AMERICAN MUSEUM Novitates PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY CENTRAL PARK WEST AT 79TH STREET, NEW YORK, N.Y. 10024 Number 3023, 34 pp., 16 figures September 11, 1991 Evolution of the Aeluroid Carnivora: Viverrid Affinities of the Miocene Carnivoran Herpestides ROBERT M. HUNT, JR.! CONTENTS ACGStrae bgt, eR SE or Be, pen aber PRR no, BUM Wie toe apthieeo IR tee ts hs ceamraguatecy 2 Introduction ........... 0.2: e ee eee ees RN A a, A ASA Tiel Ree re en 2 Acknowledgments c1.)..0 lve soa oa 2 Sealer ade yates HEN ee oe ee PE bare see 2 Abbreviations © ..6 48. file Ren dca ae eee a Fi cbt tes Leuslarce gin atte tee Gee 4 Geologic and Geographic Setting .............. 0. cece eee ee eee eee eee nen es 4 Basicrania of Herpestides (2: ect c ce eo aw 6 04 8 ee wh als le SA Oe he TN ae aly 5 Auditory Region of Herpestides antiquus ......... 000i e een e ene ees 6 Comparative Morphology of the Aeluroid Middle Ear ........... 0.0... cc cece eee 12 Auditory Ossicles in Aeluroids ............ 0.0 ce cece cee n eee eee ne ne eens 21 Phylogenetic Affinities of Herpestides .......00 0 ccc cence ence een eans 26 The Problem of Hyaenid-Herpestid Relationship .............. 0.0... c cece eee eee 30 Antiquity of the Viverrid Auditory Region ............. 0... ccc ccc cee eet eens 32 CONCHISIONS: Hof Be hy SABA hh tc tetilanout dae: Reale ks twsenceiven go bueschake wie Dey et ete gaa ial aet ete 8 32 FRE TET ET COS ee he eos Se hte Re EUR chan rn dat, Dare | PAE a hot Oe eee es ue he ai WSs eS aahy 33 ' Research Associate, Department of Vertebrate Paleontology, American Museum of Natural History; Curator, Division of Vertebrate Paleontology, University of Nebraska, Lincoln, NE 68588-0514. Copyright © American Museum of Natural History 1991 ISSN 0003-0082 / Price $4.30 2 AMERICAN MUSEUM NOVITATES NO. 3023 ABSTRACT Although the time of origin of viverrid and hyaenid carnivorans has not been clearly docu- mented in the fossil record, their theater of evo- lution has long been established by a mid-Ceno- zoic fossil distribution entirely confined to the Old World. Recent examination of the basicranial morphology of important early aeluroid crania from Europe and Asia significantly alters earlier views of viverrid and hyaenid origins. The early Miocene carnivoran Herpestides antiquus, consid- ered a potential ancestral hyaenid or herpestid in earlier studies, is identified as a true viverrid on the basis of a large sample of skulls of both ju- veniles and adults from Aquitanian sediments of the Allier basin, France. The basicranial mor- phology of Herpestides has attained the modern viverrid grade of development in the early Mio- cene (European Neogene mammal zone MN 2a), and suggests that diversification of the Viverridae was in progress by this time. In ongoing work to be published elsewhere, the mid-Miocene Asian carnivoran Tungurictis spocki, long regarded as a viverrid, is identified as an early hyaenid, follow- ing preparation and restudy of the auditory region of the genoholotype cranium from Tung Gur, Mongolia. These discoveries indicate that sepa- ration of the modern aeluroid families as discrete lineages had been accomplished by the beginning of the Neogene in the Old World, and that diver- sification within these families must have been initiated in the early to mid-Miocene. INTRODUCTION Among the superb carnivoran fossils re- covered from the Aquitanian deposits of the Allier basin, France, are abundant cranial and postcranial remains of a small civetlike ae- luroid, originally described by Blainville (1842) under the name “Viverra” antiqua. As additional remains were discovered, they were placed by later European students (Po- mel, 1853; Filhol, 1879; Schlosser, 1890; Vi- ret, 1929) in several species allocated to the extant genus Herpestes. However, in a thor- ough, recent review of this material, Beau- mont (1967) affirmed its distinctness from Herpestes and created the genus Herpestides for this Aquitanian carnivoran lineage, rec- ognizing only a single species highly variable in size and dental morphology, Herpestides antiquus (Blainville). Herpestides is the oldest Eurasian or Af- rican aeluroid carnivoran represented by nu- merous skulls with intact basicrania, includ- ing auditory bullae (fig. 1). A number of these fossil skulls retain osseous basicranial struc- ture as well preserved as in the living animal, and both juveniles and adults are represent- ed. My intent was to examine an adequate sample of the crania of Herpestides in order to determine its basicranial and bullar mor- phology; to discuss the ontogenetic pattern of bulla formation; and to attempt to estab- lish the higher-level systematic relationships at the family level. In September 1989, I was given the opportunity to study the sample of complete and partial crania of Herpestides at the Musée Guimet d’Histoire Naturelle in Lyon, and at the Naturhistorisches Museum, Basel, representing about 12 individuals, which forms the basis for this report. I did not examine the question of the num- ber of species attributable to Herpestides from the St.-Gérand region. It is sufficient to note that Beaumont (1967), who has studied the nature of dental variation in this carnivoran, determined that dental and cranial materials Suggest a range of variation somewhat greater than that found in most living aeluroid spe- cies (fig. 2). The amount of variation in dental and cranial dimensions, however, is in keep- ing with geographically variable populations of a lineage sampled over a short interval of time, as pointed out by Beaumont. Signifi- cantly, the degree of variation observed in the dentition, both in terms of morphology and size, is not evident in the basicranial morphology, which displays a high degree of uniformity. Thus, regardless of the number of species of Herpestides finally determined from the Aquitanian sediments of the Allier basin, the fossils indicate a morphologically uniform, closely related assemblage referable to a single genus. ACKNOWLEDGMENTS I wish to express my appreciation to Dr. Michel Philippe for the numerous courtesies 1991 HUNT: VIVERRID AFFINITIES OF HERPESTIDES 3 Fig. 1. Cranium and associated lower jaw of the viverrid Herpestides antiquus (Blainville), MGL St.- G. 3066, from the Aquitanian, Allier basin, France. Natural size. Fig. 2. Rostra of a large (left, NMB 6373) and a small (right, NMB 12379) individual of Herpestides from Montaigu-le-Blin, Allier basin, France, demonstrating the range in size attributed to H. antiquus by earlier workers. On dental traits, both are adults. Scale bar in this and all subsequent figures is 1 cm in length. 4 AMERICAN MUSEUM NOVITATES NO. 3023 extended me during my research atthe Musée ™ depression in mastoid produced by cau- Guimet d’Histoire Naturelle in Lyon, and to dal entotympanic Drs. Pierre Mein and Marguerite Hugueney a et rae for their assistance in the collections of the A Seti witgor petrosal Faculté des Sciences, Universite de Lyon.I pg groove in entotympanic for internal ca- also wish to thank M. Hugueney for recent rotid artery, leading to posterior carotid publications and discussion of the geologic foramen setting of the St.-Gérand deposits. My work __ pif posterior lacerate foramen at the Naturhistorisches Museum, Basel, Pp, pp _ paroccipital process of exoccipital benefited from conversations with Dr. Burk- PT pterygoid art Engesser and Prof. Dr. Johannes Hiirzeler R rostral entotympanic that improved my comprehension of issues basioccipital attachment for rectus capi- relevant to this study. M. Francois Escuillie, ls ventralis f Dépt. des Sciences de la Terre, Lyon, gen- °% suprameatal fossa ‘ sb septum bullae erously allowed me to examine two recently SQ Equatiiosal collected skulls of Herpestides from the lo- = ectotympanic calities of Montaigu-le-Blin and Gepiac which t fossa for the tensor tympani he will describe in his work on these carni- _— th tympanohyal vorans. I am particularly grateful to Dr. Guy V ventral process of the petrosal promon- Musser and Dr. Patricia Freeman who pro- torium vided the opportunity to study and dissect vf Vidian foramen ; the basicrania of representative aeluroid car- * apex of the caudal entotympanic fused nivorans in collections in their care. Refer- to the posterior edge of rostral entotym- ence casts of basicrania used in this study eae were produced by Head Preparator Gregory Brown, University of Nebraska State Muse- AMNH American Museum of Natural History, um. My thanks to R. M. Joeckel, Harold Bry- Mammalogy _ aS. ant, and Cliff Lemen for review of the manu- FSL Faculte des Sciences, Université de script. Lyon, France ne MGL Musée Guimet d’Histoire Naturelle, ABBREVIATIONS Lyon, France NMB Naturhistorisches Museum, Basel, A alisphenoid Switzerland a epitympanic wing of alisphenoid UNSM _ University of Nebraska State Museum, ac alisphenoid canal Zoology b epitympanic wing of basisphenoid BO basioccipital bof basioccipital flange GEOLOGIC AND GEOGRAP HIC BS basisphenoid SETTING os Oe a pee bie eae oF ee The remains of Herpestides discussed here- totympanic in were found in nonmarine early Miocene d depression in alisphenoid for anterior (Aquitanian) sediments of the St.-Gérand Te- limb of ectotympanic gion, Limagne d’Allier, central France. The EO exoccipital Limagnes are Cenozoic fault-bounded basins E caudal entotympanic (“fosses d’effondrement’’) within the Massif € epitympanic recess _ Central of France that developed contem- a nae aaa ap as exoccipital poraneously with Alpine tectonism. The ange Of septum Dullae contacting pro- —_—- | imagne d’Allier that contains the fossilif- montorium anterior to round window 4 : ; : fo fsramienvownle erous Aquitanian sediments of the St.-Gé- g postglenoid foramen rand sub-basin is a complex graben fill of late h hypoglossal (condyloid) foramen Eocene to early Miocene age, dissected by the ica tube for internal carotid artery modern Allier River flowing northward from L middle lacerate foramen the Massif Central. Aquitanian sediments in M mastoid process the vicinity of St.-Gérand-le-Puy have been 1991 commercially exploited for over a century for limestone, resulting in a network of quarries that produce the vertebrate fauna. The dis- tribution of the principal localities in the re- gion has recently been illustrated by Bucher, Ginsburg, and Cheneval (1985), and a map of the quarries in the St.-Gérand sub-basin near the villages of Montaigu-le-Blin, St.-Gé- rand, and Langy appears in Cheneval and Hugueney (1985). Fossils of Herpestides that I have studied in the Museum at Lyon are simply attributed to “St.-Gérand,” and lack a more specific lo- cality designation. As is well known, fossils designated by the term ‘“‘St.-Gérand”’ could have been found in any of the limestone quar- ries of the region (Viret, 1929; Cheneval and Hugueney, 1985). However, the crania from the Naturhistorisches Museum, Basel, are all attributed to Montaigu-le-Blin, with the year of collection appended. Today several active as well as inactive quarries occur near the village of Montaigu-le-Blin (Cheneval and Hugueney, 1985), but it is uncertain at which of these sites the crania were found. Recent studies of the Aquitanian sediments near Montaigu-le-Blin allow an improved under- standing of their depositional environment (Bucher et al., 1985; Hugueney, 1984; Don- simoni and Giot, 1977). The depositional setting of the vertebrate remains are described in a recent study of “Les Pérards”’ quarry near Montaigu-le-Blin by Bucher, Ginsburg, and Cheneval (1985), and my subsequent remarks are based on their illuminating publication and an excursion guide by Cheneval and Hugueney (1985). De- positional environments are primarily flu- vio-lacustrine, and record the waxing and waning of local carbonate-rich lakes within the Allier basin. Terrestrial vertebrates (in- cluding viverrid carnivorans) occur in green marls with abundant shells of the pulmonate gastropod Cepaea moroguesi deposited be- tween columnar lacustrine stromatolites (phryganid algal bioherms). These green marls lack internal stratification and have been in- terpreted as mudflows that swept littoral de- bris (including vertebrate bones) downslope onto the floor of these lakes, surrounding and influencing the growth pattern of the stro- matolite community. Vertebrates within the green marls include both aquatic and terres- HUNT: VIVERRID AFFINITIES OF HERPESTIDES 5 trial reptiles and birds; there are abundant remains of diverse terrestrial mammals (di- delphid marsupials; insectivores; bats, lago- morphs and rodents; viverrid, felid, and mustelid carnivorans; small artiodactyls). Of particular interest is the occurrence of vertebrate bones in brown marls with hy- drobids. These brown marls are similar to the green marls in that they also infiltrate between stromalitic mounds, but in the brown marls the associated vertebrates are primarily aquatic (freshwater fishes and aquatic birds); mammals are represented only by abundant remains of viverrid carnivorans, a situation not understood at the present time. However, similar lack of diversity appears in contem- poraneous fluvio-deltaic biocalcarenite in which only partial skeletons ofa single family of aquatic birds have been found. Such sit- uations reflect local ecologic and taphonomic biases. Vertebrate bones have been found within the stromatolites themselves, but are rare. In . ‘Les Pérards” quarry near Montaigu, only aquatic vertebrates have been found within these structures. Bentonites composed of montmorillonitic clay associated with these deposits indicate contemporaneous volcanism, but volcano- genic crystals for radiometric dating have not been recovered from the bentonites (Bucher et al., 1985). Was Herpestides an aquatic carnivoran? How can the abundance of its remains be explained in these deposits? Its skeletal struc- ture does not possess any evident specializa- tion that might indicate restriction to an aquatic setting; the animal may have been a capable swimmer but this assumption 1s probably not necessary to account for its good representation in lacustrine sediments. Rath- er, the abundance of Herpestides suggests that it frequented lake margins, and its dental morphology and postcranial skeleton indi- cate that it was an active predator, probably taking small terrestrial and aquatic verte- brates living in these perilacustrine environ- ments. BASICRANIA OF HERPESTIDES Complete or partial crania of 12 individ- . uals were examined in the collections at Lyon and Basel. Of these, four are represented by complete skulls (two with associated lower jaws), seven are posterior crania (only one with associated upper teeth), and one is the skull of a small individual lacking only the occiput. These fossils comprise the hypodigm under study and are listed and briefly de- scribed below: (1) MGL St.-G. 3065. — Complete skull (basilar length, 89.6 mm) without lower jaws, but retains in the maxillae both right and left P3—4 and M1. This is the only skull in the MGL collection that preserves a complete and intact auditory bulla. The complete left bulla includes both ectotympanic and caudal entotympanic; the rostral entotympanic is not visible. The right bulla lacks only the ventral part of caudal entotympanic. Length of the complete bulla from the anteriormost part of ectotympanic to posterior limit of the caudal entotympanic is 19.4 mm (exclusive of the paroccipital process). (2) MGL St.-G. 3066. — Complete skull (basilar length, 102.8 mm) associated with both lower jaws. Upper dentition includes right C, P2-4, M1-2; left C, P3-4, M1-2 (rarely does a maxilla retain M2). Lower den- tition includes right c, p2—4, m 1-2; left c, p3- 4, m1. A well-preserved skull with intact ba- sicranium, in which the auditory bullae lack only the ventral parts of the caudal entotym- panics. (3) MGL St.-G. 3067. — Posterior crani- um, including part of the frontal region but lacking rostrum and dentition. The left au- ditory region (including the petrosal) and the occiput are missing; the petrosal and ecto- tympanic chamber of the right auditory re- gion are preserved. Of interest because it rep- resents the largest individual in the MGL sample (estimated basilar length, ~ 110 mm). (4) MGL St.-G. 3068. — Juvenile posterior cranium (indicated by an open basisphenoid- basioccipital suture), without rostrum or dentition. Both auditory bullae are absent, suggesting that in juveniles ecto- and ento- tympanic elements are loosely attached to the basicranium. Both petrosals are present and well preserved. (5) MGL St.-G. 2009. - Nearly complete skull (estimated basilar length, ~80 mm), lacking occiput and basicranium, represent- ing a very small individual. Caution should be exercised in designating this specimen as one individual because, although the rostrum AMERICAN MUSEUM NOVITATES NO. 3023 has been attached to the posterior cranium with plaster, there is no certain contact be- tween the two parts. The rostrum includes the left P3—4. (6) FSL unnumbered. — Complete skull with basicranium and both auditory bullae intact, currently undergoing preparation and study by M. Escuillie, Université de Lyon. (7) NMB S.G. 11583. - Posterior cranium, lacking rostrum and dentition. Basicranium with right and left auditory regions very well preserved. Auditory bullae are represented only by the right anterior ectotympanic and rostral entotympanic, and the left rostral en- totympanic; caudal entotympanics on both sides are missing. Montaigu-le-Blin, 1921. (8) NMB S.G. 12377. — Partial posterior cranium, with left auditory region only, but with most of bulla missing. Only a dorsal remnant of caudal entotympanic remains in place. Associated with left and right maxillae with P2—4 present on each side. Montaigu- le-Blin, 1921. (9) NMB S.G. 6890. — Posterior cranium, lacking rostrum and dentition, but with both auditory regions preserved. Bullae are absent except for a small caudal entotympanic frag- ment on the right, and a detached ectotym- panic. Montaigu-le-Blin, 1913. (10) NMB S.G. 2935.-— Posterior cranium, lacking rostrum and dentition, but with both auditory regions well preserved. Bullae are entirely lacking. A juvenile with open basi- sphenoid-basioccipital suture. Montaigu-le- Blin, 1913. (11) NMB S.G. 7244. ~ Posterior cranium, lacking rostrum and dentition, and without the occiput. Both petrosals are present, but bullae are absent. A juvenile with open basi- sphenoid-basioccipital suture. Montaigu-le- Blin, 1932. (12) NMB S.G. 11407. ~ Complete skull and associated lower jaws. Both auditory regions are present, but only the left auditory bulla is exposed. Skull somewhat crushed. Montaigu-le-Blin, 1920. AUDITORY REGION OF HERPESTIDES The sample of Herpestides crania from the St.-Gérand region represents an age spectrum from juveniles to mature adults (neonatal in- dividuals are not present). The morphology of the auditory region can be determined at 1991 HUNT: VIVERRID AFFINITIES OF HERPESTIDES 7 Fig. 3. Basicrania of Herpestides juvenile (A, NMB 2935) and adult (B, NMB 11583) from Montaigu- le-Blin in ventral view. In B, small triangles indicate the perimeter of attachment for the caudal ento- tympanic in the right posterior auditory region; both right and left caudal entotympanics have been removed from the basicranium. Note that in the adult (B) the ectotympanic and rostral entotympanic are not fused to surrounding bones, and the sutures between epitympanic elements forming the roof of the posterior auditory region remain open. For abbreviations in this and all subsequent figures, see p. 4. Stereopairs. several ontogenetic stages by examining a se- ries of these basicrania in which the auditory bullae range from complete, to broken open, to fully detached from the skull. As a result, the internal structure and geometry of the middle ear and the composition of the bulla can be studied in detail. Several skulls also possess intact auditory regions for which ex- ternal basicranial morphology can be deter- mined. In all respects, the auditory regions observed in this sample of Herpestides crania demonstrate a morphology typical of extant Viverridae. PETROSAL-MASTOID In both juveniles (figs. 3A, 4A) and adults (figs. 3B, 4B) the petrosal displays a charac- teristic form most similar to that of living viverrid carnivorans such as Civettictis civet- 8 AMERICAN MUSEUM NOVITATES NO. 3023 Fig. 4. Comparison of juvenile (A, NMB 2935) and adult (B, NMB 11583) auditory regions of Herpestides in ventrolateral view, same individuals as in figure 3, both from Montaigu-le-Blin. Note the basioccipital flange (bof) and the pocketing of the lateral part of basioccipital (asterisks) that develop in adults during ontogeny. Stereopairs. ta. The structural geometry of the middle ear region, including the configuration of the bul- la elements, is largely determined by a robust promontorium that forms a transverse ridge dividing the middle ear into anterior and pos- terior chambers. A prominent ventral pro- cess of the promontorium forms a medial extension of this ridge, buttressing the lateral face of the basioccipital immediately poste- rior to the basisphenoid-basioccipital suture (fig. 3A). This process is as well developed in Herpestides as in living aeluroid carnivorans, in which it is a key synapomorphy of the group (Hunt, 1989). Both the anterior and posterior faces of the promontorium slope steeply dorsad away from the transverse ridge that forms the pro- montorial apex. On the anteromedial face of the promontorium rests an ossified rostral entotympanic element that contributes to the bulla’s medial wall. The posteromedial face of the promontorium is deeply impressed by an inflated caudal entotympanic element in juvenile and adult. Lateral to this pro- nounced indentation the promontorium bulges outward behind the round window as a knoblike rugose process. The posterior edge of the petrosal is prolonged as one or two thin sheetlike epitympanic processes of bone that contribute to the roof of the posterior cham- ber; the thin dorsal surface of the caudal en- totympanic element is closely applied to these epitympanic processes. The tegmen tympani is identified as a prominent bony shelf of the petrosal that ex- tends anterior and lateral to the promonto- rium; its ventral surface bears a sharply de- marcated tensor tympani fossa, epitympanic recess, and facial canal, described by Beau- mont (1967). Whether the tegmen is in fact 1991 a composite structure formed by both a true tegmen and by additional epitympanic pro- cesses of the petrosal, as described in some eutherians by MacPhee (1981), is unknown, but for descriptive purposes the entire shelf anterior as well as lateral to the promonto- rium is herein termed the tegmen. In juve- niles, an unfused suture between the petrosal, sphenoid, and squamosal clearly defines the limits of the tegmen (fig. 3A). The anterior portion of the tegmen is deeply pocketed by a medially placed tensor tympani fossa and a laterally placed epitympanic recess. The pe- trosal-squamosal suture runs anteroposteri- orly through the epitympanic recess, hence its lateral part lies in the squamosal and its medial portion in the tegmen. Posteromedial to the epitympanic recess the facial canal emerges dorsal to the oval window in the lateral portion of the tegmen; the nerve fol- lows a groove in the petromastoid, turning laterad to exit the cranium via the stylomas- toid foramen. A prominent fossa lies postero- medial to the facial canal and dorsolateral to the round window in the posterior part of the tegmen. In several juveniles a suture between the mastoid and tegmen may be present (fig. 3A), as in certain living viverrids. In Viverricula and Civettictis there is a persistent suture be- tween the lateral face of petrosal and what appears to be a fused squamosal-mastoid plate lateral to it. In herpestids, felids, and hyae- nids a distinct mastoid is present that be- comes attached to the petrosal in early de- velopment, the mastoid later fusing to squamosal. In viverrids, it seems probable that the squamosal-mastoid fusion occurs first, forming a squamous lamina that covers the petrosal without fusing to it. SQUAMOSAL In the basicranium the sutural boundaries of the squamosal are well defined, particu- larly in juveniles (fig. 3A). The basicranial portion of the squamosal posterior to the postglenoid process contributes the antero- lateral part of the auditory region to which the ectotympanic crura are attached. On the posterior slope of this process near its base is a reduced postglenoid foramen which, al- though variable in diameter in different in- HUNT: VIVERRID AFFINITIES OF HERPESTIDES 9 dividuals, is always quite small, indicating a vestigial postglenoid vein. Posterodorsal to this foramen is a deep pit for insertion of the anterior crus of ectotympanic. Medial to the pit the squamosal is grooved by the Glaserian fissure for the chorda tympani branch of the facial nerve. The fissure is situated imme- diately lateral and parallel to the anteropos- teriorly aligned alisphenoid-squamosal su- ture. The squamosal forms the roof of the bony external auditory meatus directly posterior to the pit for the anterior crus of ectotympanic. In juveniles and adults the roof of the meatus is impressed by a deep suprameatal fossa, as well developed as those observed in living procyonids, presumably incorporating a prominent pars flaccida of the tympanum. Dorsal to the suprameatal fossa the medial face of the squamosal is pocketed by the epi- tympanic recess (there were no auditory os- sicles preserved in any of the crania under study). The suprameatal fossa is excavated entirely in the squamosal; its posterior wall is formed by the posttympanic process of the squamosal applied against the anterior face of the mastoid, the two bones uniting to form the mastoid process. The mastoid-squamosal suture must fuse early in ontogeny because, even in juveniles, a distinct sutural contact between the mastoid bone and the posttym- panic process is difficult to identify, despite an open sutural contact between squamosal and tegmen. ALISPHENOID-BASISPHENOID The epitympanic wings of alisphenoid and basisphenoid contribute the roof of the au- ditory region anterior to the petrosal. In ju- veniles a visible basisphenoid-alisphenoid suture demonstrates that the basisphenoid wing completes the roof medial to the tensor tympani fossa and a somewhat larger alisphe- noid wing is situated anterior to the fossa. The basisphenoid wing closes the antero- medial corner of the auditory region; the ros- tral entotympanic is applied against its lateral face, and it is perforated in its anterior part by the middle lacerate foramen for entrance of the internal carotid artery into the cranial cavity. The pterygoid (Vidian) canal for the Vidian nerve (made up of the greater super- 10 AMERICAN MUSEUM NOVITATES ficial and deep petrosal nerves) can be seen as a narrow channel on the surface of the basisphenoid wing (fig. 4B). The ectotympanic’s anterior part indents the epitympanic wing ofalisphenoid, forming a prominent elliptical depression in adults. The size, depth, and orientation of impres- sions in alisphenoid and squamosal made by the anterior face of ectotympanic and the tip of its anterior crus, respectively, provide im- portant information on the configuration of the anterior bulla chamber in a variety of fossil carnivorans in which the ectotympanic has been lost from the skull. The basisphenoid-basioccipital suture is open in juveniles and young adults, and is fused in mature animals (compare fig. 3A, B). BASIOCCIPITAL-EXOCCIPITAL The basioccipital buttresses the greater part of the medial wall of the auditory bulla, whereas the exoccipital supports its posterior wall. During ontogeny in Herpestides, these two bones are significantly modified to a greater degree than any other basicranial bones bordering the auditory region. This modification results from progressive onto- genetic growth and encroachment of the cau- dal entotympanic on the basi- and exoccip- ital. Figures 3 and 4 indicate the extent of this modification during bulla growth. In the ju- venile (figs. 3A, 4A) the lateral edge of ba- sioccipital is not produced, but in the adult (figs. 3B, 4B) this edge is greatly extended ventrad, and the rugosities on the basioccip- ital’s ventral surface for attachment of the rectus capitis ventralis muscles are much more developed. When seen in lateral view (fig. 4), the edge of basioccipital that borders the mid- dle ear cavity can be divided into an anterior part in direct association with the ventral promontorial process of the petrosal, and a posterior part behind this process. In the ju- venile (fig. 4A), the anterior and posterior parts are not deeply indented by the caudal entotympanic, nor is the edge of basioccipital strongly extended ventrad. However, in the adult (fig. 4B), the anterior part is indented by the forwardly growing apex of caudal en- totympanic; the posterior part shows a prom- NO. 3023 inent elliptical depression for a medial ex- pansion of caudal entotympanic into basioccipital; and the edge of the basioccip- ital is ventrally extended to buttress the en- larging caudal entotympanic chamber. There is no epitympanic wing of the basioccipital: instead the basioccipital is entirely involved in supporting the medial wall of the bulla, and in consequence its lateral margin be- comes modified by contact with the growing caudal entotympanic element. Similarly, the exoccipital is also in contact with the expanding caudal entotympanic and becomes modified in shape during the pro- cess of ontogenetic growth. The exoccipital forms the posterior wall of the auditory re- gion, extending ventrad as the paroccipital process (fig. 3). This posterior wall becomes thinned and tightly appressed against the rap- idly inflating caudal entotympanic chamber of the bulla. The expansion of caudal ento- tympanic is so pervasive that the adjacent mastoid bone becomes impressed and slight- ly displaced by caudal entotympanic growth. Furthermore, the exoccipital develops an epi- tympanic wing that extends dorsal to the cau- dal entotympanic and fills in the roof of the posterior chamber. This epitympanic wing retains an open sutural contact with the mas- toid and with the posterior epitympanic wing of the petrosal, even in adults (fig. 3B). ROSTRAL ENTOTYMPANIC The rostral entotympanic in adults (figs. 3B, 4B) is situated in the anterointernal cor- ner of the auditory region, where it forms a small wedge of bone between the anterior face of the petrosal promontorium and the epi- tympanic wing of the basisphenoid. Approx- imately triangular in lateral view, its apex is directed dorsad toward the tensor tympani fossa. Its anterior edge fits against the basi- sphenoid where a small ridge separates it from the pterygoid canal, and its posterior edge is applied to the petrosal. The ventral margin of rostral entotympanic is in contact and fused to the ectotympanic and caudal entotympan- ic (fig. 4B), but even in adults it does not fuse to either petrosal or basisphenoid; these su- tures remain open. The medial edge of the ectotympanic contacts and fuses to the ven- 1991 HUNT: VIVERRID AFFINITIES OF HERPESTIDES 11 tral edge of rostral entotympanic (thictic state, Hunt, 1987). The anterior apex of the for- ward-growing caudal entotympanic element contacts and fuses to the posteroventral cor- ner of rostral entotympanic (fig. 4B). The ventral margin of rostral entotympanic is ex- posed on the bulla surface and can be rec- ognized even after it has fused with ectotym- panic (fig. 3B). The anteroventral corner of rostral ento- tympanic borders on the middle lacerate fo- ramen. The foramen is entirely within the basisphenoid and transmits the internal ca- rotid artery to the cranial cavity. Directly in front of the middle lacerate foramen is a tiny Vidian foramen for the nerves of the ptery- goid (Vidian) canal. ECTOTYMPANIC Relations of the chambered ectotympanic are demonstrated in figures 3B, 5A, B. In figure 3B the posterior half of the right ec- totympanic in an adult is missing, thus it is possible to observe the relation of the ante- rior crus and limb of ectotympanic where they contact squamosal and alisphenoid bones. Note the inflated volume of the anterior chamber of the auditory bulla formed by ec- totympanic, in contact with the small rostral entotympanic and petrosal, and the location of the suprameatal fossa outside the tympan- ic cavity proper. The crista tympanica is slightly inclined inward from the parasagittal plane, indicating the near-vertical orienta- tion of the tympanum. The crista is in direct linear continuity with the medial edge of the suprameatal fossa, thus the fossa is properly situated to receive the pars flaccida of the tympanum. The medial rim of ectotympanic is in con- tact with and fused to the rostral entotym- panic (thictic condition, Hunt, 1987). The external auditory meatus of the ectotympanic is not laterally prolonged as a bony tube, thus it conforms to the pattern found in living viverrids, and differs from that in most her- pestids and hyaenids in which a bony meatal tube develops. The complete ectotympanic bone (fig. 5) is known in two large individuals with basilar skull lengths of ~ 10-11 cm. In figure 5A the ectotympanic is shown in relationship to the petrosal, the entire caudal entotympanic hav- ing been removed. In figure 5B the ectotym- panic appears in relation to the dorsal part of the caudal entotympanic (and the petro- sal). Note in figure 5A the relatively inflated or expanded ectotympanic with robust anterior crus seated in the squamosal, and a more gracile posterior crus resting on the posttym- panic process of the squamosal. Internal to the tip of the posterior crus the triangular end of the tympanohyal can be seen entering the auditory region where it fuses with the crista parotica of the petrosal. Closure of the pos- terolateral wall of the ectotympanic chamber is accomplished by a prominent septum bul- lae. The bilaminar nature of the septum formed by both ectotympanic and caudal en- totympanic contributions is evident in figure 5A where a thin entotympanic was applied to the ectotympanic surface. The dorsal edge of the septum bullae is produced into a flange that rests on the ven- tral surface of the petrosal promontorium. The thickened flange makes a strong pro- montorial contact (fig. 5B) that produces a facet on the surface of the petrosal immedi- ately anterior to the round window. The lin- ear contact of the septum with the promon- torium is an aeluroid trait (Hunt, 1989), and is the result of the spatial relationship of the ectotympanic’s posterior limb to the pro- montorium in early ontogeny (see subsequent discussion). The close apposition of ectotympanic and petrosal, together with the fusion of ectotym- panic to rostral entotympanic, segregate an anterior chamber of the auditory bulla from a posterior chamber formed by caudal en- totympanic. The relative volume and dimen- sions of the two chambers are illustrated in the adult of figure 5B. By analogy with living viverrids of nearly identical bulla configu- ration, the posterior chamber of the Herpes- tides bulla grew forward into the anteroin- ternal corner of the auditory region (fig. 5B), migrating along the medial wall of the ecto- tympanic, and thereby extended the length of the septum bullae craniad. This pattern of ontogenetic growth, in which caudal ento- tympanic penetrates into the anterointernal 12 AMERICAN MUSEUM NOVITATES comer of the auditory region, is typical of viverrids and felids, not hyaenids or herpes- tids. CAUDAL ENTOTYMPANIC The caudal entotympanic gradually en- larges during ontogeny to produce a volu- minous posterior chamber of the auditory bulla (figs. 5B, 6). This process of growth and remodeling causes the caudal entotympanic to impinge upon and indent the adjacent sur- faces of the basioccipital, exoccipital, mas- toid, and petrosal. The floor and sidewalls of caudal entotympanic expand to form a large hypotympanic cavity. As growth progresses, the ventral process of the petrosal (V) be- comes enclosed and hidden from view be- tween the caudal entotympanic and basioc- cipital (fig. 5B), and hence is not visible in figure 6C (left auditory region). The external form of the caudal entotym- panic is typical of viverrids (fig. 6). The prin- cipal axis of inflation runs from the antero- medial to posterolateral surface in ventral view. The direction of ontogenetic growth is forward into the anterointernal corner of the auditory region (fig. 6A, C). It is particularly noteworthy that overgrowth of the ectotym- panic by the caudal entotympanic takes place in the same manner as observed in living viverrids (fig. 6A, B). This degree of anterior penetration by caudal entotympanic, and its overgrowth of ectotympanic, is not observed in any herpestid or hyaenid. In hyaenids and herpestids the caudal entotympanic is re- stricted to the posterior part of the auditory region. INTERNAL CAROTID ARTERY The path of the internal carotid artery as it travels through the auditory region is for- tuitously preserved in several Herpestides crania in which the bulla has been broken to reveal the internal anatomy of the middle ear cavity. In the adult (fig. 6A, C) the artery enters the auditory region behind the ventral process of the petrosal, immediately lateral to the prominent ridge on the basioccipital for attachment of the rectus capitis ventralis muscle. The artery passes between the lateral edge of basioccipital and the caudal entotym- panic, and is enclosed in a conspicuous bony NO. 3023 tube formed by the caudal entotympanic el- ement. It travels anterolaterad within the en- totympanic tube, which opens on the ventral apex of the promontorium (figs. 5B, 6C). From this point the artery ascends the an- terior face of the promontorium (the pro- montorium is slightly grooved), travels for- ward along the lateral face of rostral entotympanic, and then descends to enter the middle lacerate foramen in the basisphenoid (figs. 4B, 7B). This U-shaped arterial course and its re- lationship to surrounding bullar and basicra- nial bones are typical of viverrids (fig. 7). In figure 7A the path of the internal carotid through the auditory region of the bush civet Civettictis civetta exemplifies the viverrid condition (Hunt, 1987, 1989). In Herpestides antiquus the artery follows the same course (fig. 7B), and manifests the same relation- ships to surrounding basicranial structures. This arterial pattern distinguishes Herpes- tides from living and fossil herpestids in which the internal carotid follows a straight antero- posterior course within the medial wall of the auditory bulla (perbullar course, Hunt, 1989: fig. 4). The herpestid character state is de- rived, a synapomorphy uniting the genera of that family. The construction of the petrosal, auditory bulla, and surrounding basicranial architecture in Herpestides thus differs from the basicranial morphology of living and fos- sil herpestids, but conforms closely to that of the Viverridae. COMPARATIVE MORPHOLOGY OF THE AELUROID MIDDLE EAR In order to adequately evaluate basicranial morphology in Herpestides, it is necessary to describe and analyze two particularly diag- nostic attributes of the auditory region in the living aeluroid families: (a) the ontogeny of the auditory bulla, especially the relationship between ectotympanic and petrosal in early ontogeny, and (b) the stage of evolution of the ossicular chain, as demonstrated by the form and orientation of the auditory ossicles. In the first case, the relationship of ectotym- panic to petrosal promontorium determines the geometry of the anterior bulla chamber, and also appears to influence the final struc- tural relationship between caudal entotym- 1991 HUNT: VIVERRID AFFINITIES OF HERPESTIDES 13 Fig. 5. Relationship of ectotympanic to petrosal in Herpestides: A, posterolateral view of right auditory region (MGL St.-G. 3067) of large individual; caudal entotympanic, basi- and exoccipital removed; white triangles indicate line of attachment of caudal entotympanic to petrosal and mastoid; B, posterolateral view of left auditory region (MGL St.-G. 3066, see also fig. 1) of another large individual in which the dorsal part of caudal entotympanic is in place. Note contact between the ectotympanic flange and the promontorium just anterior to the round window in both individuals; this contact produces a small characteristic facet on the promontorium. Stereopairs. panic and ectotympanic. In the second, the morphology of the auditory ossicles of viver- rids and felids reflects a more primitive stage of evolution than the ossicles of herpestids and hyaenids. As a result, the viverrid and herpestid auditory patterns are sharply de- fined, and Herpestides, in its adult morphol- ogy and in what can be determined of its auditory ontogeny, is identifiable as a true viverrid. Fig. 6. The auditory bulla of Herpestides (MGL St.-G. 3065): A, high lateral oblique view of basi- cranium, showing intact left bulla (black triangles indicate the line of fusion between ectotympanic and caudal entotympanic), and broken right bulla opened to show internal structure of the posterior chamber, including course of internal carotid artery (dashed line) as it enters the middle ear; B, lateral view of auditory bulla showing inflation of caudal entotympanic relative to ectotympanic as in viverrids; C, posteroventral view of the basicranium, demonstrating penetration of caudal entotympanic into the anterointernal corner of the auditory region. Stereopairs. 1991 HUNT: VIVERRID AFFINITIES OF HERPESTIDES 15 Fig. 7. Comparison of the auditory regions of (A) the viverrid Civettictis civetta (UNSM 14114), Kibwezi, Kenya, and (B) Herpestides antigquus (NMB 11583), Montaigu-le-Blin, France. Dashed lines in A and B indicate the path of the internal carotid artery from its point of entrance into the auditory region until it enters the cranial cavity at the middle lacerate foramen. Stereopairs. The course of ontogenetic development in the auditory region of aeluroids is perhaps best understood in terms of the different on- togenetic pathways adopted by viverrids and herpestids: I intend to describe a sequence of early ontogenetic stages from a representative viverrid and herpestid in order to explore the developmental basis for these differences. Because the felid pattern of bulla devel- opment is nearly identical to that of viver- 16 AMERICAN MUSEUM NOVITATES rids, and has been previously described at length (Hunt, 1974, 1987, 1989), I will not review it here. The pattern of felid bullar on- togeny differs from that of viverrids only in the more elaborate inflation of the caudal en- totympanic and its diagnostic encroachment on surrounding basicranial bones. As part of this process, the felid caudal entotympanic penetrates into the anterointernal corner of the auditory region, inserting itself between ectotympanic and rostral entotympanic (bra- dynothictic state, Hunt, 1987); this insertion does not occur in viverrids. In the Aquitanian sediments of the Allier basin, both viverrid and felid patterns can be recognized. Com- parison of these contemporaneous forms in- dicates that Herpestides is not an early felid. Insight into the development of the hyae- nid bulla and ossicular chain is limited by lack of knowledge of early ontogenetic stages. However, following discussion of viverrid and herpestid morphologies, I review what is known of the hyaenid auditory region, and argue that there is sufficient knowledge of hyaenid bulla and ossicular structure to con- fidently exclude Herpestides from the Hy- aenidae. BULLA ONTOGENY IN HERPESTIDAE In an ontogenetic series of crania of Her- pestes (Xenogale) naso from various localities in Zaire, a progression from an incipient stage of bulla formation to the fully formed adult bulla defines the herpestid pattern of devel- opment. The earliest available ontogenetic stage is a skull (AMNH 51092, Niangara, Zaire) with basilar length of 30.9 mm in which the tips of the principal cusps of DP3—4 are just form- ing in the dental crypts, and the only ossified bulla element is the ectotympanic (fig. 8C). The ectotympanic crescent lies in the hori- zontal or frontal plane parallel to the basi- cranial axis. Its anterior and posterior crura are attached to the squamosal. The ventral edge of the squamosal between the two crura is bent inward as a lamina forming a floor for a large epitympanic recess. The recess contains the robust heads of malleus and in- cus (a deeply pocketed epitympanic recess within the squamosal is typical of herpestids). The anterior ectotympanic crus is anchored NO. 3023 to the edge of the squamosal, directly lateral to the gonial process of the malleus, whereas the posterior crus simply overlaps its lateral surface at the level of the posttympanic pro- cess of the squamosal. Because the ectotympanic is essentially horizontal, its anterior limb is pressed against the roof of the tympanic cavity, which is formed by the epitympanic wings of basi- and alisphenoid. Its posterior limb is applied to the petrosal promontorium slightly posterior to the round window. At this stage, the pos- terior part of the tympanic cavity is closed by a sheet of connective tissue attached to the posterior margin of ectotympanic, and extending from that margin to the edges of the basioccipital, exoccipital, and mastoid bones. The malleus and incus within the epitym- panic recess are relatively large and already ossified, and the posterior process of the incus and anterior process of the malleus that joint- ly determine the axis of rotation are devel- oped (the rotational axis of the carnivoran malleus-incus complex is a straight line drawn through the posterior process of the incus and the tip of the anterior process of the malleus). In this and all subsequent developmental stages, even in adults, a line drawn parallel to the manubrium of the malleus intersects the rotational axis of the malleus-incus at a high angle. Furthermore, the axis of ossicular rotation is nearly horizontal (here defined as a line parallel to the basicranial axis), and is not markedly inclined. The tensor tympani fossa at this stage remains open and has no bony floor. In the next available stage of development, a skull (AMNH 51093, basilar length, 42.7 mm) from Faradje, Zaire, the DP3-4 crowns are fully formed and partially erupted, and the three separate bulla elements are present as unfused ossifications. The ectotympanic crescent has widened considerably, especially in its medial part, tilting the crescent (and eardrum) slightly laterad. Although the ec- totympanic’s posterior limb rests on the pos- terior slope of the promontorium, the pro- montorium remains largely within the circumference of the ectotympanic (hence within the bulla’s anterior chamber). The ro- tational axis of the auditory ossicles does not alter its alignment, and remains nearly per- 1991 pendicular to the manubrium of the malleus. At this stage the tensor tympani fossa now is largely encapsulated by bone, and the exter- nal auditory meatus between the ectotym- panic crura has begun to close by progressive ossification of the rim of the aperture. Ap- pressed against the posterior margin of ec- totympanic and the posterior surface of the promontorium is a small caudal entotym- panic chamber, only weakly ossified and slightly inflated, enclosing a volume that is perhaps half that of the anterior chamber. At this early stage of development the en- trance of the internal carotid artery into the middle ear cavity is already blocked by os- sification of the bulla wall. Subsequently, the artery becomes isolated within a bony tube in the medial wall of the bulla (perbullar state, Hunt, 1987), a condition found in all living herpestids but in no other living aeluroids. A slightly more advanced stage of auditory development (fig. 8B) is present in two skulls (AMNH 51609, basilar length, 46.1 mm; AMNH 51089, basilar length, 45.2 mm) from Medje, Zaire, in which DP3—4 are erupted to a somewhat greater degree than these same teeth in the 42.7 mm stage. The bulla shows the initiation of caudal entotympanic infla- tion, enlarging in ventral and medial direc- tions, whereas the ectotympanic is about the same size. The amount of inclination of the ectotympanic crescent and eardrum remain the same, but closure of the external auditory meatus by bone has progressed, and this pro- cess continues during ontogeny until the mea- tal aperture is completely filled (or nearly so) by bone in the adult. This new bone forms a ventral floor below the eardrum. The rostral entotympanic has increased in size, following its initial appearance between the 30.9 and 42.7 mm stages discussed above. Dissection of the bulla reveals a bilaminar septum bullae that will fuse into a single par- tition in the adult. The plane formed by the septum bullae at the juncture of ectotympan- ic and caudal entotympanic is somewhat pos- teriorly inclined (see Hunt, 1989: fig. 5). The caudal entotympanic chamber of the bulla is confined to the posterior part of the auditory region behind the promontorium and it re- mains so in the adult. In adult herpestids the caudal entotym- panic chamber continues to inflate relative HUNT: VIVERRID AFFINITIES OF HERPESTIDES 17 to the anterior chamber, however the direc- tion of inflation is primarily ventrad, and the caudal entotympanic does not penetrate the anterior part of the auditory region as it does in viverrids and felids. In a number of her- pestid species the ectotympanic crescent and eardrum remain nearly horizontal; in other species such as Herpestes auropunctatus, the ectotympanic plane rotates outward, but the pronounced anterolateral tilt found in viver- rids (and felids) never occurs. BULLA ONTOGENY IN VIVERRIDAE Using an ontogenetic series of skulls of Viverricula indica, it is possible to examine equivalent stages of basicranial development in a typical viverrid (fig. 8A), and identify differences relative to herpestid ontogeny. At the earliest available stage, the skull (AMNH 59935, Hainan, China, basilar length 38.2 mm) shows the tips of DP3—4 cusps just forming in the alveolar crypts. The only os- sified bulla element is the ectotympanic. The ectotympanic crescent does not lie in the hor- izontal (or frontal) plane parallel to the ba- sicranial surface as in herpestids. The plane of the ectotympanic is tilted forward and out- ward (anterolaterad) relative to its orienta- tion in herpestids, the result of displacement of the ectotympanic’s posterior limb by ven- tral protrusion of an enlarged petrosal pro- montorium. Because the ectotympanic cres- cent is unable to fully surround the enlarged petrosal, its posterior limb is captured by the ventrally extended promontorium. This dis- placement causes the posterior limb to lie across the apex of the promontorium, slightly anterior to the round window. As a result, a substantial part of the promontorium lies be- hind the posterior limb, where it is covered by connective tissue forming the posterior wall of the tympanic cavity (in herpestids, more of the promontorium is enclosed by the ectotympanic). This connective tissue sheet is attached at the posterior margin of ecto- tympanic and extends to the mastoid and oc- cipital bones just as in herpestids. In a later developmental stage (AMNH 60065, Hainan, China, basilar length 51.8 mm) in which DP3-4 are formed but un- erupted, the ectotympanic has been enlarged by the addition of bone to its medial margin, Fig. 8. Petrosal-ectotympanic relationship of a viverrid and herpestid in early ontogeny. A, Basi- cranium in ventral view of viverrid Paradoxurus hermaphroditus (AMNH 59933, female neonate, Hainan, China). The caudal entotympanic has been removed from the right auditory region to dem- onstrate the petrosal-ectotympanic relationship: note large amount of petrosal exposure posterior to ectotympanic (compare fig. 8B); B, Basicranium in ventral view of herpestid Herpestes (Xenogale) naso (AMNH 51609, male neonate, Medje, Zaire). The caudal entotympanic has been opened on the left side to show petrosal nearly completely overlapped by ectotympanic: there is almost no petrosal exposure posterior to ectotympanic (compare A); C, Basicranium in ventral view of herpestid Herpestes (Xenogale) naso (AMNH 51092, male neonate, Niangara, Zaire). The ectotympanic nearly encompasses the petrosal promontorium at this very early developmental stage in herpestids. 1991 creating an anterior bulla chamber. A small but well-defined rostral entotympanic ossi- fication has appeared on the anterior slope of the promontorium, and has made contact with the medial rim of ectotympanic by a narrow isthmus of bone. Ossification of cau- dal entotympanic has progressed about three- quarters of the distance from the rear of the bulla element toward the anterointernal cor- ner, migrating along the tympanic membrane medial to ectotympanic. The ossification front is a smooth convex margin that has reached a point just anterior to the ventral process of the promontorium. In front of this margin (and posterior to the ectotympanic-rostral entotympanic complex), a small portion of the tympanic membrane remains unossified. Two skulls (AMNH 45504, Fukien, China; AMNH 107605, Bali; basilar lengths about 52-54 mm), with DP3—4 nearly fully erupted, indicate a slightly later stage in which the three ossified bulla elements have joined, but fusion is only incipient, so that the bounda- ries of each element remain evident. There is no significant change in orientation or in- flation of the ectotympanic or anterior cham- ber, and the posterior limb of ectotympanic remains in contact with the surface of the promontorium anterior to the round win- dow. As a consequence, the posterior part of the enlarged promontorium containing the round window continues to protrude into the posterior chamber of the bulla (fig. 8A). The caudal entotympanic has completed its anterior growth and is entirely ossified. Its rounded anterior margin has made contact and fused with rostral entotympanic and the medial edge of ectotympanic, and its degree of inflation is visibly greater than in the pre- vious developmental stage. The application of caudal entotympanic to the posteromedial rim of ectotympanic has produced a bilam- inar septum bullae about 1.5 mm in height. In all developmental stages of Viverricula indica, the manubrium of the malleus is “bent forward’ (strongly inclined anteromesad), placing the long axis of the manubrium at a very low angle relative to the rotational axis of the auditory ossicles (I refer to the incli- nation of the manubrium in the plane of the ossicular rotational axis in lateral view, not to medial deviation of the manubrium which can occur independently in mammals). This arrangement occurs in other viverrids and in felids, but contrasts with the herpestid con- HUNT: VIVERRID AFFINITIES OF HERPESTIDES 19 dition in which the long axis of the manu- brium maintains a high angle (> 60°) relative to the rotational axis. This difference between the neonatal mallei of herpestids and viver- rids persists in the adults, and its origin and explanation are of considerable interest. We can determine that the “‘bent”’ viverrid manubrium does not result from the change in orientation of the ectotympanic during on- togeny because, in prenatal stages in which the ectotympanic lies nearly in the horizontal plane, the manubrium has already adopted its alignment relative to the rotational axis. In fact, the “bent” manubrium is plesiomor- phic for aeluroids, and probably for Carniv- ora (see below); it is an attribute of the prim- itive therian malleus. Furthermore, in viverrids the rotational axis of the malleus-incus complex is tilted slightly upward (1.e., the attachment of the incus to the posterior wall of the epitympanic recess is dorsal to the point of attachment of the anterior process of the malleus), deviating from the horizontal to a greater degree than the herpestid axis of rotation. Felids display the same ectotympanic—cau- dal entotympanic growth pattern as viverrids during ontogeny, including displacement of the posterior limb of ectotympanic by an en- larged promontorium and, as a result, appli- cation of the posterior ectotympanic limb to the promontorium anterior to the round win- dow (fig. 9). Felids and viverrids also share the “bent” orientation of the manubrium, and the slightly tilted ossicular rotational axis. In these aspects of bulla ontogeny, and ossicle form and orientation, the two families are extremely similar. These features were con- firmed in an ontogenetic series of the do- mestic cat, and also in a comparable series of the African wild cat, Felis silvestris, and in juveniles and adults of almost all living felids. BULLA ONTOGENY IN HYAENIDAE The hyaenid auditory region is distin- guished by a unique ontogenetic pattern of bulla development in which the ectotympan- ic contributes an enlarged anterior bulla chamber that extends backward, ventral to the more confined posterior chamber formed by caudal entotympanic (Hunt, 1974, 1987, 1989). The earliest known hyaenid basicrania from the mid-Miocene of Asia already have developed this bulla configuration (Qiu et al., 20 AMERICAN MUSEUM NOVITATES NO. 3023 Fig. 9. Petrosal-ectotympanic relationship of a felid in early ontogeny. A, Basicranium in ventral view of newborn domestic cat Felis, demonstrating the large portion of the petrosal promontorium posterior to the ectotympanic, as in viverrids (compare fig. 8A); B, same specimen as A, medial view of auditory bulla, showing the considerable exposure of the promontorium posterior to the ectotympanic. Black triangles indicate ectotympanic-caudal entotympanic suture. 1988; Hunt, 1989). I know of no represen- _ onstrate that the fundamental hyaenid bulla tative ontogenetic series of the hyaenid basi- _ pattern was already developed in the earliest cranium; however, the few juvenile stages of | ontogenetic Stages presently identified (e.g., living hyaenids that have been studied dem- Hunt, 1974: fig. 35). It seems likely that the 1991 transverse lamina horizontal % ectotympanic HUNT: VIVERRID AFFINITIES OF HERPESTIDES 21 Fig. 10. Evolution of the malleus-incus complex in therian mammals: A, primitive therian; B, transitional stage; C, derived. Diagrams depict right ossicular complex in lateral view, dorsal to top, anterior to right. Key trends include change in inclination of rotational axis, reduction of malleal at- tachment to ectotympanic, reorientation of the manubrium, and relative size increase of the incus. Living aeluroid carnivorans belong to stages B and C. hyaenid bulla ossifies in the neonatal animal in much the same configuration seen in young juveniles. Even in the aberrant hyaenid Pro- teles, the basic morphology of the hyaenid bulla can still be recognized, despite the fail- ure of the anterior chamber to extend back- ward under the posterior chamber (Hunt, 1974). Hyaenids share with viverrids and felids a “bent” manubrium and an inclined ossicular rotational axis. However, hyaenids and her- pestids lack the prominent pocketed anterior lamina of the malleus found in viverrids and felids that connects the neck of the malleus with the anterior process. The anterior lam- ina of herpestids and hyaenids is strongly re- duced. AUDITORY OSSICLES IN AELUROIDS OSSICULAR ORIENTATION IN MAMMALS Ossicular patterns contribute an additional dimension to an understanding of auditory evolution in Carnivora. In living aeluroid carnivorans, morphology and orientation of the auditory ossicles are closely correlated with bulla morphology, particularly with ec- totympanic placement. This information al- lows a reasonable prediction of the mor- phology and orientation of the auditory ossicles in Herpestides antiquus, and also sup- plies evidence for a hypothesis describing the evolution of the ossicular system in aeluroids. Incorporation of the mammalian postden- tary bones into the auditory region has been much better understood in recent years (AI- lin, 1975; Kermack and Mussett, 1983). The orientation and probable functional relations of the angular (ectotympanic), articular (mal- leus), quadrate (incus), and stapes in primi- tive therians have been determined both from fossils and from anatomical studies of living eutherians and metatherians. Fleischer (1973, 1978) has proposed a series of stages in the evolution of the auditory ossicles and ecto- tympanic that progresses from a primitive therian morphotype (Ausgangstyp) via a transitional arrangement (Ubergangstyp) to a final derived condition in which the ossicles are freely mobile within the middle ear space (fig. 10). All these stages can be found among living therian mammals, and their functional and anatomical attributes have been the ob- ject of extensive investigation (Fleischer, 1978; Hunt and Korth, 1980). From my observations of eutheres and me- tatheres, and those of Segall (1969, 1970, 1971), it is possible to identify three prop- erties of the ossicular chain that change through time. Each of these properties seems to be able to evolve independently of the oth- ers: (1) Altered orientation of the rotational axis of malleus-incus. — The plesiomorphic ori- entation of the ossicular rotational axis in early mammals is slightly tilted or inclined (fig. 10A), having an approximately antero- posterior alignment (Allin, 1975: pl. 3, fig. 10; Segall, 1969: fig. 1; Fleischer, 1973: fig. Ze AMERICAN MUSEUM NOVITATES 58B). The axis passes forward and downward from the point at which the incus (quadrate) contacts the periotic bone, thence through the gonial process of the malleus (articular) along its line of attachment to the ectotympanic (angular). This primitive orientation of the ear ossicles is seen in marsupials such as Di- delphis and Caenolestes, where the axis of rotation makes an angle of about 20 to 23° with the horizontal (Segall, 1969), and is also known in placentals with primitive ossicular configurations, where one finds similar amounts of axial inclination, ranging from about 20 to 42° (Segall, 1971). In living mar- supials and placentals, the rotational axis ei- ther is maintained in an inclined orientation, or has transformed by rotation into a more horizontal alignment (fig. 10B, C). Thus an inclined axis of rotation reflects a more ple- siomorphic ossicular alignment relative to derived states that approach or attain the hor- izontal. (2) Loss of malleal attachment to ectotym- panic. — The plesiomorphic mammalian os- sicular chain (fig. 10A) incorporates a large U-shaped malleus (one leg formed by the ma- nubrium, the other by the gonial process) firmly attached to the ectotympanic, and a relatively small incus (Allin, 1975: pl. 3, figs. 10, 11; Fleischer, 1978: figs. 1, 8). This type, or a close approximation, is known in living marsupials, monotremes, and placentals, and its ontogenetic development (for example, in certain marsupials) shows a clear correspon- dence to the postdentary bones of cynodont reptiles and early mammals (Kermack and Mussett, 1983). An ossicular chain of this kind exhibits considerable torsional stiffness due to the plesiomorphic bony fusion of mal- leus to ectotympanic. As the ossicular ap- paratus evolves from primitive (fig. 10A) to more advanced states (fig. 10B, C), the mal- leus often becomes more loosely attached to ectotympanic, and as a result the degree of stiffness is proportionately reduced. This looser attachment is achieved by reduction of the bony process anchoring the malleus to ectotympanic to either a ligamentous con- nection, or if osseous, one that is very thin and flexible (living aeluroid carnivorans all maintain a thin, flexible, bony attachment). Thus a primitive eutherian malleus will have a prominent gonial process firmly attached NO. 3023 to ectotympanic, and a well-developed trans- verse lamina (fig. 10A); both the process and the lamina become much reduced in deriv- ative eutherian mallei (fig. 10B, C). (3) Realignment of the manubrium relative to the axis of rotation. — Fleischer (1978) in- dicated that the long axis of the manubrium can alter its alignment relative to the rota- tional axis of the malleus-incus complex. In primitive therians the manubrium is gener- ally aligned parallel to the rotational axis (fig. 10A). In many modern lineages with freely mobile ossicular chains, the long axis of the manubrium is nearly perpendicular to the ro- tational axis (fig. 10C), a derived character state in therian mammals. Thus we can identify plesiomorphic and derived morphological states of the auditory ossicles involving orientation of the rotation- al axis, nature of the attachment of malleus to ectotympanic, and orientation of the ma- nubrium. We can apply this knowledge to the ossicular morphology and orientation of the living aeluroids. As a result, we find that os- sicular patterns are apparent at the family level, and that both plesiomorphic and de- rived conditions exist among aeluroids that can be used to constrain evolutionary hy- potheses. OSSICULAR ORIENTATION AND MORPHOLOGY IN AELUROIDS Viverrids and felids have an ossicular ori- entation and geometry that reflect a transi- tional stage of ossicular development (fig. 10B). The axis of rotation is tilted forward and downward (fig. 11C, D), inclined 25 to 30° from the horizontal (determined as a line drawn parallel to the basicranial axis). Thus the viverrid-felid alignment approximates the axial orientation found in a living therian such as Didelphis. The manubrium is oriented at an angle of about 15 to 25° relative to the ossicular rotational axis (fig. 12C, D), hence similar to primitive manubrial alignments that parallel this axis, and unlike derived ma- nubria that lie nearly perpendicular to the axis. The malleus of viverrids and felids is firmly yet delicately attached by thin, flexible bone to the ectotympanic in both neonates and adults. The lamina of the malleus is well de- 1991 veloped, and its lateral face is characteristi- cally pocketed (fig. 12C, D); viverrids and felids preserve this plesiomorphic configu- ration of the lamina, whereas the lamina is strongly reduced or lost in herpestids and hyaenids (figs. 12A, B). The pocketed lamina is a vestige of the well-developed transverse lamina of the malleus found in primitive therians (fig. 10A). Alignment and morphology of the ossicles in viverrids and felids thus appear to differ from a primitive eutherian ossicular pattern (fig. 13) only in (a) reduction of the gonial process of the malleus to produce a more flexible bony attachment (the anterior pro- cess, fig. 12C, D) of malleus to ectotympanic; (b) a slightly greater deviation of the long axis of the manubrium from the axis of ossicular rotation; (c) size increase of incus relative to malleus. There also may have been a small (but uncertain) amount of declination of the rotational axis toward the horizontal, based on the supposition that the primitive therian axis was more steeply inclined, but there is no reason why the present inclination in vi- verrids and felids might not closely approx- imate the primitive state for carnivorans. Herpestids on the other hand have an os- sicular arrangement similar to the freely mo- bile type of Fleischer (fig. 10C). The axis of rotation is anteroposteriorally aligned with- out significant deviation from the horizontal (figs. 11B, 13E). The manubrium is more nearly perpendicular to the axis of rotation (60° to 80°, fig. 12B). The malleus is delicately attached to the ectotympanic by a fragile, much reduced anterior process (a bony con- nection, however, still persists in neonates and adults). The lamina is strongly reduced, and only a minute vestige ofa pocket remains (fig. 12B). Herpestids appear to have evolved the most derived ossicular geometry and ori- entation among the living aeluroid carnivo- rans (fig. 13). Hyaenids share aspects of both the viver- rid-felid and herpestid end-members in terms of ossicular form and axial orientation. Hyaenids retain an alignment of the rota- tional axis similar to that of viverrids and felids, inclined approximately 33-37° from the horizontal (fig. 11A). However, the at- tachment of malleus to ectotympanic is quite delicate (it is a thin osseous connection in HUNT: VIVERRID AFFINITIES OF HERPESTIDES 23 adults), the lamina is strongly reduced as in herpestids, and only a vestige of the pocketed lamina remains (figs. 12A, 13B). It is evident that hyaenids have not passed through a her- pestid grade of ossicular orientation. The in- clined rotational axis of hyaenids seems more plesiomorphic than the nearly horizontal axis of herpestids, but both groups have under- gone significant reduction of the pocketed lamina of the malleus. The fact that viverrids and felids share near- identity in their ossicular structure, whereas herpestids are much different and arguably more derived, does not necessarily substan- tiate the sister-group relationship that I have proposed for viverrids and felids on the basis of bulla ontogeny (Hunt, 1987). The orien- tation of the ossicular rotational axis and the form of the malleus in viverrids and felids are plesiomorphic (fig. 13), and could simply reflect the absence of significant change in these features over time. However, the fact that viverrids and felids share the enlarged petrosal promontorium and the concomitant tilting of the ectotympanic, correlated with the forward-growing caudal entotympanic el- ement, constitutes more reliable grounds for the proposed sister relationship between these two families. Herpestids stand apart from this viverrid-felid dichotomy because of their de- rived bulla ontogeny and their ossicular ori- entation and morphology: they cannot be an- cestral in their present form to any other group of living aeluroids. Living hyaenids are sim- ilarly derived in the form of their malleus, and so cannot be ancestral to viverrid and felid carnivorans in which the condition of the malleus is more plesiomorphic. However, the similarity of hyaenid and herpestid mallei is not necessarily evidence of close relation- ship, because it could represent parallel evo- lution, a result of the independent develop- ment of a more mobile ossicular chain in the two groups. The close correlation between bulla mor- phology and ossicular geometry and orien- tation in living aeluroids suggests a plausible reconstruction of the ossicular chain in Herpestides antiquus. One can predict that this Aquitanian viverrid possessed a slightly inclined rotational axis; the incus will be small relative to the size of the malleus; the ma- nubrium of the malleus will lie at a low angle 24 AMERICAN MUSEUM NOVITATES BASICRANIAL AXIS NO. 3023 BASICRANIAL AXIS 1991 HUNT: VIVERRID AFFINITIES OF HERPESTIDES BASICRANIAL AXIS AXIS 25 BASICRANIAL Fig. 11. Inclination of the ossicular rotational axis within the middle ear of aeluroid carnivorans, measured as the angle of intersection with the basicranial axis (an estimate of the horizontal). A, Hyaenid Hyaena brunnea (UNSM 16442); B, herpestid Herpestes auropunctatus (AMNH 232811); C, viverrid Genetta sp. (UNSM 14275); D, felid Lynx rufus (UNSM 14674). Lateral view, ventral to top, anterior to left (hyaenid in ventrolateral view). 26 AMERICAN MUSEUM NOVITATES manubrium short crus AXIS OF ROTATION AXIS OF ROTATION NO. 3023 Fig. 12. Orientation of the manubrium relative to the ossicular rotational axis in aeluroid carnivorans (measured as the angle between the long axis of the manubrium [m] and the rotational axis [ar]). A, Hyaenid Hyaena brunnea (UNSM 16442); B, herpestid Mungos mungo (AMNH 118859); C, viverrid Genetta sp. (UNSM 14275); D, felid Panthera tigris (UNSM 15484). Lateral view, ventral to top, anterior to left. Viverrids and felids retain a plesiomorphic form of the malleus in which the pocketed lamina is retained as a vestige of the prominent transverse lamina of early therians. The lamina is reduced in hyaenids and nearly absent in herpestids. to the axis of ossicular rotation; the malleus will retain a well-developed lamina, probably pocketed as in living viverrids; the connec- tion between the malleus and ectotympanic will be osseous, not ligamentous. Viverrids, among living aeluroids, retain the most ple- siomorphic ossicular chain, one which is markedly different from the derived ossicular morphology found in herpestids. PHYLOGENETIC AFFINITIES OF HERPESTIDES Basicrania attributed to Herpestides in the Basel and Lyon collections, including several belonging to complete skulls with associated dentitions, all invariably possess a common morphology of the auditory region. Both in- tact and crushed skulls document the asso- ciation of these basicrania (and auditory bul- lae) with dentitions undoubtedly referable to Herpestides antiquus (figs. 14, 15). The basicranial morphology of Herpestides is readily comparable to basicranial patterns found in living viverrids. Among the features common to Herpestides and viverrids are: (1) petrosal form and orientation; (2) configu- ration and ontogenetic growth pattern of the caudal entotympanic relative to other bulla elements and surrounding basicranial bones; (3) form, size, and structural relations of the three bulla elements (rostral and caudal en- totympanics, ectotympanic); (4) path of the internal carotid artery in the auditory region. Herpestides has been considered a herpestid (Viret, 1929) or ancestral hyaenid (Beau- mont, 1967; Beaumont and Mein, 1972; Hunt, 1989) in earlier studies. Petter (1974) believed that the auditory bulla was of the herpestid type. My opinion, initially based on illustrations in the European literature, changed when I was able to examine the fos- sils. The preceding review of herpestid and hyaenid auditory regions points to the ab- sence of any distinguishing features in the Aquitanian aeluroid that would ally it with 1991 TS HORIZONTAL HORIZONTAL HUNT: VIVERRID AFFINITIES OF HERPESTIDES 27 Fig. 13. Evolution of the ossicular complex in aeluroid Carnivora (B through E each presumed to be independently evolved from A). A, Early therian with marked axial inclination and plesiomorphic ossicle morphology (developed transverse lamina of malleus, small incus relative to malleus); B, hyaenid with moderate inclination and derived ossicle morphology (reduced lamina of malleus, incus enlarged relative to malleus); C, D, viverrid and felid with moderate inclination and plesiomorphic ossicle mor- phology (pocketed lamina of malleus, incus small); E, herpestid with low inclination and derived ossicle morphology (strongly reduced lamina, incus large relative to malleus). Lateral view of right middle ear, dorsal to top, anterior to right. either of these families. Here I briefly sum- marize the important distinctions between Herpestides and herpestids/hyaenids. Herpestids differ in their auditory structure from viverrids in having a differently shaped petrosal, characterized by a particularly prominent bony capsule anterior to the pars cochlearis that houses the tensor tympani muscle. The muscle is nearly encapsulated by bone in adults (a trait unknown in other ae- luroids), and emerges through a restricted, laterally directed aperture about the same di- ameter as the round window (see Hunt, 1989: fig. 4B). This construction of the tensor tym- pani fossa is unknown in living viverrids, fe- lids, and hyaenids, and is considered a de- rived trait of herpestids on the basis of outgroup comparison with other carnivo- rans. An encapsulated tensor tympani fossa is not found in Herpestides. The caudal entotympanic forms the pos- terior chamber of the auditory bulla in ae- luroid carnivorans. In herpestids, the caudal entotympanic remains restricted to the pos- terior part of the auditory region during on- togenetic growth. In living herpestid species, inflation of the posterior (caudal entotym- panic) chamber takes place during ontogeny 28 AMERICAN MUSEUM NOVITATES NO. 3023 Fig. 14. Association of Herpestides antiquus dentitions with typical viverrid basicrania occurs in a number of skulls from the Aquitanian of the Allier basin, France (NMB 11407, Montaigu-le-Blin). but is primarily directed ventrad and to the sides, never forward. However, in contrast to the orientation in herpestids, the caudal en- totympanic chamber grows forward in viver- rids and felids, penetrating into the anteroin- ternal corner of the auditory region, often overgrowing the anterior chamber of the bul- la. Herpestides shows a pattern of caudal en- totympanic growth comparable to that of viverrids. No living or fossil herpestid has a bulla pattern even remotely similar to that seen in Herpestides. The placement of the ectotympanic, and its relation to the petrosal, also distinguishes herpestids. Because the petrosal promonto- rium is relatively small and almost fully en- closed (more so in some species than in oth- ers) within the perimeter of the ectotympanic crescent in early developmental stages, the herpestid ectotympanic lies flat against the basicranium in the horizontal plane. How- ever, in viverrids and felids the ectotympanic crescent fails to enclose the promontorium as completely as in herpestids, apparently be- cause the promontorium is quite large and protrudes below the basicranial surface. As a result the posterior limb of ectotympanic comes in contact with the apex of the en- larged promontorium (just anterior to the round window), causing the ectotympanic to tilt forward and outward so that it initiates its growth from a position much different than that found in herpestids. The orientation of the ectotympanic in both juvenile and adult Herpestides demonstrates a viverrid-like, rather than herpestid-like, ectotympanic-pe- trosal relationship. In Herpestides the ecto- tympanic has been tilted forward and out- ward by an enlarged promontorium, and the posterior limb of ectotympanic, unable to en- close the promontorium, is applied to the petrosal surface anterior to the round win- dow. A hallmark of the herpestid auditory region is an ectotympanic oriented nearly in a fron- tal plane in juveniles and even in adults (ec- totympanic inflation causes slight outward rotation in some species). As a consequence the herpestid tympanum faces nearly directly ventrad, and were it to remain so in the adult, the anterior bulla chamber would be floored by the eardrum. To avoid this, herpestids 1991 HUNT: VIVERRID AFFINITIES OF HERPESTIDES 29 Fig. 15. The dentition of Herpestides antiquus has been attributed by Beaumont (1967) to a single, highly variable species. The upper teeth, although more plesiomorphic, show similarities in occlusal pattern to those of the African bush civet, Civettictis civetta, which may be a lineal descendant. close the auditory meatus of the ectotym- panic with bone, thereby creating an osseous floor beneath the eardrum. In viverrids and felids the more parasagittal orientation of the ectotympanic, both in early developmental stages and as a further result of ectotympanic inflation, does not result in closure of the auditory meatus by bone; rather, the meatus remains open as a prominent aperture into adult life (Hunt, 1989: fig. 2). In both juvenile and adult Herpestides, the bony external au- ditory meatus is open and shows the viverrid orientation; there is no evidence of the uniquely derived herpestid condition of the meatus. It seems probable that the different ori- entation of the ectotympanic crescent relative to the promontorium in herpestids and viver- rids/felids is the basis for the different con- figurations of the caudal entotympanic cham- ber in these two groups. The penetration of the caudal entotympanic into the anterointer- nal corner of the auditory region in viverrids and felids is correlated with outward and for- ward displacement of ectotympanic in early ontogeny. Because the ectotympanic is not displaced in herpestids, the caudal entotym- panic cannot penetrate into the anterointer- nal auditory region and, asa result, it remains within the posterior part of the auditory re- gion. Herpestides demonstrates the viverrid caudal entotympanic configuration, not the herpestid. Enclosure of the internal carotid artery in the medial bulla wall (perbullar course) in all herpestids is a unique synapomorphy of the family not found in other aeluroids. The ar- tery in Herpestides does not follow such a course but instead takes a transpromontorial path very similar to that of viverrids such as Civettictis (fig. 7). This transpromontorial course 1s probably plesiomorphic for Aelu- roidea (Hunt, 1989). The hyaenid auditory region differs fun- damentally from that of Herpestides in the configuration of anterior and posterior bulla chambers. In living and fossil hyaenids (ex- cept Proteles) the bulla’s anterior chamber, formed primarily by ectotympanic, extends backward to cover the posterior chamber. The posterior chamber formed by caudal ento- tympanic remains small in volume relative to the enormous anterior chamber. In fossil hyaenids there is no evidence that the pos- terior chamber of the hyaenid bulla was ever any larger than in living species, and the an- tiquity of this hyaenid bulla pattern extends to very primitive forms (Tungurictis, Percro- cuta) in the mid-Miocene. This hyaenid pat- tern is the antithesis of the Herpestides pat- 30 AMERICAN MUSEUM NOVITATES tern, in which the caudal entotympanic grows forward and becomes the dominant chamber. Hyaenids plausibly originate from an aelu- roid morphotype in which the caudal ento- tympanic was never enlarged and primarily confined to the posterior auditory region. Because the dominant chamber of the hyaenid bulla corresponds to the subordinate chamber of the viverrid-felid bulla, and vice versa, it is evident that these ontogenetic growth patterns represent divergent solutions to the problem of middle ear enclosure. The hyaenid pattern and the pattern identified in juvenile and adult Herpestides lie on inde- pendent developmental trajectories. Thus a relationship of the Aquitanian aeluroid to hyaenids can be rejected on the basis of their distinguishing and contradictory bulla morphs. THE PROBLEM OF HYAENID-HERPESTID RELATIONSHIP The placement of the three auditory bulla elements in living hyaenids and herpestids suggested that the adult hyaenid bulla pattern might be derived from the configuration adopted by these elements in the neonatal herpestid bulla (Hunt, 1987, 1989: fig. 5). This remains a plausible hypothesis open to future testing, however, few other features of herpestid and hyaenid morphology lend sup- port to this view. In my initial analysis of aeluroid carnivorans, some derived traits ap- peared to link hyaenids and herpestids (Hunt, 1987: fig. 19, node 4) but the nature of the evidence for a sister-group relationship was not compelling: (a) restriction of caudal en- totympanic to the posterior auditory region; (b) blunt nonretractile claws; (c) an invagi- nated anal pouch with glands or sacs; (d) an absent or vestigial aural bursa; (e) external auditory meatus prolonged as a bony tube. As an additional complication, herpestids share a large number of autapomorphies that tend to obscure their relationships with other aeluroids. In this context, the morphological similarity of herpestid and hyaenid mallei merits comment. Both families have mallei with a reduced lamina and a short anterior process, contrary to felids and viverrids in which a well-developed pocketed lamina is present. Is this similarity inherited from a NO. 3023 common ancestry, or is this a case of parallel evolution? If one accepts the morphocline polarity presented in figure 10 for the evolution of the auditory ossicular chain in eutherian mam- mals, then a malleus with a reduced lamina represents a derived character state. The hy- pothetical morphocline of figure 10 is sup- ported by the taxonomic distribution of primitive and derived ossicular states among living mammals: the presence of a malleus with well-developed transverse lamina and gonial process occurs in all monotremes, nearly all metatherians, and in eutherians such as tenrecs, erinaceids, soricids, solenodon- tids, and bats; the configuration of the au- ditory ossicles (fig. 1OA) with which such a malleus is associated is generally regarded as the plesiomorphic therian state (Fleischer, 1978). Mallei with reduced laminae (fig. 10B, C) are widely distributed across many orders of mammals, including both caniform and feliform Carnivora. Such a distribution is likely to result from repeated instances of par- allel evolution, in each case directed toward the development of freely mobile ear ossicles having low mass, little stiffness, and reduced frictional resistance (Hunt and Korth, 1980). Consequently, independent evolution of the mallei seen in living hyaenids and herpestids must be entertained as a hypothesis equally as plausible as derivation of these mallei from a common ancestor. Such mallei can be pre- dicted as the frequent product of selective pressures directed toward a more mobile os- sicular mechanism in various eutherian lin- eages. Hence, hyaenids and herpestids cannot be linked solely on the basis of the form of the malleus. Are there any other features of the auditory region that suggest affinity be- tween the two groups? The few species of hyaenids and herpestids that have been available for dissection of the auditory region show similar petrosal morphs. The promontorium appears to possess a characteristic shape, surmounted by a tall, bladelike ventral promontorial process. However, it has not been possible to survey petrosal shape in a large enough sample to determine the taxonomic distribution of this morph, and in the interim there remains a paucity of compelling synapomorphies to unite the two families. Thus, figure 16 depicts hyaenids and herpestids as independent ae- PLEISTOCENE PLIOCENE NANDINIA HYAENIDS HERPESTIDS FELIDS VIVERRIDS ® * e e e - STEM AELUROIDS CENOZOIC MODERN MODERN ARCHAIC ARCHAIC BASICRANIAL BASICRANIAL BASICRANIAL BASICRANIAL PATTERNS PATTERNS PATTERNS PATTERNS INFERRED INFERRED Fig. 16. Evolution of basicranial patterns within the aeluroid Carnivora (modified from Hunt, 1989). The modern basicranial patterns characteristic of the living families emerge during the late Oligocene to early Miocene interval in the Old World, and persist relatively unchanged for the remainder of the Neogene within each family. This revised chart incorporates new information on Eurasian aeluroids discovered since the publication of my initial hypothesis (Hunt, 1989: fig. 13): (a) transfer of early Miocene Herpestides to Viverridae, and (b) indication of a more ancient and independent derivation of hyaenids and herpestids from the stem aeluroid group. Early or mid-Miocene basicrania of modern grade are now known for all aeluroid families except Herpestidae. A, Stenogale julieni; B, Palaeoprion- odon lamandini; C, Herpestides antiquus, D, Tungurictis spocki; E, Panthera onca; F, Civettictis civetta; G, Nandinia binotata; H, Crocuta crocuta; J, Galidia elegans. 32 AMERICAN MUSEUM NOVITATES luroid lineages that may in time prove to be derived from a common ancestry exclusive of that for felids/viverrids. ANTIQUITY OF THE VIVERRID AUDITORY REGION Prior to examination of the Aquitanian samples of Herpestides, my earlier research indicated that the oldest confirmed viverrid auditory regions belonged to Plio-Pleistocene fossils from the Siwaliks of southern Asia and the Langebaanweg locality in South Africa (Hunt, 1989). Other fossils that I was able to study in the British Museum (N.H.) from Oeningen in Europe and from the Miocene Siwaliks suggested that viverrid auditory regions of modern grade possibly extended to the mid-Miocene, but there was no con- firming evidence. Now, however, because the basicranium of Herpestides is essentially that of a modern viverrid, the Lyon and Basel samples extend the record of viverrid basi- crania to the early Miocene (European Neo- gene mammal zone MN2a). We may now anticipate that true viverrids were not only present but probably diverse in the Miocene of the Old World, and that the viverrid au- ditory pattern probably developed in Oligo- cene ancestors. The oldest viverrid auditory region, how- ever, appears to belong not to Herpestides, but to the small Quercy aeluroid Palaeo- prionodon lamandini, whose skull was illus- trated by Teilhard (1915: pl. 9, fig. 10). Ihave not examined the skulls figured by Teilhard in the collections of the Paris Museum, but his photograph indicates an auditory region with large petrosal promontorium, and an ectotympanic applied to the promontorium in the manner of viverrids and felids, where- by the posterior limb rests on the promon- torium just anterior to the round window. Furthermore, although there is no caudal en- totympanic preserved in Palaeoprionodon, it is clear from Teilhard’s figure 10 that a small caudal entotympanic was probably present, occupying the posterior auditory region, and very likely extending forward a short distance along the medial side of the ectotympanic crescent. Comparison of Teilhard’s photo- graph with the skull of the living Prionodon suggests that the Quercy aeluroid had not only a caudal entotympanic of similar size and NO. 3023 form, but that the auditory regions may have been nearly identical. An important aspect of the auditory region of Palaeoprionodon is that the mastoid and exoccipital bones surround- ing the caudal entotympanic are quite thin and narrow in ventral view (Teilhard, 1915: pl. 9, fig. 10), and do not form a wide bony shelf as in the primitive aeluroid Nandinia. This configuration of mastoid and exoccipital reveals a posterior auditory region much like the modern viverrids Prionodon and Poiana. In addition to Palaeoprionodon, there is another Old World Oligocene aeluroid car- nivoran that often has been regarded as a viverrid: the Quercy Stenoplesictis cayluxi. The auditory region of Stenoplesictis, al- though undoubtedly aeluroid, is not certainly viverrid if an assessment is based on the su- perb skull described by Piveteau (1943) from the collection of the Faculté des Sciences, Marseilles. Here the ectotympanic is much more inflated than in Palaeoprionodon, and apparently encompasses the petrosal pro- montorium. This ectotympanic configura- tion would seem to preclude assignment to either viverrids or felids, and it appears to be too strongly inflated to belong to a herpestid. Moreover, the caudal entotympanic chamber must have been relatively small. This sug- gests that the Marseilles cranium referred to Stenoplesictis is either an early aeluroid lin- eage that terminates without living descen- dants, or possibly represents an ancestral hyaenid in which the bulla chambers presage the modern configuration. CONCLUSIONS 1. The auditory region of the Aquitanian (early Miocene) aeluroid Herpestides antiqu- us is of the viverrid type, and cannot be re- ferred to Herpestidae or Hyaenidae as pre- viously believed. 2. Basicranial and auditory bulla patterns of viverrids and felids had evolved by the early Miocene in Europe. The auditory bullae of the viverrid Herpestides antiquus and the felid Proailurus lemanensis, both from the Aquitanian of the Allier basin, France, are essentially of modern grade, and demonstrate that the two families existed as independent lineages by that time. 3. Viverrids and felids exhibit parallels in the pattern of ontogenetic development of 1991 their auditory structure that suggest they are sister groups, e.g., in the relationship of ec- totympanic to petrosal promontorium, the geometry of the petrosal, and the pattern of bulla development. 4. Orientation and morphology of the au- ditory ossicles in aeluroids indicate that these carnivorans display transitional to derived states of the ossicular mechanism relative to a eutherian morphotype. Among aeluroids, viverrids and felids retain the most plesio- morphic patterns, whereas hyaenids and her- pestids exhibit more derived states. Viverrids and herpestids are extremely different in os- sicular morphology, one more item in a steadily accumulating body of evidence in- dicating the marked distinction between the two groups. 5. The hyaenid bulla pattern appears in the mid-Miocene fully developed in at least two distinct lineages of widely divergent adaptive type (Percrocuta, Tungurictis), in- dicating that the origin of the pattern predates the divergence of these lineages, and there- fore is probably of early Miocene age or older. 6. Because aeluroids do not appear in the fossil record until mid- to late Oligocene, fol- lowing the “Grande Coupure”’ event in Eur- asia, and because basicranial patterns of modern grade first appear in viverrids, felids, and probably hyaenids in the early Miocene, the development of these patterns may take place over a relatively brief interval within the later part of the Oligocene. By inference, the herpestid pattern probably also originates during the late Oligocene to early Miocene interval. 7. Aeluroid basicranial types can be iden- tified primarily from the structural relation- ships of auditory bulla, petrosal, and auditory ossicles. These patterns on present evidence must develop within the mid-Cenozoic Oli- gocene interval from about 34 to 24 Ma. Once established, they persist as stable morphs for the remainder of the Cenozoic history of these carnivoran groups. Stasis in patterns of au- ditory morphology characterizes the aeluroid Carnivora during the Neogene. REFERENCES Allin, E. F. 1975. Evolution of the mammalian middle ear. J. Morphol. 147: 403-438. HUNT: VIVERRID AFFINITIES OF HERPESTIDES 33 Beaumont, G. de 1967. Observations sur les Herpestinae (Vi- verridae, Carnivora) de l’Oligocéne su- périeur avec quelques remarques sur des Hyaenidae du Néogéne. Arch. Sci. Ge- néve 20(1): 79-107. Beaumont, G. de, and P. Mein 1972. Recherches sur le genre Plioviverrops Kretzoi (Carnivora, 7?Hyaenidae). C.R. des Séances, SPHN Genéve 25(3): 383- 394. Blainville, H. M. D. de 1842. Des Viverras. Ostéographie ou descrip- tion iconographique comparée du sque- lette et du systéme dentaire des mam- miféres récent et fossiles pour servir de base a la zoologie et a la géologie. Vol. II, Carnivora. Bucher, H., L. Ginsburg, and J. Cheneval 1985. Nouvelles données et interprétations sur la formation des gisements de vertébrés aquitaniens de Saint-Gérand-le-Puy (Allier, France). Geobios 18(6): 823- 832. Cheneval, J.. and M. Hugueney 1985. Excursion 4 la carriére ““Les Pérards”’ (Saint-Gérand-le-Puy; Aquitanien de l’Allier), 21 Septembre 1985. Excursion guide: Table ronde internationale sur lévolution des oiseaux d’aprés le té- moignage des fossiles, 13 pp., 6 figs. Donsimoni, M., and D. Giot 1977. Les calcaires concrétionnés lacustres de l’Oligocéne supérieur et de l’Aquitanien de Limagne (Massif central). Bull. Bur. Rech. géol. min., Paris 1(2): 131-169. Filhol, H. | 1879. Etude des mammiféres fossiles de Saint- Gérand-le-Puy (Allier). Bibliotheque de l’Ecole des Haute Etudes (Sect. Sci. Nat.) 19(1): 1-252. Fleischer, G. 1973. Studien am Skelett des GehG6rorganes der Sdugetiere, einschliesslich des Menschen. Sdugetierkd. Mitteilungen (Munchen) 21: 131-239. Evolutionary principles of the mam- malian middle ear. Adv. Anat. Em- bryol. Cell Biol. 55(5): 1-70. Hugueney, M. 1984. Evolution du paléoenvironment dans le Tertiare de Limagne (Massif central, France) a partir des faunes de mam- miféres. Geobios. Mém. spécial 8: 385- 391. Hunt, R. M., Jr. 1974. The auditory bulla in Carnivora: an an- atomical basis for reappraisal of carni- vore evolution. J. Morphol. 143: 21-76. 1978. 34 AMERICAN MUSEUM NOVITATES Evolution of the aeluroid Carnivora: significance of auditory structure in the nimravid cat Dinictis. Am. Mus. Novi- tates 2886: 74 pp. 1989. Evolution of the aeluroid Carnivora: significance of the ventral promontorial process of the petrosal, and the origin of basicranial patterns in the living fam- ilies. Am. Mus. Novitates 2930: 32 pp. Hunt, R. M., Jr., and W. Korth 1980. The auditory region of Dermoptera: morphology and function relative to other living mammals. J. Morphol. 164: 167-211. Kermack, K. A., and F. Mussett 1983. The ear in mammal-like reptiles and early mammals. Acta Palaeontol. Po- lonica 28(1-2): 147-158. MacPhee, R. D. E. 1981. Auditory regions of primates and eu- therian insectivores. Contrib. Primatol. 18: 1-282. 1987. Petter, G. 1974. Rapports phyletiques des viverrides (Carnivores fissipedes). Les formes de Madagascar. Mammalia 38(4): 605-636. Piveteau, J. 1943. Etudes sur quelques mammiféres des Phosphorites du Quercy. Ann. Paléon- tol. 30: 63-72. Pomel, A. 1853. Catalogue méthodique et descriptif des vertébrés fossiles décoverts dans le bas- NO. 3023 sin hydrographique supérieur de la Loire et de l’Allier. Paris. Qiu, Zhanxiang, Ye Jie, and Cao Jingxuan 1988. Anew species of Percrocuta from Tong- xin, Ningxia. Vertebr. Palasiatica 26(2): 116-127. Schlosser, M. 1890. Die Affen, Lemuren, Chiropteren, In- sectivoren, Marsupialier, Creodonten und Carnivoren des europaischen Ter- tidrs, Part III. Beitrage Paldontol. Os- terreich-Ungarns Orients 8: 401-407. Segall, W. 1969. The auditory ossicles (mallus, incus) and their relationships to the tympanic in marsupials. Acta Anat. 73: 176-191. Morphological parallelisms of the bulla and auditory ossicles in some insecti- vores and marsupials. Fieldiana Zool. 51: 169-205. The auditory region (ossicles, sinuses) in gliding mammals and selected rep- resentatives of non-gliding genera. Fieldiana Zool. 58: 27-59. Teilhard de Chardin, P. 1970. 1971. 1915. Les carnassiers des Phosphorites du Quercy. Ann. Paléontol. 9(3-4): 103- 192. Viret, J. 1929. Les faunes des mammiféres de l’Oli- gocéne supérieur de la Limagne Bour- bonnaise. Ann. |’Univ. Lyon, n. ser., 47: 1-328. Recent issues of the Novitates may be purchased from the Museum. Lists of back issues of the Novitates, Bulletin, and Anthropological Papers published during the last five years are available free of charge. Address orders to: American Museum of Natural History Library, Department D, Central Park West at 79th St., New York, N.Y. 10024. THIS PUBLICATION IS PRINTED ON ACID-FREE PAPER.