (Fig. 3) Origin: As a flat sheet on the nuchal ligament posterior and deep to M. interscutularis. Relations: Superficial to Mm. rhomboideus, splenius and serratus ventralis; posterior to M. parotido-auricularis. Cleido-occipitalis ventralis (Figs. 3-6). Origin: On the mid-anteroventral wing of the atlas. Relations: Ventral to M. splenius mastoideus and serratus ventralis and dorsal to M . cleido-mastoideus. * Cleido-mastoideus (Figs. 3-6, 10). This distinct muscle, which lies posterior and deep to the parotid gland, is the second major division of M. brachioce- phalicus. Origin: Just ventral to that of M. splenius mas- toideus (Figs. 15,16). Relations: Anteroventral to Mm. cleido-occipita- lis ventralis and splenius mastoideus; deep to the parotid gland; dorsal to M. sternomastoideus. Rhomboideus (Figs. 3, 4). This is a relatively small, thin, flat muscle with two points of origin which occupies the dorsomedial portion of the neck near the nuchal ligament. The two heads have received the following designations: Rhomboideus cervicalis (Fig. 5). Origin: Along the nuchal ligament, but not quite reaching the occiput of the cranium. The muscle passes posterolaterally to join the fibers of the second part. * Rhomboideus capitis (Fig. 5). Origin: As a thin slip for about Vi” along the postzygomatic crest just below the posterolateral corner of the nuchal crest. The site of origin is deep to that of M. parotido- auricularis (Fig. 1 1 ). Relations: The two slips merge posterolaterally and are deep to M. cleido-occipitalis dorsalis, super- ficial to M. splenius mastoideus and are bound by fascia to the small dorsal slip of M. serratus ventralis. Splenius (Figs. 3, 4). An extensive muscle on the dorsal and lateral surface of the neck, the fibers of which are directed anterolaterally. The muscle is undivided posteriorly but separates into two distinct heads anteriorly, M. splenius capitis and M. splenius mastoideus. The mastoid portion is in turn separated into a dorsal, mastoid, portion and a ventral portion, M. splenius cervicis. 8 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 temporalis cleido-occipitalis dorsalis rhomboideus splenius serrotus ventrolis cleido- occipitalis ventrolis ^rotoido-Quriculoris cleido-mastoideus ■hyoideus-'-™----^^ masseter lateralis parotid gland superficiolis Figure 3. Dicotyles tajacir, superficial musculature of posterior portion of the cranium after removal of M. cutan- eus. About three-fifths natural size. sterno-hyoideus masseter lateralis superficial cleido-mastoideus parotoido-auricularis parotid gland cleido-occipitalis ventrolis serrotus ventrolis rhomboideus nucha! ligament splenius mylo-hyoideus genio-hyoideus trachea sterno-mostoideus Figure 4. Dicotvies tajacir, ventral view of head showing superficial muscles after the removal of M. cutaneus. About one-half natural size. 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OE DICOTYLES TAJACU 9 cleido-occipitalis ventralis cleido-mastoideus sterno-mastoideus masseter lateralis omo-hyoideus profundus / sterno-hyoideus pterygoideus medialis parotoido-auricularis temporalis masseter lateralis superficialis depressor labii inferioris splenius capitis rhomboideus cervicalis rhomboideus capitis splenius mastoideus splenius cervicis serratus ventralis Figure 5. Dicotyles tajacii; slightly deeper dissection of the left lateral aspect of the head. About one-half natural size. * Splenius capitis (Fig. 5): The insertion is on the upper half of the rugose rim of the occiput below and parallel to the nuchal crest (Fig. 1 1). * Splenius mastoideus (Fig. 5): This branch inserts on the mastoid region of the zygomatic arch below the external auditory meatus and dorsal to the post- glenoid process (Figs. 10, 11). Splenius cervicis (Figs. 5, 6): The insertion is on the tip of the wing of the atlas. Serratus ventralis (Figs. 3-6). This muscle mass consists of two parts, the dorsal part being more slender than the ventral. These probably correspond to the M. serratus cervicis branch of the serratus group and thus originate farther forward than in the horse, but about the same as in the ox and pig (see Sisson and Grossman, 1953, pp. 299, 353, 364). Another possible interpretation is that the dorsal part corresponds to M. serratus cervicis while the larger, ventral, part corresponds to M. serratus thoracis which would, in this case, attach quite high on the neck. Origin: Dorsal part: along the anteroventral sur- face of the tip of the atlantal wing. Ventral part: along the ventral surface of the transverse process of the axis and cervical vertebrae 3-5; deep surface of the posterior corner of the atlantal wing. Relations: Ventral to Mm. splenius mastoideus and splenius cervicis; deep to M. cleido-occipitalis ventralis; superficial to M. semispinalis capitis. * Semispinalis capitis (Fig. 6). This muscle forms a deep segment of the neck musculature and attaches to a correspondingly deep part of the occiput. In other animals, two muscle masses, a dorsal M. biventer cervicis and a ventrolateral M. complexus or M. complexus major may be differentiated. In the collared peccary the two divisions are not readily distinguishable. The fibers form a relatively thin sheet, the bundles of which are directed postero- laterally from the nuchal crest. Insertion: On the nuchal crest below M. splenius capitis from midline to posterolateral corner of nuchal crest (Fig. 1 1). Relations: Superficial to Mm. obliquus capitis anterior and obliquus capitis posterior, rectus capitis dorsalis major and minor; deep to Mm. rhomboideus, splenius and longissimus capitis. * Sterno-mastoideus (Figs. 3-5). A massive muscle between Mm. cleido-mastoideus and masseter later- alis superficialis. The muscle tapers markedly toward its tendinous insertion on the mastoid areajust dorsal to the postglenoid process on the crest which curves dorsolaterally from that process toward the base of 10 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 the external auditory meatus. The insertion area is bounded dorsally, laterally and ventrally by that of M. cleidomastoideus (Figs. 10, 11). Relations: Inserted between Mm. cleidomastoid- eus and M. masseter lateralis superficialis; superficial to M. omo-hyoideus; deep to parotid gland and M. parotido-auricularis. * Longissimus capitis et atlantis (Fig. 6). This muscle is composed of a dorsal (capitis) and ventral (atlantis) portion. The dorsal slip is much more massive than the ventral. Both lie deep to Mm. splenius mastoideus and splenius cervicis. Insertion: Dorsal portion; On the mastoid region of the cranium below the external auditory meatus deep to the dorsal half of the insertion of M . splenius mastoideus (Fig. 1 1). Ventral portion: On the poster- ior surface of the wing of the atlas. Relations: Deep to Mm. splenius mastoideus and splenius cervicis; superficial to Mm. semispinalis capitis and obliquus capitis anterior. * Obliquus capitis anterior (Figs. 6, 7). This muscle arises from the anterior and dorsal surface of the wing of the atlas and apparently divides into two heads; one extends vertically, the other anteriorly. Origin: On the anterior and dorsal surface of the wing of the atlas. Insertion: (1) the dorsal head inserts on the post- zygomatic crest below the posterolateral corner of the nuchal crest; the insertion area lies deep to but parallels that for the origin of M. parotido-auricularis (Fig. 1 1). (2) The more ventral head inserts on the mastoid area ventral to the external auditory meatus, deep to the insertions of Mm. longissimus capitis, splenius mastoideus, sterno-mastoideus and cleido- mastoideus. The insertion area of ventral head of M. obliquus capitis anterior extends medially to the squamosoexoccipital suture and ventrally to the dorsolateral base of the paroccipital process (Figs. 10,11). Relations: Superficial to Mm. rectus capitis dor- salis major and minor and M. rectus capitis lateralis; deep to the muscles listed above. Obliquus capitis posterior (Fig. 7). This horizontally oriented muscle lies between the dorsal surface of the axis and the posterior dorsal surface of the atlas. Origin: On the spine and lateral surface of the transverse process of the axis. Insertion: On the posterior dorsal surface of the wing of the atlas. Relations: Dorsal to all but the spine of the axis; ventral to M. rectus capitis dorsalis minor; deep to M. semispinalis capitis. * Rectus capitis dorsalis major (Fig. 7). A thick muscle which extends from the axis to the occiput in contact with the nuchal ligament. This is one of the deepest muscles of the occiput. Origin: Along the dorsal spine of the axis. Insertion: On nearly all of the occipital surface deep and medial to the attachment areas of the other occipital muscles. The muscle extends down along the upper fourth of the foramen magnum, then dorsally medial and parallel to the postzygomatic crest (Fig. 1 1). Relations: This is the deepest muscle of the dorsal neck region. At its lower lateral corner, it is covered by M. rectus capitis dorsalis minor; it is overlain dorsally and laterally by M. semispinalis capitis. * Rectus capitis dorsalis minor (Fig. 7). A smaller muscle lateral and ventral to the above. Origin: On the lateral surface of the spine of the axis. Insertion: On the medial surface of the post- zygomatic crest between the insertion of preceding and that of M. obliquus capitis anterior (Fig. 1 1). Relations: Deep to M. obliquus capitis anterior; lateral to M. rectus capitis dorsalis major. * Rectus capitis lateralis. A short muscle which lies between the median anterior surface of the atlas and the base of the paroccipital process. Origin: On the medial half of the anterior surface of the wing of the atlas. Insertion: Along the posterior surface and base of the paroccipital process and the posterior and ventral surface of the condyloid sulcus (Fig. 10). Relations: Lateral to M. rectus capitis ventralis minor; deep to M. obliquus anterior (ventral por- tion); anteromedial to M. obliquus capitis anterior. * Rectus capitis ventralis major (Figs. 7, 8). An elongate muscle applied to the ventral aspect of the cervical vertebrae and basilar region of cranium; the muscle divides into two slips posteriorly. Origin: On the ventral surface of transverse pro- cesses of cervical vetebrae 3 and 4. Insertion: Along the basisphenoid and basioccipi- tal bones of the cranium, medial and slightly antero- medial to the edge of the bulla (Fig. 10). Relations: Deep to Mm. cleido-mastoideus and sterno-mastoideus; ventral to M. serratus ventralis; dorsal to Mm. sterno-thyroideus, thyro-hyoideus, ventricularis; ventral to M. rectus capitis ventralis minor. * Rectus capitis ventralis minor. This is a slender muscle lying dorsal and slightly posterior to the preceding. Origin: On the ventral arch of the atlas. Insertion: This is not clearly differentiated from that of M. rectus capitis lateralis posteriorly. The insertion area lies along the ventral surface of the basioccipital, medial to the bulla and anteriorly nearly to the basioccipito-basisphenoid suture (Fig. 10). Relations: Posterior and dorsal to insertion of M. 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU pterygoideus medialis masseter lateralis superficialis longissimus atlantis sterno-hyoideus omo-h splenius cervicis serratus ventralis rectus capitis ventralis major ventricularis sterno-thyroideus thyro-hyoideus temporalis cleido-occipitalis ventralis semispinalis capitis cleidomastoideus masseter lateralis profundus obliquus capitis anterior longissimus capitis Figure 6. Dicotyles tajacii; left lateral view of the posterior portion of the head showing some of the deeper muscles of the throat and neck. Slightly larger than one-half natural size. temporalis digastricus ? masseter medialis depressor labii inferior! masseter lateralis superficialis rectus capitis dorsalis major rectus capitis dorsalis minor obliquus capitis anterior obliquus capitis posterior rectus capitis ventralis majar ventricularis sterno-thyroideus crico-thyroideus pterygoideus medialis thyro-hyoideus Figure 7. Dicotyles tajacu; left lateral view of the posterior portion of the head showing mainly the deep muscula- ture of the throat and neck. About one-half natural size. 12 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 rectus capitis ventralis major medial to M. rectus capitis lateralis. Ventral cervical and hyoid musculature * Mylo-hyoideus (Figs. 4, 8). A flat sheet covering the ventral aspect of the throat anterior to M. sterno- hyoideus. Origin: From the medial surface of the mandible at the rear of the symphysis, continuing posterodor- sally along a slightly roughened line to a point just below the rear of the last molar (Fig. 13). Insertion: On a median raphe from the symphysis to the hyoid bone anterior to Mm. sterno-hyoideus and omo-hyoideus. Relations: Deep to Mm. cutaneus; anterior to Mm. sterno-hyoideus and omo-hyoideus; superficial to M. genio-hyoideus. * Genio-hyoideus (Fig. 4). A long slender muscle which extends longitudinally between the symphysis and hyoid bone. Origin: In the genial pit just posterior to the symphysis on the medial surface of the mandible in common with the origin of Mm. genio-glossus and the anterior tip of mylo-hyoideus. (Fig. 13). Insertion: On the lingual process of the hyoid bone. Relations: Deep to Mm. mylo-hyoideus and thyro- hyoideus; superficial to M. crico-thyroideus. * Genio-glossus (Fig. 8). A fan-shaped muscle be- tween the symphsis and the base of the tongue. Origin: On the medial surface of the mandible in common with the Mm. genio-hyoideus and mylo- hyoideus (Fig. 13). Insertion: Fibers fan out from the origin to insert posterodorsally in the base of the tongue. Relations: Deep to M. genio-hyoideus The following muscles could be differentiated in the peccary but are not treated further as they make no important connections to the cranium. Several have been figured. Omo-hyoideus; sterno-hyoideus;* stylo-hyoideus; thyro-hyoideus; hyo-glossus; crico-thyroideus; * stylo-glossus; * palatinus; * levator palati; * tensor palati; * occipito-hyoideus. Mandibular Muscles Masseter lateralis. This muscle is divided by a sheet of fascia into superficialis and profundus portions. * Masseter lateralis superficialis (Figs. 3-7). This muscle forms the bulk of M. masseter as seen in lateral aspect. Origin: Thinly along the zygomatic arch, generally in association with a sharp crest, posterodorsal to the glenoid fossa; then anteriorly to a point above M^ (Figs. 9,10). Insertion: On the border of mandible from the base of the condyle to the anterior end of the post- digastric sulcus. On the ventral border, this muscle extends across that sulcus to the medial surface, blending to a slight degree with M. pterygoideus medialis (Figs. 12,13). Relations: Deep to M. cutaneus, superficial to M. masseter lateralis profundus. * Masseter lateralis profundus (Figs. 5, 6). Origin: pterygoideus medialis masseter lateralis superficialis digastricus mylo-hyoideus genio-glossus stylo-hyoideus hyo-glossus rectus capitis ventralis maj Figure 8. Dicotyles tajacir, deep ventral musculature of the throat and neck. About three-fourths natural size. 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OF DICOTYLES TAJACU 13 temporalis levator labii superioris proprius levator nasolabialis dilator naris lateralis masseter lateralis superficialis masseter lateralis superficialis masseter lateralis profundus pterygoideus medialis pterygoldeus lateralis Figure 9. Dicotyles tajacir, left lateral view of the cranium showing the attachment areas of the major muscles. About one-half natural size. Deep to the preceding, on the ventral surface of the zygomatic arch medial to the crest of the arch (Fig. 10). Insertion: In the masseteric fossa of the mandible (Fig. 12). Relations: Deep to M. masseter lateralis super- ficialis; deep to M. ?masseter medialis; posterior to Mm. dilator naris lateralis; depressor rostri and depressor labii inferioris. * ?Masseter medialis (Fig. 7). This muscle may be present, although poorly differentiated from the lateral portion of M. temporalis. Origin: Along the medial edge of the ventral portion of the zygomatic arch. Insertion: In the coronoid fossa on the lateral surface of the mandible. (Fig. 12). Relations: Ventral to the main body of M. tempo- ralis; deep to M. masseter lateralis profundus. * Temporalis (Figs. 1-3, 5-7). This large muscle occupies the temporal fossa of the cranium, passes medial to the zygomatic arch and inserts on the coronoid process of the mandible. Origin: From the postorbital process of the frontal posteriorly along the parietal crest; the anterior edge of the nuchal and postzygomatic crests; the anterior surface of the bony covering of the external auditory meatus; anteriorly along the dorsal edge of the medial surface of the zygoma to the postorbital process of the Jugal; from this the line of origin continues anteroventrally, following the curve of the maxilla to the maxillary tuberosity above M^, then posteriorly along the ventral edge of the zygoma to the glenoid fossa; thence along the squamosal to the anterodorsal edge of the bulla medially to the infratemporal crest and dorsally to the base of the postorbital process of the frontal (Fig. 9). Insertion: On the dorsal tip of the coroid process by a strong tendon (Figs. 12,13). Relations: This is the major muscle in the temporal fossa; medial to M. ?masseter medialis ventrally. * Digastricus (Figs. 7, 8). A long thin double bellied muscle situated obliquely between the paroccipital process and the anterior medial surface of the man- dible. Origin: Along the ventromedial border of the paroccipital process superficial to M. stylo-hyoideus (Fig. 10). Insertion: On the medial surface of the mandible just posterior to the symphysis, dorsal to the ventral border of the mandible (Fig. 13). Relations: The muscle passes deep to M. ptery- goideus medialis as a tendon; lateral to M. mylo- hyoideus; superficial to Mm. genio-hyoideus, hyo- glossus, stylo-glossus and stylo-hyoideus. * ?Occipito-mandibularis This muscle apparently lies between the paroc- cipital process and the posterior edge of the man- dible. It is poorly differentiated from M. digastricus. 14 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 masseter lateralis superficialis masseter lateralis profundus depressor rostn dilator naris lateralis cleido-mastoideus sterno-mastoideus splenius mastoideus ^ digastricus |fp^^^^^__obliquus capitis anterior pterygoideus lateralis pterygoideus medialis rectus capitis ventralis rectus capitis lateralis rectus capitis major ventralis minor Figure 10. Dicotyles tajacu; ventral view of left half of the posterior portion of the cranium showing the attach- ment areas of the major muscles. About three-fourths natural size. ^Pterygoideus medialis (Figs. 5-8). This muscle is the counterpart of M. masseter lateralis superficialis on the medial surface of the mandible. Origin: Along the ventral surface of the alisphen- oid in the pterygoid fossa, lateral to the hamular process of the pterygoid bone (Figs. 9,10). Insertion: From the medial edge of the coronoid process of the mandible opposite the rear of M3, down the anterior end of the angle of the mandible, along the ventral surface of the angle and posterior side of a dorsally projecting spine, then up to the ventral edge of the condyle (Fig. 13). Relations: The area of insertion is thin, forming a band peripheral to that of M. pterygoideus lateralis on the medial surface of the mandible. * Pterygoideus lateralis. Origin: The lateral edge of the ventral surface of the alisphenoid dorsally into the pterygopalatine fossa and anteriorly to a point above (Figs. 9,10). Insertion: On the medial surface of the mandible central to the area circumscribed by the insertion of the preceding (Fig. 13). Relations: This muscle forms the bulk of the pterygoid musculature. Variation in Muscle Scars It is often tempting to reconstruct the musculature of an extinct animal from its fossilized hard parts. A number of observations made during this study may serve to encourage caution in such exercises. One hundred thirty-five specimens of Dicotyles tajacu in the collections of the Department of Mammalogy, the American Museum of Natural History, were examined with regard to individual variation. Al- though minor variations in cranial osteology could be found, marked departures from the general mor- phologic plan of Dicotyles were not observed, even though geographic range of the collection is wide (Table 1). In general, specimens from Colombia seemed to be somewhat smaller and more lightly constructed, and had less well defined attachment areas for such muscles as M. digastricus and M. mylo-hyoideus than was the case for specimens from other areas. Moreover, one-fourth of the Colombian specimens were Juvenile (deciduous premolars still functional). Considering the collection as a whole, minor variations in the configuration of the cranium include the degree of anterior tapering of the rostrum as seen from above, the strength of the facial crest, the development and posterior extent of the crest marking the dorsal limit of M. buccinator in the diastemal region, the degree of roundness of the cross section of the infraorbital foramen, the strength of the canine buttresses, the depth of the concavity 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OE DICOTYLES TAJACU 15 developed on the dorsolateral surface of the zygo- matic arch below the orbit and the relative breadth of the cranium across the zygomatic arches at the post- glenoid processes. In the mandible, the strength of the attachment area for M. temporalis, the degree of rugosity in the masseteric and pterygoid fossae, the depth of the digastric fossa, the relative size of the postdigastric sulcus and the degree of rugosity of the origin of M. mylo-hyoideus are the major features, related to muscle attachments, in which variation was noted. Of the 47 muscles described in the preceding section 36 attach either to the cranium or to the mandible. Of these, only 1 1 are associated with sufficiently distinctive structures that their areas of attachment may be readily determined. In most cases, these structures are such integral features of the cranium as to be essentially immune to signifi- cant individual variation. The areas of attachment for the 1 1 muscles in question are as follows: 1 . M . temporalis: The temporal fossa in which this muscle originates is clearly defined in all cases. Even at its anterior end, the temporal fossa is separated by a slight vertical ridge (infratemporal crest) from the orbital and pterygopalatine fossae. M. temporalis inserts on the dorsal tip of the coronoid process of the mandible (Figs. 1 2, 1 3). This structure was easily discernible in all cases although the ventral limit of the muscle on the lateral surface of the mandible is not determinable on the bone. The cavity on the coronoid process which lies dorsal to the masseteric fossa and which is often known as the temporal fossa of the mandible may house M. masse- ter medialis in the peccary. 2. M. rnasseter lateralis superficialis: The site of origin of this muscle, taken as the edge of the sharp, ventrally directed crest at the ventrolateral edge of the zygomatic arch, is distinctly marked in all speci- mens examined. As indicated in Figure 9, a few fibers of this muscle attach above and posterior to the glenoid fossa, but this site of attachment was not well marked on any of the crania examined. The insertion of M. rnasseter lateralis superficialis was clearly visible only in the dissected specimen. A few of the specimens in the American Museum approached, but did not equal, this degree of rugosity. 3. M. rnasseter lateralis profundus: The muscle originates in an elongate, relatively deep, ventrally facing concavity located on the ventral edge of the zygomatic arch just medial to the origin of the preceding muscle. The origin of M . rnasseter lateralis profundus was easily determinable in all specimens examined although in no case was its site of insertion separable from that for M. rnasseter lateralis super- ficialis . 4. M. buccinator: This muscle originates, on the cranium, in an elongate, laterally facing, concavity. This concavity or fossa, as it might be termed when well developed, lies just dorsal to the alveolar border from the base of the canine to the level of M* or M^. In some specimens, the crest which defines the dorsal edge of the fossa reaches only to the last premolar. In no instance was the posterior limit of the origin of M. buccinator, i.e., above the last molar on the underside of the maxillary base of the zygomatic arch, clearly indicated. The site of origin of this muscle on the mandible was never found to be indicated in any way. rhomboideus capitis parotido-auricularis obliquus capitis anterior longissimus capitis splenius mastoideus sterno-mastoideus cleido-mastoideus splenius capitis semispinalis capitis rectus capitis dorsalis major rect. capit. dorsal, minor Figure 1 1. Dicotyles tajacir, posterior view of left half of the occiput showing the attachment areas of the major muscles. About three-fourths natural size. 16 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 5. M. rectus capitis dorsalis major: This muscle inserts in the occipital fossa (Fig. 11) which was found to be well developed in all except the most immature crania (those with none but deciduous teeth). The ventral and ventrolateral limits of the insertion of this muscle were never well defined. 6. M. rectus capitis dorsalis minor: This muscle inserts on a tear-drop shaped area above the base of the paroccipital process. A thin, distally tapering process is occasionally developed at the lateral edge of this insertion area. Of the 1 35 specimens examined the configuration of the insertion of M . rectus capitis dorsalis minor could be adequately determined on only 34 crania. Twelve of these crania were of male individuals, 5 were female and in 17 the sex was not determined. 7. M. pterygoideus medialis: The site of origin occurs on the lateral surface of the hamular process of the pterygoid and adjacent ventral surface of the alisphenoid, reaching posteriorly along that surface to the anteromedial corner of the bulla. In the sense that this flat ventral strut of alisphenoid and associ- ated pterygoid are always well defined, the origin of M. pterygoideus medialis is always visible in the peccary. 8. M .pterygoideus lateralis muscle originates roughly from the pterygopalatine fossa, the dorsal limit of which corresponds to a strong ridge which extends posteriorly from the roof of the sphenopala- tine foramen toward the optic foramen. Inasmuch as the pterygopalatine fossa is usually well defined in Dicotyles the area of origin for M. pterygoideus lateralis is also determinable. As mentioned above, the site of insertion of the muscle could not be distinguished from that of M. pterygoideus medialis in any of the crania in the American Museum of Natural History. 9. M. rectus capitis ventralis major: This muscle inserts on the basilar eminences of the basioccipital and basisphenoid. These eminences are usually visible, although the exact limit of the muscle attach- ment is not determinable. 10. M. digastricus: The muscle inserts in the digastric fossa on the medial surface of the mandible (Fig. 1 3). The fossa is not always well defined, but in many cases the configuration of the attachment area is indicated by an elongate slightly rugose area on the surface of the bone just anterior to the postdi- gastric sulcus. The insertion area for M. digastricus could be determined for 81 of the 135 specimens examined; three of these were immature individuals, 20 were male, 21 were female and in 37 cases sex was not determined. 1 1. M. mylo-hyoideus: This muscle originates in a narrow band 2 to 3mm. wide which traverses the medial surface of the mandible from the rear of the symphysis to a point slightly below the rear of M3. The surface of the attachment area is often slightly roughened, but may also be indicated as a slight “break” in the medial surface of the mandible. The site of origin of M. mylo-hyoideus was discernible in 91 of the 135 speciments; 3 of these were immature individuals, 17 were male, 16 were female and in 55 the sex was not determined. The results of the above considerations show that only a small percentage of the total number of mus- cles attaching to the cranium or mandible of the collared peccary leave sufficiently distinctive scars or rugosities that the area of attachment may be discerned with reliability. The muscles which are associated with definitive structures seem to be those involving mastication and the support of the head upon the neck. The structures to which the muscles attach are major features of the cranium. temporalis ?masseter medialis masseter lateralis superficialis mass. lat. profundus Figure 12. Dicotyles tajacu; lateral view of left mandible showing the attachment areas of the major muscles. About three-fourths natural size. 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU 17 temporalis depressor labii inferioris pterygoideus /j medialis pterygoideus lateralis masseter lateralis digastricus genio-glossus genio'hyoideus superficialis mylo-hyoideus Figure 13. Dicotyles tajacir, medial view of left mandible showing the attachment areas of the major muscles. About three-fourths natural size. but other, equally distinctive cranial structures seem to support more than one muscle. It is convenient, and perhaps more meaningful, to correlate a given crest or fossa with the group of muscles which are associated with it. Such a tabulation as the following may be conveniently designed; Cranium Occipital Aspect (Fig. 1 1) Occipital fossa: Mm. rectus capitis dorsalis major and minor. Nuchal crest: Mm. splenius capitis, rhomboideus capitis, semispinalis capitis. Postzygomatic crest: Mm. parotido-auricularis, obliquus capitis anterior (dorsal head). Mastoid region: Mm. longissimus capitis, splenius mastoideus,sterno-mastoideus,cleido-mastoideus, obliquus capitis anterior (ventral head). Paroccipital process: Mm. obliquus capitis anterior, rectus capitis lateralis, digastricus (Fig. 10), occipito-hyoideus. Lateral Surface (Figs. 9,10) Lacial crest: Mm. depressor rostri, dilator naris lateralis. Zygomatic arch: Mm. masseter lateralis superficialis and profundus. Pterygopalatine fossa: M. pterygoideus lateralis. Buccinator fossa: M. buccinator. Temporal fossa: M. temporalis. Ventral Aspect (Fig. 10) Pterygoid fossa: M. pterygoideus medialis. Condyloid sulcus: Mm. rectus capitis lateralis, rectus capitis ventralis minor. Basilar eminences: M. rectus capitis ventralis major. Mandible Lateral Aspect (Fig. 12) Masseteric fossa: Mm. masseter lateralis superficialis and profundus. Coronoid process and fossa: Mm. ?masseter medialis and temporalis. Medial Aspect (Fig. 13) Pterygoid fossa: Mm. pterygoideus lateralis and medialis. Digastric fossa: M. digastricus. Posterior surface of symphysis: Mm. genio-glossus, genio-hyoideus, mylo-hyoideus. Mylo-hyoid line: M. mylo-hyoideus. CRANIAL OSTEOLOGY OF THE LIVING PECCARIES The descriptions that follow are not meant to be minutely exhaustive but rather to be a comparison of the salient features of the crania of the two species. A detailed description of the cranial elements of Dicotyles tajacu may be found in Rusconi (1929). 18 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 The paragraphs are numbered to facilitate compari- son with similarly numbered paragraphs in the description of Tayassu. Collared Peccary, Dicotyles tajacu (Linnaeus) Cranium: 1. In dorsal and ventral view (PI. 1) the greatest width of the cranium (Table 2) is usually measured across the zygomatic arches at the level of the ante- rior end of the glenoid fossa. The facial crest con- tinues anteromedially from the arch and generally fades out along the rostrum posterodorsal to the canine buttress. In some specimens the facial crest is shorter and more obliquely oriented than in others, but in no case is the ventrolateral edge of the maxillary visible from dorsal view as it is in Tayassu (PI. 3, Fig. 1). 2. The rostrum, that part of the cranium anterior to the zygomatic arches, is narrow and relatively deep. It tapers slightly anteriorly in dorsal view, but its outline is largely unbroken except for the promi- nent canine buttress. The profile of the dorsal surface of the rostrum, as seen in anterior view, is convex. Farther posteriorly, in the frontal area, the supra- orbital foramina lead anteriorly as the supraorbital- nasal canals; these converge shortly after leaving the foramina (Pis. 1, 2), but diverge again, so that the greatest width between them is measured anterior to the frontal crests, about at the level of P^. Anteriorly, the canals continue onward, above the canine but- tresses, to the rear of the narial notch. The latter is conspicuously more rectangular, in profile, than in Tayassu. The canine buttresses (PI. 1, Fig. 2; PI. 2, Fig. 2), developed over the roots of those teeth, are prominent, with a strongly flange-like flat anterior surface. Posterior to the buttresses the lateral wall of the rostrum is excavated to a variable degree below the facial crest, the ventral border of this concavity being formed by the linear ridge which corresponds to the upper limit of the buccinator musculature. The buccinator fossa, which extends posteriorly from the rear of the canine to a point above M* or, rarely, M^, faces laterally. 3. In ventral view (PI. 1, Fig. 3; PI. 2, Fig. 3), the two pairs of incisors (the third having been lost during phylogeny) are arranged in a V-shaped array with a diastema of about 1 5 mm. occurring between F and the canine. The incisive foramina are rela- tively large; their combined width being almost one- half of the total width of the rostrum at F. Posterior to these foramina, the surface of the palate is moder- ately rugose; palatine grooves seem to be developed from about P^ onward, but their course is somewhat obscured and subsidiary openings in the palatal surfacemay occur farther posteriorly. A longitudinal median keel is often present, particularly between the canine and M-'^. 4. Sutures are visible only in immature individu- als of the collared peccary, but in less fully grown individuals (deciduous premolars functional, erupting) it can be seen that the premaxillaries (PI. 4, Figs. 1, 2; PI. 5, Figs. 1, 2) bound the narial aper- ture laterally and extend posterodorsally to a point slightly (ca. 7.5 mm.) anterior to the greatest diver- gence of the supraorbital-nasal canals. The nasal bones are sharply restricted (PI. 4, Fig. 1; PI. 5, Fig. 1) between the premaxillaries, but still form the roof of the narial aperture. The maxillo-nasal suture is almost straight as it continues posteriorly from the premaxillary to contact the frontal. Ven- trally the premaxillo-maxillary suture enters the palate through the incisor-canine diastema and ex- tends forward to the lateral edges of the incisive foramina. 5. On the top of the cranium the frontal (PI. 4, Fig. 1; PI. 5, Fig. 1) and parietal elements have an apparently normal configuration. Depressions on the dorsal surface of the frontal above the postorbital processes are usually present, even if only to a very small degree. The fronto-nasal contact leads antero- laterally to join the maxillary; the suture of the latter with the frontal then curves posteroventrally for about 9-10 mm. to meet the jugal (PI. 4, Fig. 2; PI. 5, Fig. 2). The fronto-jugal suture extends posteriorly and slightly ventrally to the small lacrimal tuberosity and crosses over into the orbit, passes through the lacrimal foramen, continues ventrally and, lastly, anteriorly to enter the sphenopalatine foramen. The latter lies just dorsal and medial to the larger maxil- lary foramen. Just medial to the latter and ventral to the sphenopalatine foramen is a third, small, verti- cally elliptical aperture which probably represents the posterior palatine foramen. The frontal leaves the sphenopalatine foramen in contact, for a short distance, with the palatine, but after about 14 mm. passes posterodorsally, in contact with the larger orbitosphenoid. The frontal finally completes its circuit just dorsal to the infratemporal crests where the fronto-parietal suture meets the orbitosphenoid. 6. The posterior sutures of the parietal are lost in the rugosities of the nuchal and postzygomatic crests; the parietosquamosal suture extends across the temporal fossa, near its middle, to meet the alisphenoid just posteroventral to the infratemporal crest. Continuing almost vertically, the squamoso- alisphenoid suture extends downward, crosses over the ventral, pterygoid, crest of the alisphenoid to disappear near the anterodorsal base of the auditory bulla. 7. The bulla (PI. 1 , Fig. 3; PI. 2, Fig. 3; PI. 5, Fig. 3) is bulky, globular, but moderately acuminate anteromedially; it is composed of cancellous tissue internally and bears an anteromedially directed groove for the tympanohyal along its ventral surface. Foramina found peripheral to the bulla are: an- 1968 19 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU teriorly, the foramen ovale; medially the median lacerate foramen; posteriorly the posterior lacerate and (just labial to it) the stylomastoid foramina. Lateral to the bulla and just behind the postglenoid process is the postglenoid foramen. A small aperture lateral to the foramen ovale probably represents the alisphenoid canal. 8. Posterior to the bulla, the squamoso-exoccipi- tal suture descends along the posterior surface of the canal of the stylomastoid foramen, traverses the anterior base of the paroccipital process, and leads dorsally and slightly medially to join the parietal at the postzygomatic crest. 9. Anteriorly, the squamosal contacts the Jugal at the anterior edge of the glenoid fossa. The bone is excavated laterally and flares outward below and anterior to the orbit. The postorbital process of the jugal may approach, but does not meet that of the frontal. There is nothing unusual regarding the relations of the jugal to the surrounding bones. The ventral surface of the jugal, and part of the adjacent maxillary, supports the origin of the masseteric musculature. The area of attachment is an elongate surface, beginning at the anterior end of the glenoid fossa, which broadens in its anterior third before terminating behind the site of origin of the depressor snout muscles (PI. 1 , Fig. 3; PI. 2, Fig. 3). The point at which these two muscle masses meet (along and medial to the facial crest above or M^) is often marked by a pointed or spinous process. The sites of origin for both of these muscle groups face generally ventrally and are largely hidden in lateral view by the strong, thick zygomatic crest which projects somewhat ventrally. The site of origin for M. masse- ter lateralis profundus in particular is somewhat higher, relative to the level of the alveolar border (and to*that of the origin of M. buccinator) than in Tayassu. 10. In lateral view the facial crest follows the ventral edge of the jugal below the orbit and curves smoothly upward above infraorbital foramen toward the point of maximum divergence of the supra- orbital-nasal canals. In most specimens of Dicotyles the anterior curvature of the facial crest is slight. As mentioned above, the crest is nearly straight, but oblique, in dorsal view. Medial to the crest, the site of the origin of the depressor snout muscles is an anteriorly tapering ventrally facing concavity. 1 1. Within the orbit, the /acnm«/ is slender dorsal- ly, but broadens ventrally and does not enter the maxillary foramen. In facial aspect the lacrimal is slightly exposed below the lacrimal tubercle. 1 2. The maxillary lies on the lateral surface of the snout and face. Its contacts with other bones of the cranium are rather ordinary and will not be des- cribed further. On the lateral surface of the bone, the infraorbital foramen lies over P^ or, in some in- stances, Ml (PI. 1, Fig. 2; PI. 2, Fig. 2). In cross section the foramen is wide, and nearly circular. Below the foramen, the ridge indicating the dorsal extent' of M. buccinator diminishes in magnitude as it continues posteriorly. In ventral view, the edges of the palate are constricted conspicuously relative to the strong canine buttresses (PI. 1, Fig. 3; PI. 2, Fig. 3). The cheek teeth are slightly bowed outward, the greatest point of separation being between the inner edges of Mi. In overall trend, the cheek teeth also diverge progressively from front to back, the distance between the anterolingual alveoli of M^ usually being greater than that for the lingual alveoli of P2. 1 3. Even in young specimens, themaxillo-palatine suture is not visible. Otherwise, the slightly concave palatal surface is distinguished by the longitudinal median keel, mentioned above, and by a number of small nutrient foramina which dot the bone just medial to cheek tooth alveoli. Posterior to M^, a small, semicircular posterior palatine notch is incised between the maxillary tuberosity and the pterygoid ramus of the palatine bone. Between the last molars the surface of the palatine is nearly flat, being marked in some specimens only by a narrow median longi- tudinal groove which deepens posteriorly toward the choanal fossa. Posteriorly, the palatine overlaps the pterygoid which is inserted between the latter and the alisphenoid. In lateral view, the palatine, in contact with the alisphenoid, extends dorsally into the pterygopalatine fossa and then anteriorly, first in contact with the orbitosphenoid, and second, with the frontal, into the sphenopalatine foramen. 14. The pterygoid is a thin, ventrally tapering element which is inserted between the palatine and alisphenoid bones near the anterior end of the rather narrowly V-shaped choanal fossa. The pterygoid does not extend laterally into the pterygopalatine fossa. 1 5. The posterior contact of the alisphenoid with the frontal has already been discussed. Just below the infratemporal crest, the alispheno-frontal suture enters the large anterior lacterate-rotundum foramen which is situated just medial to the infratemporal crests. The smaller optic foramen is situated antero- dorsal to the anterior lacerate-rotundum foramen. The alispheno-orbitosphenoid suture leads anteriorly out of this foramen to contact the palatine about 30 mm. farther on and lies ventral and medial to a rather sharply developed linear eminence which leads posteriorly from the roof of the sphenopalatine foramen and fades out toward the optic foramen. The development of this eminence, the bony cover- ing of the ethmoidal sinus, is variable; the lateral surface is generally broadly ridgelike, but occasion- 20 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 ally it may be flat with a prominent but moderately broad, ventrally projecting crest. In either case, this eminence forms the dorsal border, and the flared ventrolateral edge of the alisphenoid forms the ventral border, of the elongate pterygopalatine fossa. This fossa contains the origin of M. ptery- goideus lateralis. M.pterygoideusmedialis originates along the flat ventral surface of the alisphenoid, anterior to the bulla. 16. Between the bullae, the basioccipito-basis- phenoid contact is not visible, but the basilar emi- nences are probably developed across this suture. The eminences (PI. 1, Fig. 3; PI. 2, Fig. 3) are a pair of prominent spindle shaped, longitudinally elongate structures which support the insertion of M. rectus capitus ventraljs major. 17. Posterolateral to the eminences and anterior to the condyles are the condyloid foramina. Lateral to these, the paroccipital processes, formed largely of the exoccipitals, project posteroventrally, tapering toward their tips. The latter are about on a level with the tips of the postglenoid processes, which lie slightly above that of the cheek teeth. 1 8. Dorsally, on the occipital surface, the contact between the exoccipitals and the supraoccipital is not determinable. Medial to the external auditory meatus and just below the tip of the postzygomatic crest, a small process is variably developed. It is associated with the insertion of M. rectus capitis dorsalis minor. The external auditory canal and meatus is apparently backed by the squamosal. The mastoid is not visible externally. Dorsally and medi- ally the flange-like edges of the lambdoidal and postzygomatic crests bound the occipital fossa which chiefly houses the insertion of M. rectus capitis dorsalis major. Mandible: 19. In general the profile of the mandible is long and low (PI. 5, Figs. 4, 5). The symphysis curves upward from below P2, then is procumbent anterior to the canine. The incisors extend forward from the anterior tip of the symphysis, the first and second pairs forming a slightly concave occlusal shelf ante- rior to the third. There is no diastema between I3 and the canine. The latter projects almost vertically from its alveolus and is slightly splayed laterally (PI. 6, Fig. 3). The diastemal crests behind the canines are narrow, but not sharp, and rise slightly toward the anterior edge of P2. Mental foramina lie below the diastemal crest, one or two below P2 and another about midway between P2 and the canine. Posterior to this the horizontal rami are narrow with steep lingual and labial surfaces. A slight convexity on the labial surface of the body of the ramus begins below Ml and rises posteriorly toward the temporal fossa of the coronoid process. Below the premolars, the ventral edge of the ramus is essentially straight, but the postdigastric sulcus forms a broad, shallow con- cavity between the anterior portion of the mandible and the angle. The peripheral border of the angle is broadly rounded; the condyle sits slightly anterior to the posterior border of the angle. 20. Anterior to the condyle, the temporal fossa occupies the labial surface of the coronoid process and is separated by a rather low ridge from the insertion area of M. masseter lateralis (PI. 5, Figs. 4, 5). As mentioned previously, this so-called tem- poral fossa may house, at least in part, fibers of M. masseter medialis. In dorsal view (PI. 6, Fig. 3) the condyle is somewhat tear-shaped with a pointed lingual tip and a broader labial extremity. The anterior edge of the ascending ramus forms a prog- gressively widening shallow sulcus toward the rear of M3; the sulcus houses the origin of M. depressor labii inferioris and (in part) M. buccinator, the fibers of the two muscles being inseparable at this point. The cheek teeth are aligned in a linear fashion, each tooth row diverging slightly from the other from front to back. The dorsal surface of the symphysis rises from the genial pits at an angle of about 30° ; in cross section the dorsal surface of the symphysis is narrowly concave. 2 1 . The paired genial pits are low, on the posterior surface of the symphysis, and lie just above a poorly developed, median, genial spine. On the lingual surface of the mandible, a narrow, flat, smooth to variably roughened line can often be followed poster- odorsally from the genial pits, above the digastric fossa to just below the rear of M3. This is the line of attachment of M. mylohyoideus. Just below this, and below M1-M3 the lingual surface of the mandible is excavated by the concave, anteriorly tapering, digastric fossa. Posterior to the latter, the usually flat, but variably roughened lingual surface of the angle contains the insertion of the pterygoideus muscles. Above this, the mandibular foramen sepa- rates the insertion areas of the pterygoideus muscles (below) and the temporalis muscles (above). 22. In ventral view (PI. 6, Fig. 1) the horizontal rami diverge from the symphysis at an angle of about 15°. The anteroventral tip of each angle is inflected medially out of line with the overall trend of the ramus, although the degree of inflection is variable. The ventral border of the postdigastric sulcus is narrow, but not sharp. Anteriorly, below the pre- molars, the essentially flat ventral surface slants slightly dorsolaterally. White Lipped Peccary, Tayassu pecari Fischer The numbering of the paragraphs in this section corresponds with that in the preceding section. Features not explicitly mentioned below can be 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU 21 considered to generally resemble their counterparts in Dicotyles. In the following description, points of difference between the two forms are emphasized. Cranium: 1 . As in Dicotyles the greatest width of the crani- um is measured (Table 8) at the anterior edge of the glenoid fossa, but the zygomatic arches, although essentially straight, taper anteriorly more conspic- uously (PI. 3, Figs. 1,3) than in the collared peccary. In dorsal view (PI. 3, Fig. 1 ), the ventrolateral edge of the maxillary projects laterally beyond the facial crest in Tayassu. 2. Instead of being excavated, the rostrum expands laterally, partially concealing the mass of the canine buttress which is, in addition, less robust than in Dicotyles. Along the undersurface of this linear swelling, which probably conforms to an expanded maxillary sinus, the attachment area of M. buccina- tor is developed as an elongate, ventrally facing concavity which tapers posteriorly to P'* or (PI. 3, Figs. 2, 3). The dorsal surface of the rostrum is generally flatter in Tayassu than in Dicotyles. The supraorbital foramina are situated similarly in the two genera, but in Tayassu, the canals almost always diverge immediately upon leaving the foramina; occasional convergence (PI. 3, Fig. 1) is of lesser magnitude and occurs nearer the foramina than in Dicotyles. The anterior extremities of the supraor- bital-nasal canals are not as well defined as in the collared peccary, and the posterior profile of the narial notch (PI. 3, Fig. 2) is conspicuously more acuminate in Tayassu. The canine buttress is not only less robust than in Dicotyles, but its vertical dimension is less. 3. In ventral view (PI. 3, Fig. 3), the incisor pairs are arrayed in a V-shaped configuration as in Dico- tyles, but the diastema between the incisors and canines is about 20 mm. The incisive foramina are relatively smaller than in Dicotyles, their combined width being only about one-third the width of the rostrum at F. Posteriorly, the surface of the palate is smooth and flat although it may bear a series of obliquely transverse grooves and ridges which reflect corrugations in the soft palate. The narrow palatine grooves emerge opposite the rear of the canines. A median longitudinal keel is not present. 4. Sutures are often visible in individuals with a complete but little worn dentition. The premaxil- laries (PI. 4, Figs. 3, 4) tend to be slender and more evenly acuminate posteriorly in Tayassu; the width of the nasals is not restricted between the premaxil- laries as seen in Dicotyles. The combined line of contact between the nasals and the premaxillaries is straight. It diverges posteriorly only before contac- ting the frontal. The premaxillo-maxillary contact on the palatal surface is as in Dicotyles. 5. In dorsal view (PI. 4, Fig. 3) the contact between the frontal and parietal elements in Tayassu is much like the condition in Dicotyles except that in the former genus the fronto-parietal suture reverses direction and projects farther anteriorly at the mid- line. The fronto-nasal contact resembles that of Dicotyles, but the short maxillo-frontal suture seems to be directed more vertically in Tayassu. Whereas the maxillo-jugal suture in Dicotyles curves gently posterodorsally behind the facial crest to contact the frontal, the same suture in Tayassu is sharply angulate (PI. 4, Fig. 4). 6, 7 and 8. The description of Dicotyles given in these paragraphs would not have to be significantly altered to apply to Tayassu. 9. In contrast to Dicotyles, the jugal is not exca- vated laterally and does not flare outward below the orbit in Tayassu. The differences in the maxillo- jugal suture in the two genera have been noted above. The site of origin of M. masseter lateralis super- ficialis and profundus faces ventrolaterally in Tayas- su; the zygomatic crest is not broad and does not project ventrally as strongly as in Dicotyles. The origin of the depressor snout muscles occurs mark- edly dorsal to that for the masseteric muscles in Tayassu and lies at a level conspicuously higher than of the alveolar border. 10. The facial crest is not smoothly continuous with the crest of the zygomatic arch, but diverges sharply upward (PI. 3, Fig. 2; PI. 4, Fig. 4). The profile of the cranium in dorsal view is usually more sharply constricted anterior to the zygomatic arch in Tayassu than in Dicotyles. The area of origin of the depressor snout muscles in the former genus is oriented anterodorsally; it is broadly triangular and faces nearly directly anteriorly in sharp contrast to the condition in Dicotyles. 1 1 . Tayassu and Dicotyles are generally similar in the features described in this paragraph. 12. On the lateral surface of the maxillary, the infraorbital foramen lies above P"* or M^. The cross section of the foramen is narrow and slit-like. The configuration of the attachment area for M . buccina- tor has been described in paragraph 2. Tayassu also differs from Dicotyles in that the palate is not sharply constricted in the diastemal region. Furthermore, while the cheek teeth diverge slightly posteriorly, the tooth rows are not as bowed outward as in Dicotyles, and are set closer together relative to the width of the cranium. 13 and 14. The palatine and pterygoid bones of Taya.ssu are much like those in Dicotyles except that a broad, moderately deep ovoid depression occurs on the palatal surface between the last molars and extends posteriorly toward the anteroventral edge of the choanal fossa. MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 22 15. Many of the features of this paragraph are similar in the two genera. The orbitosphenoid, how- ever, has an elongate slender neck, in Tayassu which extends into the anterior portion of the pterygopala- tine fossa below the crest which forms the dorsal border of the latter. In Dicotyles this crest is slightly more ventral in position and the posterior portion of the orbitosphenoid is larger and broader. In Tayassu the attachment area for M. pterygoideus medialis is broader and flatter than in Dicotyles. 16. Between the bullae, the basioccipito-basis- phenoid contact is a nearly straight transverse line. The basilar eminences (PI. 3, Fig. 3) are not as strongly developed, not as linear and not as readily differentiated from each other as in Dicotyles. 17. The condyloid foramina are similar to those of Dicotyles. 1 8. The exoccipito-supraoccipital suture emerges from the anterolateral corner of the foramen mag- num and continues dorsolaterally to reach the squa- mosal just ventral to the squamoso-parietal suture. The small process which may be developed in asso- ciation with M. rectus capitis dorsalis minor in Dicotyles has not been observed to occur in Tayassu. The occipital fossa, bounded by the lambdoidal and postzygomatic crests is not as broad as in Dicotyles. Mandible: 19. The profile of the mandible in Tayassu (PI. 5, Fig. 6) is generally longer and shallower below P2 than in Dicotyles. The profile of the symphysis is not as strongly curved upward, the dorsal and ventral borders are usually more nearly parallel, the con- cavity below the canine is not as pronounced and the first two pairs of incisors emerge from the symphysis at a somewhat steeper angle than in Dicotyles. The diastemal crests tend to be sharper in Tayassu, and below these, the mental foramina are slightly pos- terior to their position in Dicotyles. In Tayassu the horizontal rami are more massive and swollen, particularly just below the alveolar border on either side of the mandible. The ventral profile of the mandible in Tayassu is not as straight below the premolars to the rear of M2 as in Dicotyles. 20. The features described in this paragraph are comparable in Tayassu and Dicotyles except that in the former the cheek tooth rows are set relatively closer together. 21. The major difference here between Tayassu and Dicotyles is that the ventral border of the man- dible tends to be deflected inward below the digastric fossa in the white lipped peccary. 22. The degree of divergence of the rami in Tayas- su is about the same as that in Dicotyles, but the angle of the mandible is not inflected (PI. 6, Fig. 2) in the former genus. The ventral border of the postdi- gastric sulcus thickens anteriorly in Tayassu in con- trast to the uniformly narrow edge of this sulcus in Dicotyles. Variation: Inspection of the more than fifty crania of Tayassu pecari in the Department of Mammalogy of the American Museum of Natural History indicates that this genus is subject to a kind and degree of osteological variation similar to that which was noted for Dicotyles tajacu. Each species shows little variation, other than general robustness, which can definitely be correlated with sex (compare male and female canine dimensions. Table 1 0). With respect to ontogeny, it is interesting to note that features of the dentition and various sutural relationships are the most useful means for separating immature crania (with only the deciduous dentition in place) of the two species. Only in crania of older individuals do the characteristic crests and ridges of the rostral and adjacent areas become expressed to a degree ap- proaching the adult condition, and from the time that the first or second permanent premolars are fully erupted, such characters are essentially com- pletely developed. Comparison of Tables 2 and 6 with Tables 8 and 12 shows that in 12 out of 16 dimensions, the range observed for the Tayassu sample overlaps that ob- served for Dicotyles. In three of the 1 2 instances (i.e., width between M^ alveoli, depth below P2 and the width between the alveoli of M3) the degree of over- lap is quite large. In two out of the twelve cases the degree of overlap is quite small (F to condyle length and Ii to condyle length). In the remaining four cases (least rostral width, height from condyles to nuchal crest, breadth between the postorbital pro- cesses, and depth from the coronoid process to the angle of the mandible) the observed ranges for each sample do not overlap. In most cranial and mandi- bular dimensions the mean calculated for the Di- cotyles sample is 10 to 20 per cent smaller than that calculated for Tayassu. Important exceptions to this are: the length of the upper post-canine diastema {Dicotyles is 30 per cent smaller), the least width of the rostrum {Dicotyles is 47 per cent smaller), and the length of the lower post-canine diastema {Di- cotyles is 27 per cent smaller). The mean depth of the mandible below P2 is 95 per cent of that below M3 in Dicotyles as compared with 89 per cent in Tayassu. While this difference is small it confirms the impression gained when the measurements were being made that the mandibular ramus is pro- portionally shallower anteriorly in Tayassu. 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU 23 DENTAL CHARACTERISTICS OF THE LIVING PECCARIES Dicotyles tajacii Upper Dentition Deciduous teeth (PI. 5, Fig. 3). As in the case of the permanent premolar dentition, molar cusp termin- ology is used for convenience, but not necessarily to indicate homology. The deciduous first incisors are spatulate, small versions of the permanent tooth, but compared with its successor, dE is simpler, and lacks the basal inner cingulum. Variation encountered with respect to these teeth chiefly involves degree of robustness. Also, in some specimens a slight increase in the size of dE relative to that of dP has been observed. The deciduous canine is an elongate, slightly compressed and barely recurved tooth which pro- trudes from its alveolus at a much shallower angle than does the nearly vertical permanent tooth. The deciduous tooth is unmarked by ridges or crests and tapers gradually apically. The deciduous second premolar is triangular in outline, the apex of which is anterior. A pair of labial cusps are the major features of the crown, the second, or metacone, being set somewhat postero- labial to the paracone. A subsidiary, conical proto- cone is incipiently developed immediately postero- lingual to the latter. Posterior to this is a lower, but more distinct, conule. A posterior cingulum traverses the heel and curves around the lingual side to reach the base of the protocone. Anterior to the latter, a cingulum extends to the narrow tip of the tooth. Infraspecific variation has been observed in the following features: the degree of triangularity of the basal outline; the presence or absence of an oblique crest extending anterolingually from the base of the metacone; the degree of prominence of the heel area lingual to the metacone; the position of the proto- cone relative to the paracone; the size of this and the following tooth relative to that of dP”*. The deciduous third premolar is more molariform than the preceding and has a well developed para- cone, metacone and hypocone. The protocone is still in a relatively incipient stage and lies along the lingual base of the paracone. The V-shaped trans- verse valley is unobstructed; the rather pyramidal cusps are not connected into lophs. The anterior and posterior cingula are relatively short; small labial and lingual cingula occur at the transverse valley. The posterior moiety is wider than the anterior moiety. In some specimens, the protocone is situated more posterolingually, a small conule occurs be- tween the protocone and metacone, and the posterior cingulum is continuous around to the anterior edge of the transverse valley on either side of the tooth. In contrast to the preceding tooth, dP‘‘ is com- pletely molariform and is a slightly smaller replica of ML In general, the cusps of the deciduous dentition were observed to be conical and rather bunodont, although in a number of instances teeth were noted to have cusps which were steeper and more pointed. Penuanent teeth: The first incisor (Pis. 1 , 2; Table 3) is a robust spatulate tooth with a smooth convex outer surface and concave inner surface, bordered by a basal cingulum. In unworn teeth, one or more weak crests extend from the cingulum to the apex along the mid-posterior surface. The second incisor is smaller than P, although the sizes of the two teeth are variable with respect to one another. Also, P has a convex outer surface and an inner basal cingulum, but, in contrast to P, the high shearing crest is lo- cated above the longitudinal midline of the tooth rather than being recurved lingually. In unworn teeth the apex of the crest may be posteriorly re- curved. The shearing crest of P was seen to be incip- iently divided into a larger anterior, and a smaller posterior, conule in only a few specimens. The canine (Pis. 1, 2) is slightly recurved, and extends nearly vertically from its alveolus. In cross section, the outline of the tooth is triangualr with the apex posterior. The flat anterior surface is em- phasized as wear progresses. A slightly recurved groove, flanked by low ridges, extends apically along the mid-labial surface; a similar configuration is usually found on the labial surface as well, although the degree to which these grooves and ridges are developed on either surface is variable. As shown in Table 4, the mean length and width of the canines of male individuals is slightly greater than for females. This sexual dimorphism, while not marked, appar- ently causes the rather high values for the coefficient of variation (V) when calculated for the whole sample. The second premolar (Pis. 1, 2, PI. 7, figs. 5-7 is subtriangular in outline with a narrow, but rounded anterior tip and a broad heel. A prominent paracone is separated by a narrow transverse valley from the smaller metacone. The labial cingulum is absent, but the anterior cingulum extends around to the lingual side. The posterior cingulum extends around to the posterolabial base of the metacone. A protocone of variable size lies posterolingual to the paracone and in some instances a subsidiary swelling of the cingu- lum occurs anterior to the protocone. A small conule may be developed between the metacone and the protocone, or the metacone may be reduced and incorporated into the anterior member of a pair of anteriorly concentric crests which traverse the heel 24 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 (PI. 2, Fig. 3). Other variations observed in in- clude its size relative to P^ and the degree to which the paracone and protocone are emphasized to the detriment of the elements of the heel. As shown in Table 5, the range of the dimensions of P^ overlaps that for P3. However, observations made on 80 specimens revealed that within each individual, P^ is definitely smaller than P^. Whether the width measurement is taken through the paracone or the metacone is dependent upon the position of the protocone relative to those cusps and the degree of development of the anterolingual cingulum. In 41 out of 52 cases, the greatest width of P^ was mea- sured through the metacone; in 1 1 cases the mea- surement was made through the paracone. The third premolar is more molariform, with the protocone being only slightly less well developed than the paracone and larger than the metacone. Occasionally, however, the bulk of the metacone equals that of the paracone (PI. 7, Fig. 6). A small cusp usually occurs lingual to the metacone; this should be designated as the metaconule because in some cases, a distinct hypocone is developed from the lingual cingulum. A small protoconule is usually present. An anterior cingulum is continuous from the base of the protocone to the labial side of the paracone and a posterior cingulum is similarly developed around to the labial base of the metacone. As in P2, the heel can be differentiated into a pair of parallel anteriorly concave lophs, the posterior of which curves around to the rear of the protocone, and in this case the metacone is reduced to little more than a slight swelling. The dimensions of all P^’s measured are summarized in Table 5. In one out of 80 cases was P^ found to be longer than the adja- cent P'*. In none of these individual specimens was P3 observed to be wider than P'*. The point of mea- surement of the greatest width of P^ is governed by the same variables as in P^. In 39 of 54 specimens the greatest width was measured through the para- cone; in 12 specimens through the metacone and in 3 specimens the measurement was the same for either alternative. The last premolar is essentially similar to, but larger than P^. The configuration of the cusps is the same, but the cusps are larger. As in P^, P'* is broader across the anterior moiety than across the posterior moiety although the degree of reduction of the latter is variable. Other variations are similar to those encountered in P^. Dimensions for P** are also given in Table 5. As for P^, P'* was found to be longer than the succeeding tooth of the series (M^) in only one out of the 80 specimens. On the other hand, in 24 of the 80 specimens, P'* was wider than the adjacent M’. The site at which the maximum width of P'* is measured varies chiefly with the prominence of the cusps and cingulum in the posterolingual quadrant of the tooth. Thus, in 45 out of 54 instances, the maximum width of the tooth was measured through the paracone, in 6 through the metacone and in 2 cases the widths for the two alternatives were equal. The first molar is subquadrangular in outline. All four major cusps are well developed, the paracone and metacone being slightly anterior to their lingual counterparts. A sinuous transverse valley separates the anterior moiety from the posterior moiety, and is somewhat interrupted, in labial profile, by the metaconule. Short basal crests from the antero-and posterolingual corners of the protocone meet sim- ilarly short crests from the paracone. Those con- tributed by the anterior base of each cusp unite in the low protoconule which is basally fused to the anterior cingulum. The latter extends from the protocone around to the anterolabial base of the paracone. Basal crests similar to those of the proto- cone are developed from the hypocone, the anterior of these being the metaconule. The metacone is simply conical. A posterior cingulum extends from the hypocone to the posterolabial base of the meta- cone. Labial and lingual cingula are present or absent. The second molar is larger than the first, has a strong labial and an incipient lingual cingulum, but is otherwise as ML In M^, the cingula are better developed and the tooth tapers more conspicuously posteriorly than in or M^. The second molar is usually the largest of the cheek teeth. The ranges for the dimensions of the upper molars are given in Table 5; some overlap between analo- gous dimensions of successive teeth can be observed. On an individual basis, however, was never observed (in 80 specimens) to be longer than while was longer than in only 1 1 cases. The length of equalled that for in only two of the 80 specimens. Regarding the molar widths, was wider than in only one out of 80 individuals, but M2 was wider than M^ in 72 cases. The width of the molars is usually greatest across the anterior moiety. The reverse was true in only one of 53 cases for M^ and was never observed for M^ or M^. The width of the posterior moiety equalled that of the anterior moiety only once in 50 cases for each of the upper molars. As far as other features are concerned, the upper molars of Dicotyles tajacu often show increased hyposodonty and sharpness of the cusps, and empha- sis and anterolabial elongation of the protoconule and metaconule. In a number of instances, labial cingula become more strongly developed and occa- sional increase in the breadth of the posterior cingu- lum has been noted. In a few specimens, M^ tapers conspicuously posteriorly (PI. 7, Fig. 5) and extreme 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU 25 examples of this have been observed in M^. An extra, anteriorly concave crest has been found on between the base of the metacone and the posterior cingulum. Because the lengths of the upper cheek teeth are subject to ontogenetic intertooth wear, the widths of the teeth would seem, from theoretical considerations, to be the most stable dimensions available for comparison with other samples. How- ever, inspection of Table 5 reveals that except for the coefficient of variation (V) for the width of any given member of the cheek tooth series is greater than that for the length. In general, the dimensional variability of the premolars is greater than for the molars. As would be expected, the length of has the greatest variability of the molar dimensions. Lower Dentition Deciduous teeth: The deciduous incisors are essen- tially small models of the mature teeth, with dla being the largest. The third incisor is considerably simpler than its permanent replacement. As in the upper dentition, the deciduous canine is a simple, apically tapering slightly recurved tooth which emerges from its alveolus at a shallower angle than does the permanent tooth. It is also splayed outward more than the latter. In dP2 a large main cuspid is located slightly anterior to the central position on the crown and is bounded anteriorly by a small conulid and poste- riorly by a simple talonid. Both the size of the ante- rior conulid and the complexity of the talonid are variable. A crest begins at the apex of the main cuspid, passes down along the posterolingual edge and curves labially across the base to join the talonid at the posterolabial corner of the tooth. In a few cases, incipient bifurcation of the main cuspid has been observed. In dPs, which is larger and more elongate than dP2, the central cuspid is transversely bifid with a lingual connection to the talonid. The latter is com- posed of two or more cuspids, aligned transversely. The anterior basal conulid of the tooth is better developed than in dP2 and is associated with one or more accessory conulids. In some individuals, the accessory conulids are stronger and the tooth tapers abruptly rather than gradually anterior to the central pair of cuspids. The three transverse pairs of cuspids of dP4 seem to be typical of peccaries, as well as of other artio- dactyls. The anterolingual cuspid is conical, but its labial counterpart bears a short basal posterolingual crest which terminates at the transverse valley between the first and second pair of cuspids. In some specimens, a crest leads anterolingually from the cuspid under discussion and terminates at a low, median anterior conulid. In the second, or central row of cuspids, the labial protoconid bears two short basal crests, one of which leads anterolingually, the other posterolingually. Each meets a corres- ponding crest from the metaconid. In the final row, the lingual cuspid, or entoconid, is conical, but the hypoconid bears antero- and posterolingually direc- ted crests, the latter being formed into the hypoconu- lid. A posterior cingulum extends between the hypoconid and entoconid. In dP4, individual varia- tion involves sharpening of the cuspids, accentua- tion of the accessory conulids and the abrupt, rather than gradual, anterior tapering of the tooth. Permanent teeth: The first two incisors (Pis. 5, 6) are procumbent, elongate teeth which combine to form a slightly concave spatulate surface. Posterior to this, I3 is a short low tooth with a slightly oblique sectorial crest. Differences in actual size are the main form of variation in the incisor dentition. The third incisor is often reduced and simple. Dimen- sions of I1-3 are presented in Table 3. The canine has a triangular occlusal outline with a concave posterior surface. The lingual surface is smooth but a weak, recurved groove, bounded ante- riorly by a low ridge, ascends the labial surface. Variation involves the actual size of the tooth and the degree to which the lateral grooves and ridges are developed. Compared with that of Tayassu, the canine of Dicotyles is relatively more hyposodont and considerably slenderer. The first premolar of the permanent series, P2 (Pis. 5, 6, 7, Figs. 1-4), is constructed essentially as its deciduous predecessor with the exception that labial and lingual crests on the posterior surface of the main cuspid are usually better developed. The main cuspid may or may not be twinned, and in the single condition, its cross sectional mass may be relatively great. In some individuals which are us- ually, but not always, members of the subspecies known as Dicotyles tajacu angulatus (PI. 7, Fig. 1), the premolar teeth are large and quite hypsodont with the cuspids of the talonid being greatly reduced relative to those of the trigonid. Considering D. tajacu as a whole, the talonid is composed of two or three conulids arranged in a transverse row; a median conulid usually lies in front of these and may develop into a transversely extended arm. Absence of this conulid results in a basin bounded anteriorly by the trigonid and posteriorly by the cuspids of the talonid. The basin may be open lin- gually, but it is usually closed off labially. In 81 instances, P2 was neither longer nor wider than P3 in the same individual although in the total range of the sample (Table 7) the dimensions for the two teeth do overlap. The width of the trigonid varies relative 26 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 to that of the talonid to the extent that in 1 1 of 54 cases the maximum width of P2 was measured across the trigonid. In three of these cases the two alterna- tive widths were equal. Except for being relatively broader, and having a more distinct separation of the labial and lingual components of the main cuspid and of the anterior basal conulids, P3 is much like dPa. Variations in P3, which are similar to those observed in the pre- ceding tooth, include relative expansion of the heel, proliferation of its cuspids (PI. 7, Fig. 3) as well as a considerable increase in the height of the main cuspid. In 81 observations, P3 was never seen to be longer than P4. The frequency with which the tri- gonid is wider than the talonid in P3 is about the same as for P2. Thus, in 55 observations, only 10 instances were noted in which the trigonid was wider and in two instances the trigonid and talonid were of equal width. The permanent P4 is, of course, strikingly differ- ent from dP4. The former is a submolariform tooth with components which may be identified as proto- conid, metaconid, hypoconid and entoconid. The cuspids of the anterior moiety are higher and more strongly developed than those of the posterior moiety. A small conulid usually lies athwart the bases of the protoconid and metaconid anteriorly, while posteriorly, each of these cuspids gives rise to a short basal longitudinal crest. These crests, in conjunction with a smaller pair contributed by the hypoconid and entoconid, obstruct the transverse valley. Between the hypoconid and entoconid a small hypoconulid (PI. 7, Figs. 2, 4) usually occupies the posterior midline of the tooth, but may come to lie behind the hypoconid. The crest which extends along the posterior surface of the protoconid may become enlarged to the detriment of its lingual counterpart; this is associated with an alteration of the cuspids of the talonid into a cross-shaped con- figuration. Another variation on this theme is the emphasis and transverse elongation of the posterior crest from the protoconid, the loss of its lingual partner, the reduction of the anterior crest from the entoconid and emphasis and anterolingual elonga- tion of that from the hypoconid (PI. 7, Fig. 4). A conulid may develop from the labial cingulum at the transverse valley; a crest-like continuation of the posterior cingulum usually leads from this conulid to the hypoconulid. The configuration just described is associated with an increase in the width of the talo- nid (PI. 7, Figs. 2, 3). In contrast to the upper denti- tion, P4 is only rarely longer or wider than the adjacent Mi, the incidence for this being, respective- ly, four and three cases out of 8 1 . In one additional instance, P4 was as long as Mj. The degree of molari- zation of P4 closely approaches that of Mi and in 5 1 out of 53 instances the width of the trigonid was observed to be greater than that of the talonid. In none of these instances were two alternative widths equal. The ranges of the dimensions of P4 in terms of the total sample are summarized in Table 7. In Ml the four major cuspids are well developed. Between those of the anterior moiety and the poste- rior moiety is the sinuous transverse valley. The tooth has a strongly bilobate occlusal outline because neither a labial nor a lingual cingulum is present at the ends of the valley. Anterior and posterior cingula are short. The protoconid gives rise to a short, an- terolingually directed, basal crest which ends at the anterior cingulum; the metaconid gives rise to a short posterolabially directed basal crest which terminates at the transverse valley. Directly posterior to this is a short crest from the base of the hypoconid. The entoconid is subconical with a slight labial basal expansion toward the hypoconid. A small, conical hypoconulid merges with the posterior cingulum, and is connected, to a variable degree, to the hypo- conid. The second and third molars are constructed as in Ml except that in M3 the hypoconulid is expanded posteriorly with a large terminal cuspid and a vari- able number of smaller conulids anterior to it. The heel of M3 is usually narrow, but it may be squared off and broad (pi. 7, Fig. 4). Usually, there is a single median conulid posterior to the hypoconid and entoconid; behind this are a pair of larger conulids each of which may be joined by crests to the anterior conulid (pi. 7, Fig. 2). In other cases two median conulids are aligned behind the “hypolophid” and in some instances, three conulids are transversely ar- rayed across the tip of the heel, being separated from the “hypolophid” by a single small conulid. The last molar is usually narrower than M2, but this is var- iable. Other variations in the configuration of the lower molars include minor additional complexities, the addition of labial cingula, the presence of a distinct conulid at the anterolingual base of the metaconid, and an increased hypsodonty of the cuspids. The ranges of the dimensions of the lower molars are presented in Table 7. Within individual specimens (81 cases) Mi was never longer nor wider than M2, and M2 was never longer than M3. How- ever, M2 was wider than M3 in 68 cases. The talonid of Ml is usually (50 of 58 cases) wider than the trigonid. In M2, however, the trigonid was wider than the talonid in only 28 of 57 cases. In one in- stance, the widths of the two parts of the tooth were equal. In M3 the trigonid is always wider than the talonid. As indicated in Table 7, the dimensional variability of the premolar series is somewhat greater than that for the molar series with the exception of the length of M3. 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU Tayassii pecari Upper Dentition Deciduous teeth: The deciduous upper dentition of Tayassu pecari is substantially like that of the more robust variants of Dicotles tajacu with the qualifi- cation that the premolars are absolutely larger and, in detail, more complex. In dP^ accessory conules are found along the posterior surface of the paracone and on the posterolingual half of the heel. In dP^ an accessory conule anterior to the protocone is well differentiated. Permanent teeth: The first incisor (PI. 3) is essen- tially like that of Dicotyles except for being larger and stouter. The second incisor of Tayassu resembles that of certain members of Dicotyles in which the tip of the tooth is recurved so that it lies over the lingual portion of the crown rather than over the labial half of the tooth (see Tables 9, 10 for dimen- sions). The second premolar (Pis. 3, 7, Fig. 8) of Tayassu generally resembles that of Dicotyles and is subject to about the same sorts of variation. In some speci- mens the anterior tip may be considerably narrower than the posterior portion, but in others the anterior end of the tooth is essentially blunt. In no case, however, was an increase in the height of the proto- cone and paracone observed to approach the relative elevation of these cusps in certain individuals of Dicotyles. The dimensions of the total sample for P2 are given in Table 1 1 . Individually, P^ is neither longer nor wider than P^ (50 observations). The third premolar is larger, generally more complexly wrinkled than in Dicotyles, but the gen- eral orientation of the cusps is about the same. In most cases the hypoconid is strongly developed from the lingual cingulum, but in some specimens it is reduced. The presence of a posterolabial cingular conule was often noted. See Table 1 1 for a summary of the dimensions of P^. In only one of 50 instances was P^ noted to be longer than the adjacent P"*; P3 was uniformly narrower than P"^. The fourth premolar is larger than the third and generally more molariform; the hypoconid is more uniformly developed. In detail, the enamel surface is more finely crenulate than in Dicotyles. A summary of the dimensions of P^ is given in Table 1 1. In contrast to Dicotyles a survey of 50 individuals revealed only one case where P'* was wider than M’ and none where P"* was longer. In one instance, the width of P'* equalled that of MT The upper molars of Tayassu are generally con- structed as in Dicotyles, but are larger and were often observed to have a strong continuous labial cingu- lum. In some cases, the metaconule was emphasized and elongated anterolabially. was observed to be extremely short in a few cases, a condition not usually found in Dicotyles. The dimensions of the upper molars are summarized in Table 1 1. In only one case out of 50 was M* found to be longer than M2. The width of M* was never seen to be greater than for M^. was longer than in only four of 48 cases, and wider in 43 cases; the two teeth were equally wide in only one instance. Lower Dentition Deciduous teeth :The deciduous incisors and canines of Tayassu are not conspicuously different from those of Dicotyles, other than in terms of size. Tayassu differs from Dicotyles in the more con- spicuously bifid main cuspid and broader, more complex heel of dP2. In similar fashion, the crown of dPs is more complex than in the collared peccary, and in addition, the tooth is broader and relatively shorter in Tayassu and the accessory conulids tend to be more distinctly developed. Other than in actual size and in the detailed complexity of the enamel surface, dP4 of Tayassu is not markedly different from that of Dicotyles. Permanent teeth: The permanent lower incisors (Pis. 5, 6) of Tayassu are small relative to the size of the mandible. As in the case of the upper canines, their lower counterparts are more robust, but rela- tively lower crowned than in Dicotyles. The dimen- sions of the lower incisors and canines are given in Tables 9 and 10. As in the case of its deciduous predecessor, the permanent P2 of Tayassu differs from comparable teeth in Dicotyles by the conspicuously bifid nature of its main cuspid (PI. 6, Fig. 5). Also, as seen in the deciduous dentition, P3 and P4 of Tayassu are less hypsodont than in Dicotyles. In particular, the anterior conulids and other accessory conulids of the premolars of Tayassu are less prominent than in Dicotyles. Otherwise, the permanent cheek tooth dentitions of both genera are substantially similar. As mentioned previously both upper and lower cheek tooth series are set relatively closer together in Tayassu than in Dicotyles. Dimensions of the lower cheek teeth of Tayassu are given in Table 13. In general, the sample shows less variability in tooth dimensions than does that of Dicotyles (Table 7); this is particularly evident in the case of the pre- molars. Overlap in dimensions occurs throughout the series when the total sample for Tayassu is con- sidered. Within individual specimens, however, the following relations apply. In 50 individuals P2 < P3, P3 < P4, P4 < Ml, and Mi < M2. In all cases, M2 was also shorter than M3, but in only 1 4 instances was M2 narrower than M3. In two cases the widths of the two teeth were equal. 28 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 Some of the more pertinent variations observed in the cheek teeth of Tayassu include: P2, non-bifurcate main cuspid, reduction in width relative to that of the molars; P3, reduction of the small conulid along the posterior midline of the main cuspid, increase in the width of the heel, development of two transverse lophids on the heel; P4, enlargement of the middle conulid on the heel to a longitudinally elongate lophid with concomitant reduction of the single conulid usually present between the trigonid and the rest of the talonid cuspids, increase in the width of the heel; M1-M3, increase in the relative width of the teeth and simplification of their enamel patterns by means of a decrease in the surface crenulations and commissures, posterior prolongation and extreme tapering of M3. In general the dentition of Tayassu is not as variable as that of Dicotyles. While the rather wide range of dental variation in Dicotyles hinders the development of a clear-cut statement of the differences between the former genus and that of Tayassu, it would seem that on the whole, P4 of the collared peccary is higher crowned relative to Mi than in the white lipped peccary, and, further, that in the former the height of the trigonid relative to the height of the talonid of P4 is greater than in Tayassu. In both genera P4 erupts after M2 and before M3. Another subtle, but generally useful, distinction between Dicotyles and Tayassu is the generally sharper, more steeply sided cusps of teeth in both the upper and lower dentitions of Dicotyles. In Tayassu, the sides of the paracone, metacone, protoconid and hypoconid in particular, usually slope gradually rather than abruptly from base to tip at a moderate, rather than steep, angle and the bases of these pairs of cusps tend to be more closely ap- pressed to each other at the transverse valley. The dental dimensions for the measured sample of Tayassu pecari usually overlap the ranges of the analogous dimensions for Dicotyles tajacu (compare Tables 3, 4, 5 and 7 with Tables 9, 10, 1 1 and 13). In some (e.g., width of T, width of P, length of Ii, and width of both the upper and lower canines) the degree of overlap is exceptionally large. Those dimensions in which the range of overlap is rather small include: length of P2, width of P^, width of P4, length of M*, width of M^, and width of M^. Dimensions in which no overlap is observed between the two samples are:length and width of Mi, length of M2, and width of M3. Comparison of the mean values of the dental dimensions indicates that the teeth of Dicotyles tajacu are usually between 10 and 20 per cent smaller than those of Tayassu pecari. Exceptions to this include: width of P^ (Dicotyles is 22 per cent smaller than Tayassu), width of M^ (Dicotyles is 22 per cent narrower), width of (Dicotyles is three per cent smaller), width of Mi (Dicotyles is 22 per cent narrower), width of M2 (Dicotyles is 22 per cent narrower), and width of M3 (Dicotyles is 24 per cent narrower). Thus, while many dental dimensions of the Tayassu sample might represent an allomorphic increase relative to Dicotyles, some, e.g., the dimensions of the canines, do not. SUMMARY OF THE SALIENT DIFFERENCES BETWEEN DICOTYLES AND TAYASSU Dicotyles tajacu 1. Cranium relatively low, wide and short. 2. Rostrum excavated above and anterior to in- fraorbital foramen; posterior border of canine buttress well defined. 3. Narial notch broad posteriorly. 4. Dorsal surface of rostrum generally narrow, convex. 5. Supraorbital-nasal canals converge markedly anteriorly, then diverge; canals well defined anteriorly. 6. Canine buttress large. 7. Buccinator fossa faces laterally, extends back to above M' or M^. 8. Combined width of incisive foramina nearly one-half of the total width of the rostrum at P. 9. Palatal surface is rugose with subsidiary palatal Tayassu pecari 1 . Cranium relatively high, narrow and elongate. 2. Rostrum expanded ventrolaterally anterior to infraorbital foramen; posterior border of canine buttress not as well defined. 3. Narial notch acuminate posteriorly. 4. Dorsal surface of rostrum generally broad, relatively flat. 5. Supraorbital-nasal canals do not converge mark- edly anteriorly; canals not well defined anterior- ly- 6. Canine buttress small. 7. Buccinator muscles do not originate in a fossa; attachment area faces ventrally below swollen maxillary sinus, extends back above P**. 8. Combined width of incisive foramina only about one-third of the total width of the rostrum at P. 9. Palatal surface is generally smooth; median keel 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU 29 foramina and a persistent median keel; palatal surface between the last molars is generally flat, or with only a slight, narrow, depression. 10. Width of the nasals is restricted between the premaxillaries. 1 1. Fronto-parietal suture does not project markedly anteriorly near midline. 12. Maxillo-frontal suture curves posteroventrally. 13. Maxillo-jugal contact curves posterodorsally. 14. Lateral surface of the jugal is excavated; the bone flares outward below the orbit. 15. Height of the zygomatic arch posterior to the postorbital process is considerably less than that anterior to the process. 16. Site of origin of M. masseter lateralis profundus on the ventral side of the jugal is moderately expanded anteriorly and faces ventrally. 17. Site of origin of depressor snout muscles, below the facial crest, is shallow and is usually only slightly above the level of the origin of the mas- seteric muscles. 18. Facial crest is smoothly continuous with the zygomatic crest. 19. Infraorbital foramen lies above or and has a subcircular cross section. 20. Pterygoid surface of the alisphenoid is relatively narrow. 2 1 . Basilar eminences are generally elongate, spindle shaped. 22. Diastema is short, less than the length from P2 to PL 23. Mandibular profile is sharply concave below the canines; the incisors project more anteriorly. 24. Postdigastric sulcus is elongate, lies below M2 to beyond M3. 25. Digastric fossa is relatively shallow. 26. Ascending ramus rises well behind M3. 27. Angle of the mandible is slightly inflected. 28. Tooth rows are relatively far apart; teeth are smaller, canines are slenderer and tend to be more elongate; cusps of the teeth are sharper; P4 is higher crowned than Mi. 29. Overall size smaller. is absent; palatal surface between the last molars bears an elongate, ovid moderately deep sulcus. 10. Width of the nasals is not restricted between the premaxillaries. 1 1 . Fronto-parietal suture projects markedly an- teriorly near midline. 12. Maxillo-frontal suture passes vertically. 13. Maxillo-jugal contact is angulate, shaped as an inverted L. 14. Lateral surface of the jugal is flat and not flared outward. 15. Height of the zygomatic arch posterior to the postorbital process is nearly the same as that anterior to the process. 16. Site of origin of M. masseter lateralis profundus on the ventral side of the jugal is broadly ex- panded anteriorly and faces ventrolaterally. 17. Site of origin of depressor snout muscles is a deep pocket which lies considerably dorsal to the origin of the masseteric muscles. 18. Facial crest diverges sharply upward from the zygomatic crest. 19. Infraorbital foramen lies above P"* or M^ and has a flat, slit-like cross section. 20. Pterygoid surface of the alisphenoid is broad. 21. Basilar eminences are poorly distinguishable from each other, and are not linear. 22. Diastema is long, equal to or slightly longer than the length from P^ to P^. 23. Mandibular profile below the canines is gener- ally smoothly convex; the incisors project more dorsally. 24. Postdigastric sulcus is shorter, lies below M3 only. 25. Digastric fossa is relatively deeper, with the ad- jacent ventral edge of the mandible produced inward. 26. Ascending ramus passes by the rear half of M3. 27. Angle of the mandible is not inflected. 28. Tooth rows are relatively closer together; teeth are larger, canines are stouter and less elongate; cusps of the teeth are blunter, broader; height of P4 is about equal to that of Mi. 29. Overall size larger. THE ANCESTRY OF DICOTYLES AND TAYASSU Although various inferences as to the habits of the North American fossil peccaries of Quaternary and late Tertiary age have been drawn from observations on the living species, it is intriguing that no close relative of, for example, Dicotyles tajacu, has yet been found in the late Cenozoic fossil record of this continent. The pioneer investigations in paleonto- logy by Leidy, Cope, Marsh and others resulted in the naming of a number of Tertiary “species” of Dicotyles, but these have since become relegated to 30 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 such genera as Cynorca, Mylohyiis, Platygonus and Prosthennops. Although Pleistocene remains prob- ably referable to one or the other of the Recent genera have been found in South America (Rus- coni, 1931), species names have not been formally applied. Dicotyles traunmillleri Spillman, 1949, from the late Pliocene of Peru may be a valid species, but it is known from a proximal fragment of a hum- erus and a worn out left upper third molar. At first glance, much of the fossil record seems to be of little use in deciphering the past history of the Recent genera. The primary problem to be investigated, then, is; What paleontological evidence can be brought to bear on the question as to whether or not the living peccaries constitute valid genera? As pointed out above, the major differences between the living genera of peccaries are found largely in their crania, the teeth being distinctive mainly in a subtle way. Furthermore, the occipital and temporal regions of the two forms do not differ appreciably although the post-dental portion of the cranium of Tayassu is shorter relative to its dental portion than in Dicotyles. In general, Dicotyles tcijacu can be said to be a relatively short and heavily snouted peccary in which the site of the origin for the buccinator muscles is well expressed laterally, the masseteric muscles originate from an elongate, but relatively narrow, ventrally facing surface of the jugal and the site of origin of the depressor snout muscles occurs anterior to, but at about the same level as that of M. masseter. On the other hand, Tayassu pecari may be roughly characterized as a moderately long and slender snouted peccary with a ventral rather than lateral orientation of the attach- ment area for M. buccinator, an elongate but broad, ventrolaterally facing site of origin for M. masseter which lies relatively well below that of the depressor snout muscles. Outside of the possibility, based on the increased hypsodonty of their premolar dentition, that some more northern races of Dicotyles are possibly less omnivorous than the individuals of Tayassu seem to be, it is difficult to correlate the morphological differences just summarized with any drastic behav- ioral differences. Whatever the mechanical and behavioral differences may be, however, it seems relatively certain that the above two suites of charac- ters reflect the derivation of Dicotyles and Tayassu along somewhat different adaptive zones. The fossil record indicates further that two such adaptive zones extend as discrete evolutionary fields as far back as early Pliocene time. In the following section the broader aspects of various late Tertiary tayassuid generic lineages will be discussed. A more detailed investigation of the.se lineages at the specific level will be the subject of future work. Because it seems likely that the content of such established generic names as Platygonus and Prosthennops will be altered as a result of ensuing work, mention of a particular species of, for exam- ple, Prosthennops, should not be construed as bind- ing on any other of the currently known species of the genus. In Prosthennops niobrarensis (early Pliocene or possibly late Miocene of Nebraska) we find an animal with a moderately long, slender snout, small canine buttresses and, more importantly, an elong- ate, flat ventrolaterally facing surface on the ventral aspect of the zygomatic arch similar to that in Tayas- su. The temporal region of the Nebraska fossil is relatively short and the post-canine diastema is about as long as the premolar dentition, as in Tayas- su. On the other hand, the maxillary sinus is not swollen, so that while a strong buccinator fossa is not developed, the presumed site of origin for the buc- cinator muscles faces laterally, as in Dicotyles. Further discussion of the other species of Prosthen- nops, some of which are obviously too specialized to be of interest here (P. crassigenis Gidley, 1904) or are known chiefly from dental evidence, is beyond the scope of this report. It is sufficient for present purposes to indicate that, broadly speaking, Pros- thennips niobrarensis is a suitable ancestral proto- type of Tayassu. Other peccaries with crania of the same basic character as that summarized for P. niobrarensis are found in Mylohyus nasiitus (Leidy, 1868) (see Lundelius, 1960) from the later Pleis- tocene of North America and in Platygonus ( Brasili- ochoerus) stenocephalus (Rusconi, 1931, pp. 160- 1 63, and Winge, 1 906, pi. 6) from the latest Pleisto- cene or earliest Recent of South America. The three species just mentioned not only share most of the features enumerated for P. niobrarensis but also have a broad depression just dorsal to the anterior end of the zygomatic arch and anterior to the orbit. It seems likely that this depression contained the origin of the depressor snout muscles. The depres- sion is shallow in P. niobrarensis but in both My- lohyus and Platyonus (B) stenocephalus it is backed by a distinct ridge. To judge from the illustrations in Lundelius (1960) this ridge is oriented at a moder- ately shallow angle and extends anterodorsally in Mylohus nasutus. In Platygonus ( B.) stenocephalus, however, the ridge is aligned nearly vertically, even more steeply than in Tayassu. Available information on the dentition of the South American genus is rather scanty. It appears that the type specimen was aberrant to the extent that a supernumary tooth is present between M' and on each side of the palate. Although the crowns of the teeth illustrated by Winge ( 1906, pi. 6) are rather worn (particularly the lowers) the dentition of Platygonus (B.) steno- 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU 31 cephalus seems to be less specialized than that of Mylohyiis, and thus more like the cheek teeth of Tayassii. In spite ot the various differences shown by the Brazilian genus, such as its larger size, greater elongation of the snout and the positioning of the glenoid fossa directly beneath the orbits, there can be little doubt that this animal descended from the same general prototype as produced Tayassu pecari. The latter appears to be more generalized in terms of its rather bunodont dentition, shorter snout, shorter post-canine diastema and more robust incisors than either Mylohyiis or the Brazilian genus (but note that the incisors of the latter are not known). On the other hand, Tayassu differs from these, and all other currently known genera of peccaries in the conspic- uous expansion of its maxillary sinus. The fact remains, however, that Tayassu, Mylohyiis and the Brazilian peccary seem to be more closely related to each other than to any other fossil representative, and that the ultimate ancestry of all three probably passed through a stage similar to Prosthennops niobrarensis. The determination of the paleontological asso- ciates of Dicotyles tajacu is not as straightforward as in the case of Tayassu. But, using the broad charac- terization of the collared peccary as set out at the beginning of this section, it can be seen that there is a close basic resemblance between the Recent genus and various North American species of Platygonus. In particular, the cranium which Skinner (1942) referred to Platygonus alemanii Duges, 1 887 shows a number of basic similarities to Dicotyles tajacu. These include the relatively deep, moderately short snout, the ventrally facing site of origin for M. masseter lateralis profundus, the lip-like lateral boundary of this attachment area, the straight, anteriorly directed facial crest and the well devel- oped horizontal bony ridge defining the dorsal limit of the buccinator fossa. Although it is rather short, the facial crest extends somewhat anteroventrally, in marked contrast to the sharply anterodorsal config- uration noted in Tayassu and Platygonus (Brasili- ochoenis) stenocephaliis. In a recent review Slaugh- ter (1966) suggests that the number of late Pleisto- cene species of Platygonus proposed in the literature may be reduced to Platygonus compressus Le Conte, 1 848, Platygonus vetus Leidy, 1 882, and Platygonus cumberlanciensis Gidley, 1920. P. compressus in- cludes Hyops depressifrons Le Conte, 1 848, Dicoty- les costatus Le Conte, 1 848, Euchoerus macrops Leidy, 1 853, PlatygonusleptorhiniisVYWWs.ion, 1 894, and probably P. alemanii Duges, 1887. Simpson (1949) indicated that Platygonus setiger Hay , 1920, and Platygonus francisi Hay, 1920, are also snyo- nyms of P. compressus and that Protochoerus pris- maticus Le Conte, 1 848, is a nomen vanum. Platygonus cumberlanciensis includes P. interme- dins Gidley, 1920, and is characterized as having a long, slender snout, relatively long postcanine dias- tema, broadly flaring zygomatic arches, and aglenoid fossa which is usually positioned below the level of the occlusal plane. Inasmuch as the morphological summary given above for P. alemanii could apply equally well to P. compressus (see Simpson, 1949, figs. 6, 7), the latter can be taken to represent peccaries which differ from P. cumberlandensis and resemble Dico- tyles tajacu having a shorter, deeper snout, less strongly flaring zygomatic arches, shorter postcanine diastemata, and a higher position of the glenoid fossa. In this regard, P. compressus may be said to represent a less specialized level of organization than does P. cumberlandensis. The type cranium of P. vetus unfortunately does not portray many of the features cited above, but to judge from the low diastema-cheek tooth index (54) it might fall into the short snouted group, being at least a collateral species relative to P. compressus. Slaughter (1966) cites the large size of the dentition of P. vetus as setting it apart from the other two species of the genus, noting that it might be more closely allied to P. cumberlandensis. My interpretation, based on meager information as to the probable construction of the cranium, differs from his, but resolution of this point is not critical to the present study. In view of the similarly low (ca. 40) diastema- cheek tooth index calculated length, post-canine diastema ^ length, cheek tooth series for Platygonus texanus Gidley, 1903, this early Pleistocene form may also be a member of the lin- eage leading toward P. compressus. The material preserved for Platygonus bicalcaratus Cope, 1893, from the early Pleistocene of Texas, is insufficient for evaluation in the present context. I have not seen the actual material of Platygonus pearcei Gazin, 1938, from the late Pliocene Hagerman fauna of Idaho, but to judge from the illustrations (Gazin, 1938, Fig. 2) this is a long-snouted species with widely flaring zygomatic arches and is probably a member of the lineage leading toward P. cumber- landensis. This preliminary survey suggests that the genus Platygonus contains at least two major lines of descent in North America; that leading toward P. cumberlandensis is traceable into the late Pliocene while the P. compressus stem may carry back as least as far as the early Pleistocene on present evi- dence. Of the two, the P. compressus group seems to be the more conservative and thus shows closest affinity with Dicotyles tajacu. The earliest currently known record for the direct ancestry of the latter 32 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 may be represented by D. traunmidleri from the late Pliocene of Peru. A South American branch of Platygomis is found in Platygonus (Parachoerus) carlesi which, along with a derivative subspecies, is known from deposits of middle to latest Pleistocene age (Rusconi, 1931, pp. 150-159, 231-238, Pis. 3-6, 16-18 Figs. 26, 3 1 , 32). From Rusconi’s description and figures, the cranium of Platygonus (P.) carlesi resembles that of Dicotyles in the following points: anterior constriction ofthe supraorbital-nasal canals, decrease in vertical height of the zygomatic arch posterior to the postglenoid process, the excavation of the lateral surface of the maxillary posterior to the canines and anterior to the zygomatic arch, the smooth continuation of the facial crest at a shallow angle anterodorsally from the zygomatic crest, the narrow, elongate, ventrally facing site of origin of the masseteric muscles, the slightly rather than strongly curved profile of the ventral edge of the zygomatic crest as seen in lateral view, the strong canine buttresses and the absence of an expanded maxillary sinus. In most of these points, the Argen- tine genus also resembles the North American spe- cies of Platygonus, but almost all the characters listed constitute differences when compared with Tayassu, Mylohyus, Platygonus (Brasiliochoerus) stenocephalus and Prosthennops niobrarensis. Con- tinued association of Brasiliochoerus stenocephalus with Platygonus, which is not only misleading but seems also to be inconsistent with the evidence summarized in this discussion, should be rectified at some future time. A more detailed consideration of the affinities of the various South American peccaries will be under- taken at a later date, but for the present it is sufficient to note that Platygonus (Parachoerus) carlesi, Dico- tyles tajacu and the North American species of Platygonus are more closely related to each other than either is to any of the other genera and species discussed herein. Although the crowns of the teeth of Platygonus (P.) carlesi are worn down, the denti- tion of a later subspecies, Platygonus (P.) carlesi wagneri (Rusconi, 1931, Pis. 16-18) seems to be somewhat less hypsodont and zygodont than in the North American species of Platygonus. In these terms, the dentition of the Argentine subspecies approaches the condition of some undescribed early Pliocene peccaries in the Frick Collection of the American Museum of Natural History. The cheek teeth of the latter specimens can be readily separated from the teeth of contemporaneous representatives of Prosthennops and, at the same time, are suitably constructed to serve, in a broad way, as the ancestral prototype of Dicotyles and the various species of Platygonus. Dentally, at least, Dicotyles seems to be a remnant of an earlier level of organization and, except for the strong facial crest, its cranium is more generalized than that of Platygonus. Cranial material of the pertinent specimens in the Frick Collection is unfor- tunately not preserved. The cranium of Dyseohyus fricki (late Miocene of California) is substantially like that of Prosthennops niobrarensis and is thus an unsuitable ancestor for Dicotyles unless substantial modifications, for which there is no current evi- dence, took place in the intervening steps. The other potential ancestor of Dicotyles, Cynorca (Miocene of various North American localities), is known mainly by its dentition which is somewhat more primitive, particularly in the degree of molarization of the premolars, than that of Dyseohyus. Discovery of cranial material of Cynorca, which is strategically placed in time and space, might cast new light on the origin of Dicotyles. The prevailing uncertainties notwithstanding, the proposal that Dicotyles tajacu and Tayassu pecari are members of independent, long-lived lineages seems to be corroborated by the available fossil evidence; morphologically and paleontologically, the separation of these two peccaries at the generic level seems to be not only eminently reasonable, but warranted by the facts and inferences presented herein. In summary, Cynorca and Dyseohyus are the only known genera which are suitable ancestors for the Pliocene, Pleistocene and Recent peccaries. The lineage which ultimately produced Tayassu pecari is broadly associated with Mylohyus and “Platy- gonus” (Brasiliochoerus) stenocephalus. The com- mon line of descent for these three peccaries can probably be traced through Prosthennops niobra- rensis and thence to Dyseohyus. The oldest known records of Tayassu are found in the late Pleistocene or early Recent of South America and direct fore- runners of this genus certainly have not yet come to light in Pliocene or Pleistocene deposits of North America. The probability that Tayassu underwent most of its evolution in Central and South America is relatively great. A similar statement can be made for Dicotyles, whose earliest known occurrence is in the late Plio- cene of Peru. The point in space and time when Dicotyles diverged from its phyletic associate, Platy- gonus, cannot yet be documented and inferences as to the ultimate origin of the Dicotyles — Platygonus branch of the peccary family tree cannot be as readily made as for Tayassu and its associates. Both the white lipped and collared peccaries possess a relatively large number of conservative characters and hence are less divergent, than the other more specialized members of their respective lineages, from the basic morphological level repre- 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU 33 sented by Dyseohyus and Cynorca. The living pec- caries have thus outlasted their more progressive relatives. On the other hand, the morphological proximity of Dicotyles and Tayassii to the ancestral fount of the later Tertiary and Quaternary Tayas- suidae has tended to obscure the fact that their adaptive separation has indeed been of a duration and magnitude which warrants generic distinction. CONCLUSIONS The nomenclature of the modern peccaries has been confused for many years. The following chronologic summary (Table 14) indicates the progression of the various names. Since the names of Frisch (1775) do not apply to zoological nomenclature by reason of their being non-Linnaean (see Hershkovitz, 1948) the earliest valid generic name for any peccary seems to be Tayassu Fischer, 1814. The genotypic species for this genus is clearly T. pecari, the white lipped peccary. All later names for the white lipped species are pre-empted by, and are synonymous with, this binomial and, according to the conclusions of the current study, the genus Tayassu is monotypic, as far as the Recent fauna is concerned. Fischer (1814) also proposed the name Tayassu patira for the collared peccary, but as indicated above, the genus Tayassu is not available for this usage and the species T. patira is predated by Sus tajacu Linnaeus, 1758. The next available name for the collared peccary is Dicotyles Cuvier, 1817. The usage of the name Dicotyles has had a complex history, most of the difficulty seeming to have resulted by the “designa- tion” by Gray (1868) of D. labiatus (white lipped peccary) as the genotypic species. Thus, Dicotyles came to be considered as synonymous with Tayassu Fischer, 1814, although the former name was held to be esthetically preferable by a number of workers and was utilized by many North American paleonto- logists and neontologists in the late 19th and early 20th centuries. At one time, a proposal was put before the International Commission on Zoological Nomenclature to have the rules suspended in favor of Dicotyles, but this was turned down (opinion 90; see Schenk and McMasters, 1948, p. 58). Now that the living peccaries are treated as separate genera, Dicotyles should again become available for evalua- tion. From Cuvier’s text (1817, pp. 237-238) D. torquatus is clearly the genotypic species of Dicoty- les and refers to the collared peccary. D. torquatus not only has page priority over D. labiatus (Cuvier’s name for the white lipped peccary), but is more thoroughly diagnosed than the latter and seems to have been central to Cuvier’s concept of the genus. There can be little doubt that Dicotyles torquatus refers to a different species of peccary than does Tayassu pecari. In light of the current study, the two species also represent distinct genera. Tayassu pecari Fischer, 1814, refers to the white lipped peccary. The collared peccary would be designated as Dicoty- les torquatus Cuvier, 1817, were it not for the fact that Cuvier’s specific name, but not his generic name is pre-dated by Tayassu patira Fischer, 1814, and Sus tajacu Linnaeus, 1758. The collared peccary must be known, therefore, as Dicotyles tajacu (Lin- naeus, 1758). NOTE ADDED IN PROOF During the time this paper was in page proof, Mr. William A. Low, Dept. Zoology, Univ. British Columbia, brought to my attention a paper by Hershkovitz (P., 1963. Proc. Biol. Soc. Wash., 76; 85-88) in which Link (D.H.F., 1795. Beitrage zur Naturgeschichte. Rostock und Leipzig, 2; 1-126) is given credit for Tayassu pecari, the name for the white lipped peccary. I have not been able to obtain a copy of Link’s article as yet and have not been able to make a decision on this point. Although this nomenclatural technicality should be pursued further, it alters neither the broader conclusions set out above nor the name of the collared peccary 34 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 Literature Cited the Texas Pleistocene. Bull. Texas Mem. Mils.. 1: 1-40. Cuvier, G. 1817. Le regne animal distribue d’apres son orga- nization. Tome 1. Paris, Masson, 540 p. Frisch, J. L. 1775. Das Natur-System der vierfussigen Thiere, in Tabellen, darinnen alle Ordnungen, Geschlechte und Arten, nicht nur mit bes- timmenden Benennungen, sondern bey- gesetzten unterscheidenden Kennzeichen angezeigt werden, zum Nutzen der erwach- senen Schuljugend. Glogau, 30 p. Fischer, G. 1814. Zoognosia tabulis synopticus illustrata. Pt. HI. Moscow, Nicolai S. Vsevolozsky, 732 p. 1817. Adversaria Zoologica. Mem. Soc. Imperial Nat. Moscow, 5: 357-446. Frenchkop, S. 1955. Sous-ordre des Suiformes. In Grasse, P. P., [ed.], Traite de Zoologie, 17: 509-535, Paris, Masson et Cie. Gazin, C. L. 1938. Fossil peccary remains from the upper Pliocene of Idaho. J. Washington Acad. Sci., 28: 41-49, 3 figs., 1 tab. GilUT. 1902. Notes on the names of the genera of pec- caries. Proc. Biol. Soc. Wash. 15: 38-39. Gidley, J. W. 1920. Pleistocene peccaries from the Cumberland Cave deposit. Proc. U.S. Natl. Mus., 57: 651-678. Gray, J. E. 1868. Synopsis of the species of pigs (Suidae) in the British Museum. Proc. Zool. Soc. Lon- don, p. 17-49. Hall, E. R., and K. R. Kelson 1959. The mammals of North America. The Ronald Press Co. New York, 1083 p. Hershkovitz, P. 1948. Names of mammals dated from Frisch, 1775, and Zimmerman, 1777. J. Mammal., 29: 272-277. Illiger, C. 1815'. Ueberblick der Saugthiere nach ihrer Ver- theilung fiber die Welttheile. Abhandl. Akad. Wiss., Berlin: 39-160. Linnaeus, C. 1758. Systema naturae. 10th ed., vol. 1. Stock- holm, 824 p. Lundelius, E. L. 1960. Mylohyus nasutu.s. Long-nosed peccary of Merriam, C. H. 1901. Description of four new peccaries from Mexico. Proc. Biol. Soc. Wash. 14: 1 19-124. Miller, G. S., Jr. 1924. List of North American Recent Mammals. U.S. Natl. Mus., Bull. 128: 1-673. Miller, G. S., Jr., and R. Kellogg 1955. List of North American Recent Mammals. U.S. Natl. Mus., Bull. 205: 1-954. Reichenbach, A. B. 1835. Bildergalerie der Thierwelt. Heft 6. E. Ponicke and Sohn. Leipzig. Rusconi, C. 1929. Anatomia craneodental de los Tayassuinos vivientes (Pecaries). Anales Cientifica Ar- gentina, 107: 66-138. 1931. Las especies fosiles Argentinas de pecaries (“Tayassuidae”) y sus relaciones con las del Brasil y Norte America. Anales Museo Nac. hist. nat. “Bernardino Rivadavia,” 36: 121-241. Schenk, E. T. andJ.A. McMasters 1948. Procedure in Taxonorny. Stanford Univ. Press, Stanford, 72 p. Simpson, G. G. 1949. A fossil deposit in a cave in St. Louis. Amer. Mus. Novitates, no. 1408: 1-46. Sisson, S., and J. D. Grossman 1953. The anatomy of the domestic animals. 4th ed. W. B. Saunders Company, Philadelphia and London, 972 p. Skinner, Morris F. 1942. The fauna of Papago Springs Cave. Bull. Amer. Mus. Nat. Hist., 80: 143-220. Slaughter, B. H. 1966. Platygonus compressus and associated fauna from the Laubach Cave of Texas. Amer. Midi. Nat., 75: 475-494. Spillman, F. 1949. Contribution a la paleontologia del Peru. Una mamifauna fosil de la region del rio Ucayali. Publicationes del Museo Hist. Nat. “Javier Prado,” ser. C, 1: 1-40. Tate, G. H. H. 1939. The mammals of the Guiana Region. Amer. Mus. Nat. Hist. Bull., 76: 151-229. Winge, Herluf 1906. Jordfunde og nulevende Hovdyr {Vngulata) fra Lagoa Santa, Minas Geraes, Brasilien. Copenhagen, Museo Lundii, 3: 1-239. 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU 35 PLATE 1 Dicotyles tajacii torvus 3. A.M.N.H. (M.) No. 92838, adult cranium. 1. Dorsal view. 2. Lateral view. 3. Ventral view. Approximately X Vz 36 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 PLATE 2 Dicotyles tajacu angiilatus 9. A.M.N.H. (M.) No. HIM, adult cranium. 1. Dorsal view. 2. Lateral view. 3. Ventral view. Approximately X V2 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OL DICOTYLES TAJACU 37 PLATE 3 Tayassu pecari heehei A.M.N.H. (M.) No. 142981, adult cranium. 1. Dorsal view. 2. Lateral view. 3. Ventral view. X Vi J8 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 Fn.ATE 4 1, 2. Dicotyles tajacu torvus A.M.N.H. (M.) No. 73598, immature cranium with dl‘-2, dC‘, dP^-^, ME 1. Dorsal view. 2. Lateral view. 3, 4. Tayassu pecari heehei -. A.M.N.H. (M.) No. 140534, subadult cranium with canine and P2-M2. 3. Dorsal view. 4. Lateral view. X '/a 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OF DICOTYLES TAJACU 39 PLATE 5 1, 2, 4. Dicotyles tajacu angiilcitus. 1, 2. A.M.N.H. (M.) No. 23868. 1. Dorsal view of immature cranium. 2. Lateral view of immature cranium. 4. A.M.N.H. (M.) No. 22674 9. Right mandible. Lateral view. 3, 5. Dicotyles tajacu torvus. 3. A.M.N.H. (M.) No. 73598 -. Ventral view of immature cranium with dP-^, dC*, dP^-^, ML 5. A.M.N.H. (M.) No. 92838 S. Adult mandible, lateral view. 6. Tayassu pecari heehei -. A.M.N.H. (M.) No. 142981. Adult man- dible, lateral view. All X V3. 40 MEMOIR OF THE SOUTHERN CALIFORNIA ACADEMY OF SCIENCES VOL. 7 PLATE 6 I. Dicotyles tajacn anfinlatiis 9. A.M.N.H. (M.) No. 22674. 2. 4, 5. Tayassu pecari heehei -. A.M.N.H. (M.) No. 142981. 3. Dicotvies tajacn towns d'. A.M.N.H. (M.) No. 92838. I, 2. Adult mandibles. Ventral view. 3, 4. Adult mandibles. Occlusal view. 5. Right lower cheek teeth. Occlusal view. 1-4, X '/2. 5, X I. 1968 THE CRANIAL MYOLOGY AND OSTEOLOGY OE DICOTYLES TAJACU 41 PLATE 7 \. Dicotyles tajacn angidatiis 9. A.M.N.H. (M.) No. 22764. 2-7. Dicotyles tajacii torvus. 2. A.M.N.H. (M.) No. 92838