Front Cover
 Diversity of canidae in the pliocene...
 Literature cited
 Back Cover

Group Title: Pliocene Canidae of Florida
Title: The Pliocene Canidae of Florida
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00001522/00001
 Material Information
Title: The Pliocene Canidae of Florida
Series Title: Bulletin of the Florida State Museum
Physical Description: 273-308 p. : illus. ; 23 cm.
Language: English
Creator: Webb, S. David ( Sawney David ), 1936-
Publisher: University of Florida
Place of Publication: Gainesville
Publication Date: 1969
Subject: Canidae, Fossil   ( lcsh )
Paleontology -- Pliocene   ( lcsh )
Paleontology -- Florida   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Bibliography: Bibliography: p. 306-308.
General Note: Cover title.
Statement of Responsibility: by S. David Webb.
 Record Information
Bibliographic ID: UF00001522
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aleph - 000299486
oclc - 00609703
notis - ABS5925
lccn - 72632276
 Related Items
Other version: Alternate version (PALMM)
PALMM Version

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
        Page 273
        Page 274
        Page 275
        Page 276
        Page 277
        Page 278
        Page 279
        Page 280
        Page 281
        Page 282
        Page 283
        Page 284
        Page 285
        Page 286
        Page 287
        Page 288
        Page 289
        Page 290
        Page 291
        Page 292
        Page 293
        Page 294
        Page 295
        Page 296
        Page 297
        Page 298
        Page 299
        Page 300
        Page 301
        Page 302
        Page 303
        Page 304
    Diversity of canidae in the pliocene of Florida
        Page 305
    Literature cited
        Page 306
        Page 307
        Page 308
    Back Cover
        Back Cover 1
        Back Cover 2
Full Text
S. David Webb

Numbers of the BULLETIN OF THE FLORIDA STATE MUSEUM are published at irregular intervals. Volumes contain about 300 pages and are not necessarily completed in any one calendar year.
Walter Auffenberc, Managing Editor Oliver L. Austin, Jr., Editor
Consultant for this issue: Richard H. Tedford
Communications concerning purchase or exchange of the publication and all manuscripts should be addressed to the Managing Editor of the Bulletin, Florida State Museum, Seagle Building, Gainesville, Florida 32601.
Published 31 December 1969
Price for this issue $.70

Synopsis: A new genus and species of canid, Carpocyon limosus is described from the Hemphillian (Middle Pliocene) Bone Valley Formation. Though showing strong omnivorous tendencies, it is not closely related to cynarctine eanids; nor is it close to Actiocyon Stock which is here synonymized with Alope-cocyon Camp and Vanderhoof. Carpocyon cwtpidatus (Thorpe), new combination, from the late Barstovian (Late Miocene) of Nebraska, represents an earner stage of evolution and suggests derivation of Carpocyon from Tomarctus.
Five species of Osteoborus are now known from Florida: (1) Osteoborus dudleyi (White), new combination, is a large, progressive species from the Bone Valley fauna. Pliogulo White is considered a synonym of Osteoborus Stirton and Vanderhoof. Osteoborus crassapineatus Olsen is placed in synonymy with O. dudleyi (White). (2) O. ore, new species, is a very small but advanced form from Withlacoochcc River site 4A; and two less progressive species occur together in Mixson's Bone Bed, (.3) O. galushai, a new medium-sized species, near O. cyonoides, and (4) a very large species near O. validus. The latter also occurs at McGchee Farm. (5) A species larger and even more primitive than O. galushai also occurs at McGchee Farm and at Hogtown Creek site II. Until it can be characterized by more adequate material, it is referred to O. galushai.
Table of Contents
Introdtjctiox____________________ _________________ ________...................- ---------------------- 274
Acknowledgments _______ _____________________ ---------------------- .....-...... 274
Abbreviations .........._______ ---------........-......__________.......------------ 274
Systematica __________.....____________ ________....._________ ---------------------------------------- 275
Subfamily Caniuae Gill ------.............. ----------------------------- ------------------------ 275
Carpocyon new genus __________________ -------------.................... ------------------ 275
Carpocyon limosus new species ... --------.....__________________________________________ 276
Subfamily Borophaginac Simpson _________. _________....................________ - 281
Osteoborus Stirton and Vanderhoof ....__________________________________________________ 281
Osteoborus dudleyi (White) new combination ____________........_............._. 281
Osteoborus ore new species_____......_............. .................________________________ 287
Osteoborus galushai new species .... ____________.........._______________ 296
Osteoborus cf galushai................................................................___________ 299
Osteoborus validus (Matthew and Cook) ......______________________________..... 302
Canid Diversity in Florida____________________.................______....._..................._______ 305
References Cited ________........___.....___________________________________ ... ______________ _. 306
1The author is Assistant Curator of fossil vertebrates, Florida State Museum, and Assistant Professor of zoology, University of Florida, His principal research concerns the Pliocene mammals of Florida. Manuscript received 4 November 1968.
Webb, S. David. The Pliocene Canidae of Florida, Bull, Florida State Mus., vol. 14, No. 4, pp. 273-308.

Florida provides almost the only record of Pliocene terrestrial life in the eastern United States. This is so tantalizingly scant that the Florida State Museum recently has undertaken extensive paleontologi-cal exploration of the state's Pliocene terrestrial deposits. This is the ninth contribution resulting from this program.1
Only two species of Pliocene canidae have been described from eastern LTnitcd States, as against six Miocene species. Even the Pliocene age of these taxa was doubted because of their uncertain stratigraphic provenience within the Bone Valley District (Olsen, 1956b; White, 1942). It can now be shown that the two previously named taxa are synonyms and that they are late Hemphillian (Middle Pliocene) in age. Furthermore, one new genus, three new species, and two referred species can now be added to the record of Pliocene canidae in the eastern United States.
The revival of interest in Pliocene deposits in Florida began in 1962 when Clayton Ray, my predecessor at the Florida State Museum, initiated excavations at the MeGehee Site. In 1963 and 1964 the Frick Corporation undertook joint sponsorship of this work with the Florida State Museum, After my arrival in 196*4, the study of Pliocene deposits was broadened to include large scale exploration for new sites, as well as continued excavation at the MeGehee Site. The Frick Corporation continued partial sponsorship of these operations through 1965. Subsequently this program has been supported by NSF grant GB 3862.
During these studies it became clear that the classic Bone Valley collections were confused strati graphically and that precise field studies were required. The phosphate companies in the Bono Valley District have been most cooperative, not only offering us access to every section in every pit, but also placing valuable equipment at our service. Particular acknowledgment is due the International Minerals and Chemical Corporation for partial support of our work in the area during 1967 and 1968.
In 1939 through 1941 Ted Galusha led field parties from the Frick Laboratory of the American Museum of Natural History in intensive excavation of the Mixson's Bone Bed. I am grateful to Ted Galusha and other personnel in the Frick Lab for permission to study the canidae from Mixson's.
My understanding of carnivore phylogeny has been enhanced considerably by conversations with Donald E. Savage, Richard H. Tedford, and John E. Mawby,
All measurements are given in mm, unless otherwise noted. Institutions that have loaned materials for this study arc abbreviated as follows:
iPrevious contributions are Brodkorb (1963); Webb (1964 and 1966); Auf-fenberg (1966); Rose and Weaver (1966); Weaver and Robertson (1967); Hirschfeld and Webb (1968); Webb and Tessman (1968),

CWTPanhandle Plains Historic Museum, Canyon, Texas FAMFrick Laboratory, American Museum of Natural History FGSFlorida Geological Survey, Tallahassee, Florida
MCZMuseum of Comparative Zoology, Harvard University, Cambridge, Massachusetts
UCMPUniversity of California, Museum of Paleontology, Berkeley, California
UFUniversity of Florida, Florida State Museum, Gainesville Florida UMMPUniversity of Michigan, Museum of Paleontology, Ann Arbor, Michigan
YPMYale Peabody Museum, Yale University, New Haven, Connecticut Certain Florida Pliocene localities are briefly described in this and other papers. More complete geographic descriptions and stratigraphic data are recorded in the locality catalog, Florida State Museum.
System atics
Family Cakidae Gray Subfamily Caninae Gill
Carpocyon new genus
Genotype.Carpocyon limosits new species, Bone Valley Fauna of Florida, late Hemphillian age.
Etymology.Greek: Karpos, fruit; Kyon, dog; in reference to the presumed significance of fruit in the diet of these canids.
Referred Species.Cynodesmus cuspidattis Thorpe (1922), collected 1873 by Otto Harger (no exact date with type specimen YPM 12788), Niobrara River Fauna of Nebraska, late Barstovian age. Presumably collected from the lowermost (Crookston Bridge) member of the Valentine Formation (Skinner, et aL> 1968) on the basis that the bone is hard and tan and the teeth molasses-colored, as is typical of this fauna.
Generic Diagnosis.Medium-sized eanid, larger than Procyon lot or; Dentition bhint, tending to wear heavily. Molars enlarged and premolars relatively reduced. P' with posterior accessory cusp and tending to develop strong lingual cingulum. P' triangular with strong parastyle, low weak paracone-metacono blade, protocone broad with no distinct crests on parastyle or paraconc, weak lingual cingulum. M1 massive, rectangular, 70 to 75% wider than long, paraconc and metacone suboqual, no stylar cusps, continuous labial cingulum, pro-toconule absent, strong metaconule subequal to and symmetrical with protocone, hypocone a blunt enlargement of lingual cingulum. M2 relatively large, structurally similar to M1. M,.t present in lower jaw.

Carpocyon limosus new species Figure 1, Table 1 Etymology.Greek: limosus, hungry.
figure 1. Stereoscopic views of palate of Carpocyon limosus new genus and species. Ilolotype, uf 12069. A. Left P:*-M>. B. Right P'-M-. Xat-. ural Size.
Type Honrzox And Locality.Palmetto Washer: Hexameryx Locality (UF locality catalog) about 7 miles southeast of Brewster, Polk County, Florida. In light gray medium phosphatic sands from a phosphate mine. In the same matrix and in the same individual dragline dump were found Nannippus ingenuus (Leidy), Neohipparion

Table 1. Measurements1 of Holotype of Carpocyon limosus (in mm)
Tooth Length Width
pa 9.6 6,0
15.6 9.1
Ml 11.4 14.2
M2 9.1 13.2
iMeasurements of left jaw and right jaw agree on comparable teeth. Length-width measurements taken across most widely separated points of enamel. On M.2 "width" is actually an antcrolabial-posterollnffual measurement.
phosphorum Simpson, and Hexameryx simpsoni White, indicating a Hemphillian (Middle Pliocene) age.
Diagnosis.P:1 expanded lingimlly with distinct lingual root, P* with strong parastyle, short bulbous protocone, continuous lingual cingulum. M1 with strong labial cingulum, lingual moiety nearly as wide as labial; M2 80% as wide as M\ posterolingually elongated.
Description of Type.The third premolar is anteroposteriorly elongate. There are four cusps in line: a large protocone, a smaller posterior accessory cusp, and in line with these, an anterior and a posterior cingular cusp. A weak labial cingulum extends the length of the tooth. A strong continuous lingual cingulum connects the anterior with the posterior cingular cusp and from the latter a weak cingulum extends along the posterolabial side of the tooth. The lingual cingulum broadens considerably near its middle, and this broad area is supported by a separate lingual root.
The fourth upper premolar is triangular in shape. The shearing blade is weakly developed, the paracone and metacone relatively short and blunt. The blade is oriented in a nearly anteroposterior direction as in modern canines, not obliquely as in more primitive forms. A strong parastyle lies at the anterior end of the carnassial blade. The protocone remains a distinct rounded cusp separated by a shallow valley from the parastyle and paracone. A weak but continuous lingual cingulum extends from the protocone to the posterior edge of the metacone blade. On the left carnassial shearing wear has eradicated most of this cingulum.
The molars are relatively large and heavily worn. The first upper molar is roughly rectangular in shape and nearly 25% wider than long. The paracone and metacone are broad low cusps of about the same size. No stylar cusps are present, but there is a broad continuous external cingulum. The pTOtocone lies directly lingual to the metacone. These cusps, though broad, arc low and readily wear to dentine. Replacement dentine fills the pulp cavities that become exposed under

these cusps. The hypocone is also broad and low. It readily wears down and connects with the anterolingual cingulum to form a long lingual talon.
The second upper molar resembles the first, but it is a little smaller and its lingual side is skewed posteriorly giving the tooth a subtrape-zoidal shape. The paracone and metacone are smaller, lower, and more closely appresscd than in the first molar. The hypocone is relatively large and occupies a position wholly posterior to that of the metacone. As in the first molar, there are strong labial and anterolingual cingula. Ms is absent.
The position of the infraorbital foramen cannot be determined from the type specimen. The anterior root of the zygomatic arch lies above the first molar. The jugal is 15 mm deep immediately posterior to its contact with the maxillarv.
RelationshipsCarpocyon limosus is considerably younger than Carpocyon cuspidatits (Thorpe) of late Barstovian age. The canoid genera to which Carpocyon bears a close resemblance are Tomarctus, Cynarctus, and Actiocyon ( = Alopecocyon). Tomarctus (including Tephrocyon) ranges in age from the Arikareean tlu*ough Clarendonian (Downs, 1956; Olsen, 1956a; Wilson, 1960; Macdonald, 1963); Cyn-arctus (excluding Cynarctoides) is recorded from the Barstovian and Clarendonian (McGrew, 1937, 1938; Hall and Dalquest, 1962), and Actiocyon ( Alopecocyon) is known from the late Clarendonian of California (Stock, 1947).
In Carpocyon the broad low cusps, relatively large, nearly square molars, and heavy wear on the teeth indicate adaptation to a varied diet that probably included many fruits and berries. In this aspect of its dentition the genus resembles procyonids, but in every fundamental feature, such as the shape of the carnassial and the five-cusped pattern of the molars, the dental pattern is canid.
In both its omnivorous aspect and its basically canoid pattern, the dentition of Carpocyon more nearly approaches that of certain species of Cynarctus. Recent studies of the cranial features of Cynarctus and its relatives by Dahr (1948), Hough (1948) and Galbreath (1956) demonstrate clearly that the cynarctincs are canids, convergent in some features of their dentition with procyonids, but not, as had previously been supposed, closely related (o procyonids or even ursids. Cynarctus (and even the older Cynarctoides) had in many respects become more procyonid-like than Carpocyon. Differences in detail are considerable, as indicated in Tabic 2, Evidently Carpocyon bears no special relationship to cynarctines. Resemblances to that group

Table 2, Comparisons of Carpocyon, Upper Dentition with That of Certain
Other Canids
P+ parastyle strong, protocone low, anterior, close to blade, lingual cingulum low, narrow
M1 rectangular, wider than long,
metaconule and protocone subequal,
labial cingulum broad and continuous
parastyle absent, protocone large, bulbous, far lingual from blade, lingual cingulum high and broad
rectangular, longer than wide (even in Cynarctoides),
metaconule larger than protocone,
labial cingulum broken or absent
M2 moderately enlarged greatly enlarged
parastyle weak, protocone long, low anterior, far lingual from blade, grades into low, but broad, cingulum
triangular, wider than long,
metaconule much weaker than protocone,
labial cingulum moderately developed
greatly reduced
are either features common to most canids or are convcrgently produced by similar omnivorous habits.
Actiocyon Stock (1947) from Barstovian deposits in Cnyama Valley, California is here synonymized with Alopecocyon Camp and Vanderhoof (1940) from Vindobonian deposits in France. The close resemblance between these genera was pointed out to mc by D. E. Savage. Viret's (1933) earlier name Alopecodon was preoccupied by Broom's use of it in 1908 for a mammal-like reptile. Dc Beaumont (1964) suggests that Alopecocyon is a descendant of Broiliana, and a mustelid not a canid. Unfortunately, the absence of a braincasc for Alopecocyon makes this a difficult point to pursue.
Alopecocyon is characterized by its short, nearly quadrangular upper carnassial and the low connate cusps of the upper molars. These procyonoid features suggest, at first glance, possible relationship to Carpocyon limosus, but the resemblance is remote, as Table 2 shows. Moreover the Barstovian species of Carpocyon, C. cuspidatus, is even more distinct from contemporaneous Alopecocyon than is the Hemphillian species.
The genus Tomarctus more nearly approaches Carpocyon in the fundamental features of its dentition, and certain species of that genus occur early enough to be ancestral to Carpocyon. The genus Tomarctus Cope (including Matthew's Tephrocyon), typified by the Barstovian species T. bremrostris, has been given a central place in canid

phylogeny. It is generally believed to constitute the 'main line" of canid evolution through most of the Miocene, giving rise to both the hyaenoid dogs, Aeluwdon and Osteoborus, and to Canis and related modern genera. Confusion of several species of Tomarctus with those of Cynodesmtis has gradually been resolved (Simpson, 1932; McGrew, 1935; White, 1942; Green,' 1948; Downs, 1956; Olsen, 1956a; and Macdonald, 1963), so that a consistent definition of each is now established.
The principal features indicating the probable ancestry of Tomarctus to Carpocyon are the following;
a) stepped premolars (with strong posterior accessory and cingular cusps);
b) nearly anteroposterior orientation of carnassial blades;
c) strong parastyle on upper carnassial in many early species, for example, T. brevirostris Cope (genotype) and T, rurestris (Condon), though absent in later species such as T. euthos (McGrew );
d) protocone on upper carnassial low but distinctly separated anteriorly;
e) molars with metaconule large, about equal to and symmetrical with protocone (Olsen, 1956a);
f) molars lacking protoconule (this feature is probably correlated with the preceding);
g) talon of first upper molar tending to become wider anteropostc-riorly and square, exemplified in T. paulus Henshaw, T. temera-rhis (Leidy), certain specimens referred to T. kelloggi (Mer-riam), and especially in T. cf, brevirostris described by Gazin from Skull Spring, Oregon,
The omnivorous features adumbrated in certain species of Tomarctus arc carried to greater extremes in Carpocyon, In that genus the animals became larger with blunt-cusped teeth. The upper molars were enlarged with a relatively weak paracone, broad hypocone, strong lingual cingulum, and very broad external cingulum. M~ also became large and square. The parastyle of the upper carnassial was further developed with the shearing blade tending to wear more bluntly. While the molars were enlarged, the anterior premolars were relatively reduced.
It is evident that Carpocyon limosus represents the culmination of adaptive trends that began in some large Hemingfordian species of Tomarctus. Certain Barstovian species such as T. temeraritis (Leidy) and the Skull Spring Tomarctus described by Gazin (1932) may

represent a continuation of this stock. The first clearcut embodiment of these trends is Carpocyon cuspidatus (Thorpe) of late Barstovian age.
In addition to the type palate of C. cuspidatus, also in the Yale collection is a lower jaw fragment collected by the same collector, assigned the same number, and preserved in the same manner as the two upper jaw fragments constituting the type. Thorpe (1922) mentions it as "a fragment of a jaw without teeth." The significance of this fragment, if properly associated, is that it indicates the presence of a long two-rooted M2 at least 12 mm long and a rather elongate Ms at least 6 mm long behind the lower carnassial. The jaw is 20 mm deep below M2. These teeth further indicate the relationships of Carpocyon to the main line of canid evolution.
The principal advancements by Carpocyon limosus over C. cuspidatus are the following:
a) P heavier cusped, and broader;
b) lingual cingulum heavier, supported by added lingual root;
c) P4 parastyle stronger;
d) protocone less projecting, more bulbous;
e) anterior end of lingual cingulum added;
f) MJ talon squared up (11.0 long in C. limosus, compared to 10.2 in C. cuspidatus);
g) M- enlarged, elongated posterolingually.
Each of these features tends to produce a heavier grinding battery. With the exceptions of (b) and (e), they augment adaptations already observed in Carpocyon cuspidatus. The direction of these trends, and the fact that they were intensified during the Barstovian through Hemphillian interval suggests their initiation in an early Barstovian species of Tomarctus.
Subfamily Borophaginae Simpson Osteoborus Stirton and Vanderhoof Osteoborus dudleyi (White) new combination
Figure 2, Table 3
Pliogulo dudleyi White, 1941. Osteoborus crassapineatus Olsen, 1956b.
Holotype.MCZ 3688, a skull lacking right zygoma and all teeth except RP3.

bulletin florida state ml'sel'm
vol 14
ficwie 2. Oiteoborus dudleyi. a. uf 10335, left p-m. b. fcs v 5644, l.i-lt iii.iiniilil)'. iliilnl\p.- of O. cnuutipinealua. nnlural size
Type Locality and Horizon.Phosphate pits near Mulberry, Polk County, Florida. Exact locality unknown. From Bone Valley Gravels.
revised diagnosis.Moderately large species of Ostcoboru* with short face, high-vaulted cranium and deep jaws with massive canines.
table 3. measurement* or the upper dtvrmov or Chteoborut Dmlleui
(in nun)
O. cyouotde*
(> dudleyi ucmp
upper holotypc up uf (krgert)
dentition mcz 3088 10335 12401 30103. 30101
p length x width 12,3x7.1 11.8 x 7.0
1" lenfth x width 27.6x14.0 27.5 x 11.0
m* length x width it)' n 21' 175x23.4 16.1 x 18.3 16.7x23.1
m1 length x width 0.1 x 12.0 mix 11:
l"-m' min. length 38.3 38.7
1" m min. length 42.2

V1 absent, P:J two-rooted, P4 with posterior accessory cuspid and posterior cingulum compressed but distinct, P4 with strong parastyle, distinct protocone and posterolingual cingulum, Ma with strong meta-conid and short bicuspid heel. Closely comparable to O. cyonoides, but differing in larger size, higher vaulted skull, and stronger parastyle on P4. Comparable in size to O. hilli, but less progressive in premolar reduction and simplification of the upper carnassial.
Referred Material.FGS V-5644, right mandible with canine and P4-M1; broken behind Mi and lacking symphysis, type of O. crassapin-eatus, from the American Agriculture Chemical Company phosphate pits at Pierce, Polk County, Florida, UF 10335, right maxillary fragment with PA-M2 and UF 12401, RM\ both from Palmetto Washer (UF locality catalog) near Pierce, Polk County, Florida.
Synonymy.In 1941 White based a new genus and species of large carnivore on a nearly complete but essentially edentulous skull from somewhere in the vicinity of Mulberry, Although he recognized it as borophagine, he made it the type species of a new genus, Pliogulo. The generic name selected by White and the fact that most of his comparisons were with the wolverine has led some to suppose it a mustelid. Olsen (1956b) described a borophagine lower jaw from the same area as a new species of Osteoborus. Though he suggested Pliogulo White might be a synonym of Osteoborus, he failed to discuss the possibility of his specimen being referable to White's species. Study of the type specimens of both these species, as well as additional new materia], makes it clear to me that only one borophagine species occurs in the Bone Valley Formation, Osteoborus dudleyi (White).
Description of Type.The type skull of Osteoborus dudleyi compares closely with those of other species of Osteoborus, particularly O. cyonoides, Hemphill Fauna (Matthew and Stirton, 1930) and O. validus, Higgins Fauna (Johnston, 1939a). The two latter species represent a more progressive stage of evolution, as indicated by the high arched cranial vault and shortened face, than O, pugnator and especially O. littoralis. The top of the skull is even higher and more rounded in O. dudleyi than in O. cyonoides and O. validus.
It is not surprising that differences between Osteoborus skulls of a progressive type and those of Borophagm should be so subtle in view of the direct phylogcnetic relationship indicated by dental studies. Direct comparison of the skull of Osteoborus dudleyi with skulls of Borophagus diversidens from the Blanco and Cita Canyon faunas of Texas reveals certain differences in detail. The most striking differ-erence is the broad postglenoid process in O. dudleyi, which extends

far lateral to the external auditory meatus, whereas in B. diversidens the process is narrow and tightly pressed against the auditory tube. Also the postglenoid foramen faces more anteriorly in B. diversidens than in O, dudleyi, These comparisons strongly suggest that White's borophagine pertains to Osteoborus.
The only previously known Bone Valley material representing the upper dentition is the type skull of Osteoborus dudleyi, which has the alveoli for most of the teeth, but only RP3 actually present Additional material now includes an upper jaw with cheek teeth and an upper molur. In most respects the upper dentition of O. dudleyi compares closely with that of O. cyonoides as represented by the UCMP sample from Coffee Ranch Quarry (Hemphill Fauna) in Texas. The sample from the Coffee Ranch Quarry was used in this study as a control to indicate the range of variation in features of Osteoborus. The principal difference between 0> dudleyi and O. cyonoides is the larger size of the Bone Valley srjeeies.
P3 in the type of O. dudleyi is well worn, but retains indications of three weakly separated cusps aligned antoropostcriorly. The middle cusp is the largest and was presumably the highest. A narrow lingual cingulum occupies the lingual side of this tooth. The anterior end of P3 lies medial to Pa as in O. cyonoides rather than lateral to it as in Borophagus pachyodon. P3 overlaps P- only slightly. P1 lies almost directly behind the canine rather than overlapping it.
The upper carnassial is represented in the maxillary, UF 10335. The protocone protrudes rather far rnigiially and rises to a distinct cuspule. Although this cuspule is not common, it appears as a variant in the Hemphill sample of O. cyonoides. The parastyle is connected by a strong crest to the protocone. This crest is better developed than in any specimens observed from the Hemphill Fauna. The paracone and metacone contribute about equally to the carnassial blade, although the paracone is taller and more robust than the metacone. A weak cingulum occurs on the lingual side ol the metacone.
In both first upper molars the protoconal shelf is rather narrow with a sharp posterolingual angle. The external cingulum is very weak. A distinct anterolingual cingulum below and immediately anterior to the protoconule occurs in both Bone Valley specimens. However the Coffee Ranch sample of Osteoborus cyonoides demonstrates that in M1 all cf these features may vary greatly within a population (Matthew and Stirtoix 1930, Plate 28).
M2 is rather large but offers nothing in the way of distinctive features.

The only known lower jaw of this species is a right mandible with C and P^Mi, from near Pierce, Bone Valley District, Florida, described by Olsen (1956b) as the type of O. crassapineatus, here syn-onymized with O. dudleyi.
The dimensions of the lower canine in the Bone Valley mandible are remarkably large. Olsen considered this one of the major diagnostic features of the Bone Valley Osteoborus, although later in the same paper he notes that this "may be just an anomalous individual that would not be representative of the borophagine population from Bone Valley Formation/*
In the large sample of jaws from the Coffee Ranch Quarry, it is apparent that canine size is subject to considerable individual variation. It is also possible that a unit increase in the body size of these canids brought about an exponential increase in the size of the canine. In O. hilli from Axtel local fauna in Oklahoma (Johnston, 1939b) which is closely comparable in size to O. dudleyi, the lower canine ranges in size from an anteroposterior dimension of 12.3 mm and a transverse dimension of about 10 mm in the type specimen, to dimensions of 18.8 x 12.5 in CWT 1756. Thus the observed range of variation in a few specimens is quite wide and at the upper end approaches the dimensions of the Bone Valley jaw (19.7 x 15.3). In a specimen of O. validus from Higgins, Texas described by Johnston (1939a) the dimensions of the lower canine are 19 x 17.5 mm.
The very heavy wear on the canine is surprising in view of the light wear on the protoconids of P,, and Mt. The canine is worn most heavily on the labial side, but is also heavily worn on the dorsal surface and the lingual side. Presumably the lingual surface wears against the large upper third incisor.
The symphysis in the Bone Valley mandible is extraordinarily long and deep. The symphyseal region extends as far back as the anterior root of P4, whereas in most other Osteoborus specimens the posterior edge of the symphysis lies below P3, The jaw deepens from the molar to the symphyseal region in the Bone Valley jaw, in contrast to smaller Osteoborus jaws that usually become more slender anteriorly. Nevertheless the jaw proportions are quite variable in modern wolves and their is little doubt that they varied considerably in populations of Osteoborus as well. Presumably the massive symphysis in the Bone Valley jaw is functionally related to the powerful canines.
The appearance of the alveoli in the Bone Valley jaw suggests that Pi was absent and shows that P3 was doubled rooted. The number of roots on P2 cannot be judged as the critical area was broken away.

In P4 the principal cusp is high and inclined posteriorly. An antero-posteriorly compressed posterior accessory cusp separates the principal cusp from the posterior cingulum in the fashion characteristic of Osteo-borus. The posterior cingulum is also compressed and is best developed on the labial side. In typical Osteoborus fashion P4 is set well labial to the carnassial and the jaw is bowed outward at this point.
The lower carnassial exhibits a short bicuspid heel and a distinct metaconid. The heel is well worn exposing a flat unenameled surface. By contrast the high protoconid is only lightly worn.
Relationships.A critical nomenclatural point is the synonymy of "Pliogulo" dudleyi based on a skull and Osteoborus crassapineatus (misspelled crassipineatus under the type figure) based on a mandible. An extreme typologist could not be satisfied on this point because as yet no upper and lower jaws of this species have been found in association and the two type specimens certainly did not belong to a single individual. Yet that each of these specimens belongs to the genus Osteoborus can be demonstrated, and their similar size and provenience makes it highly probable that they came from the same population of Osteoborus. Agreement in size between the two specimens may be shown directly by comparing certain dimensions that are functionally correlated and indirectly by comparing each specimen to corresponding elements in closely related species that are represented by adequate samples.
In Osteoborus and, indeed, in late Ccnozoic can ids generally 9 the lower carnassial tooth is very close to 75% of the combined lengths of the upper carnassial and the first upper molar of the same individual. Clearly this relationship must be functionally controlled within narrow limits. In the type mandible of O. crassapineatus Mx is 31.0 mm long. In the type skull of Pliogulo dudleyi the minimum length of P1 and M1, established by the alveoli, is 38.3 mm; thus the lower carnassial of O. crassapineatus is no more than 81%most probably a few percent lessof the length of the corresponding teeth of Pliogulo dudleyi, This is a reasonably close comparison for different individuals. Similarly the length Pi-P3 inclusive is slightly greater than PT-P3 in most individual canids. In O. crassapineatus the length of Px-P3 is 28.5 while in P. dudleyi the corresponding length of the uppers is 26,7. Furthermore, the distance from the lower canine to the posterior end of Mi is 79.9 in O. crassapineatus; while in P. dudleyi the distance from the canine to the middle of M1 is 79.0. Thus the two critical type specimens compare closely in functional dimensions. It should be

apparent from the previous description that each specimen is referable on its own merits to Osteoborus. Hence the probability that the two species in question are synonymous is great. Following the rules of priority and the subjective generic and specific synonymies just indicated, the proper combination is Osteoborus dudleyi (White).
O. dudleyi compares most closely with O. cyonoides from which it differs in its consistently larger size. This is particularly evident in the huge canines and massive jaw. The skull is more highly vaulted than in O. cyonoides, and the parastyle of the upper carnassial is stronger than in any of the Osteoborus samples from the Hemphill (Coffee Ranch).
0. dudleyi agrees in size with O. hilli from Axtel fauna in Oklahoma (Johnston, 1939b). O. hilli is among the most progressive species of the genus (see following discussion of O. ore) and shows several advances beyond the stage of O. dudleyi. The last lower premolar is higher and more compressed posteriorly in O. hilli than in O. dudleyi. In the upper carnassial O. dudleyi retains a distinct protocone, a posterolingual cingulum, and a strong parastyle, whereas in 0. hilli (as in Borophagus) all of these features are nearly suppressed.
Osteoborus ore new species Figures 3, 4, Tables 4, 5, 6
Figuiie 3. Mandible of Osteoborus ore new species. Holotypc, UF 13180. Occlusal and lateral views. Natural Size.

Vol. 14
FIGURE 4. Maxillae of Osteoborus ore new species. A. UF 12314, occlusal view. B. UF 12313, occlusal and lateral views. X1V4

Upper Dentition UF 12313 UF 12314 UF 12350 UF 12316 UF 12317
C length x width 1.1.2 x 7.4 10.0 x 7.6
P2 length x width 8.4 x 4.8 7.6 x 4.2 7.2 x 4.01
length x width 11.0 x 5.9 10.8 x 5-5 11.5 x 6.01
P4 lengths width 19.5x11.8 18.3 x 9.9 19.4 x 9.51
M1 length x width 12.6 x 17.4 12.8 x 16-9
C-Ps length 29.0 27.6
C-P^ length 48.0 47.2
C-M- length 58.5 60,0
Lower Dentition UF 12320 UF 12326 UF 12319 UF 12321
C length x width 11.6 x 9.7
P., length x width 4.6 x 6.0
P3 length x width 7,9x5.0
P4 length \ width 14.1 x 8.5
M1 length 211
M, length x width 10.4x6.9 10.0x6.8
C-P4 length 44.2 ll1
Mx-M3 length 38
P4-M3 length 52
C-M3 length 761
Depth below P4 24
Depth below Ml 25
Etymology.Mytholgical. Orc=& ruthless carnivorous creature serving the forces of evil in the Third Age of Middle Earth (see Tolkien, 1965, especially Chapter 3, part 2).
iioLOTYPE.LTF 13180, right mandible with P.-M2 and alveoli for C, Pi and P^.
Material.UF 12313, left maxilla with P-M1; UF 12314, right maxilla with P^M1; UF 12315, right maxilla and premaxilla with canine tooth and alveoli for I1 and P; "3; UF 12317, two upper carnassial teeth; UF 12350, left maxilla with P1-?1; UF 14782, P3; UF 12319, anterior mandibular fragment; UF 12320, right mandible with P2j PR and Ma; UF 13181, left mandible with P-Mi; UF 12321, left mandible with M2; UF 12326, left mandible with alveoli for P,-Pa; UF 14783, right mandi-
Table 4. Measurements of the Dentition of Osteoborus ore (in mm)

ble with Ml,; UF 14785, right mandible with Px and P3; UF 14786, three mandibular fragments; UF 14781, P,; UF 14784 three M/s; UF 12323, atlas; UF 12324, two axes; UF 12332, lumbar vertebra; UF 12333, two humeri; UF 12327, four radii; UF 12328, 3 ulnae; UF 12325, two pelves; UF 12329, femur; UF 12330, two tibiae; UF 12331, 17 metapodials, Collected by Robert Allen, Norm Tessman, and Kent Ainslic, May, 1967.
Type Horizon and Locality.Withlacoochee River, locality 4A, (UF locality catalog), about 8 miles southeast of Dunnellon, Marion County, Florida. Fossil vertebrates preserved in massive green clay, filling a sinkhole developed in Inglis member of Ocala Limestone (Late Eocene). This is the second report on parts of the fauna (Hirsclrfield and Webb, 1968); other elements clearly indicate a Hemphillian age.
Diagnosis.The smallest known species of Osteoborus, with relatively shallow jaws and delicate canine teeth. Anterior premolars greatly reduced; Pt and PA present or not. Length P1_a 65% of length P'-M1. Y.> single rooted. P2 and P3 without distinct accessory cus-p(id)s. P4 with strong parastyle, weak protocone connected by two low ridges to parastyle. M1 paracone much stronger than metacone; hypocone narrow and strongly skewed posterolingually; nearly continuous labial cingulum. P, with posterior accessory cuspid and high, wide posterior cingulum tightly appressed, advanced well beyond cyonoides and approaching pugnator in these features; anterior slope crested with cingular cusp at base. M1 witli weak metaconid, relatively shorter talonid, with basin closed in by hypoconulid much more than in cyonoides. M2 elongate, with strong labial cingulum, greatly reduced trigonid, lacking entoconid.
Description of Material.P1 is represented in both UF 12313 and 12315 by a single alveolus 3.0 mm in diameter, but in UF 12314 no alveolus for P1 appears. P2 and Ps are low short teeth, closely spaced and obliquely set in the jaw, each supported by two unfused roots. The crown of P2 consists of a principal cusp less than 4 mm high from which oblique descending crests reach the ends of the crown. On the lingual side is a weak but continuous cingulum. P3 is similar in shape but larger. The principal cusp is about 6 mm high. Small anterior and posterior accessory cusps are indicated by swellings on the longitudinal crests.
The upper carnassial retains a strong parastyle that rises to nearly half the height of the paracone-metacone blade. It wears very heavily, A small anterior protocone is connected to the parastyle by two thin

Atlas vertebra maximum width anterior width centrum maximum depth centrum dorsal length centrum ventral length centrum
Axis vertebra length dorsal crest length centrum width across zygapophyses anterior width centrum
First Lumbar Vertebra length centrum posterior width centrum posterior depth centrum
Humerus maximum distal width maximum distal articular width height anconeal fossa
maximum length
proximal transverse/anteroposterior diameter
distal transyerse/antero-posterior diameter
Ulna proximal transverse anteroposterior diameter depth of humeral notch distal transverse/antero-posterior diameter
minimum depth ilium shaft maximum height acetabulum
Femur length
maximum width distal end maximum depth distal end
proximal transverse/anteroposterior diameter
distal transverse/antero-posterior diameter
UF 12323 ca. 78 36 24 16 11
UF 12324a 50 41 32 28
UF 12324b ca. 55
UF 12332 23 20 12
UF 12333a 48 29 17
UF 12333b 43 28 12
UF 12327a UF 12327b UF 12327c UF 12327d 124
20/15 21/15 27/17
UF 12328a 14 25 19
UF 12328b
UF 12328c
UF 12325a 19 23
UF 12325b 19
UF 12329 ca. 110 25 30
UF 12330a 35/36
UF 12330b
Table 5. Skeletal Measurements of Osteoborus ore (in mm)

Metatarsal IV UF 1233In UF 1233lo UF 12331 p
length 52 58 60
proximo] transverse/antero- 8/12 10/12 9/13 posterior diameter
Metatarsal V UF 1233lq UF 12331r
length 44 oa. 49
proximal transversc/antero- 9/10 8/10 posterior diameter
discontinuous crests. In worn specimens the protocone appears as a mere wear surface on the basal swelling of the tooth. The carnassial blade wears along two surfaces: bluntly on the tops of the paracone and metacone, and to a lesser extent vertically on the lingual side in the usual sectorial fashion. Both wear surfaces are marked by very deep striae in the four available camassials. A strong posterolingual cingulum lies at the top of the shearing surface.
The first molar presents a relatively high paracone and by contrast a very low metacone. In the slightly worn specimen the paracone rises 5.0 mm above the labial cingulum, whereas the metacone is only 2.7 mm higher. The labial cingulum is moderately developed, but is cut by the valley between the two principal cusps. The lingual portion of the tooth is relatively narrow with the hypocone strongly skewed posteriorly. The lingual part of this tooth most nearly resembles the slenderest, most skewed variants in the Hemphill (Coffee Ranch) Fauna from Texas (Matthew and Stirton, 1930, plate 28, figure b), but both specimens here are even more extreme.
As in Osteoborus generally, a cingulum extends from the hypo-cone around the anterolingual side of the tooth, and fades out at a point well labial to the protocone. The protocone is unusual in its relationships. The anterior crest from the protocone produces two strong cuspnles and then forms a strong cingulum that borders the anterolingual slope of the paracone and joins the anterior end of the labial cingulum.
An analogous cingulum, without cuspnles, runs from the metacon-ule posterolabially to the posterior end of the labial cingulum. These ridges are much heavier in Osteoborus than in Canis lupus or C. latrans.
The mandible of O. ore is relatively shallower than in other species. In keeping with its small size this species has rather delicate canines.
The lower premolars are closely crowded. Pi is present in one lower jaw (UF 12319), but two other jaws lack alveoli for that tooth.
Table 5. (cont.)

Pa is single rooted, though a groove on the lingual side indicates that it was derived by the fusion of two roots.
Pn presents only a low principal cusp from which longitudinal crests descend anteriorly and posteriorly. There are no accessory cusps or cingula. The tooth is supported by two roots.
The principal cusp in P} presents a strongly convex anterior profile, in this respect resembling Borophagus more than other species of Osteoborus. This convex anterior slope is produced by a strong anterior crest and by an anterior cingular cuspid at its base. The posterior accessory cuspid is weak, but distinct. The posterior cingulum is strongly compressed anteroposteriorly and the posterior edge of P4 is square cut as in Borophagus and Osteoborus progressus Hib-bard (1944).
The trigonid and talonid of M2 are equal in length. The cusps of the trigonid are low, but distinct. The paraconid occupies the middle of the anterior edge of the tooth, and the metaconid lies directly opposite the protocone. A broad triangular cingulum squares up the anterolabial corner of the tooth. The talonid forms a broad shallow concavity marked only by a low hypoconid. The entoconid is absent in both specimens, one unworn, in this sample.
The skeletal remains of O. ore differ from those of Canis lupus only in proportions and in minor osteological features. The anterior part of the atlantal arch is very heavily built, the odontoid process of the axis is unusually stout, and the zygapophyses are wide and very sturdily constructed.
The limb elements are heavy jointed. The distal end of the humerus in particular is much broader than in similar sized modern Canis. One specimen has a well developed entepicondylar foramen that is absent in a smaller specimen. The radius and metapodials are much shorter and yet broader than the corresponding elements in Canis lupus. The short broad limb structure has received comment in larger species of this genus (Matthew and Stirton, 1930); it seems even more remarkable in so small a species as Osteoborus ore.
Discussion.Osteoborus ore is the smallest of the species referred to either Osteoborus or Borophagus. It falls distinctly below the range of variation in the large sample of O. cyonoides from the Hemphill Fauna of Texas, and it is about one-third smaller than the small California species, O. diabloensis Richey and O. littoralis Vanderhoof. An upper jaw from the Upper Snake Creek beds, referred by

Cook and Macdonald (1962) to Borophagus sp., approaches but does not equal this species in smallness of its dimensions.
The nearest comparison in size, as well as in other features, is with O. progressus Hibbard (1944) from the Saw Rock Fauna of Kansas. The comparable teeth are lower P3, P4, Mi and M2, and in these O. ore is 20 to 25% smaller than O. progressus.
A small canid maxillary from the Upper Snake Creek beds was referred by Cook and Macdonald (1962) to Borophagus sp. Although its premolars are comparable in size to those of Osteoborus ore, the critical features are inadequately known to determine its relationships. It may well belong in Osteoborus near O. secundus.
The premolars of Osteoborus ore are crowded and greatly reduced. In this respect O. ore is one of the most advanced species known (along with O. hilliy O. pugnator, and O. progressus), and is comparable with some species of Borophagus. For example in the two upper jaws of O. ore the length of P1"1 is 62% and 68% respectively of that of P^-M1, and in Borophagus solus the length of P1-3 is 64% of that of PMVI1. In O. hilli this percentage is about 60%. In most species of Osteoborus this percentage lies between 70% and 80% (Table 6), the premolars being less reduced.
Other progressive features of the premolars of Osteoborus ore are the complete suppression of the accessory cusps or cuspids on P2, Po, and P3 and their near absence on PK, weak protocone on P4, single rooted P2, P4 with convex anterior slope bearing a crest and cingular cuspid, and with a compressed square-cut posterior cingulum. The very strong wear on the parastyle of the upper carnassials, noted above in three of four specimens, is accomplished by the robust principal cusp of P4. As this cusp becomes enlarged still further (in Borophagus) the parastyle against which it grinds "fades" into a long vertical crest on the anterior slope of the paracone of the upper carnassial. O. ore approaches Borophagus in the heavy wear on the parastyle of P4, but the parastyle remains strong.
The upper first molar also represents a progressive state in Osteoborus evolution. In the great inequality between the paracone and metacone and in the narrow skewed talon this tooth closely resembles that of Borophagus. The last lower molar is relatively reduced in size and lacks the entoconid, both advanced features for Osteoborus.
Most of these features noted in ore represent advanced stages in the evolution of Osteoborus into Borophagus. A number of features, such as the degree of premolar reduction, fusion of roots on Pa, narrow skewed structure of M1, and the loss of the entoconid on M2, suggest

Table 6. 1/pfer Dentition Measurements or Certain Borophagine Species
(in mm)
O. dudleyi MCZ 3688 O. hilli Holotype O. cyonoides UC 30115 O. littoralis Holotype O. diabloensis Holotype O. ore (mean two jaws)
Canine length 19 17.5 16 14.5 10.0
Canine width 12 11 10 ca. 7 7.6
P1-3 length 26.7 26.0 25.5 29.8 30.0 20.1
Pi-M1 length 38.3 39.6 35.8 37.0 37.4 30.9
Ratio P1-3 70 66 71 80 80 65
Borophagus pachyodon B. pachyodon B. matthewi B. solus
Holotype UC 32464 Holotype Holotype
Canine length 18 19 20 18.5
Canine width 11 13 13 11.4
P13 length 27.1 25.9 27.5 23.9
P+-M1 length 37.2 42.3 36.5 37.3
Ratio P1"3 73 61 75 64

that Osteoborus ore might be the proximate ancestor of Borophagus, along with O. secundus and O. progressus. However the retention of a
of Borophagus. Moreover the anterior crest and cingular cuspid of are peculiar features not observed in Borophagus or other Hemphillian material of Osteoborus. These features are best developed in Aeluro-don and therefore represent primitive characters of the borophagine tribe.
In most respects O. ore is a progressive species of Osteoborus. While not so likely ancestral to some or all species of Borophagus as arc O, hilli or O. progressus, it is nearly as advanced. Presumably this species is from the later half of the Hemphillian age,
Osteoborus galushai new species Figure 5, Table 7
Etymology.Named in honor of Ted Galusha of the Frick Laboratory for his extensive contribution to knowledge of the carnivores and to vertebrate paleontology in general,
Hototyptc.Mandible with nearly complete left and right dental series, FAM 61671 and 61672.
Type Locality and Horizon.Alachua Clay, Mixson's Bone Bed sinkhole, northeast of Williston, Levy County, Florida. Associated with the Hemphillian faunule described by Leidy and Lucas (1896), and Simpson (1930).
Referred Material.All from the type locality: left maxillary fragment with P-, P4 and alveoli for C and P3, FAM 61675; right maxillary fragment, FAM 61557; left maxillary fragment with M1 and M2; FAM 61677; maxillary fragment with P-Ma and alveoli for P1 and P2, FAM 61676; upper canine, FAM 61678; zygomatic arch, FAM 61674a; right ramus with Mn, FAM 61673; right ramal fragment with M2, FAM 61674; and lower canine, FAM 61673h.
Diagnosis.Close to O. cyonoides in size and other features Pi present; anterior premolars widely spaced. Differs from O. cyonoides in the higher cusp on F2 and the higher cusps and distinct heels on P^ and P^; Pt with slender upright principal cusp and strong posterior accessory cusp. M3 elongate and strongly cuspidate as in O. cyonoides.
Description.The single rooted first premolars are represented by alveoli in the mandible. They are separated from the canines by diastemata of 2 mm. The succeeding premolars are loosely spaced, not

FlGUBE 5. Mandible of Osteoborus galushai new species. Holotype, EAM 61671 and 61672. Occlusal and left lateral views. Natural Size.
crowded or overlapping as in advanced species of Osteoborus. Though P2 and P;, lack distinct accessory cuspids, the principal cusp is high and slender, not bluntly rounded as in O. secundus. A distinct posterior heel remains in each tooth. In P., the principal cusp is slender and nearly vertical as compared with the bulbous backswept cusp in 0. secundus and other more borophagine species. A sharp crest marks its anterior slope. The posterior accessory cusp is high and not compressed against the posterior cingulum. A lingual cingulum is continuous.
mi agrees closely with the same tooth in (). cyonoides. The meta-conid is strong and the talonid relatively long. The talonid basin is divided by a low hypholophid into a large anterior and a smaller posterior basin. M3 is relatively long and thus agrees with O. cyonoides and differs from more advanced species such as O. secundus. It bears

Table 7. Measurements of Dentition of Osteoborus galushai (in mm)
Dentition FAM 61676 FAM 61675 FAM 61677
P2 length X width 9.5 x 5.9
P3 length X width 11.9 x 5.8
P4 length X width 23,5 x 10,7 24,8 x 12.6
M1 length X width 16.3 x 19.2 16.5x19,7
M2 length X width 9.1 x 13,7 11.2x13,6
P1-?1 length 51.8 52.3
P^M2 length 69.5
Lower Dentition FAM 61671 FAM 61672 FAM 61673 FAM 61674
F2 length X width 8.2 x 4.7 8.3 x 4.7
P^ length X width 9.6 x 5,4 9.6 x 5.4
P4 length X width 15.9 x 9.3 15.9 x 9.3
M1 length X width 25.6 x 9.5 25.9 x 10.7
Mn Trigonid x Talonid 19.4 x 6.4 17.6 x 8,0
M2 length X width 12.3 x 7,9 13.9 x 8.3
P2-4 length 35.4 35.1
P4-M2 length 52.7 ca. 57
Depth jaw below P4 29 28
a distinct trigonid and a longer basined talonid as in O. cyonoides. Ms is represented by a single elongate alveolus.
Two mental foramina occur on the labial side of the jaw about 12 mm below the anterior ends of P4 and P3 respectively.
The upper premolars of O. galushai are not so closely spaced as in many species of Osteoboms. P1 is relatively large, but single rooted. The alveolar diameter in FAM 61675 measures 5 mm, P2 and P3 are double-rooted elongate teeth. No anterior accessory cusp is evident in P2, whereas in P3 it is a distinct cusp nearly as large as the posterior accessory cusp. The principal cusp in P2 is higher than in O. cyonoides and bears stronger anteroposterior crests. Its unworn height in FAM 61675 is 5.7 mm. P3 agrees closely with that in O. cyonoides.
Two of the three upper carnassials of O. galushai differ little from those of O. cyonoides. The third (FAM 61675) differs markedly in the structure of the protocone. In it the protocone forms a relatively bulbous protuberance from the anterolingual corner of the tooth. Three distinct cuspules descend one beneath another in the protoconal area; the largest is lowermost and touches the parastyle. This character must be regarded as an atavistic peculiarity of one specimen, rather than a diagnostic feature of the species.
M1 docs not differ significantly from those of O, cyonoides. In one specimen of (FAM 61677) the labial moiety is wider and the meta-

cone is relatively larger than in observed specimens of O. cyonoides. However a second specimen (FAM (51676) agrees in every detail with specimens of O. cyonoides.
REi-ATio\siiiPs.-().s7r<>/;o/m- galushai from Mixson's most nearly resembles (). cyonoides from the Fdson Beds in Kansas. The principal differences are the presence of Pi and the widely spaced premolars. Further differences appear in the more slender, higher cusped, less boropbagine structure of the premolars, and suggest a slightly earlier stage of evolution than (). cyonoides.
Osteoborus cf. O. galushai Figure 6, Table 8
material.UF 12304. right mandibular ramus with M: and roots or alveoli for C through M UF 12312, RM,; UF 12307, LC; UF 12305, RP'; UF 12306, LMl; UF 15129, RM,; UF 12309, thoracic vertebra. UF 12310, left metacarpal V; UF 12311 left metacarpal II; and
Ficcre 0. Mandible of Osteoborus cf. O. galushai. A. and B. Occlusal and lateral views, UF 12304, right mandible with M. C. UF 12312, right M X l'/i. All natural size.

o o
Table 8. Compahative Measurements of Dentition of Osteoborus cf. galushai
(in mm)
Upper Dentition
P4 length x width M1 length. M2 width
Lower Dentition
length x width
Mj length x width M2 length x width Length between C and P4 Length between C and M,
Depth Jaw Below P4 Depth Jaw Below M1
UF 12305 McGehee
281 x 14.2
UF 12402 Hogtown
17.81 24.2
O. cf. galushai
UF 12304
11.51 16.51 29.71 14.3x10.1
UF 12312
32.8 x 12.8
O. dudleyi FGS V5644
17.8 x 12.8
30.9 x 12.6
CWT 2419
18.4 x 12.8 30.1 x 12.6
Osteoborus hilli CWT 1558
17.2 x 12.0 30.5x12.8
CWT 1643
19.3 x 14.1 32.1 x 13.6
c r t-1
f O
1-3 >

UF 10990, left calcaneum, all from MeGehee Farm Locality, and UF 12402 RM1 from Hogtown Creek II Locality.
Horizons and Localities.MeGehee Farm Locality, 5 miles north of Newberry, Alachua County, Florida. In fluviatile clay, sand, and gravel deposits of the Alachua "Formation." Early Hemphillian age (Webb, 1964; Ilirschfeld and Webb, 1968). Hogtown Creek II Locality on western edge of the city of Gainesville, Alachua County, Florida; Hemphillian fossils, including Geochelonc alleni described by Auffenberg (1966), occur in fluviatile phosphatic sands and gravels.
Description.The MeGehee jaw is slender and tapers in the symphyseal region, thus contrasting with the Bone Valley jaw of massive ramus and deep symphysis, Such differences may be taxonomieally significant, but they may also correlate with individual differences in size, sex, and age.
No alveolus for Pt appears in the jaw of O. cf galushai from MeGehee. P2 was evidently single-rooted with an alveolus about 5 mm in diameter. P3 was double-rooted and about two-thirds the size of P4. Such a large P^ suggests a rather primitive stage of Osteoborus evolution. On the other hand, one of the Hemphill (Coffee Ranch) jaws of O. cyonoides (UCMP 30113) has a P3 proportionally as large as the MeGehee form. P4 is set well labial to Ml3 and the jaw is bowed labially in the manner characteristic of Osteoborus. Unfortunately the crown of P< is not represented.
The lower carnassial tooth, UF 12313, agrees in size with the roots in the jaw of UF 12304 and presumably represents the same species. It is rather well worn, but still retains a distinct broadening at the metaconid. The carnassial cusps are worn flat as in boropha-gines generally.
In the MeGehee mandible M2 is well developed. A distinct para-lophid occurs, and the metaconid lies opposite the protoconid. The cusps of the talonid are no longer distinct, but rather a well developed lophid encircles the heel and joins the protoconid. The tooth closely resembles My of O. cyonoides, except that it is larger. There is no alveolus for M3 in this specimen.
The first upper molars from Hogtown Creek and MeGehee are heavily built in characteristic borophagine fashion. Adaptation to bone crushing is emphasized in the MeGehee specimen by the presence of an unusually broad, rounded protoconai shelf, but this character shows great variation in MJ in large samples of Osteoborus.
Relationship.In most features the sample of Osteoborus from MeGehee and Hogtown resembles that from the Hemphill Fauna in

Texas referred to O. cyonoides. However the measurements of the McGehce specimens fall above or at the upper edge of the range of the Hemphill sample and the premolars are larger and more widely spaced. The same differences distinguish the McGehee-Hogtown form from the type of O. cyonoides (Martin) from the Edson fauna of Kansas. In the larger anterior premolars, the McGehee-Hogtown sample resembles O. galushai from Mixson's Bone Bed.
In size the specimens from McGehee and Hogtown Creek agree more closely with the somewhat larger O. hilli and O. dudleyi. However the Florida specimens are clearly more primitive on the basis of the relatively long slender jaw, the relatively large P3, and the strong metaconid on the lower carnassial. The first two characters also suggest a possible relationship to the California species, O, littoralis (Vanderhoof), though M1, the only directly comparable element, is nearly square in Q, littoralis and transversely elongate in the McGehce and Hogtown populations. Possibly this material represents a larger, perhaps earlier, sample of Osteoborus galushai. Final determination must await better specimens.
Osteoborus validus (Matthew and Cook) Figure 7, Table 9
MATKRiAL.-^Right mandibular ramus with alveoli for C through M UF 12308; left metacarpal II, UF 12403; both from the fluvio-estuarine gravels, McGehee Farm Locality, 5 miles north of Newberry, Alachua County, Florida, early Hemphillian Age (Webb, 1964). RP4, FAM 61556, from Mixson's Bone Bed, Williston, Levy County, Florida.
Description.This new material adds little to our knowledge of O, validus but it does indicate the presence of that very large species in eastern United States during Hemphillian time.
The approximate lengths of the lower teeth in the edentulous jaw are given in Table 9. The canine alveolus shows that it was a very large tooth, but the dimensions cannot be measured accurately. The premolars are considerably reduced as compared with the lower carnassial. Those anterior to P4 are obliquely oriented, with the anterior end of each tooth overlapping the posterior end of each preceding tooth on the labial side. Px is represented by a large single rooted alveolus; the other premolars are double rooted.
The bottom edge of the ramus was sheared away in the McGehee jaw, but there can be no doubt that the mandible was very heavily constructed, with the maximum width of 26 mm just posterior to the

Figure 7. Mandible of Osteoborus validus. A. UF 12307, lower canine. B. and C.
FAM 61556, right V4, x IM. D. UF 12304, mandible with alveoli for Pj-Mj. All natural size.
symphysis and a depth below Mj of about 57 mm. The positions of the two mental foramina agree with those in the type of O. validus and in the referred jaw from the Higgins Fauna (Hesse, 1940). They lie about 21 mm below the alveolar border, one directly below the posterior root of P;), and the other below the: posterior root of Pa. The isolated from Mixson's Bone Bed has two closely appressed roots, a low anterior cingular cusp, and a heavy principal cusp. Although the posterior portion of the tooth is broken, it evidently consisted of a heavy posterior accessory cuspid and a broad posterior cingulum. The roots of the tooth have a vertical dimension of 33 mm.
rklationshups.In their original description Matthew and Cook (1909) regarded validus as a subspecies of Aelurodon haijdeni that differed from the type "in the shorter and more crowded premolar

Table 9. Measurements of Teeth and Metacarpal of Osteoborus cf. validus
(in mm).
Teeth UF 12308 FAM 61556
P1 length 61
P- length 11.51
P3 length 151
P4 length x width 211 ca. 231 x 14.2
M1 length 42]
Metacarpal UF 12403
Max. length 87.7
Prox. width 16.0
Frox. depth 20.5
Dist. width 16.5
Dist. depth 15.5
region, reduction of the tubercular teeth, and slight enlargement of the carnassial." In their study of the Coffee Ranch sample (now referred to Osteoborus cyonoides) Matthew and Stirton (1930) continued to regard validus as a subspecies of Aelurodon haydeni. Then Stirton and Vanderhoof (1933) intimated that validus was specifically distinct from A. haydeni and referable to Osteoborus, and Hesse (1940), Johnston (1939a), and Vanderhoof and Gregory (1940) supported this view. The principal basis for this generic transfer is the reduction and crowding of the premolars in Q. validus. This feature is clearly shown in the type specimen as originally noted by Matthew and Cook. This condition is not exhibited so convincingly in the Higgins Fauna sample, referred to validus by both Johnston and Hesse. In a skull Johnston (1939a) notes that "there is no crowding of the teeth, and there are short spaces between the premolars." Likewise in the mandible figured by Hesse (1940) the premolars are in anteroposterior alignment and arc separated by small gaps, although Hesse described them as "crowded together."
With respect to premolar spacing, the McGehee jaw agrees much more closely with the type of O. validus. It also agrees with the type in being slightly smaller than the jaws from Higgins.
In the type of Aelurodon haydeni the length of Yz is more than 70% of the length of P4, whereas in the Higgins, Snake Creek, and McGehee jaws, it is 60% or less. Premolar spacing is a highly vari-

able character, and one that is particularly affected by ontogenetic change. Evidently reduction in premolar size is a more reliable feature, and on this basis all the above samples are referred to validus and that species placed in Osteoborus.
Diversity of Canidae in the Pliocene of Florida
The diversity of Osteoborus species in Hemphillian deposits of Florida seems remarkable, especially if one accepts the traditional view that the land surface of peninsular Florida was restricted to a small archipelago during the Pliocene. Recent evidence shows that during some parts of the Hemphillian, the sea lay at least as low as its present level (Webb and Tcssman, 1968). Even so, four or five
Bone Valley Withlacoochee 4 A
LATE Carpocyon limosus
Osteoborus dudleyi
Osteoborus ore
Osteoborus cf galushai Osteoborus galushai'
HEMPHILLIAN Qsieooorus validus Osteoborus vaifdus
MeGehee-Hogtown Mixson's Bone Bed
Figure 8. Age and distribution of Canidae in Florida.

species in a single genus of canids is twice what is observed today in Florida, or even in North America. Nor is such localized intrage-neric diversity found in any fossil canids during the Miocene.
Two alternative explanations for the Hemphillian diversity within Osteoborus may be advanced:
1) The Hemphillian included sufficient time for a major faunal overturn, and we are sampling two or more faunal strata, or
2) the genus Osteoborus is so broadly defined that it includes several adaptively distinct lineages, and we are sampling diverse ecologies.
The evidence summarized in Figure 8 suggests that, at least with respect to Osteoborus, two temporally distinct faunal strata occupied Florida during Hemphillian time, an interval of about 6 million years. This does not mean that the later stratum evolved from the earlier, in fact this view is strongly opposed by comparisons between the known species. Rather in the later Hemphillian new immigrant species appeared as ecological replacements of the early Hemphillian species. The explanations for such faunal turnover lie beyond the scope of this paper. It is sufficient here to note the occurrence.
The second explanation for the diversity of Hemphillian Osteoborus in Florida also warrants some consideration. The morphological separation between certain species pairs is vast. Including little Osteoborus ore in the same genus as the great species of Osteoborus strains the definition of that genus, O. ore evidently occurs in an inland situation, whereas the Bone Valley species is clearly coastal. When only the early Hemphillian canid fauna is considered, the situation is less complex. Only two species are known, both of the genus Osteoborus, a large one, O. validus, and a medium-sized one, O. galushai. Similar species occur together at both McGehee and Mixson's, even though the ecologies appear to be considerably different at these sites. Thus, the early Hemphillian diversity of Osteoborus may be compared readily to that of Canis today.
Literature Cited
Auffenberg, Walter, 1966. A new species of Pliocene tortoise, genus Geochel-
one, from Florida. J. Palcontol. 40: 877-882. dc Beaumont, Gerard. 1964. Essai sur la position taxonomique des genres
Alopecocyon Viret et Simoctjon Wagner (Carnlvora). Eclogae Geol. Hel-
vetiae, 57: 829-836.
Brodkorb, Pierce. 1963. Fossil birds from the Alachua Clay of Florida. Florida
Geol. Surv., Spec. Publ. No. 2, (4): 1-17. Camp, C. L. and V. L. Vanderhoof. 1940. Bibliography of fossil vertebrates
1928-1933. Geol. Soc. Amer., Spec. Paper, 27, pp. 1-503.

Cook, Harold J. and J. R. Macdonald. 1962. New Carnivora from the Miocene and Pliocene of western Nebraska, J. Paleontol. 36: 560-567.
Dahr, Elias. 1948. On the systematic position of Fhlaocyon leucosteus Matthew and some related forms. Arkiv fur Zoologi. 41: 1-15.
Downs, Theodore. 1956. The Mascall fauna from the Miocene of Oregon. Univ. California PubL Gcol. Sci. 31: 199-354.
Galbreath, E. C, 1956. Remarks on Cynarctoides arcidens from the Miocene of Northeastern Colorado. Trans. Kansas Acad. Sci. 59: 373-378.
Gazin, C, Lewis. 1932. A Miocene mammalian fauna from southeastern Oregon. Cam. Inst. Washington Publ. (418): 37-86.
Green, Morton. 1948, A new species of dog from the Lower Pliocene of California. Univ. California Pubis., Dept. GeoL Sci., 28: 81-90.
Hall, E. R. and W. W. Dalquest. 1962. A new doglike carnivore, genus Cynarctus, from the Clarendonian, Pliocene, of Texas. Univ. Kansas Pubis., Mus. Nat. Hist., 14: 135-138.
Hesse, Curtis J. 1940. A Pliocene vertebrate fauna from Higgins> Lipscomb County, Texas. Univ. Texas Publ., (3945): 671-689.
Hibhard, Claude W. 1944. Two new mammals from the Middle Pliocene of Seward County, Kansas. Univ. Kansas Sci. Bull. 30: 107-114.
Hirschfeld, Sue E, and S. David Webb. 1968. Plio-Pleistocene megalonychid sloths of North America. Bull. Florida State Mus. 12: 213-295.
Hough, Jean R. 1948. The auditory region in some members of the Procyoni-dae, Canidae and Ursidae. Bull. Amer. Mus. Nat. Hist., 92: 67-118.
Johnston, C. Stuart. 1939a. A skull of Osteoborus validus from the early Middle Pliocene of Texas. J. Paleontol. 13: 526-530.
-. 1939b. Preliminary report on the late Middle Pliocene, Axtel Locality,
and the description of a new member of trie genus Osteoborus. Amer. J. Sci. 237: 895-898.
Leidy, J. and F. A. Lucas. 1896. Fossil vertebrates from the Alachua Clays of
Florida. Trans. Wagner Free Inst. Science, 4: 1-61. Macdonald, J, R, 1948. The Pliocene carnivores of the Black Hawk Ranch
Fauna. Univ. California Pubis., Dept. Geol. Sci., 28: 53-80, -. 1963. The Miocene faunas from the Wounded Knee Area of Western
South Dakota. Bull. Amer. Mus. Nat. Hist. 125: 143-238. McGrcw, P. O. 1935. A new Cynodesmus from the Lower Pliocene of
Nebraska with notes on the phylogeny of the dogs. Univ. California Pubis.,
Dept. Geol. Sci., 23: 305-312.
-. 1937. The genus Cynarctus. J. Paleontol. 11: 444-449.
-. 1938. Dental morphology of the Procyonidae with a description of Cynarctoides, gen. nov. Field Mus. Nat. Hist., Geol. Ser., 6: 323-339. Martin, H. T. 1930. Two new carnivores from the Pliocene of Kansas. J.
Mammal. 9: 233-238. Matthew, W. D. and Harold J. Cook. 1909. A Pliocene fauna from western
Nebraska. Bull. Amer. Mus. Nat. Hist. 26: 361-414. -, and R. A. Stirton. 1930. Osteology and affinities of Borophagus. Univ.
California Pubis., Bull. Dept. Geol. Sci., 19: 21-34. Olsen, S. J. 1956a. The Caninae of the Thomas Farm Miocene. Breviora,
(66): 1-12.
-, 1956b. A new species of Osteoborus from the Bone Valley Formation of
Florida. Florida Ceol, Surv., Spec. PubL (2): 1-5.

Rose, F. L. and W. G. Weaver. 1966. Two new species of Chrysemys (~Pseu-demys) from the Florida Pliocene. Tulane Studies Geol, 5: 41-48.
Simpson, G. G. 1930. Tertiary land mammals of Florida. Bull. Amer. Mus. Nat. Hist, 59: 149-211.
-. 1932. Miocene land mammals from Florida. Bull. Florida Geol Surv.,
10: 9-41.
Skinner, M. F., S. M. Skinner, and R. J. Gooris. 1968. Cenozoie rocks and faunas of turtle buttc, south central South Dakota. Bull. Amer, Mus. Nat. Hist. 138: 379-436,
Stirton, R. A. and V. L. Vanderhoof. 1933. Osteoborus, a new genus of dogs,
and its relations to Borophagus Cope, Univ. California Pubis., Bull. Dept.
Geol. Sci. 23: 175-182. Stock, Chester. 1947, A peculiar new carnivore from the Cuyama Miocene,
California. Bull. So. California Acad. Sci. 46: 84-89. Thorpe, Malcolm R. 1922. Some Tertiary Carnivora in the Marsh collection,
with description of new forms. Amer. J. Sci. 3: 423-455. Tolkien, J. R. R. 1965. The Lord of the Rings. 3 vols. New York; Ballan-
tine Books.
Vanderhoof, V. L. 1931, Borophagus littoralis from the marine Tertiary of California. Univ. California Pubis., Dept. Geol. Sci., 21: 15-24.
-. and J. T, Gregory. 1940. A review of the genus Aelurodon. Univ.
California Pubis., Dept. Geol. Sci., 25: 143-164.
Viret, Jean. 1933. Contribution a 1'etude des carnassiers rniocenes de la Grive St. Alban. Trav. Lab. Geol. Univ. Lyon, 21: 1-30.
Weaver, W. G. and J. S. Robertson. 1967. A re-evaluation of fossil turtles of the Chrysemys scripta group. Tulane Studies Geol., 5: 53-66.
Webb, S. D. 1964. The Alachua Formation. In Guidebook, 1964 Field Trip, Soc. Vert. Paleo., Gainesville, Florida, pp. 22-29.
-. 1966. A relict species of the burrowing rodent Mylagaulus from the
Pliocene of Florida. J. Mammal. 47: 401-412.
-. and Norm Tessman. 1968. A Pliocene vertebrate fauna from low elevation in Manatee County, Florida. Amer. J. Sci. 266: 777-811.
White, T. E. 1941. Additions to the fauna of the Florida Pliocene. Proc. New England Zool. Club, 18: 67-70.
-. 1942. The Lower Miocene mammal fauna of Florida. Bull. Mus. Comp.
Zool. 92: 1-49.
Wilson, J. A. 1960, Miocene carnivores, Texas Coastal Plain. J. Palcontol. 34: 983-1000.

v v Y Contributions to the bulletin of the Florida state museum may be in any field of biology. Manuscripts dealing with natural history or systematic problems involving the southeastern United States or the Caribbean area are solicited especially.
Manuscripts should be of medium length50 to 200 pages. Examination for suitability is made by an Editorial Board.
The bulletin is distributed worldwide through institutional subscriptions and exchanges only. It is considered the responsibility of the author to distribute his paper to all interested individuals. To aid in this, fifty copies are furnished the author without cost.
Highly recommended as a guide is the volume:
Conference of Biological Editors, Committee on Form and Style, 1964. Style manual for biological journals. (Second Edition) Amer. Inst. Biol. Sci., Washington. 117 pp.
Manuscripts should be typewritten with double spacing throughout, with ample margins, and on only one side of the paper. The author should keep a copy; the original and a carbon must be submitted. Tables and legends of figures should be typed on sheets separate from the text. Several legends or tables may be placed on a single sheet.
Illustrations, including maps and photographs, should be referred to as "figures." All illustrations are reduced to a maximum of 4-1/4 by 7-1/8 inches. Size scales, wherever they are necessary, should be incorporated into the figure.
References to literature should conform with the bibliographic style used in recent numbers of the bulletin. Spell out in full the titles of non-English serials and places of publication.
Footnote material should be kept to a minimum. However, provide copy for a footnote detailing the title, affiliations, and address of the author (see recent numbers of the bulletin).
Manuscripts must be accompanied by a synopsisa brief and factual summary (not a mere description) of the contents and conclusions, which points out the presence of any new information and indicates its relevance. In it list all new organisms described and give their ranges; indicate all taxonomic changes proposed. The synopsis, written in full sentences, should be concise, but completely intelligible in itself without references to the paper, thereby enabling the busy reader to decide more surely than he can from the title alone whether the paper merits his reading. The synopsis will be published with the paper. It does not replace the usual conclusions or summary sections. It may also serve as copy for the abstracting services.
Manuscripts and all editorial matters should be addressed to:
Managing Editor of the bulletin Florida State Museum Seagle Building Gainesville, Florida

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs