See +34 Viteds ite , a Hep aan doe. 8H, Ft Ree) RAR rr $ ay oe yf eo aatiae 4 ttiltitass a oe dyaseiey saat i os ri t “4 4A ae tee Hage teie® rt Re A ts et jen “eh bet. ee ae + ath Satie thy a ewer] ite vod pines aa wlan th i ce ea ae a fs Nohtiaiienttes ! iy P 43 Rian “ bh seh re Aree heed « ie me Tein aa ts hatcaecd ota tc tte ca i neta 4‘: 7 r\ ; sii ti a. fe Yih aes, MA + . », td des ayn it He taps iV yy; Poeersatset he ni " of as ft ae i itis Ha: Sate ait beast Listst Bian sane i inet ot an nt (esha rete i hat ad ne ‘ xt a i is . 7 * » ‘ ett fri lt wt = ~. 4 Se a Bie ——— ‘s 3 “> 3 sae TS a egeansh — aes: 5 A See SS aa ' : ( tnt iM ihe iv i He x me ae ; cara fitee ht itil erie si gee Hers i ae Aq seit Ap Ae RCS TPE EEL } HH ae SARA Sed MC NG GLA NaN LAR a Maka Ai i! Eero cra aA Mb ti th i ite iat ye Niebatp: IS dui beat itatuists i a He ate ae crt Hy Ui ah Pies Pi) i Wun Mi rit Ma ith ney arty iP} un drt 4 RRA TEN SCHR NEEDRS ASDOMILNE AASS een En Ut Sih REM eitaninet masa tye ats a bash : ABtat Rye iat} NE ate ety AD) LEB aS eet nj ‘ Nie ality: iat HH it iti i } 4 f pied aye ‘9 ithiek as Mitte eh) asi} . 4? Hp +9 4 vit? aie tik sh ih ty wih : hula iT y My Pas seedy eit ati 4 ! ‘ia fT Hatt rally i tt isl pas i: ean melt iy ty Sethe 40 SNH . f 4 Abed +* ; j ‘ wy sto tay i ; 24h MED new re Malt St Hates id 33 i ey PERSE M MH eM ataeaR OSC MAES Litt bat Ls a au apn aL aT h ino Pet sores :} ; J Hiab st r i? z 1 + Sis iy’ ohh M4 N tw iy ‘i S05 be thal ayy iy uray i " ; Aah * ne { . te i Hutt tiaests Me dei bend rer gis Ht } ‘ y i i be tt wie Poptart ‘ i a0 +s the eh aye }: Bi , ee re Seetyetr ees = = Festitetess Satats prdstseisites =¥33 ess —< nuh Ab Aitinatae i Pasay Hy Aa aEy bahay: ae ehh ii aii BS Wat y a try » em) f east weiith teal eas a < . a3 Hoetee + 33 2, Sees as sess: sega S53 = 253 Hi sas 335535 eee =3 oh + prtist: ae an “@ JAN 29 B59 GEOLOGY UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGI GEOLOGY Return thi S book on or before the Latest D ate stamped below. GEOLOGy LIBRARY University of Illinois Library MAY 22 MAY 9 1966 OCT 03 1994 OCT 02 W994 L161—H41 7 Pasi aaa . a a 1 — s Pec vu rh: * s me - \ SOME EARLY MIOCENE CARNIVORES “A BY f ELMER S. RIGGS CURATOR OF PALEONTOLOGY, RETIRED SEPTEMBER 15, 1942 , GEOLOGICAL SERIES. FIELD MUSEUM OF NATURAL HISTORY VOLUME 9, NUMBER 3 OCTOBER 4, 1945 PUBLICATION 573 NOTE TO LIBRARIANS The accompanying publication, Early Miocene Carnivores, by Elmer §S. Riggs, is the concluding number in Volume 9 of the Geological Series of Field Museum of Natural History. Volumes 7 and 8 of this series have also been closed, and the indexes, title pages, and lists of Contents of all three volumes will soon be issued. In December, 1948, the name of Field Museum of Natural History was changed to Chicago Natural History Museum. In the future, technical geological publications of octavo size that are © issued by this institution will bear the new name of the institution, and the general title Fieldiana, Geology. They will carry successive numerical designations that are continuous with those of volumes that have been issued in the Geological Series of Field Museum. Geological publications of quarto size will be entitled Fieldiana, Geology Memoirs, and their numerical designations will be similarly continuous with the former Memoirs Series. Technical publications of other Departments in the Museum will be treated in the same manner. 7 + hones 3 : SOME EARLY MIOCENE CARNIVORES BY ELMER §&. RIGGS CURATOR OF PALEONTOLOGY, RETIRED SEPTEMBER 15, 1942 GEOLOGICAL SERIES FIELD MUSEUM OF NATURAL HISTORY VOLUME 9, NUMBER 3 OCTOBER 4, 1945 PUBLICATION 573 THE LIBRARY OF THE NOV 7 1245 UNIVERSITY neg i53.-.... PRINTED IN THE UNITED STATES OF AMERIGA BY FIELD MUSEUM PRESS SOME EARLY MIOCENE CARNIVORES The early Miocene carnivores of the Great Plains region of North America include forms that are referred to four well-known families, the Canidae, Felidae, Mustelidae, and Procyonidae. Members of these families vary widely in numbers and in size and strength. The carnivores of this stage apparently are derived in part from forms that are known to have existed during the preceding White River Oligocene, and in part from stocks that are entirely strange to any known earlier fauna of the Great Plains region. Apparently the latter are immigrants from some remote locality. The sources of such immigrant forms are more or less conjectural. The Canidae of the Lower Miocene stage are relatively abundant and are, for the most part, derivable from stocks that were common to the Oligocene of the White River stage. The lesser canid phylum represented by Pseudocynodictis carries over from the Brule clays of the earlier epoch as a practically uninterrupted line. Likewise, the larger canid phylum represented in the Brule clays by the genus Daphoenus leads almost directly to Daphaenodon, of the Great Plains and the Great Basin regions. Other recorded genera of the early Miocene are less nearly related. The Felidae, represented in the White River stage by three well- known genera, are almost entirely absent from the records of the Lower Miocene in the Great Plains region.” This reduction of a numerous and virile group of carnivores over so large an area perhaps may be attributed to general changes in environmental conditions and also to the appearance of a compara- tively new and sturdy rival group, the large mustelines, rather than to any general elimination of the felids.? The change in the region about the Black Hills from conditions of flood-plain deposition to one of wind-borne sands with occasional lagoon and channel deposits, may well have been accompanied by changes of temperature and of food supply resulting in important animal migrations. 1Seott (1937) points out this sequence in the John Day beds of Oregon; a 7 Setiggs (P14797) from the Gering of Nebraska is evidence of a similar sequence there. 2 Nimravus sectator Matthew (1907), Lower Rosebud, South Dakota, is a definitely identified form; another is reported as a ‘“‘felid.”’ 3 It is well known that at least two genera of Oligocene felids continued in the John Day stage and elsewhere in the Great Basin region. 69 70 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 At the beginning of Miocene time the Mustelidae present a more elusive problem than any of the other families. From the White River series have been described two genera of mustelines, Mustelavus and Bunaelurus,'! both rather advanced in the structure of the sec- torial teeth and in the reduction of the anterior premolars. They also have several basicranial characters that designate them as mem- bers of the musteline family. The Lower Miocene formation has produced a variety of small mustelines, most of them having definite family characteristics, but many of them a little weaker in dental development. The crowns of the molar and premolar teeth are not so high and the accessory cusps are more reduced and often worn away in older individuals. Some of the smaller forms are definitely progressing toward the genus Martes, which has been recognized from the Upper Miocene. There have been reported also a number of intermediate-size mustelines, comparable to the virile stock repre- sented by the holotype of Oligobunis crassivultus? of the John Day formation. Nearly related to the holotype of this genus in structural characters, in relative strength of jaw and dentition, and of a size fitting them to prey upon larger animals, are the sturdy forms that have been referred to the genera Aelurocyon and Megalictis. These animals are too large and too highly specialized to have been derived from the known mustelines of the White River Oligocene; therefore I regard them as immigrants. Of the smaller forms included under the genus Promartes, some nearly complete skulls and a large part of an associated skeleton are described in this paper. Two referred species are discussed. The larger form, Aelurocyon, is described from an articulated skeleton. Procyonids have been recognized from the Lower Miocene since 1899. Two genera have been described, Phlaocyon,*? from Colorado and Nebraska, and Aletocyon,! from the Harrison beds of eastern Wyoming. The first was believed to be, in a general way, ancestral to Procyon and the last not far from the ancestry of Azlurus. Ina recent preliminary paper (1942) I introduced the genus Zodiolestes as a member of the Procyonidae. The holotype of this form, con- sisting of an articulated skeleton, is described in detail here. 1 Bunaelurus Cope (1883); Mustelavus Scott and Jepsen (1936). 2 Various smaller species which recently have been referred to oligobunis are in this paper referred by me to the genus Promartes. 3 Matthew (1899, p. 19). 4 McGrew (1941), on the basis of dental characters, believed Aletocyon to be nearer to the procyonines than to the pandas. EARLY MIOCENE CARNIVORES T1 The drawings used to illustrate this paper are made in projection, either by use of the pantograph or by direct projection. As such, they should be essentially accurate. The scale of reduction, in a few instances, can not be indicated by a simple fraction. In those instances the scale of reduction is indicated by a near fraction and marked as “‘approximate.”’ For exact data, see measurements. I wish to express my obligation to Professor G. W. Bain of Amherst College, to the late Walter Granger of the American Museum of Natural History, to Mr. Leroy Kay of the Carnegie Museum, and to Dr. Claude Hibbard of the University of Kansas for the privilege of studying important specimens in those institu- tions; to Mr. Bryan Patterson of this Museum for the benefit of conclusions derived from the study of the auditory region in carni- vores; to Mrs. Margaret Hough for special studies in the structure of the ear; and to the Department of Zoology of this institution for the continued privilege of using specimens for study and comparison. Drawings used in illustration are by my classmate, the late Sydney Prentice of the Carnegie Museum, and by the late Carl F. Gronemann, Staff Illustrator, and Mr. John C. Hansen, Department Artist, both of Field Museum. Family Mustelidae Promartes Riggs This genus was proposed (1942) to receive certain smaller species of mustelines, some of which had been referred to Oligobunis Cope. Promartes olcotti Riggs Promartes olcotti Riggs, Field Mus. Nat. Hist., Geol. Ser., 8, p. 59, 1942. The holotype of this genus (F.M. No. P15178) consists of a broken skull with mandibles almost entire, and more than half the skeleton. The specimen was collected by Mr. T. F. Olcott, who was a member of a Museum expedition to Nebraska and Wyoming in 1906. An interesting feature of the occurrence was that it was found in association with the jaws of a species of Palaeocastor and embedded in the body of a horizontal branch of a Daimonelix' in situ. 1 The two specimens were recorded by the collector as ‘‘carnivore and rodent.”’ When the collection was being prepared in the laboratory, the rodent was recog- nized and the section of the corkscrew with its contents was labeled ‘‘fragmentary rodent”’ and returned to storage. Many years afterward the carnivore was recog- nized and prepared. 72 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 _ SKELETON The skull of Promartes olcotti (figs. 23, 24) may conveniently be compared with that of the recent species, Martes actuosa. It is some- what shorter and broader than that of the recent forms, the palate is broader, and the basicranium somewhat shorter. The palate is Fic. 23. a, detail view of holotype, Promartes olcotti. f.l.m., foramen lacerum medium; ™.da.e., meatus auditorius externus; f.l.p., foramen lacerum posticum; c.f., carotid fora- men; p.0.p., paroccipital process. )b, lateral view, mandible of the same. more deeply excavated beside the upper carnassial teeth; the poster- ior nares open farther forward; the mastoid process is broader antero- posteriorly, and more prominent. The auditory bullae are well expanded; they are more laterally compressed and their anterior surfaces approach more closely to the posterior margins of the glenoid fossae. There is no inferior lip to the meatus; the sulcus tympanicus is somewhat larger than that of the recent species. A sharp crest extends above the opening of the meatus connecting the posterior margin of the zygomatic arch with the mastoid process. There is a small paroccipital process which is free from the bulla at its extrem- ity. M2 is present but greatly reduced. The mandible is much more massive and deeper in the symphysis than that of Martes actuosa (fig. 23). The dentition in the holotype is worn at the apices of the premolars and carnassial. The first pre- EARLY MIOCENE CARNIVORES 73 molars both above and below should be considered as vestigial; one of the lower pair is absent from the holotype and one of the upper pair is absent from the paratype. P# is tubercular in form and placed closely beside the canine; P2 and P® are strong, functional teeth, increasing in size and closely set together. P+ is a typical musteline carnassial, strong and low-crowned; the protocone is Fic. 24. Skull of paratype, Promartes olcotti. a, lateral view, x 34; b, palatal view, X 1; c, dorsal view, X 1. c mesial and advanced in position. M+ has a long narrow crown set almost transverse to the cranial axis; the second molar is much reduced in size but preserves two distinct cusps. P; is reduced to a small tubercle, present in one mandible and absent in the other. The succeeding premolars are too much worn 74 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 to show minute details; Ps has a well-developed secondary cusp. The talonid of M; is concave, with a slightly elevated lateral margin; Ms retains two distinct roots. In describing the bones of the skeleton it is convenient to use the fisher, Martes penanti, as a basis of comparison. Vertebrae.—The axis, three cervicals, eight dorsals, five lumbars, one sacral, and parts of five caudal vertebrae belong to the holotype Fic. 25. Atlas, comparative series. From above downward: Bas- sariscus, Zodiolestes, Gulo, Aelurocyon. All xX 4. (fig. 25). The centrum of the atlas is not longer than in one lumbar; it is broad and flattened at the anterior end and has a moderate keel on the inferior surface. The odontoid process is relatively short and decurved near the extremity. The cervicals are short and broad; the neural canal is widely open. The centra of the dorsals are rounded on the inferior surface as in the lumbars. The first sacral centrum has an articular surface concave on the superior and convex on the inferior surface; the arch is low and the spine is a mere median crest. Other vertebrae of the sacrum are lacking. Four broken caudal vertebrae offer no evidence of variation from the early marten type. The scapula in P. olcotti (fig. 26) is distinctive. In it, the fossa for the teres major and minor is so far developed as to be separated EARLY MIOCENE CARNIVORES 75 from the fossa of the subspinatus by a secondary spine. This struc- ture is evidently an adaptive character and is found in varying degree in other related species. In the fisher there is only a slight thicken- ing of the axillary margin near its upper extremity. The metacro- mion is apparently placed low on the spine, although the process itself is broken away. The anterior lip of the glenoid fossa is short; the coracoid process is broken at the extreme end. The humerus (fig. 27) is perhaps the most distinctive element in the postcranial skeleton. In this species it is a moderately stout bone, strongly curved in the shaft antero-posteriorly and having a broad deltoid area which covers the anterior surface of the shaft and extends below the middle. The supinator ridge is more promi- Fic. 26. Scapulae: a, Promartes olcotti, XK 14; b, Martes penanti, X 34. nent than that in the fisher, extending along more than a third of the shaft. The inner condyle is likewise more strongly developed than in the fisher and, unlike it, extends below the margin of the trochlea. There is no trace of a postcondylar fossa, a character slightly marked in the recent martens but prominent in Aelurocyon and Megalictis (see pp. 89, 95). The inner condyle has an even greater development in the American badger, Taxidea americana, where it appears to be an adaptation for burrowing. The ulna is equal in length to the humerus, as restored (fig. 27). It has a relatively long olecranon, is narrow antero-posteriorly and much less flattened in its transverse diameter than that of the fisher. The shaft is laterally curved and bears a sharp crest on the mesial surface of the distal end. The radius is likewise curved anteriorly in the shaft; it bears two distinct tubercles at the point of insertion for the tendon of the 76 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 biceps. The articulation for the ulna extends two-thirds of the way around the head, showing that the manus was capable of supina- tion. The fore foot in this specimen consists of a scapho-lunar, four metacarpals, and a number of phalanges, including unguals. The entire foot from the scapho-lunar to the tip of the ungual in the third digit is somewhat longer than the radius of this animal. The scapho-lunar is short in the axial direction of the foot and is strongly Fic. 27. Promartes olcotti, holotype; bones of the fore leg. a, humerus, front view; b, humerus, side view; c, ulna; d, radius. All x 1. convex proximally; the facet for the radius extends down to the margin of the trapezoid. The facets for the trapezium and the trapezoid are distinct. The first metacarpal is much stronger than that of the fisher, being almost as heavy in the shaft as the second and the third metacarpals of this animal. The first phalanges are moderately curved; the second are not modified for retractility. The unguals are hooded but rather narrow and straight. Of the pelvis, only the posterior half of the ilium and the entire pubis with included acetabulum are preserved. These parts, in comparison with recent species, offer no distinctive characters. The femur, a little longer than the humerus as restored, is slender and straight in the shaft (fig. 28). The head and great trochanter are EARLY MIOCENE CARNIVORES 77 inclined forward on the shaft, making a decided curve in the vicinity of the lesser trochanter. This character is not observed in Martes americana or in the fisher, although it is present to some degree in the holotype of P. gemmarosae. The facet for the patella extends but little above the superior surface of the condyles; the latter are rather widely separated. The tibia (fig. 28) is about one-tenth shorter than the femur and is strongly curved in the shaft, laterally as well as antero-posteriorly. The lateral condyle is much larger than the mesial one; it is strongly Fic. 28. Promartes olcotti, holotype; bones of the hind leg. a, femur, front view; b, femur, side view; c, tibia. All approx. x 1. convex antero-posteriorly but slightly concave in the lateral direc- tion. The mesial condyle is concave in both directions. This struc- ture indicates a freedom of rotation in the knee joint. Hind foot.—The astragalus is marked on the proximal surface by a wide, shallow groove, the lateral surface of which is angular and more elevated than the mesial; the latter is rounded and termi- nates anteriorly in a crest which extends along the neck toward the center of the head. The head is much compressed vertically and the capitular facet is indented at the superior margin. As in recent mustelines, the facet for the sustentaculum is advanced to a position 78 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 on the inferior surface of the neck and is connected with the capitular facet by a short crest. The calcaneum presents no distinctive characters other than that the facet for the cuboid is rounded, somewhat oblique to the axis of the bone and that the lateral margin of the facet is more prominent than the small infero-mesial tubercle. The remainder of the foot consists of a fifth metatarsal and some phalanges; the unguals of the fore foot show unmistakable evidence of being hooded. The general characters of this animal indicate a size and strength considerably greater than that of Martes americana. The skull is 7 mm. longer and is much broader in the facial region. The leg bones MEASUREMENTS (In millimeters) No. P15178, holotype SKULL Length of upper dental series, canine to last molar..... 34.3 Breadthacross crowns-of first.molars... «54.70 e ee 32 LOWER JAW Bia len gt iron rok ole ok ae eae ee RO Meee 59.8 Length of lower dental series P;-Ms................. 32.7 Lower second: molar'to: condyles oat, mata ose ce ook 19.2 Height of coronoid process above inferior margin....... 27 SCAPULA Greatest length: 453: ie Bee a ae ee ke 52 HUMERUS Breadth distalcender.c. oes aot eee ee ns 2 aes 20 ULNA Igengthtover alll ts- icc a spade se a ne ee ere de 70 Length of olecranon from lip of sigmoid notch......... 12 RADIUS Greatest:axiallength=. i. she ee rt cer tee ae 52.5 Ischium, length from margin of acetabulum........... 21 FEMUR Length, head to margin of inner condyle.............. 73.3 TIBIA Greatest. lengthy cits get th on a a eee 64.2 ASTRAGALUS Greatest:lengthi:5. 500 638 eal ne eta eee ak 13 CALCANEUM Greatest leneth ss aa oa a a See eee 18 Length metacarpal Tit...) yc oes en te be ea eae be 19 Lengthametatarsal ill yon e mead (eet eee ee ee 26 EARLY MIOCENE CARNIVORES 79 are perhaps one-fourth stronger, the scapula and the pelvis are cor- respondingly stronger. The development of the deltoid area and the supinator and pronator attachments of the humerus indicate an arboreal animal; the great area for insertion of the teres major especially points to arboreal habits. This character I have found in similar development only in certain species of marmosets, and to an even higher degree in the great sloths. The genus Promartes appears to be ancestral to Martes, species of which are known from upper Miocene to Recent time. The transition might be accomplished by such changes in the dentition as elimination of the vestigial M? and broadening of the mesial lobe of M+. This change would be possible from the fossil form P. alcotti to such recent species as M. americana, which has a cor- respondingly reduced premolar series; in M. penanti the first pre- molar is actually stronger than in the fossil species cited. Other characters offer no obstacle to such a transition. Promartes lepidus Matthew Oligobunis lepidus Matthew, Bull. Amer. Mus. Nat. Hist., 23, p. 194, 1907. Horizon.—Lower Rosebud, Lower Miocene, South Dakota. This species was described in 1907, based upon the facial region of a skull with associated mandible and nearly entire dentition. The species was referred to Oligobunis Cope, this in turn based upon an incomplete skull and mandible. The Field Museum collection includes a posterior half of skull, P12155, with one sectorial tooth and an associated pair of mandibles with dentition complete excepting the incisors. The locality is Raw Hide Butte, Wyoming, and the horizon apparently the upper levels of buff sand, designated by Peterson as Upper Harrison beds. This specimen agrees with the holotype of P. lepidus in so far as they have parts in common. The Field Museum specimen is therefore referred to this form, and the species is transferred to Promartes. From this specimen and from the holotype, which I have been permitted to examine, the following specific characters are derived: Length of skull in specimen P12155 (estimated), 95 mm.; cranial index (estimated), 31.5; M2 tubercular and one-rooted; bullae well expanded and paroccipital process closely applied to it; lower dental series but little curved laterally and interlocking with the upper series. Py present but vestigial and slightly displaced mesially from the dental series; Pz and Ps have basal cingula on the posterior margins, Pz has the posterior accessory cusp. In My the talonid is 80 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 concave at the crown with lateral margin slightly elevated; the apex of the paracone is slightly lower than that of the metaconid; Mz is a functional, bilobate tooth. The mandible is 65 mm. in length, and the dental series 41 mm. in length. The ramus is appreciably stronger than that of P. olcotti, the dentition correspondingly stronger. ‘The coronoid process is broad but does not overhang the condyle as in Oligobunis crassi- vultus. Other differences are that the teeth are in nearly straight line and are not set oblique to the axis of the ramus. Promartes vantasselensis Loomis Oligobunis vantasselensis Loomis, Amer. Jour. Sci., (5), p. 321, 1932. Horizon.—Harrison beds, Lower Miocene, Wyoming. This species is based upon a skull and mandibles almost entire. Through the courtesy of Amherst College the holotype of the species Fic. 29. Promartes vantasselensis. Skull of holotype refigured; x 1. was lent for study (fig. 29). In the interest of accuracy it has been refigured.!. The skull is very similar in proportions to skulls of recent American martens, notably M. americana and M. caurina. Com- parison will be made with the latter species. It has a similar out- line as seen from the dorsal aspect, a similar though slightly more elevated occiput, and a similar lateral view, including the charac- teristic upward curve of the zygomatic arch in the martens. The length and the general configuration of the basicranial region are similar; the length of the dental series and of the temporal arcade are almost identical. The bullae are well inflated but vary in outline 1 Attention should be called to inaccuracies in the illustrations which accom- panied the type description. The upper sectorial and P2 and P® have crowns of moderate length. The scale of the figure of skull in side view should be given as a little more than 5/4, and the scale of the figure of upper dentition should be given as 2/1. See measurements, op. cit., p. 322. EARLY MIOCENE CARNIVORES 81 from the recent form; the position of the posterior nares cannot be determined in the fossil; the structure of the internal ear has been studied only in part. The floor of the meatus, which is extended into a lip in the recent species, cannot be determined from the fossil. The structure of the internal ear shows differences, but nothing that may be considered important. The molar-premolar dentition is little worn; the premolars and sectorial are more pointed, but not essentially different from the dentitions of the smaller species of the genus, especially P. olcotti. The ramus of the mandible is stronger in the fossil than in the recent species of martens. The masseteric fossa is similar in outline but deeper in the fossil; the coronoid process is equally elevated but broader in the crest, the posterior margin is slightly recurved. In the present knowledge of the fossil species, it cannot be said that P. vantasselensis is directly ancestral to any species of living martens but the general structure of the skull and the relative developments in a number of variants as cited above, lead to the conclusion that this species is in a general way ancestral to the smaller American martens. Also, the general structure of the skull and the dental formula indicate some relationship to Bunaelurus of the Oligo- cene, but the structure of the upper sectorial and of the first molar, as well as various characters of the basicranium, bar it from close relationship to that genus. It is much more closely related to Pro- martes olcotti. Promartes gemmarosae Loomis Oligobunis gemmarosae Loomis, Amer. Jour. Sci., (5), p. 317, 1932. Horizon.—Lower Rosebud, Lower Miocene, South Dakota. This species was described from a nearly entire but imperfectly preserved skeleton. The same difficulties that were encountered by others in undertaking to classify early mustelines and procyonids on the basis of dental characters apparently were realized by Loomis. The specimen is not well preserved and at the time it was described had not been removed from the slab of matrix where parts of it were concealed. Such studies as I could make, aided by further prepara- tion, have led me to transfer the species to the genus Promartes. The length and slenderness of the skull are characters common to this species and to Oligobunis darbyi Thorpe. The unusual length of the legs in this species is a marked character, but cannot be con- 82 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 sidered as of more than specific value. The structure of scapula and of pelvis is typical of the early martens. Treating this species as a viverrine on the basis of comparison with the dentition of Herpestes does not appear justifiable. Some errors appear in the figures published with the type descrip- tion. A new figure of the scapula (fig. 30) shows the suprascapular area as triangular, and that a prominent process for attachment of Fic. 30. Promartes gemmarosae, holotype. a, fore foot, x 2; 6, scapula refigured, approx. X 34. the teres major is present and extends along the distal portion of the axillary border. The fore foot is refigured in order to represent more accurately the arrangement of the carpals (fig. 30)... Measurements were given in the type description. Oligobunis darbyi Thorpe Horizon and locality —Monroe Creek beds, Lower Miocene, Pine Ridge, Nebraska. This species was described from a skull and jaws in Peabody Museum, Yale University. A cast of the specimen together with published figures shows the following characters: Skull long and nar- row, facial region short, cranial region elongated; Pz, P+ and Mz reduced to vestiges; Mz a functional tooth reduced in size and well worn in this specimen; bullae expanded, oblique in position, stylo- mastoid process closely applied. The tympanic region shows no distinctive characters, the teeth appear to be somewhat stronger 1T regret that a more detailed study of this skeleton cannot be given in this connection. This could be done only by removing it from the matrix. EARLY MIOCENE CARNIVORES 83 than those of Promartes; the coronoid process is recurved and over- hangs the condyle. Some of these characters appear to ally the species with the mustelines of larger size, such as Oligobunis cras- sivultus. Definite determination of the relationships must await a study of the tympanic region or some knowledge of the skeleton. Aelurocyon Peterson Genotype Aelurocyon brevifacies Peterson. The holotype is an incomplete skull and mandible with associated parts of the skeleton. The dentition is nearly complete but the nasals and basicranium are missing. No classification of this genus was made at the time of description, although it was compared with the mustelines, Gulo and Mellivora. In a later paper (1910) the genus was referred by its author to the Mustelidae. So far as this writer knows, no further information on this genus has been published and no other species has been referred to it. Aelurocyon brevifacies Peterson Aelurocyon brevifacies Peterson, Ann. Carnegie Mus., 4, p. 68, 1906. Three specimens collected by the Field Museum Expedition of 1906 have been referred to this species. The locality at which they were found is the north fork of Raw Hide Creek (J. M. Creek) about two miles above its junction with the west fork of that stream; the horizon apparently is Harrison beds. Credit for collecting these specimens is due to the late John B. Abbott, member of the Field Museum of Natural History Expedition of 1906 and long a member of the paleontological staff of the Museum. The first and most important specimen is an articulated skeleton (P12154) lacking the anterior half of the skull and lower jaw, most of the feet, and parts of other bones. The second is a pair of lower jaws (P12283) having the condyles, the angles, and the symphysis weathered but the rami, coronoid processes, and dentition preserved in their natural relations. The third specimen (P12152) consists of a weathered mandible with two teeth in position, together with three other sectorial and two canine teeth, the shafts of various leg bones, an os penis, and a number of metapodials and phalanges. The speci- mens were recorded as Aelurocyon species and the skeleton was long exhibited in a slab mount with parts in position as found. As no complete skull was known, identification had been regarded as uncertain. Recent comparison of the jaws with the holotype of 84 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 Peterson’s species, made possible by removing the skeleton from the original slab of matrix, has convinced this writer that the series of specimens is of the same form and should be referred to A. brew- facies. The following characterization of this genus is based upon the holotype of A. brevifacies, lent to this museum for study, and upon the specimens cited above: Mustelines of relatively large size; upper and lower jaws massive; sagittal crest prominent; dentition It, Ci, P34, M?; Pi and M2 vestigial; lower carnassial with reduced internal cusp;' coronoid process recurved and overhanging the condyle; otic bullae moder- ately inflated; paroccipital process free from bulla and prominent; superior arch of atlas elevated and overhanging the condylar facets of same; scapula with axillary process; humerus with strong deltoid crest and inner condyle extending below the trochlea; inner condyle of humerus excavated posteriorly by postcondylar fossa; metacarpus and metatarsus short and stout, unguals not preserved. Vertebral formula: 7, 14, 6, 4, ?; ilium convex on lateral surface, inferior border massive; os penis uniformly curved and exceeding femur in length. SKELETON The skull of this specimen (P12154; fig. 31) consists of the posterior half, including the cranium, basicranial region with bullae broken, the base of the right arch and the right mandible as far as the fourth premolar tooth. The parts common to the two specimens have a similar structure, and the measurements, as far as they can be compared, agree quite closely with those of the holotype of this species. The cranium is long and rather low; the sagittal crest is narrow and sharp, extending forward to a point opposite the middle of the temporal arcade. The mastoid process is prominent and laterally directed; the auditory bullae, preserved in part only, appear to have been rather broad and low asinGulo. There is no alisphenoid canal. The paroccipital process is prominent, extending backward to a point opposite the lateral extremity of the condyle. The auditory meatus was apparently extended into a rather elongate tube, though the extremity of that structure is broken. The postglenoid process firmly encloses the condyle of the mandible at its mesial extremity. The basioccipital region is broad and flat, showing no evidence of a median keel. ‘This character is described by Peterson (1910) as reduced. The teeth of specimens in Field Museum are too much worn to show this structure. EARLY MIOCENE CARNIVORES 85 The mandible in specimen P12154 includes only the posterior half of the right side with Mz in position (fig. 31). The condyle is elon- gated transversely and but little elevated above the angle. The latter is produced into a slender point which extends beyond the condyle; Fic. 31. Aelurocyon brevifacies. a, restored skull and mandible, No. P12154; b, right mandible of pair, No. P12283. Both xk \%. the coronoid process is broad and its posterior margin overhangs the condyle. The masseteric fossa is deeply excavated and bounded anteriorly by a sharp ridge which extends upward as far as the crest. Mandibles No. P12283, length approximately 145 mm., are very similar in structure and retain most of the teeth (fig. 31). Py is not preserved in any specimen though it may have been present; Ps is likewise missing from all of the Museum specimens though present in the holotype. The canine is strong but much worn at the crown. Ps and Pz are worn at the crown; a vestigial paraconid is present on Pz. Mrz is likewise much worn and present in left ramus only. 86 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 Vertebrae.—The vertebral column, as found, was articulated and continuous from cranium to sacrum with most of the ribs in position. The vertebrae had suffered somewhat from being dissolved by perco- lating waters and in places from weathering at the surface. Missing parts have been reconstructed by comparison with other mustelines and in part with the canid, Daphoenus vetus, of similar size. The vertebral column as a whole is somewhat longer than that of Daphoe- nus; the lumbar region is relatively less strong (fig. 37). The atlas (figs. 25, 32) is similar in size to that of the gray wolf, Lupus nubilis. Its structure is nearest to that of Zodiolestes. The superior arch is more massive in proportion to size than that of Gulo. There is a rugose prominence at the crest and an emargination of the anterior border similar to that observed in the atlas of Zodiolestes and in Felis concolor. The arch projects well above the margins of anterior articular facets as in the latter form; at the median line it projects forward beyond them. The neural opening is almost cir- cular; the inferior arch is narrow as in Gulo. The anterior external foramina open laterally into wide fossae. The margin of the lateral process is notched for the passage of the vessels but not bridged over to form a foramen. The inferior surface of the root of the lateral process is excavated by another large fossa which gives entrance for the arterial canal. The canal passes obliquely upward and back- ward and emerges on the superior surface beside the margin of the posterior facet. The structure of the superior arch and the position of the foramina are similar to those in the holotype of Zodiolestes. In the axis of this specimen, only the arch with postzygapophyses, the greater part of the spine and the odontoid process are preserved. The spine is high and relatively short, overhanging the postzyga- pophyses and terminating anteriorly in a thickened tuberosity. The odontoid process is rather short and truncated (fig. 32). The cervical vertebrae are known from molds in the matrix and from the fourth, which was preserved. The centra were broader than long and moderately concavo-convex at the articulations. The inferior surface of the fourth centrum is marked by a median and two lateral keels. The lateral process in the sixth is well rounded in the shaft but little recurved distally. The spine on the seventh, as shown by a fragment in the matrix, was rather high and tapering. The first and second dorsal vertebrae have these transitional characters: The anterior zygapophyses in the first are widely spaced as in the cervicals; they face obliquely inward though presenting almost plane articular surfaces. The posterior pair are a little more closely EARLY MIOCENE CARNIVORES 87 apposed and are plano-convex in form, embraced by the plano- concave facets of the succeeding vertebra, in a manner similar to the articulations of the lumbar vertebrae. In the second dorsal vertebra the transition from cervical to dorsal type of articulation is com- pleted, the posterior zygapophyses having the usual closely apposed and essentially plane facets. A similar transition is observed in a specimen of Gulo, but in Taxidea the facets between these vertebrae are essentially plane. In Zodiolestes, in which the structure of atlas is similar to that of the specimen under description, the transition from cervical to dorsal type of articulation is accomplished in the first dorsal vertebra. The transverse process of the first dorsal is strong and subtriangular in section; the facet for the rib articulation Fic. 32. Atlas of Aelurocyon brevifacies, No. P12154. a, anterior view; b, posterior view. Both x 4. is concave and directed obliquely outward and downward. The capitular facet covers the antero-lateral half of the centrum. The length of spines in the anterior series of the dorsal vertebrae cannot be determined from these specimens. The anticline is at the twelfth vertebra. The spines from the thirteenth dorsal to the fifth lumbar are broad and inclined slightly forward. The zygapophyses throughout the lumbar series are widely spaced and strongly inter- locking. Such lumbar centra as are preserved, notably 4 and 6, are smooth on the inferior surface and much broader than long. No reliable evidence as to other processes can be derived from this specimen. The sacrum is so poorly preserved as to give rise to some uncer- tainty. It has been restored as having four vertebrae in its composi- tion. Parts of four vertebral centra are imbedded in the original matrix. The first and second centra, as shown (fig. 37), have parts missing, but it does not appear possible that they could be combined to form one centrum; the third and fourth centra are evident. Other parts preserved are one anterior and both posterior zygapophyses, two spines, and parts of the lateral masses of the first sacral vertebra. Of the caudal series, parts of only two arches remain. 88 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 There are fourteen pairs of ribs. Between the right and left sides, sixteen rib heads are preserved, from which the structure of ten can be determined. The first rib is relatively strong, with the head set at an angle of 82 degrees to the upper margin of the shaft. The tubercular facet extends well over on the posterior surface. The scapula is relatively long and narrow as compared with other carnivores of equal strength (fig. 33, a). The spine extends obliquely Fic. 33. Aelurocyon brevifacies, No. P12154. a, scapula; 6, humerus, lateral view; c, humerus, anterior view. All x 4. across the suboval blade; a prominent process for attachment of the teres major muscle arises from the upper third of the axillary border, forming a broad process. A rather strong but simple acromion process projects over the proximal articulation. The glenoid facet is broad and suboval in outline, the margins forming a rounded angle antero-laterally. The coracoid process is reduced to a tubercle sur- mounted by a short, medially inflected hook, as in Gulo. The EARLY MIOCENE CARNIVORES 89 posterior margin of the blade turns outward in the middle third of its length and continues in the upper third to form a secondary spine, separating the axillary process from the subscapular fossa. This spine is almost as prominent relatively as that in Promartes olcottt. The pelvis is unfortunately incomplete (fig. 34). The ilium is distinguished by having the lateral surface of the blade uniformly convex. The inferior margin of the blade is thickened and massive, meeting the lateral surface in a decided angle; the tubercle for attach- ment of the rectus femoris is unusually prominent. The ischiatic Cond. fossa a b Fic. 34. Aelurocyon brevifacies, No. P12154. a, humerus, posterior view, distal third in detail; 6, pelvis in lateral view. Both x 4. tubercle has been restored in this specimen from related forms but. was apparently deflected laterally, giving to the ischium a strongly concave outline as seen from above. The pubic arch is broken away close to the acetabulum. The humerus of this form is highly characteristic and deserves a detailed description (fig. 38, b,c). The right humerus is almost entire and but little distorted; the left is somewhat weathered at the proxi- mal end but has been restored from the opposite one. This bone is rather short and stout, strongly curved in the antero-posterior direc- tion and well expanded at the distal extremity. The great tuberosity does not extend above the head; the lesser tuberosity has the form of a crest, oblique to the direction of the shaft and similar to that of Gulo. The deltoid area is broad at the proximal end, extending well over on the lateral surface. The crest is strongly marked and 7 ‘90 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 extends over the entire middle third of the shaft. It consists of two diverging, curved, and rugose lines which join in their lower exten- sion to form a single crest. These attachments for the deltoid muscles are relatively stronger than those observed in the felid Hoplophoneus of the Oligocene, or in Taxidea. The attachment for the triceps muscle is marked by two rugose fossae on the lateral and the mesial surfaces of the shaft posterior to the greater and lesser tubercles, respectively. The antero-superior boundary of the lateral fossa is marked by an oblique crest; that on the mesial surface -excavates the posterior surface of the lesser tuberosity. The supinator ridge extends above the lower extremity of the deltoid crest as in Gulo, but is relatively less prominent than in Taxi- dea. The inner condyle is prominent and rugose, extending below the trochlea as in Promartes. The entepicondylar foramen is oval in outline as is common among mustelines, having its center opposite the superior margin of the trochlea. The epitrochlear fossa is deep but not perforate; a second recess, which for lack of other name may be designated as the postcondylar fossa, excavates the inner condyle opposite the superior margin of the articular surface, as may be seen to some extent in Taxidea and in smaller martens of recent age (fig. 84). In addition to its much greater size, the humerus of Aelurocyon differs from that of Promartes in the greater extension and the lateral expansion of the deltoid crest and in the presence of the postcondylar fossa. It differs chiefly in size from a larger specimen which will be described below as belonging to a species of Megalictis. Aelurocyon is clearly a burrowing animal as may be seen by comparison of the humerus and the scapula with the well- known recent burrowing form, Taxidea taxus. The ulna of this species has a somewhat greater length than those of Daphoenus vetus or Hoplophoneus primaevus, and is about one- fourth stronger (fig. 35). The olecranon is but little longer than the breadth of the great sigmoid notch; it is strongly inflected as in Gulo and the inferior margin is turned inward in the form of a prominent tubercle. The styloid process bears a strong, rounded terminal facet; that for articulation with the radius is small and elevated. The tubercle for interosseous ligament is elongated and terminates well above the facet. The radius is rounded in the shaft and well expanded at the articular ends. The head is somewhat flattened antero-posteri- orly; the contact with the ulna is outlined on the lateral surface by a low ridge, indicating a limited capacity for rotation. The insertion for the supinator tendon is likewise but faintly marked. The distal EARLY MIOCENE CARNIVORES 91 facet is broad antero-posteriorly and laterally placed, leaving a large meso-terminal surface for ligamentary attachment. Fore foot—Only the scapho-lunar and metacarpal II are pre- served. The metacarpal is a short, stout bone as measurements will indicate (26.0 mm.), and deeply concave at the proximal articulation. Fic. 35. Aelurocyon brevifacies, No. P12154. a, lateral view of ulna and radius; b, anterior view of femur; c, anterior view of tibia and fibula. All approx. x 4. Femur.—In this specimen both femora are preserved but neither has the great trochanter or the lateral wall of the digital fossa entire (fig. 35). The femur is somewhat longer than that of a male specimen of Daphoenus vetus and is stronger in the shaft and broader in the extremities. The head is appreciably larger in diameter; the pit for the ligamentum teres appears on the postero-mesial surface of the head somewhat as in Taxidea, rather than on the mesial surface as in the Oligocene canids and felids. The lesser trochanter is conical in 92 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 form and is directed posteriorly; a rugose line extends from it obliquely down the shaft. The same position of the lesser trochanter is observed also in the femur of the contemporary Zodvolestes; in the smaller Promartes olcotti the trochanter is directed postero-mesially as it is in Daphoenus and in the recent Cryptoprocta. Linea aspera are not well defined on the femur of Aelwrocyon nor are they found as well marked on other Miocene mustelines as they are in the Oligo- cene Daphoenus. This may be due to imperfect preservation. The condyles are equal in size and the intercondylar notch is relatively narrow; the facet for the patella is short and broad as in the short- legged machairodonts. The tibia is appreciably shorter than that in a specimen of Daphoenus, but broader at the articular ends and heavier in the Fic. 36. Aelurocyon brevifacies, No. P12154. a, astragalus, superior view; b, caleaneum. Both x 4. shaft (fig. 85). The cnemial crest is prominent; the surface immedi- ately below isrounded. At the middle of the shaft and on the antero- lateral surface there is a slight rugosity, observed also on the tibia of Promartes olcotti but not in carnivores of other families. The internal malleolus is short and truncated; the distal facet does not cover the entire end of the bone as in the Oligocene canids, but leaves a wide margin for ligamentary attachment. The fibula is of equal strength to that of Daphoenus. An interest- ing feature of this specimen is that the right tibia and fibula had been fractured during the life of the individual. The tibia in the lower fourth had thrust past at the fracture and had partly healed but a permanent exostosis remained. The fibula had fractured in the shaft above the middle but had successfully healed and the external malleolus was left projecting beyond the astragalus. As a result, the astragalus and the caleaneum had become diseased and in part atrophied. The left astragalus has lost the head by weathering (fig. 36). This bone differs but little in proportions from that of Daphoenus. The trochlear facet does not extend quite around the proximal end, indi- cating that the animal was subdigitigrade; the head is deeper in the vertical dimension and its facet is joined with the facet for the sus- tentaculum by a narrow bridge, as observed in Zodiolestes. 9/1 X faaoqe [lejep Ul painsy se ‘EgzZZId PUB FEIZId SUOUTIDGdS WOL] PolOysel UOJ@[aYG “savovfuaiq Uohivoinjay “LE “O1T Peewee. oves + seed ioe 93 94 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 Of the calcaneuwm, only the diseased bone is available for descrip- tion (fig. 836). Itis short and stout, presenting a massive and rugose extremity for attachment to the tendon of achilles and a rounded and concave facet for the cuboid. Metatarsals I, II, and III are preserved. The first is much reduced in size; length 29 mm. The proximal end is enlarged and bifurcate posteriorly, presenting a small facet for Mt. II. The latter is a stout bone but so broken at the anterior end that it offers no charac- ters for description. Mt. III is somewhat longer (length 48 mm.) than II, stout in the shaft and oblique in the anterior articulation. MEASUREMENTS (In millimeters) HUMERUS AKialvlengthies wets hates io Leora Oh a aR yRe rere kt ee hp eh ee 178 ULNA TON SEN oe. 8 phoresis 180 RADIUS Ten thy seo Van yen ne ee Seg ee espe ee era: 142 FEMUR Greatestilengthias restored... sume ee ee 201 Breadth across:condy lesen ee ee eee 41 TIBIA Greatest: lengthieag cee ts eee cet eo cauaeiny rene 3 169 FIBULA TON Sth 9 oot ees gs ae eer ee aE ES ee nes 156 CALCANEUM TONS CDA eet OS tas Med ane Shae Are A Rn tee, & Aen 54 Megalictis Matthew Megalictis ferox Matthew Megalictis ferox Matthew, Bull. Amer. Mus. Nat. Hist., 5, pp. 175-204, 1907. This is the largest known form of the Mustelidae. The genus is based upon two specimens designated by Matthew as holotype and paratype. The first includes a fragmentary skull with upper sectorial and upper and lower molars. The paratype includes the humerus, the tibia, and most of the fore and hind feet. A fragmentary specimen in the Field Museum collections (P121385) includes the proximal and distal ends of the humerus and the femur with anterior half of the ulna and a number of vertebral centra. These make it possible to add some characters to the knowl- edge of this great musteline. EARLY MIOCENE CARNIVORES 95. The humerus, as restored and figured, is somewhat longer than that of the paratype (fig. 38). The head and the great tuberosity are equal in height; the bicipital groove is broad and shallow as in Aelurocyon. The attachments for the triceps muscle consist of deep pits, bounded superiorly by two sharp crests which continue back- ward from the inferior margins of the capitular surface. The distal end of the humerus is most characteristic, as in other mustelines. The mesial condyle extends below the trochlea in a rugose promi- nence. The olecranal fossa is moderately deep, and extending from it mesially is the postcondylar fossa. The latter opens downward along the margin of the articular surface and is not so deep propor- tionately as in a specimen of Aelurocyon described above. The anterior fossa is broad and shallow; the supinator ridge is more. rounded and less prominent than that in the smaller form. The ulna is known from the anterior half only (fig. 38). There are no important differences between this bone and that of Aelwrocyon except size. The femur, so far as may be determined, has general charac- teristics similar to those of Aelwrocyon (fig. 38). The head is well rounded, with a pit for the ligamentum teres placed near the center of the articular area, but opening downward. The great trochanter rises to a level with the head; its lateral surface is broad and rugose to a point below the level of the lesser trochanter. The digital fossa excavates the posterior surface of the great trochanter, opening - upward in a narrow cleft. The lesser trochanter is relatively small. The condyles are almost equal in size and are separated by a narrow intercondylar notch. The surface for the patella is broad and does. not extend above the superior margin of the condyles. This specimen was found in a small residual deposit of the Upper Harrison beds, as described by Peterson, lying directly upon an eroded surface of the Daimonelix beds. Allowing for a consider- able interval of time between the two formations, these two large mustelines, Megalictis and Aelurocyon, might have been succes- sors in the same line. The rapid development of these larger forms, apparently from animals of such type as Oligobunis crassi- vultus of the John Day beds or Paroligobunis of the Lower Miocene of Nebraska, indicates an unusually rapid expansion in this line. The great strength and the apparent voracious nature of the beasts may account for their supplanting the sabertooth cats in the Great Basin area during Lower and Middle Miocene time. Their disap- pearance was apparently as sudden as their development was rapid. 96 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 Upper Miocene deposits have yielded no evidence of these sturdy and destructive invaders. MEASUREMENTS (In millimeters) No. P12135 HUMERUS Breadth across head and great tuberosity............. 63.5 Greatest: breadthjof distal:ends. 3. jn. 80 oe eee 67.0 FEMUR Breadth: across COndViesige oe, ae ee eee 56.5 ULNA Length of olecranon beyond margin of sigmoid notch, PLOJECUON ee Aaa ee eee nae toes 44.0 Family Procyonidae Three genera of Lower Miocene carnivores have been referred to this family. These genera may be divided into two groups: (1) Forms having broad, multicuspidate molars, imperfectly differ- entiated sectorials, and an alisphenoid canal include Aletocyon and Phlaocyon. (2) Those with narrow-crowned molars, well- differentiated sectorials, and no alisphenoid canal include Zodiolestes and are most nearly related to the living Bassariscus. Zodiolestes Riggs This genus of carnivores was named and described in a pre- liminary paper (Riggs, 1942) published by this Museum a short time ago. The specimen upon which it was based has been in the Museum collections for a number of years and has offered a problem in classification. Its dental formula and its tooth structure seemed to place it in the Mustelidae. The full premolar dentition, with functional first pair, and the longer facial region were noted as incon- sistent with known mustelids but were taken to indicate a somewhat less specialized development than species described from the Lower Miocene or those now included in the genus Promartes. As more complete specimens made it possible to determine the basicranial structure in the smaller mustelines of the period, it became evident that the relatively broad and short basicranium of the doubtful specimen could not be aligned with any of those species. Similar distinctions barred it from any grouping with known canids of the Oligocene or the Lower Miocene. Careful comparisons were made with the recent genus Cryptoprocta, and again the _ basicranial Cc d e Fic. 38. Megalictis ferox, No. P12135. a, b, proximal end of ulna; ¢, anterior view, d, lateral view of restored humerus; e, anterior view of restored femur. All X 2/5. 97 98 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 region was found an obstacle. However, many characters of the two postcranial skeletons proved to be very similar. In this difficulty, one of my associates in the Museum! suggested that a study of the inner ear be made. This was done, comparing the ear-structures in a series of carnivores both recent and fossil, and the conclusion was reached that the ear was essentially that of a procyonid. While known fossil specimens of this family are rare and only a skull of Amphianasua was available from South American fossil faunas, the Museum was fortunate in having good comparative material of the recent genera and many points of similarity were found with Bassariscus, in skeleton as well as in the cranium and particularly in the ear. Surprising as this conclusion is—finding an almost typical mustelid dentition in association with a procyonid basicranium and ear—the study has been so thoroughly done that the result may be accepted as offering little room for doubt. An added consideration is that the cephalo-cranial index is identical with that of such primitive canids as Pseudocynodictis and close to that of the procyonids. This affords good reason for removing the holotype of Zodiolestes from its former determination as a mustelid. Also, the structure of the deltoid crest and the development of the inner condyle in the humerus of this form are more nearly like those of Cryptoprocta and Bassariscus than any of the known Miocene mustelids. For convenience, the generic characterization of Zodiolestes, published in an earlier paper as referred to above, is repeated here. Small carnivores, size and proportions similar to those of Cryptoprocta. Denti- tion mustelid-like; formula 3, +, 4, $, all teeth functional; parametacone and meta- conid of carnassials moderately developed . . . sagittal and lambdoidal crests promi- nent; alisphenoid canal absent; basicranium broad and short. Tympanic region procyonid-like, agreeing with living genera in structure and especially in the presence of a fossa in superficies meatus of squamosal and in flooring of medial part of fossa muscularis major by periotic; differing in that tympanic does not form a bony external auditory meatus; paroccipital process closest to that of Bas- sariscus but stouter and more posteriorly directed. To the above it may be added that the scapula has an axillary process in the superior fourth of axillary margin; the humerus has a simple but elongate deltoid crest, inner condyle not excavated by a second posterior condylar fossa and not extending below the trochlea. The following description of the auditory structures in this genus is given by Margaret Hough (1944): 1 Mr. Bryan Patterson, Assistant Curator of Paleontology. Fic. 39. Skull and mandible of Zodiolestes daimonelixensis, No. P12032. a, lateral view, approx. X 9/10; b, palatal view, x 9/10. 99 100 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 Carotid canal formed by the tympanic along the medial wall of the bulla; a foramen stylomastoideum primitivum is present and the facial canal must have been ligamentous in its external third; tympanohyal lodged in a groove on the surface of the bulla. Relationships.—There is no doubt whatever that Zodziolestes should be classified with the Procyonidae on the basis of the ear region. Zodiolestes daimonelixensis Riggs Zodiolestes daimonelixensis Riggs, Field. Mus. Nat. Hist., Geol. Ser., 8, p. 59, 1942. This species was named for the peculiar spiral form Daimoneliz, in which the holotype of this species was found embedded. The skeleton was found coiled about in a lifelike position at the middle of the spiral. Clearly this spiral form was, at the time the animal entered, an opening in the sand in which the animal found harborage (figs. 44, 45). SKELETON In structure the holotype of this species has no close parallel among known mammals. The dentition is of a primitive canid- mustelid type. The size and the general outline of the skull are nearest to Cryptoprocta of Madagascar. The basicranium has the shortness characteristic of canids and procyonids, the auditory region is closest to the latter groups; the postcranial skeleton is of the size and general proportion of Cryptoprocta. Comparisons are therefore made with a number of genera, especially Bassariscus and Crypto- procta. The outlines of the facial and the supracranial regions of the skull are similar to those of Cryptoprocta; the temporal arcade is longer, the cranium correspondingly shorter (figs. 39, 40). The arches are equally strong; there is no postorbital process on the jugal; the bridge over the infraorbital canal is broader than in the recent form. The premaxillaries are narrow and extend almost as far as the anterior extremity of the frontals (fig. 40). The nasals taper posteri- orly and terminate above the anterior rim of the orbit. The teeth are low-crowned, straight in alignment and inter-mesh with the opposing series. The basicranial region is broad; the auditory bullae are well rounded, widely spaced from the median line and center with the auditory meatus as in Aletocyon and in Bassariscus. The posterior nares open just back of the last molars. EARLY MIOCENE CARNIVORES 101 The mandibles are moderately strong in the ramus, the inferior out- line is gently curved. The coronoid processes are broad and slightly recurved but their posterior margins do not overhang the condyles as in Oligobunis crassivultus and Aelurocyon. The masseteric fossae are deep and are bounded anteriorly by sharp ridges which extend as far as the superior crest of the coronoid process. The entire structure of the skull and mandible is finely modeled, in decided contrast with the heavy structure of the recent Procyon. Neverthe- less, the short and broad basicranium, the lipless meatus, the rounded Fic. 40. Zodiolestes daimonelixensis, No. P12032. a, detail of nasal re- gion; b, axis, lateral view; c, axis, posterior view. All xX %%. bullae centering with the meatus, the detached paroccipital process and the structure of the auditory region, all point to basic procyonid relationship. Vertebral column (fig. 43).—The presacral series of vertebrae, with the exception of some of the processes, is preserved entire. As the skeleton was articulated when found, and was later prepared and photographed in that position, there can be no doubt as to the series being intact (fig. 45). The vertebral formula is C-7, D-13, L-7, S-3, C-?. The column as a whole is somewhat lighter in the cervical and the anterior dorsal series than in Cryptoprocta but in the lumbar series it becomes heavier and the processes become stronger. The series of cervical vertebrae, measured over the centra, is 8 mm. shorter than that of Cryptoprocta. The atlas is not so broad in the arches and is lighter throughout (fig. 25). The anterior atlantar foramen opens laterally on the surface as in Bassariscus, but not in a fossa. The lateral margin of the transverse process is rounded where the vertebral artery passes over it, but not indented. 102 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 The posterior opening of the arterial canal, like the anterior opening, may be seen from the dorsal view. In the latter characteristic the atlas of Zodiolestes is nearer to that of Aelurocyon than to any of the recent procyonids or the American mustelids. Of the axis, only the centrum, the odontoid process, and the transverse processes are preserved (fig. 40). The centrum has a keel on the inferior surface which is bifurcate posteriorly. The odontoid process is tapering in outline and decurved at the extremity. The succeeding five cervicals are similar in size and in general characters to those of Cryptoprocta. In the fourth and fifth a pair of small, pointed processes spring from the posterior margin of the neural arch mesad to the postzygapophyses; similar processes were probably present on the second and third, but those portions of the arches are not preserved. The neural spines throughout these five vertebrae are long and tapering (fig. 43). The dorsal vertebrae are thirteen in number and of the type common to small and active arctoid carnivores. The series is 8 mm. shorter than the same in Cryptoprocta. The neural spines are equally long and more massive in proportions than in that form. Through- out the series they are broader antero-posteriorly. The tenth marks the change in type of postzygapophyses; the eleventh has a short spine. The succeeding dorsal vertebrae have increasingly broad and anteriorly inclined spines, grading imperceptibly into those of the lumbar series. The first two transverse processes of the dorsal vertebrae are laterally directed and stronger than those of the suc- ceeding units. With the fifth, a small, hook-like process appears at the antero-lateral extremity and continues in the succeeding units as far as the tenth. A corresponding posterior process is present on the fifth, sixth, seventh, and tenth. Only two of the mid-dorsal centra may be described as keeled. The lumbar vertebrae, as a series, are 10 mm. longer than the same series in the specimen of Cryptoprocta. The spines of the anterior three are broader and more expanded at their extremities; the last four are similar to those of the recent form. The transverse processes are less curved, somewhat longer, more laterally directed and in the last three are expanded at their extremities. The last vertebra is much broader across the postzygapophyses. The sacrum is made up of three co-ossified vertebrae. It is equal in length to that of Cryptoprocta, and is narrower and more tapering posteriorly; the centrum of the last vertebra is not so large. These characters indicate that the tail was not so long or so powerful EARLY MIOCENE CARNIVORES 103 as in the recent animal. The neural spines increase in height from anterior to posterior. The second and third are separate as in Bassariscus. The scapula is similar in size to that of Cryptoprocta (fig. 41); the base of the spine is longer, the anterior border is convex below and slightly concave above, as in Bassariscus; superiorly it is rounded, the axillary border somewhat shorter. The superior axillary margin is produced into a process for attachment of the teres major as in the latter form. The blade of the spine inclines posteriorly so as to Fic. 41. Zodiolestes daimonelixensis, No. P12032. a, scapula; 6b, pelvis and sacrum. Both x 34. enclose the infra-scapular fossa. The coracoid process is proximal; the acromion and metacromion are closely joined as in Bassariscus, the two processes forming a bar, oblique to the direction of the spine as is common to some mustelines. The pelvis as preserved in this specimen includes the ilium, the ischium, and the descending branch of the pubis, all of the right side, together with the greater part of the left ilium (fig. 41). Contact with the sacrum is perfect. From these important parts the pelvis has been restored by comparison with that of Nasua, the living procyonid nearest its size. The pelvis is characterized by the unusual length of the ischium from acetabulum to tubercle and by the oblique direction of the ilium, upward and laterally, as compared with the 104 FreELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 supra-acetabular margin. The crest of the ilium is uniformly rounded, showing nothing of the antero-inferior angle common to this element in both Bassariscus and Nasua. The attachment for the rectus femoris is a prominent tubercle. The humerus is 5 mm. shorter than that of Cryptoprocta; it is rounded in the shaft and curved antero-posteriorly as in Bassariscus (fig. 42). The deltoid crest is strongly marked and extends well below the middle of the shaft. The muscular and ligamentary attachments are more strongly developed throughout than in Crypto- procta, less so than inGulo. The supinator ridge rises as a sharp crest extending along the lower third of the shaft and affording another procyonid character. The inner condyle does not extend below the articular facet as in the American mustelines; the entepicondylar foramen is small, rounded, and placed low on the shaft as in Bassaris- cus and Nasua. The ulna is equal in length to the humerus of the same individual, being 2 mm. longer than that of the specimen of Cryptoprocta. The olecranon is laterally compressed as in Bassariscus, less curved in its _ dorsal outline. The dorso-lateral angle is moderately inflected but not produced into a strong tubercle as in Promartes and Aelurocyon. The styloid process is laterally compressed but prominent. The radius has lost some 15 mm. of the proximal end by erosion, as determined by comparison with the ulna. The meso-distal angle is notched, apparently for the passage of the extensor tendon. The fore foot of Zodiolestes is longer than that of Cryptoprocta but similar in strength (fig. 42). The metacarpals are from one-fifth to one-sixth longer. The scapho-lunar has its proximal articulation extended farther over upon the anterior surface, indicating a semi- digitigrade position of the foot; the distal surface is less pointed and is not notched into the magnum. Likewise, the trapezoid is not notched into the distal surface of the scapho-lunar. The magnum, on its anterior face, approaches the form of a right triangle, the longer dimension divided in its contacts between the trapezoid and the scapho-lunar. The unciform is less narrowly pointed at its proximal end and separates the elements of the proximal row more widely. The two proximal facets of the unciform are concave; that for the ulna is larger, accounting for the larger facet on the styliform process of that bone. The femur of the left side is preserved almost entire; of the right side only the proximal half remains (fig. 42). It is 6 mm. shorter than the femur of Cryptoprocta and is stronger and more curved in Fic. 42. Zodiolestes daimonelizxensis, No. P12032. a, humerus, anterior view; b, humerus, lateral view; c, fore foot; d, hind foot; e, femur. All approx. x 34. 105 POSTS oe ee me SES eT ey hey ee ee ee Pe aT ee eee were: ee 6h “SI 106 EARLY MIOCENE CARNIVORES 107 the shaft. The head is directed obliquely forward on the shaft and does not rise above the great trochanter. The tibiae are both broken, but between the two all of the ana- tomical characters can be determined. They are apparently a little shorter than the femur of the same animal. This bone as a whole is Fic. 44. Photograph showing the holotype specimen of Zodiolestes daimone- lixensis in relief on the lower half of a spiral of a daimonelix in plaster wrappings. The upper half of the spiral had been removed. Approx. x 14. considerably stronger than that of Cryptoprocta. The distal end bears a distinct facet for articulation with the fibula. The fibula is straight in the shaft, angular, and moderately expanded at the extre- mities. A spine-like process extends mesially from the proximal end just below the facet. The distal end bears a distinct facet for the tibia above that for the astragalus. 108 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 The hind foot, consistent with the strength of the pelvis and the hind leg, is larger and stronger than that of Cryptoprocta (fig. 42). The first digit is appreciably reduced, the fifth is almost as strong as the second. The third and fourth metatarsals are equal in length. The astragalus is rather deeply concave proximally; the facet for the navicular is elongate in the transverse direction, and its center is opposite the terminus of the external angle. The caleaneum has a - Fic. 45. Sectional view of Daimonelix spiral, showing the skeleton of Z. dai- monelixensis as it was found coiled and embedded in the sandy mass of the spiral. prominent tubercle on the lateral surface opposite the sustenta- culum; the anterior end bears a rounded and moderately concave facet; the mesocuneiform meets the navicular in a well-defined articulation. The unguals are hooded but not retractile. In conclusion, it may be pointed out that Zodiolestes was an animal of similar size, proportions, and muscular development to Cryptoprocta (fig. 43). It retained the full complement of premolar teeth and molars of similar structure to those of the smaller, lower EARLY MIOCENE CARNIVORES 109 Miocene Promartes. The short and broad basicranium with rounded bullae, detached and backwardly directed paroccipital processes and the procyonid-like structure of the auditory region, at once distinguish it from Cryptoprocta or from any of the American Musteli- dae. While it must be regarded as a form detached and apart from known procyonids, the basicranial characters and the structure of the inner ear definitely group it with the Procyonidae. MEASUREMENTS (In millimeters) SKULL Length, condyle to incisors.................. aie Pere wien 114 PRPBOe tlh SOPs ARCMIN Ce ean bole ae oe eek 73 Breadth across mastoid processes. ...............00000: 49 DOAPUNOL-CEHCAl SOLICR Gy see eso wae wate hee oo ola bene ae 52 Breadth across crowns of first molars................... 37 Molar II to posterior margin of glenoid fossa............ 31 Projection—posterior margin of glenoid fossa to margin of RUNING Wl eer re eR ee pik er et galore aka eee eoaS PGHOCH Ol SAQILUA! -CLOSG ir eo tiirets Clo oie pe Lee eas 53 cxneatest, BFEAGtI:OD CRANTAIN 8558 ay is Rikeraca hs oxl optecea ons 44 Anterior margin of orbit to anterior margin of canine alveo- Py RO ar al eS bie, UM Cee seve cuace Poet 31 LOWER JAW Greatest length, condyle to incisor alveolus.............. 83 Height of coronoid process above inferior margin......... 36 Occipital condyle to margin of molar II................. 36 engeth: of lower dental’ series... oekaki 51 VERTEBRA Atlas - Breadth across transverse processes. ............... 46 ‘Ss Dreauel: OL SUPETION ATCH:. A.c eto e als eset oie wed ee 8 ISTEACUN OLANICTION- TCH. frie i, Soi s aie iN cree a's okewe 4 Axis ; Length of centrum including odontoid process....... 26 Breadth of anterior‘articulation..... ..... 6.666. 660008. 18 Breadth of centrum, posterior end................. 11 Fifth cervical vertebra Axial length: Of centrums. 225 sss hardiecs 084.6 write boon 12 Breadth across prezygapophyses................... 19 Dorsal Vertebra I Axiallength-01, contrumnt th... cetera ska 11 Breadth across transverse processes................ 21 Height of spine above base of centrum.............. 33 Lumbar Vertebra I Length of centrum: is sse3e obs os Sasi sie eee 19 Breadth across prezygapophyses................... 8 Height of spine above base of centrum.............. 23 110 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 VERTEBRA—Continued Lumbar Vertebra VII Axialdlength-oficentrum 2.32 .ce en ee hee ees 19 Breadth across transverse processes................ 52 Breadth across prezygapophyses................... 19 Height of spine above base of centrum.............. 28 SCAPULA Axialjlength:. seh cpa aaa cree etree ee be ee oP 72 engtheof-axillary: bordéetasiajcnsas kod ose eee 64 Greatest bread than set es eee ee ele ay te 39 Breadth-ofcarticulaniend iwi conch a ne 16 HUMERUS Da GET 2) a greatest length’. ..oo..c oes ea cate cid ee oe Sh seh 33 AStragalis? Createst wen etn se. oi foe, op coronene ee 22 NEGtatarsal 5 eh en et eno Font erat kc en eather tes ee 29 Metatarsal’ ETy length” 2c. 550 25-2 te ths Geis wom oe 40 Metatarsal lt Baten ogi. deca it ttre te arses heehee aren 47 Metatarsal Vc length. 6 260 Sk Can eons oe sine ee 49 Metatarsalew tanvun. cot or Niehaus ieee a CONCLUSIONS From the character of known mustelines of Lower Miocene age, it appears that there were two stocks of these animals present in the Great Plains region. First, an indigenous stock descended from known Oligocene forms, of small size and apparently of active habits. These forms include the early, marten-like group, designated by the genus Promartes. It may be that other small forms will not fall within this genus. Second, a larger and stronger stock of mustelines, earliest known as Oligobunis crassivultus of the John Day formation but represented also by other strong-jawed forms such as Paroli- gobunis of the Great Plains region. Along with these animals and probably descended from the same stock, were the larger muste- lines, Aelurocyon and Megalictis, sturdy and voracious killers and strong rivals in their field of depredation. These animals appear to have driven out most of the felids and to have remained as a promi- nent group well into Middle Miocene time. Before the end of the Miocene they had disappeared and the felids, filtering back into the Great Plains region, were again becoming more numerous. The canids as a family appear to have held their own against the rivalry of the larger mustelids. The smaller line as represented by Pseudocynodictis apparently disappeared early in the Lower Miocene. The stronger canid line, in sequence to the Oligocene Daphoenus and the Lower Miocene Daphaenodon, held their own. Later the great bear-dogs appeared as dominating carnivores. With the large mustelines out of the race, and the procyonids never strong enough to play an important part, the balance of power became established between the great canids and surviving felids. Difference in habits between cat and dog families appears to have been early established, the one lying in wait or stalking while the other pursued its prey, either overtaking it by greater swiftness of foot or by driving it to exhaustion. Their rivalry, because of different hunting habits, has never been pushed to the point of exterminating either family. “SOIPIABD PlOUs[s JO SUIBZIVUT IOLI04SOd ssOlov UMBIP dUI] 04 Se[ApuUod ‘aul] UBIpeul UO painseayy t “SIB[OUL JSB] JO SUISIVW IOII04SOd SsOJO® UMBIP OUI] 0} SIOSIOUI ‘aUl] UBIpeu UO painseay, | “6681 ‘MOYUIeIY pue uewyI0M Aq UoIdOsep WO , Ajenbiqo eulw asie] A[eqelapoyy | oyeZuoj[a oezIs wnIpes{ yews pepunol pure [[eug | -el0y ouneed “quy ' 1yUD ; 9} 81pew1e}Uy 2} BIPeW19}UT api Aleyelepoyy | -wad sajippy Ul SB MOLIEN ayeleg zd Jo ulsieUl ¢d jo zd JO UlsIeU “YUe d}IS oinjns ‘yue ayisoddo sassolg Jajued ayisoddo sassoig poulusejap JON -oddo oul] UeIpeul sassolg oulzeled-o][ xe fy pa[Ba0U00 JON pa[eaouod YON eng Aq peleaou0g bisa | UI sB BI[NG Aq paeeou0D) wWINIpeu ‘oR] UsUTRIO ‘dT wor ayeredag ‘dT wy wor ayeiedag “qsod ‘oR] uswWIeloJ wWoll oyeredeg usweloj Ie[Apuovalg yuaeseid JON quaselg quesalg qyueseid JON jeuvo plousydsiy yuoulwmoid A[eye1epo fy pedoyeasp 23417 $9]8a]01p0Z ul uvYy} pedojeAsp ssa] pedojaasp Aleyelepoyy ssev0id plojse yy toorj ynq [jews peoelip sseooid ‘snjrew Moleg dal} SSBDO0Ig A[piVeMyoeqg pure vaigq ssao0id [eqidimoo01eg %8 OVE = “WU GHZ %MGLZ = “WU GZ %9Y LZ = “WU G'TE Sle = "wu gg | fyzsue] wnluesoiseg % “by = “wu 0g MSL = “WU EF MV 9 = “WU “Eg MES sy = “WU g'Tg 4yysug] eee “UU 8h X 18 09 X 16 L9 X FIT 9L X 6IT | Ul suoIsueWIp [[NYS uaulloads pallejal ‘snasiupsspg xed AjO[oy Uohv0nj, Yq adAjojoy uohvoja) 7 adAjo[oy sa7saj01poz SCINOAODOUd YUAHLO HLIM GYYVd WOO SALSA TOICOZ 112 “668T ‘MouWey, pue uewyI0M Aq UONdIOSep WoO , a0B1} ON A[UO 9081) V Bssoj Ie[Apuoojsog [[vus usuIe10,q qyeys UO MOT peovjd suljjno ul [Vag ueul -e10} Ie[Apuooideyugq 4y8ys JO o[pplur 0} Spue}xe 4seld plozjaq , UoADOIg Ul uRYyy Jepus[s alow snieuny,, qyeys JO I[pplu MOleq spuezxy 4sol0 plojjeap ‘snisuny Japiog JO YyyAINo}J IO Japioq JO YyyINO} IOI -adns uo ssev0id AlelIxy -adns uo sseo0id Alel[Ixy eindeog MOIA [BS 9AOQB WOOL} V{[GISIA JOU -10Op WIO1} a[GISIA [eUuBd JO [euvo jo Zutuedo Joleysod | preMdn AjySI[s syueseid Zuluedo JoWe\ysog ‘uew sUsWIBIOJ “YUB PUODES ON | [VUBI Jo ZuIUAdO JOLIOISOg -B10J 1OL1ajUB PUES ON sey Y}00} PeUMOID-MOT BUOIIS ‘J [euoljouny ynq peonpey | FW ‘sq ‘queseid ajnuog Buoys peonpal yonyy zW soeddg yueseid wn] qyueseid aynuoo qua -nZuld pue suos0jold YW | [eudoJUL pue 9eU0d0}01g -said au0d0jo1d Buo01js Y | Y}004 [eLIoyas JaddqQ AYABO Ay Ay ploue]3 apisyno ysnf{ spur | -Avo prous[s seyovoiddy | -Aavo ploue[s soyoroiddy auoq jesne s[equody s[ey s[equodj Jo ssav0id wodj ojyeiedas AjapiM | -UoIl Buoow ‘ayeZuo[y | s[eyuOIy YORordde you oq | ‘jue yYyoroidde ‘1apuals soLlel[Ixewolg uewldeds pasiajas ‘snosiupsspg xed Ajooy Uohson]yq adAjojoy uohsojayy adAjo]OY 8a78a)01p0z panuiyjuoj)—SQAINOAOQOUd YUAHLO HLIM GHUVdWOO SULSYTOIGOZ 113 114 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 Of the Procyonidae much less is known. We have a glimpse of two multicuspidate forms in the Lower Miocene and now we have in Zodiolestes a procyonid which has retained the sectorial function of the carnassial teeth. This appears to have been a form of ter- restrial animal quite as alert as the cat or the fox. Up to this time it is known from one specimen only; its source and its ultimate fate are as yet unknown. Transmitted September 2, 1942 REFERENCES Cop, E. D. 1881. Miocene Dogs. Amer. Nat., 15, p. 497. 1883. The Vertebrata of the Tertiary Formations of the West. Book I. Rept. U.S. Geol. Surv. Terr., xxxv + 1009 pp., 75 pls. Houau, M. J. 1944. The Auditory Region in Some Miocene Carnivores. Jour. Paleont., 18, pp. 470-479. Loomis, F. B. 1932. The Small Carnivores of the Miocene. Amer. Jour. Sci., (5), 24, pp. 316-3829. MATTHEW, W. D. 1907. A Lower Miocene Fauna from South Dakota. Bull. Amer. Mus. Nat. Hist., 23, pp. 169-219. 1899. A Provisional Classification of the Fresh Water Tertiary of the West. With lists of the mammals occurring in the formations. Bull. Amer. Mus. Nat. Hist., 12, pp. 19-75. McGrew, P. O. 1941. A New Procyonid from the Miocene of Nebraska. Field Mus. Nat. Hist., Geol. Ser., 8, pp. 33-36. PETERSON, O. A. 1906. The Miocene Beds of Western Nebraska and Eastern Wyoming and their Vertebrate Faunae. Ann. Carnegie Mus., 4, pp. 21-72. 1910. Description of New Carnivores from Western Nebraska. Mem. Carnegie Mus., 4, pp. 205-278. Riaes, E. S. 1942. Preliminary Description of Two Lower Miocene Carnivores. Field Mus. Nat. Hist., Geol. Ser., 7, pp. 59-62. Scott, W. B. 1937. A History of Land Mammals in the Western Hemisphere. Rev. ed., xiv +786 pp., 420 figs. MacMillan Company. —— and JEPSEN, G. L. 1936. Mammalian Fauna of the White River Oligocene. Trans. Phila. Acad. Sci., pt. I, pp. 1-168. THORPE, M. R. 1921. Two New Fossil Carnivora. Amer. Jour. Sci., (5), 1, pp. 477-483. WoRTMAN, J. L. and MATTHEW, W. D. 1899. The Ancestors of Certain Members of the Canidae, the Viverridae and Procyonidae. Bull. Amer. Mus. Nat. Hist., 12, pp. 109-139. INDEX VOLUME 8 Adianthidae 18 Carnivores 59 Adianthinae 13, 16, 19 Catapleura 64, 65, 69, 70, 73, 74 Adianthus 16, 17, 18, 19 arkansaw 68, 70, 71 bucatus 18, 14, 17, 19 ponderosa 64 atagonicus 13, 16, 17 repanda 64, 71 Aelurodon 75, 76, 79, 80, 81, 82, 83 Chelonidae 73 brachygnathus 79 Cochilius 24, 25 ferox 79 Coniopternium 53 haydeni 79, 82, 84 “Corkscrew carnivore’’ 62 inflatus 79, 80 Cramauchenia 14, 15 meandrinus 79 Crocodilia 27 mortifer 79, 80 Cryptodira 65 saevus 79, 80, 82, 83, 84 Cryptoprocta 61 taxoides 79 Cynarctoides 35, 36 wheelerianus 79 Ailurus 36 Dermochelidae 73 Aletocyon 33, 34, 35 Desmatochelys 68 multicuspis 33, 34, 35, 36 Desmatolagus 39, 40 Aletomeryx gracilis 27, 28 Deuterotherium distichum 14 Alligator 28, 32 megrewi 27, 29, 31, 32 Eretmochelys 72 mississipiensis 30, 32 Erinaceidae 438, 45, 46 prenasalis 28, 32 Erinaceus 44, 45, 46 sinensis 30, 32 Eusuchia 27 thomsoni 28, 32 Alligatoridae 27 Galeopithecidae 24, 25 Allognathosuchus riggsi 32 Galeopithecus 24 Amphilagus antiquus 39, 40 Galeopterus 24 Andrewsornis 50, 52 Gruiformes 50 abbotti 50, 51, 52, 53 Anseriformes 53 Heliscomys 55, 56, 57 Archaeophylus 24, 25 egoryi 57 Argyrohyrax 24 atcheri 55, 57 Aucornis euryrhyncus 52, 53 senex 55 solidus 52 . vetus 55 Austrolagomys 40 woodi 55, 56, 57 Heteromyidae 55 Baculites 7 Hyaenognathus cynoides 75 Bantuchelys congolensis 9 Hyracoidea 18 Bassariscus 34, 35, 61 Borophagus 75, 81 Interatheriidae 24 Bostryoceras 7 Bothremydidae 3 Lagomorph 37 Brachyerix 45, 46, 47 Leporidae 39, 40 Brontornithidae 52 Leptictidae 46 Litopterna 13 Caimanoidea vischeri 32 Loxornis clivus 53 Caninae 34 Lytoloma 65, 66, 73, 74 Canis 82 lupus 80, 81, 83 Macraucheniidae 138, 18, 19 lycaon 80 Macraucheniinae 14, 15, 16, 19 rufus 80 Martes americana 59 Caretta 74 retusa 59 Cariamae 50 Mesembriornis 49, 52 115 116 FIELD MUSEUM OF NATURAL HIsTORY—GEOLOGY, VOL. 8 Metechinus 438, 44, 45, 46, 47 affinities of 44 marslandensis 48, 45, 47 nevadensis 48, 44, 45, 47 Meterix latidens 48, 45, 46, 47 Mookomys 56 Mustelidae 59 Mytonolagus 39 Ochotonidae 39 Oligobunis 62 crassivultus 61 emmarosae 60 epidus 60, 62 vantasselensis 60 Oreolagus 37, 39, 40 nebrascensis 38, 39 nevadensis 37 Osteoborus 75, 77, 81 cynoides 75, 76, 77 Osteopygis 71, 78, 74 gibbi 72 Palaeoerinaceus 44, 45, 46 Palaeoscaptor 45 Paleola 37, 39 nevadensis 37, 38 ‘Parastrapotherium 52 Pelicyornis 52 Pelomedusidae 3 Phanophilus 24 Phlaocyon 338, 35 leucosteus 33, 34, 35 marslandensis 34, 35 Phororhacidae 50 Phororhacoid birds 49, 53 Phororhacoidea 50, 52 Phororhacos 50, 52 affinis 52, 53. delicatus 53 modicus 53 Phyllemys 65, 69, 74 barberi 66, 67, 68 Physornis fortis 52, 53 Placenticeras 7 Plagiarthrus 24, 25 olivus 24 Pleurodira 3 Podocnemis 1, 3, 6, 9, 10, 11 aegyptiaca 9 antiqua 9 barberi 3, 4, 9, 10, 11, 63, 64 bowerbanki 9 bramlyi 9 brasiliensis 5, 10 cayennensis 9 congolensis 9 dehmi 9 delabechei 9 dumeriliana 9 expansa 7, 9, 10 harrisi 1, 9, 10 indica 9 lata 9, 11 lewyana 9 madagascariensis 9, 10 olssoni 9, 10 podocnemoides 9 sextuberculata 9, 10 stromeri 9 unifilis 9 vogli 9 Porthochelys 72 laticeps 68 Porthocyon 75 Proadiantus 18, 14, 15, 17, 18, 19 excavatus 14, 17 gibbus 14 pungidens 14, 17, 18 Procariama 49 Procyon 38, 34, 35 Procyonidae 61 hie, palpi aneaatin a g 24 Proheptaconus 15, 17, 18 trelewense 138, 14, 17, 19 Progaleopithecus 21, 24 fissurellatus 21 tournouéri 21, 22, 23 Promacrauchenia 19 antiqua 18 Promartes 59, 60 olcotti 59, 60 Proterix 46, 47 Proterotheriidae 13, 16 Protypotherium sp. 24 Pseudadiantus 16 Pseudocynodictis 33 Psilopteridae 52 Psilopterus 52, 53 australis 53 Pyrotherium 52 Romerolagus 39 Rostrornis 53 Sinolagomys 40 Smiliornis penetrans 52, 53 Soricidae 45 Stereogenys podocnemoides 9 Sylvilagus baileyi 38 Taphrosphys 8, 10 Testudinata 3, 65 Testudo tornieri 74 Thalassemyd turtles 638, 65 Thalassemydidae 64, 65, 74 Theosodon 14, 15 Tomistoma 11 Toxochelys 74 Trionychidae 74 Trionyx 11 Typotheria 24, 25 Zodiolestes 61 daimonelixensis 61 INDEX VOLUME 9 wae? deh 70, Bg. 88, 90, 92, 95, 101, brevifacies 838, 84 Vm 38 rus 70 hlatpeves 70, 96, 100 Allomyidae 5 Allomys 4, 5, 8, 10, 11, 12, 18, 23, 26 cavatus "10, 11 liolophus 11 nitens 11 Ameiuridae 48 Ameiurus 48 Amphianasua 98 Antilocapridae 64 . Aphelops 38 Aplodontia a 6, 7, 8, 12, 18, 16, 17, 8, 2 Aplodontidae 3, 4, 5, 18, 19, 24, 26 Aplodontids 13, 19, 23, 27 Aplodontoidea 5, 24, 26 Aplodontoids 7 Archidiskodon 38, 46 Artiodactyla 63 Asinus 58, 61, 62 Astrohippus 45, 46 Bassariscus 96, 98, 100, 101, 103, 104 Bison 38, 42, 46 ret i. 33, 38, 40, 41, 42, 45, 46, comparison with Sand Draw and Broadwater faunas 39 Boreostracon 42 Borophagus 38, 40, 41, 42, 46 Broadwater fauna 338, 36, 37, 40, 41, 45, 52, 56, 64 comparison with Rexroad and Blanco faunas 39 Bunaelurus 70, 81 Burchelli 58 Caballus 58 Camelidae 63 Camelops 38, 40, 63 Canidae 538, 6 Canimartes 53 Canis cf. latrans 53 Capromeryx 38 sp. 64 Carnivora 53 Castorid 4 Castoridae 52 Castoroides 38, 41, 53 Ceratogaulus 4 Cervalces 46 Coso Mountain fauna 38, 40 Cryptoprocta 92, 96, 98, 100, 101, 102, 108, 104, 107, 108, 109 Cylindrodontidae 24 Daimonelix 71, 100 Beds 95 s Daphaenodon 69, 111 Daphoenus 69, 86, 92, 111 vetus 86, 90, 91 Dhok Pathan fauna 46 Dipodidae 24 Dipoides 38, 53 Dolichohippus 62, 63 Eligmodontia 38 Emydidae 48 Eocastoroides lanei 52 Eohaplomys 5, 12, 13, 16, 17, 23, 24, 25, 26, 27 Epigaulus hatcheri 4 Equidae 42, 47, 55 iar 41, 42, 44, 45, 46, 47, 56, 62, caballus 58, 59, 60, 61 excelsus 45 ar wee 36 emionus 59, 60 przewalskii 56 stenonis 44, 46, 56, 63 ss. 88, 44, 45 Felidae 55, 69 Felis 38 concolor 86 Gastrocopta cristata 36 Geomyidae 49 Geomys 36, 50 minor 52. parvidens 50 persimilis 50 quinni 49, 50, 52; comparison with other species of Geomys 52 tuza 49 ince aa sp. 64 Grevyi 58 Gulo 83, 84, 86, 87, 88, 89, 90, 104 Gyraulus pattersoni 36 Hagerman fauna 38, 40, 41 117 118 FIELD MUSEUM OF NATURAL HISTORY—GEOLOGY, VOL. 9 Haplodontia 4 Haplomys 11, 13, 20, 23, 26, 27 Hay Springs fauna 35, 42, 44, 46, 47, 48 Hemionus 58 Hemphillian 41, 45, 46 Herpestes 82 Hippidium 46 Hippotigris 46, 55, 62, 63 simplicidens 55 Histricomorph 4, 24 Holmesina 42 Hoplophoneus 90 primaevus 90 Humboldtidae 55 Hyaenognathus 41 Hydrochoerus 42 Hystrix 4 Ischyromyidae 24 Ischyromyoidea 24, 25 Ischyromyoids 24 Ischyromys 8, 24 Kansan fauna 42 Lepus 41 Liodontia 3, 4, 12, 13, 16, 18, 19, 26, 27, 38: ‘Loup Fork” fauna 41 Lupus nubilis 86 Lutravus(?) 41 Machairodus 38 . Mammalia 49 Mammonteus 38, 42, 46 Martes 70 actuosa 72 americana 77, 78, 80 caurina 80 penanti 72 Mastodon americanus 41 Megalictis 70, 75, 90, 95, 111 ferox 94 Mellivora 83 Menetus kansasensis 36 Meniscomys 8, 4, 5, 6, 8, 9, 10, 11, 12, 18, are 47; 26,79, 2G) 21: 22, 28, cavatus 10, 11 hippodus 4, 9, 10, 11, 12, 17 liolophus 10 nitens 10, 11 Merycodus 38 Mesogaulus 4, 5, 20, 22, 26 cf. pristinus 20 Metoreodon 38 Mimomys 88, 40, 41, 42 Mustelavus 70 Mustelidae 53, 69, 70, 71, 83, 94, 96, 109 Mylagaulidae 3, 4, 5, 7, 19, 24, 26, 27 By pass 3, 5, 12, 20, 22 evolution of dentition 19 Mylagaulodon 8, 4, 5, 8, 9, 10, 19, 20, 21, 23, 26 ef. angulatus 9, 23 in 8, 4, 5, 6,7, 8, 23; 225.26, ef. monodon 4, 20, 21, 22 sesquipedalis 4 Mylodont sloths 41 Mylohyus 63 ““Myomorphs’”’ 24 Nannippus 38, 40, 42, 46 Nasua 108, 104 Neochoerus 42 Neohipparion 38 Neotoma 38 Nimravus sectator 69 Norwich Crag fauna 48 Oligobunis 70 crassivultus 70, 80, 83, 95, 101, 111 darbyi 81, 82 emmarosae 81 epidus 79 Ondatra 38 Onochomys 388 Onohippidium 46 Osteoborus 38 Palaeocastor 71 Panthera 55 Paramyidae 24 Paramys 5, 8 Parelephas 38, 46 Paroligobunis 95, 111 Perissodactyla 55 Phlaocyon 70, 96 Pinjor fauna 44, 46 Piscis 48 Flstveonng 88, 40 sp. Pleistocene mammals 33, 34, 41, 42, 47 ranges of 43 Plesippus 36, 38, 40, 41, 44, 45, 46, 56, 63 simplicidens 55 Pliohippus 38, 45 Proboscidea 55 Procastoroides 38 sweeti 52, 53 Procyon 70, 101 Procyonidae 69, 70, 96, 100, 109, 114 Procyonids 70, 108, 111 Procyonines 7 0 Promartes 70, 71, 79, 81, 83, 90, 96, 104, 109, 111 in ra 77, 81 epidus 79 olcotti 71, 72, 74, 80, 81, 89, 92 vantasselensis 80, 81 INDEX Promylagaulus 5, 7, 8, 9, 10, 19, 21, 22, 23, 26 riggsi 5, 6, 9, 21 Prosciurus ef. relictus 11, 12, 28, 24 Prosthennops 38, 63 Protogaulus 12 Protogomorpha 24, 26 Protoptychidae 24 Pseudaplodon 26, 27 Pseudemys sp. 48 Pseudocynodictis 69, 98, 111 Reptilia 48 Rexroad fauna 37, 38, 40, 52, 53 comparison with Sand Draw and Blanco faunas 37 Rhyncotherium 38 Rock Creek fauna 41, 44 Rodentia 27, 49 Sand Draw fauna 38, 34, 37, 40, 41, 44, 48, 49, 50, 53, 54, 55, 56, 64 comparison with Rexroad and Blanco faunas 39 invertebrates from 36 San Joaquin fauna 40 Sciuravid 24 Sciuravidae 24 Sciuridae 4 Sciuromorph 27 Serbelodon 38 Sigmodon 38 Smilodon sp. 55 californicus 55 119 Spalacidae 4 Sphenophalus 38 Stegomastodon 38, 55 primitivus 55 “quarry” 36, 64 Symbos 42, 46 Tagassuidae 63 Tanupolama 40 Tatrot fauna 44, 46 Tatu 42 Taxidea 37, 87, 90, 91 americana 75 ef. taxus 53, 54, 55 Teleoceras 38 Testudinata 48 Testudinidae 48 Testudo 49 campester 49 sp. Titanotylopus 64 Trachemys 48 Trigonictis 37, 53 kansasensis 53 Ursus 41 Villafranchian fauna 44, 45, 46, 47, 48 Yarmouth 42, 48 Zodiolestes 70, 86, 87, 92, 96, 98, 100, 102, 104, 108, 114 daimonelixensis 100 spel a ‘¥ gee at a » 1 bts i, aati [=e Uae wi: oy a ae ase Sa? Paty cage atl ie 38 * nay Rar ied ey May 1 . “Tl Nh