i - I THE UNIVERSITY OF ILLINOIS LIBRARY 557 CI22 V.5 Digitized by the Internet Archive in 2017 with funding from University of Illinois Urbana-Champaign Alternates https://archive.org/details/notesonquaternar5130bova UNIVERSITY OF CALIFORNIA PUBLICATIONS BULLETIN OF THE DEPARTMENT OF GEOLOGY ANDl^KVV C. LAWSON EDITOR VOLUME V WITH 42 PLATES BERKELEY THE UNIVERSITY PRESS 1906-1910 sen C V. S' CONTENTS. PACiE No. 1. Carnivora from the Tertiary Formations of the John Day Kejrion, by John C. Merriam 1 No. 2. Some Fhlentate-like Hemains from the Maseall Beds of Oregon, by William J. Sinclair ho No. 3. Fossil Mollusca from the .lohn Day and Maseall Beds of Oregon, by Bobert E. C. Stearns 67 No. 4. New Cestraciont Teeth from the West American Triassic, by Edna Wemple 71 No. 5. Preliminary Note on a New Marine Reptile from tiie Middle Triassic of Nevada, by John C. Merriam 75 No. 6. Lawsonite, ('olnmbite. Beryl, Barite, and Calcite, by Arthur S. Eakle 81 No. 7. The Fossil Fishes of California, with Supplementary Notes on Other Species of Extinct Fishes, by David Starr Jordan 95 No. 8. Fish Remains from the Marine Lower Triassic of Aspen Ridge, Idaho, by Malcolm Goddard 145 No. 9. Benitoite, a New California Gem Mineral, by George Davis Londerback, with Chemical Analysis by Walter G. Blasdale 149 No. 10. Notes on Quaternary Felidae from California, by John F. Bovard 155 No. 11. Tertiary Faunas of the John Day Region, by John C. Merriam and William J. Sinclair 171 No. 12. Quaternary Myriopods and Insects of California, by Fordyce Griunell, Jr 207 No. 13. Notes on the Osteology of the Thalattosaurian Genus Necto- saurus, by John C. Merriam 217 No. 14. Notes on Some California Minerals, by Arthur S. Eakle 225 No. 15. Notes on a Collection of Fossil Mammals from Virgin Val- ley, Nevada, by James Williams Gidley 235 No. 16. Stratigraphy and Palaeontology of the San Pablo Formation in Middle California, by Charles E. Weaver 243 No, 17. New Echinoids from the Tertiary of California, by Charles E. Weaver 271 589279 i>a(;e Xo. 18. Notes on Echiiioiils from the Tertiary of California, by It. W. Pack 27i) No. If). Pavo californiciis, a Fossil l^eacock from the t^uaternay .\s])halt Peds of Rancho I^a Brea, by Loye Holmes Miller. .. 285 No. 20. The Skull and Dentition of an Extinct Cat, closely allieil tri Felix atrox Eeidy, by John C. Merriam 2!)I No. 21. Teratornis, a New Avian (ienus from Rancho La Brea, liy Tjoye Holmes Miller 305 No. 22. The Occurrence of Strei>sicerine Antelopes in the Tertiary of Northwestern Nevada 319 No. 23. Benitoite. its Paragenesis and Mode of Occurrence, by (leorge Davis [jouderback, with Chemical Analysis liy Walter C. Blasdale 331 No. 24. The Skull and Dentition of a Primitive I chthyosaurian from the Middle Triassic, by John Merriam 381 No. 25. New Mammalia from Rancho La Brea, by John C. Merriam.... 391 No. 26. An A])lodont Rodent from the Tertiary of Nevada, by Eustace L. Furlong 397 No. 27. Evesthes jordani. a Primitive Flounder from the Miocene of California, by .lames Zaechaeus Oilbert 405 No. 28. The Probable Tertiar}’ Land Connection between Asia and North America, by Adolph Kno])f i 413 No. 29. Rodent Fauna of the Late Tertiary Beds at Virgin Valley and Thousand Creeh, Nevada, by Louise Kellogg 421 No. 30. Wading Birds from the Quaternary As[ihalt Beds of Rancho La Brea, by Loye Holmes Miller 439 Index 44!) NOV 26 1918 UNIVERSITY OF CALIFORNIA PUBLICATIONS BULLETIN OF THE DEPARTMENT OF GEOLOGY Vol. 5, No. 1, pp. 1-64, Pis. 1-6 ANDREW C. LAWSON, Editor CAIINIVOKA FROM THE T E RTI ARY FO RM AT I ON S OF THE JOHN DAY REGION. BY •lOHN 0. MeKKIAM. CONTENTS. PAGE Introduction - 2 Occurrence - 3 Canidae 5 Canis' sp. - - 5 Tephrocyon rurestris Condon - - (i Cynoflictis (?) oregoncnsis, n. sj). 11 Nothocyon geismarianus, n. var. mollis 13 Nothoeyon lemur Cojie - 14 Nothocyon latidens Cope 1.5 Mesoeyon coryphaeus Cojie 1(5 Mesocyon brachyops, n. sp 17 Mesoeyon josephi Cope (?) 19 Indeterminate 20 Temnoeyon altigenis Cope 21 Philotrox condoni, n. gen. and s]i. 30 Felidae 37 Nimravus and Archaelurus 38 Generic and specific distinctions 38 Liudis - 4() Vertebrae 51 Pogonodon davisi, n. sp 53 Age and stage of evolution of the .lohn Day carnixore fauna 58 f'anidae 58 Felidae (51 (Conclusions .. (54 o UnivcrsUy of California I'n/jlicaliona. [ (iKOUKiY INTIfODlKI'l'ION. Ill 1899 mid 1900 field parlies from Ihe University of Uali- fornia, workin<>' under the direetion of llie writer, eolleeted over the irreater part of tlie exposed area of llie Tertiary formations in tlie Jolin Day Valley of Eastern Oreffon. In the fall of 1900 L. S. Davis and V. C. Osniont eontinned in the field after Ihe University jiarly had returned, and made additional eolleetions, particularly in the reylon of the Crooked Diver and Lo^^fan Butte, south of the John Day Basin. The collections of Osmont and Davis were purchased for the University, and the material now available may he considered representative of all the phases of the faunas of the John Day region. The new rodent and ungu- late material from these collections has already been discussed by Dr. Sinclair.^ In the following paper there are presented such additions to our knowledge of this fauna as have been obtained from a study of the carnivore material. In the ])i'ei)aration of this report the writer has been espec- ially indebted to Professor II. F. Osborn and Dr. W. D. Matthew of the American Museum for courtesies extended during the ex- amination of the Cope collection of John Day types. Dr. Mat- thew has also very kindly made numerous examinations of types for me while the material was in the process of description. Pro- fessor W. B. Scott and Di-. Marcus Pai-r have shown me the types of John Day ('aimivora at Pi'inceton University, and Dr. Sinclair has recently i-eexamined several tyjies at my recjuest. In woi'king over the subject of variation in the dejitition and skull characters of the recent Canidae, Dr. C. Hart IMerriam and Mr. Vernon Bailey have funiished most valuable data. In the examination of the Canidae, IMr. Bailey has worked over a very large series of skulls and has reported on over fifteen hundred examinations. In justice to Ihe aitist and tngravei-, it should be staled that Ihe illustrations accompanying this paper were I'eproduced from the first rou^b pi'oofs, the oi'iginal plates and drawings having- been (hsiroyid in the San Ei’ancisco lire of A])ril eighteenth. ' W. .1. Sinclair. Hull. Dopt. (Icol., T^^niv. Cal., Vol. 4, No. C. VoL. 5J M erriam . — J oh n Day Carnivora. 3 OCCURRENCE. The general stratigraphic siiecessioii of the Tertiary foiana- tions of the John Day region has been discussed by the writer in a previoiis pai)ei\- The sequence of forniatioiis recognized is as follows : — John Day River terraces Rattlesnake formation Mascall formation Columbia Ijava formation John Day series Upper John Day Middle .tohn Day Lower .tohn Day Clarno formation Upper Clarno Lower Clarno (Quaternary I’liocene Miocene Miocene .Miocene to Oligocene Eocene Of these formations, those included in the beds below the Pliocene are made up mainly of igneous materials. The John Day and Ma.scall beds are almost entirely composed of volcanic ash and tuff in various forms. Mammal remains are known from the John Day, IMascall, Rattlesnake, and Terrace deitosits. Excepting the uppermost portion of the series, the Jolin Day l)eds show a remarkal)le evenness in their stratification, and con- tain a fauna which is characteristic of dry land. In the higher strata, cross-liedding appears, with more or less interstratified gravel, and a number of fresh-water types are seen in the fauna. The greater iiortion of the series is presumably made up of slow aecunndations of ash, which fell mainly on open plains, upon which shifting shallow lakes may have existed from time to time. In the latter i)ortion of this period the topography appears to have been more diversified and the action of streams to have be- come more pronounced. The greater portion of the total thiekne.ss of the Mascall is, like the .lohn Day, made up of evenly stratified ash beds. In the lower portion of the formation there is, however, evidence of ac- cunudation of a considerable thickness of fine sediment in a body of fresh water. In this there are numerous remains of fresh- water fishes and mollusks, and large (piantities of fossil plants. The Rattlesnake beds consist mainly of heavy gravels. Asso- -Bull. Dept. GeoL, Univ. Cal., Vol. 2, No. 9. 4 Universiiy of Calif omia I’lihlicaiions. [GKOLooy dated witli these, as a very prominent and persistent feature, is a heavy rhyolite flow. Some of the beds beneath the rhyolite evidently repi-esent old alluvial floors which have been built up largely by dust accumulations. Although the Carnivora of the Tertiary faunas of the John Day region have been known through numerous types, the actual number of specimens is not large, and they may be counted as among the rarities. It is interesting to note that in the collec- tions obtained by the Tbiiversity of California, there is a consid- erable percentage of new forms, although the number of speci- mens is relatively small. This indicates that the fauna is still only impei'fectly I'cpresented in the collections. As yet, carni- vore remains are certainly known only from the John Day and i\fascall. As far as possible in our field work, the effort has been made to determine the stratigraphic horizons in which si)ecimens have been found. While the results of this discrimination have been most apparent in the study of the Ungulata, some light has also been throAvn on the history of the Carnivora. Of the known car- nivore species from this region there are still a considerable num- ber of which the geological range or occurrence is unknown. As far as has l)een determined, the following ai-e the time relations of the species. The number of specimens knoAvn to occur at each horizon is indicated by the figures in the table. Middle John Day Canis (?) sp - Tephroci/on rurestris - Cynodictis (?) oregonensis 2 Nothocyon geismarianns mollis . 1 Nothocyon letmir 3 Nothocyon latidens (?) 2 Mesocyon coryphaeus 2 Mcsocyon hrachyops Mesocyon josephi ( ?) 1 Temnocyon altigenis 2 Temnocyon ferox I’hilotrox condoni 1 Archaclnrus dchilis major 7 Nimravvs gomphodus 1 J’ogonodon davisi J’ogonodon pin ty cop is Upper John Day Mascall i(?) 1 3 (?) 1 (?) 1 (?) ] 1 VoL. 5 I Mcrri(nn. — John Day Carnivora. dANIDAI-]. 'I'he Tertuii'y bods of oasloni Oregon have furnislied a oe- markable variety of canid ty})es compared with otlier formations in America. Nearly all of these are known from the .John Day beds, only two of the eighteen species occnrrinf? in the Mascall. Althongli most of the types I'epresented exhibit some primitive characters, much variation in strnctni'e is shown. Tn some cases the dilferentiation has led to develo{)ment of considerably spe- cialized forms. In the following discussion only those species have beeii touched upon, concerning which additional evidence has been obtained. The following are the I'ecognized species occurring in the .John Day and Mascall : — Mascall. Canis (?) sp. Tephrocyon ruresiris Condon. John Day. Paradaphaenus cuspigerus Cope. Cynodictis (?) oregonensis, n. sp. Nothocyon geismarianus Cope. Nothocyon lemur Cope. Nothocyon latidens Cope. Mesocyon coryphaeus Cope. Mesocyon hrachyops, n. sp. Mesocyon josephi Cope. Temnocyon altigenis Cope. Temnocyon wallovianus Cope. Temnocyon ferox Eyerman. Philotrox condoni, n. gen. and sp. Oligohunis crassivultus Cope. Enhydrocyon stenocephalus Cope. Hyaenocyon hasilatus Cope. Hyaenocyon sectorius Cope. CANIS (?) sp. Text-figure 1. The known material representing this species consists of two well-preserved upper molars in a fragment of a jaw. The teeth resemble those of the coyotes in general form, and are tentatively referred to the genus Canis. They are a little smaller than the 6 University of California Publications. I Gkolooy molars of the livin" coyotes of this j’e^jion, and ]\P is relatively a little largrer. Fig. 1. Canis sp. No. 54.5. Mascall beds (?), Eattlesnake Creek, near Cottonwood, Wheeler County, Oregon. X 1. The outer cusps of are laterally com])re.sscd and rather sharp. The protocone is well developed, but the innei’ cusp of the cingulum is relatively small. The intermediate tubercles are both present. The outer ridge of the cingulum is well defined, but is narrow. On M- the metaeone is about as large as the paracone. The V-shaped protocone is distinct and the inner cusp or hypocone is well developed. M’, greatest antero-jiosterior diameter 9.7 M^, greatest transverse diameter 13.3 M“, greatest antero-posterior diameter C.6 M“, greatest transverse diameter 9.8 The only known specimen, No. 545, was obtained between Cottonwood Creek and Birch Creek, in the southea.st corner of Wheeler County, Oregon. (Locality No. 887, Univ. Calif. Coll. Vert. Palae. ) At this locality the Mascall formation is capped by the Rattlesnake beds. This specimen was obtained on an ex- posure of Mascall beds immediately below a Rattlesnake outcrop. It was not in the matrix, and we cannot be absolutely certain that it had not originally come from the Rattlesnake beds above. Tephrocyon, New Genus. TEPIIROCYON RURESTRIS Condoil. PI. 1, Figs. 1, 2, and 3. Canis rurcstris Condon, The Two Islands, 1902, p. 139, PI. 18. Type specimen No. 382, private collection of Professor Thomas Condon. This species is represented by a very fine skull obtained by Professor Thomas Condon many years ago in the Mascall forma- tion near Cottonwood, Grant County, Oregon. It was figured by VoL. 5] Mcrri(t}H. — John Daij Carnivora. 7 Professor Condon in his work on Die Two Isljinds iind wns after- ward kindly loajied to the antlior by liiin for eoinpai'ison and study. The skull laeks only the portion anterior to the eanines. The i)arts present are well preserved and in pood condition for study. Didinclive Characters . — Muzzle short, posterior ends of pre- maxillaries extending behind the anterior ends of the f rentals, sagittal ei'est high, inion x)rot>iinent, auditoiy bullae very large, inferior niargin of the mandil)le sti-ongly eonvex below the ante- rior side of the eoronoid process. Deiditioii f, ], if, f. P^ with incipient pj'otostyle, deuterocone Aveak'. with hi-oad in- ternal lobe, ai)parently without protocoinde. P.^, P3, and P4 with l)Osterior basal tubercles and without anterior cusps. Mj with well-developed nietaconid, heel with low entoconid and hypoco- nid. Ti-igonid of iM, with di.stinct paraconid. M, with three or four low tubercles or ridges. Skull . — The type of .skull .shown in this si)eeies is that of a short-muzzled form with greatly developed temporal muscles. The shortening of the facial region is accompanied by relatively great backward extension of the premaxillaries, which iiass the anterior ends of the frontals, and almost reach a line connecting the anterior borders of the orbits. The sagittal crest is high and sharp. The inion projects rather more than is typical of Canis. The occipital region shows strong buttresses running out on either side to the lambdoidal crest. The auditory Inillae are very large, and the jiaraoccipital processes prominent. On the mandible, the massteric fossa is deep. The inferior margin of the jaw shows a somwhat greater degree of convexity below the anterior end of the massateric fossa than is common in the typical canids. Dentition . — The dentition is in general much like that of Canis, but differs in the iiresence of a notch on the anterior side of P^, which separates an incipient protostyle from the proto- cone ; in the presence of a minute jjaraconid on M,, and of sev- eral tubercles on iMg ; and in the greater width of the lingual side of Mb s University of (Uthfornia Uiihlicatioiis. All flip proiiiohu-s pxccpliii”- tlu* fir'sl ai'p rpi)r(>spiitc(l in l)o1li jaws. Thp prpspiipp of llip aritprior otip is in(lica1(>(l in both oases by a lar^p alveolus. P.j, and 1*4 have no aTitp)-ior pns])s or tubei’ples, but all exhibit botli a posterior eusp and posterior basal tubei-ele. P- differs from all of tlie otliei' preinolars, exeeptinjr the earnassial, in ])Ossessin‘j a minute anterior eus]). P^ shows both a. i)Osterior eus]) and a ])ostei'ior basal tubei-ele. On the up- ])er earnassial, a distiuet noteh on flip anterior side of tlie proto- eoue .separates a small eusp eorres])ondin<>' I0 the pT-otostyle. The deuteroeoue is small. The first ui)j)ei- molai- is unusually broad on the lino'ual .side, thouyh the jiosterior inuei' eus]) or liypoeoue is not relatively large. A well-developed metaeonule is present, but there is ap- pai-ently no ])rotoeonu]e. Tlie external eingulum is distinctly marked. On IM" the protoeone is large and the metacone .some- what reduced. i\I, is not far from the stage of evolution seen in Canis, though the metaeonid is perhaps a little larger in Tephrocyon. On the heel the low hyiioeouid and entoeonid are distinctly separated. The trigonid of INL is less reduced than in Canis, and possesses a small paracouid. The heel of this tooth is not markedly different from that of Cauls. The hyjioconid is a di.stinet but low tubercle. On jM, there is a faint external ridge with three smaller ridges running toward the inner margin. These elevations .seem to rep- resent the hyiioeouid and the three elements of the trigonid. Affinities. — This form seems to lie geuerically separable from the canids thus far described from the iMiocene of America, but it shows affinities with Canis, and to a certain extent with Aelii- rodoii. In the Cauidae. the development of a protostyle on P* is a charactei'istic of the Aelui’odons and Simocyon, though occurring occasionally in other forms. In its incipient form it is present rarely in modern species of Canis, ]iarticularly in the milk car- nassial. In order to detei'miue the degree of variability existing in wild canid species, l)i'. C. llai't iMerriam and iMr. Yeruon Bailey have ivceufly examined a large series of skulls in the col- lections of the Biologicid Survey. The following results I (piote from a statement by I)i-. .Mei'riam. VoL. 5] ]\I errimti . — Joltn Dai) (Uirnivoyd. 9 “Til Ili(‘ big Wolf, out of 107 skulls, a faint trace of a cusp may be seen in 1 adult. “In Coyotes, out of 450 skulls, a faint trace exists in 3 adults, and a distinct cus}) is present in the milk tooth of 15 young. “In Canis (or (lerdocyon) mayelhtnicus, out of 22 skulls, no trace of a cusj) in tlie adult; a faint cus]) is iiresent in the milk tooth of 2 out of () young. “In lied Foxes, out of 244 skulls, no truce in adult; slight trace in milk tooth of 5 out of 10 young. “In Ivit and Desert Foxes, out of 70 skulls, 2 adults show trace of a cusp; 9 young have a small cusp on the milk tooth in each case. “In the Arctic Fox, out of 24 skulls, 1 adult shows a faint trace, and 5 young in milk teeth .show a faint trace. “In Urocyon, out of 241 skulls, 2 adults and 3 young in milk teeth show trace of a cusj) ; other young show no trace. “It is pi'ohahly an exaggeration to state that any of the adults show a trace of the real protostyle.” The cusp as it ai)pears in T. rurestris is stronger than in any valuation of the recent Canis which I have seen, hut it is much weaker than in Aclurodon, and should hardly he termed a true protostyle. This .species also shows some resemblance to the Aelurodons in iiF and ]\Io. It is distinguished h'omAelurodon, among other characters, by the very imperfect development of the protostyle of P^, the less reduced metacouid of M^, the larger ]\I-, and the large inner lobe of ]\P. In mo.st of the characters in which it differs from Aelurodon, it approaches Canis. Teplirocyon differs from Canis in the structure of P*, M 2 , and IMj, as indicated in the discussion of the dentition above. There are also differences in the skull. The premaxillaries pass the an- terior ends of the frontals, the sagittal crest is very prominent, the otic bullae are extraordinarily large, and the mandilile is strongly convex below the posterior molars. The genus is more primitive than the typical Aelurodons in the imperfect development of the protostyle of P'‘, and in the larger metaconid of iMj. It is more primitive than Canis in the characters of M„ and M^. The anterior molars in both jaws are near the stage of evolution seen in Canis. As the animal was a 10 University of California I’uhlicalions. [f!Eoj/)(nr rallici' lar<2:e form willi short facial 7-e"ion and lioavy-hladod car- nassials, it was prolialjly a i-apidly specializing form. Tlie trend of its specialization seems to have been toward the Aelurodon type. Aelurodon was ivpresented by .several species in the Loup Foi’k epoch, but T am not awai-e of the existence of any remains from older beds. Tlie Mascall beds have been thought to repre- sent a .somewhat eai-lier epoch than the typical Loup Fork, and it is at precisely this hoiazon that we would search for the ance.s- tral type of the Aelurodon group. While it is not necessarily true that Teprhocyon rurestris is directly anee.stral to Aelurodon, it seems to i-epresent a tendency toward specialization in that direction in Ihe Canidae, at the time when the genus Aelurodon probably had its oi-igin, and it is evidently not far removed from this line of descent. MEASUREMENTS. mill. Tjengtli, anterior end of orbit to inion 124 Length, postorbital process to inion 97 Width between orbits 38 Width across snout above middle of superior canine 38 Length of mandible from anterior side of Ik to posterior side of condyle 108 Height of mandible below protocone of M, 20 Ijength, anterior side of P' to posterior side of M' 60 Length, anterior side of Pj to posterior side of 66 Antero-posterior diameter of superior canine at base 10 P-, antero-posterior diameter 8. .5 P^, antero-posterior diameter 10 P*, antero-posterior diameter 17 P*, greatest transverse diameter 9.-5 M’, antero-posterior diameter 13 M*, greatest transverse diameter 16 antero-posterior diameter across protocoue 10 iP, antero-posterior diameter 7.7 M", greatest transverse diameter 12 Pj, antero-posterior diameter 7.5 P3, antero-posterior diameter 9 Pj, antero-posterior diameter 11.5 Mj, antero-jiosterior diameter 20 •M,, aiitero-])osterior diameter of heel 6 .Mj, antero-jiosterior diameter 11.5 Mj, antcro-jiosterior diameter of heel 4 Mj, antero-posterior diameter 5 VoL. 5] Merrimn. — John Day Carnivora. n CYNODICTIS (?) OREGONENSIS, 11. Sp. PI. 2, Pigs. 4 and 5. Cauls f/reyarius Cope, Bull. U. S. Oeol. Surv. Terrs., 1879, j). 58. (ialecymis greyarim Co])c, Tertiary Vertehrata, p. 917, PI. 08, I'dgs. 5 to 8. The specifie name oregonensis is here applied to the John Day representatives of Cynodictis (?) {(lalecynus) gregarius Cope. The Jolin Day form lias generally jia.ssed under the same name as the spieeies from the White River. It is to be distinguished from the lat ter by the eonstant presence of a posterior cusp in addition to the anterior and posterior basal tubercles on IC, the larger j\D, larger brain case, less pronounced postorbital constriction, and other characters.^. ScotC has already pointed out that “the John Day specimens, which Cope has referred to Cynodictis gregarius, have an even fuller cranium and shallower postorbital constric- tion, which should, perhaps, be reason for separating these ani- mals specifically from the White River forms.” This species is represented in our collections by several speci- mens. The most important of these is a mandible (No. 316, pi. 2, fig. 4) showing the dentition excepting the median incisors, Pj, and IMj. In addition to this there is a jaw fragment with the unworn i\Ij and P^, and several fragmentary mandibles. A frag- ment of a maxillary (No. 11791 with both molars perfectly pre- served may also belong here. In the specimens which have come under my observation the characters mentioned above are constant. Compared with the other species of the John Day fauna, the lower sectorial is also diagnostic. The dimensions of the tooth are near those of speci- men No. 10256, discussed below with N . latidens. It is distin- guished from this form by the absence of the postero-external tubercle on the base of the protoconid, by the more distinctly ® In recent correspondence regarding comparisons of specimens at the American Museum, Dr. W. D. Matthew has added the following to the above mentioned characters, distinguishing this form from C. gregarius: “Skull wider ; muzzle somewhat wider ; interorbital width somewhat greater ; pre- molars and carnassials slightly smaller; M‘ more quadrate internally, the hypocone crescent extending farther forward around the protocone ; M, much larger. ’ ’ * W. B. Scott. Canidae of the AVhite River Oligoeene, p. 368. 12 Universily of (UtUfoniia /‘iihlicdldons. I fiK,OI,(»«lY l):isiii-slia])cd talon, and l)y tlie abseneo or weak dovelopniont of the oxtornal cnn<>nlnrn, ex('C‘i)tinfj a faint ridjro below tin; para- eonid. d’lie liy])oconid and cntoconid are sbar])ly compressed I'idges connected postei'iorly, wliile in tlie form represented by No. 102r)() the entoeonid is smaller, is nearly conical and is not so closely connected with the bypoconid. On Mo tliere is a small bnt acute i)araconid. Altbou^b smaller than the other tubercles of the trifi'onid, it is nevei'tlieless distinctly developed. 'I'be en- amel of most of tlie jiremolars and molars is stronj^ly wrinkled. It seems not inijirobable that the lower jaws referred to (ia- lecjjiuis lanmr by 0o])e really belonged in C. (f) oreffonruiHis. None of the sjieeimens described by bim were in association with crania. The dimensions of his specimens are almost identical with those of oregone lists. As indicated elsewhere,'’’ there is some I'eason to believe that a number of the very small interior sec- torials showino’ the peculiar characters of lotidcns really rejire- sent leiinir. MEASUREMENTS. No. No. No. ¥T6 trs'o' TTT Length of iiiiindible, anterior side canine to posterior side P, 23.8 mm. tleight of mandible lielow protoconid of Mi .. 8 P.., antero-posterior diameter 4.5 P,, antero-posterior diameter 6.2 5.9 t Ml, antero-posterior diameter 8.2 7.9 10.5 -Ml, antero-posterior diameter of heel 2.7 2.7 3.6 skull fra.mneut (No. 584, see measurements above) showing a perfect inferior sectorial with 1\, P,. the trigonid of 5fi. and a part of the first upper molar represents an animal considerably larger than the tyiiical C. (?) orcgoncnsis. The form of the teeth is much the same as in orcgoncnsis and is (piite different from that in Jatidcns and gcismarianiis. 1 think it is very doubtful whether this form belongs in orcgoncnsis, but the amount of vari- ation shown in all the groups of dohn Day dogs is so great that I hesitate to sejiarate it until more is known concerning it. ^ Bee ]>. 1(), lliis j)ai)ei'. VoL. 5] — John Daij Carnivoro. i:;! NOTIIOCYON (iKISMAHlANlJS Cope, 11. V;l I'. MOMilS. PL 2, Fig. 1, and PI. 3, l'’ig. 1. 'I’ype siiccinuiii No. 90, Univ. Calif. Col. Vert. Palac. From tlio Middle .lolin Day Peds at Turtle Cove, John Day Fiver, Oregon. A perfectly preserved craniiiiii witli dentition represents a type ditfering somewhat from the other small eanids of the John Day. The skull is that of an adult animal with somewhat worn teeth. It is mncli larger than tlie skull of lemur and consider- ahly smaller than the typical fjeismarianus. The brain case is large and the facial region short. The orbits are smaller than in Imaur, while the auditory bullae are relatively as large as in that species. The cranium is rather sharply constricted behind the acute postorhital jiroeesses. The temporal ridges show a lyrate arrangement, though it is not so strongly marked as in lemur. The I'idges are sei)aratcd by a space from 9 to 12 mm. wide, ex- cepting over the posterior portion of the brain case, where they unite to form a low crest about 15 mm. in front of the inion. The dentition repi-esents a stage between that of Gynoclictis{f) orego- nensis and N. latidcns. seems to have had a posterior cusp, though the posterior margin is worn and only a trace of the cusp is preserved. The superior carnassial and the molars are of the broad type, Imt are narrower than in latidens. Except for the somewhat smaller size, lyi'ate aiTaugement of the temporal ridges and .somewhat narrower molars, this form is close to N. geisma- rianus. It differs from C. ( f) oregonensis in its larger size, larger brain case, nuich largei' otic bullae, longer lyrate temporal area, and l)roader teeth. This s])ecimen I'epresents a variation of the Nothocijon type which cannot be placed in any of tbe described species without appai-ently extending the limits of the group so far as to overlap some other species. The uniting of any of the types now known does not .seem advisable until more is known of intermediate stages. Neither does it seem desirable to increase tbe number of species where considerable variation is suggested. I have there- fore referred to this form as a variant of the geismarianus typo. Eater investigations may show that it should be raised to specific rank. 14 Universil}/ of California I'uhlicnliona. f (7 rmAHTT Approacliinj? Cynodklh (?) oreyoncnsis ratlier closoly, as it does, lliis fojaii makes a separation of tlie Jolin Day Cynodidis tyi)e fi'oni the moi-e specialized Nolliocyon very difficult, if it is really i)ossil)le. MEASUKEMENTS. mm. Length, ])rem;ixillary to inion 104 Length, postoi])it!il {)roecss of frontal to inion 54 Length of su|)crior dental scries 50 P“, antero-jiosterior diameter (J.2 I’*, antero-j)osterior diameter 9 P*, transverse iliameter across deuterocone 5.2 ^P, antero-posterior diameter 7.4 i\P, transverse diameter 9.2 j\P, antero-posterior diameter 4.6 M", transverse diameter 7.2 liong diameter of otic bulla 19 Width between otic bullae 7 Elevation of inion above foramen magnum 16 Width between canines 12.5 Ijength of superior dental series including canine 44 NOTIIOCYON LEMUR CopC. PI. 2, Figs. 2, 3, 6, 7, and 8. Galecyims lemur Cope, Bull. U. S. Geol. Surv., v. 6, p. 181, and Tertiary Vertebrata, p. 931. Cynodictis lemur (Scott), Trans. Amer. Phil. Soc., 1898, p. 400. Nothocyon lemur (Wortman and Matthew), Bull. Am. Mus. Nat. Hist., v. 12, p. 127. This species is represented by two crania, Ity several upper jaws with well-preserved teeth, and by a number of loose molars. Several lower seetorials and jaws discussed under the next species po.ssibly belong here also. Although no mandibles have as yet been found in association with crania, there seems to be some reason for believing that the inferior dentition of lemur is much like that of laiidcns, as is indicated in the di.scussion of latidens following. The following measurements of upiier teeth of several indi- viduals of the lemur type slunv considerable variation, but all Iioint toward this species rather than latidens. Vol. 5] Merriam.— -John Day Carnivora. 15 MEASUKKMENTS. V. hititlenn Typo N. lemvr Type'' Inn u r Ootype CJopo' No. AS" No. TX No. TX? P‘, antero-posterior diameter.... 7 7 7 7.2 M‘, antero-posterior diameter... 7 5 5.8 5.8 5.3 5.3 M', transverse diameter 8 7.8 7.8 7.4 7.4 7 M', antero-posterior diameter... 4.5 3.8 3.5 3.6 M’’, transverse diameter 5 5.2 5 5.25 In a small cranium (pi. 2, fig. 2. No. 10208) having the gen- eral form and dimensions of lemur, tlic snout is exceedingly nar- row, the temporal ridges do not unite posteriorly and the brain case is relatively very large. Tt appears even larger than in lemur, and the superior outline is strongly convex over the i)a- rietals. Another small, narrow-nosed slcnll evidently belonging in lemur exhibits a i)erfectly preserved brain cast (pi. 2, fig. 3, No. 10209) illustrating the principal features of both cerebrum and cerebellum. This specimen resembles lemur in dimensions and general form. In most respects the brain of this species is strongly similar to that of Ctj)io(liciis as figured by Scott® for C. gregarius of the White Eiver. NOTHOCYON LATIDENS CopC. PI. 2, Figs. (5 and 7. Gulecynus latidens Cojie, Bull. U. S. Oeol. Surv., Vol. 6, p. 181, 1881. Cijnodictis latidens (Scott), Trans. Am. Phil. Soc., 1898, p. 400. Nothocyon latidens (Wortman and Matthe^Y), Bull. Am. jNIus. Nat. Hist., A"ol. 12, p. 127. In our collection there is a perfectly preserved and unworn inferior carna.ssial (pi. 2, figs. G and 7, No. 88) exhibiting a feature which Cope considered characteristic of latidens, viz., the “possession of a narrow tubercle at the external base of the i)rin- cipal cusp.” There is also a very small tubercle anterior to the base of the entoconid. The length of this tooth is somewhat less than that of the type of latidens, and the heel is relatively short. “No. (1888, Amer. Mus. Nat. Hist., Pol. Vert. Palae. ’ No. 6889, Amer. Mus. Nat. Hist., Col. Vert. Palae. ® W. B. Scott. Canidae of the White Eiver Oligoeene, 1898. Trans. Am. Phil. Soc., Vol. 19, PI. 19, Fig. 12. 16 (I niL'crsiiif of (htUfornia I'liljliaitions. I (!koi/i(;y In iuiotlioi' six'cinieii (No. consislin'j of a lar^'e portion of a lower jaw with 1*4 and M,, llie external tnberele at the base of the ])i'otoeoni(l and the smaller (me antei-ior to the entoeoind are both present, '^fhe external tnberele is, bow(‘vei‘, f^reatly redncffd by wear wilbont the remainder of the tooth ai)i)earin" much worn, 'file dimensions of Ibis jaw and of the heel of Mj are eon- side)‘:d)ly less than those of the ty])e of lalideiis, and are neai-er the dimensions re(piii'ed by lemur. These relations together with the faet that the jaws desei'ibed as lemur agree with (Jjjno- diclis ( ?) ore(/ouciisis seem to indieate that the small seetorials jnst described possibly belong to loMir, and that the j)]'esenee of the extei’nal t(d)ei-ele ori the base of the jn'otoeonid is a eharae- teri.stie of lemur as well ai s of lalide.ns. MEASUIIEMENTS. No. No. Type spec . Type spec. 1 OTYc, K 8 lafidens lemur P,, aiitero-posterior diameter 6 5.5 AI,, autero-posterior diameter 8.2 7.6 8 8 Al„ antero-jiosterior diameter of heel 2.8 2.8 3..5 “.3 Depth of ramus at sectorial . 8 10.5 8 JIESOCYON CORYPHAEUS Coi)e. Temnocyon coryphaeus Cope, Proc. l*hiiad. Acad., 1879, p. 180; and Ter- tiary Vertebrata, p. 906. Mesocyon coryphaeus (Scott), Princeton College Bull. 2, No. 2, j). .18, Apr., 1890. 11 ypotemnodon coryphaeus (Evernian), Ain. Gcol., Vol. 14, p. .321, 1894. 'I'his .species is rejiresented in the collection Ity .several crania and mandibles. Specimen No. 1888 .shows the complete upper dentition, of which partictdaily the sectorial and the molars are unworn and in perfect .state of preservation. In No. 1165 the upper dentition is comjilete excepting the incisors. Both of these specimens agipe with Colic's type very cktsely in their measure- ments. Another eranium (No. 1692) representing the genus ditl'ers fi'om both M. edryptiaeus ami J/. jbsephi, and has been made the type of a new species de^^erilied below. “According lo Cope’s figure, ’Pert. Vert., I’l. 70, Pig. la. VoL. 5 ] Merriam. — John Day Carnivora. I A well-pi'pserved iii;in(lil)le (No. 1:507) sliows llie t-yi)o of doii- litioii in M. corypliaca.s. Tho sc'ctorinls nrc soinewlnit woini, so that no internal tnl)ere]e is sliown. "flic tnl)erenlai's ai'e veiy im- perfectly i)reserved. On anothei' specimen (No. lOdOfi) llie last two j)reniolars, the sectorial, and the first tnhereular are perfectly preserved. The measurements of these .jaws, as also of several detached teeth, are considerably less than those of the type of M. corypliariis, as is indicated in the measurements below. Al- though Cope states that the mandible described by him was a.sso- ciated with the smallest of the several skulls available, I find that the dentition of mandible No. 1307 in our collection fits the denti- tion of the skulls of coryphaeus, particularly that of No. 1165, very closely, aud seems also to have the dimensions required foi- the mandible of the type cranium. The measurements availal)le seem to indicate considerable variation in size among the forms of this group. MEASUREMENTS. M,, autei'o-posterior itiameter M,, antero-posterior diameter Pj, antero-posterior diameter P 3 , antero-posterior diameter Height of jaw below protoconid of M, Height of jaw below protoconid of P 2 Type No. No. No. 13 0 7 o y y 1 TT 3 6 (1 18 16..5 15 14.5 7.8 11 10 10.5 9.4 8.7 19.3 17 18 15 MESOCYON BRACHYOPS, 11. Sp. Text-figure 2. Type specimen No. 1692, Univ. Cal. Col. Vert. Pal. From the upper por- tion of the John Day series at Logan Butte, Crook County, Oregon. Facial region and palate relatively shorter than in either M. coryphaeus or M. josephi. Superior dental series somewhat crowded anteriorly, so that is very close to the canine. Supe- rior sectorial and first tubercular with dimensions near those of coryphaeus, and absolutely larger than in josephi. The lirain case is relatively larger than in coryphaeus. “See also measurements of superior dental series of coryphaeus below, j). 19. 18 Uiiiversiti/ of (Jolifornia J‘ublicalions. \ Okolooy Fig. 2. Mesocyon brachyops, n. sp. No. I(i92. John Day beds, Logan Butte, Crook County, Oregon. X %. This is evidently a time Mesoci/on near the coryphaeus type, possessing like that form both posterior cu.sp and basal tubercle on P®. The sectorial is nearly as large as in coryphaeus. P® and P- are relatively smaller. The palate is as wide posteriorly as in coryphaeus, and the width between the canines is also as great, hut the jialate of the typical coryphaeus is 15 to 20 per cent, longer. There is a moderate degree of crowding of the premolars anteriorly, in contrast to the oiien arrangement in coryphaeus and Joseph i. The size of the brain ease is relatively considerably greater than in the tyjie of coryphaeus or in a specimen of the same general dimensions in our collection. M. josephi dilfers fi'om this form in having a longer palate, a smaller superior sec- torial and tirst molar, and in lacking a posterior cusi) on P®. VoL. 5] Merriam. — John Day Carnivora. 19 The presence of this short-faced variety in close association with lonp:-faced forms of the same f?ronp is another illustration of the interesting tendency toward the evolution of short-faced forms in the John Day. MEASUREMENTS. i I I! Length, prenia-xillary to occipital condyle inclusive 134 160 Length, posterior sih(icus the talon of M 2 supports a single tubercle. The .sec- ond and third inferior molars of M. josephi are unknown, though VoL. 5] Merriam. — John Day Carnivora. 21 Olio might suspect from tlio clmracter of tlio specimen desc'rihed ;d)ove that tliey are broader than in M. coryphaeus. Tlie heel of AL in Philolrox coudoui is broader than in M. coryphaeus, hut does not sliow a develojmient of tlie cntoconid and hypoconid comparable to that seen here. Tlie form of AT, in Oliyohuuis ap- pears to he (juite different from that in this specimen, and AT,, is absent. Alg is said to he trencliant in Enh ydrocyon and is un- known in llyaeuocyou. J 4 3 6 Fig. 3. Mesocyon coryphaeus Cope. Inner side of right M,. No. 599. Upper portion of John Day Series, Haystack Valley, John Day Kiver, Oregon. X 1. Fig. 4. Indeterminate. Superior aspect of left M, and Mj. No. 072. Middle John Day, below Clarno’s Ferry, John Day Eiver, Oregon. X 11 / 2 . Fig. 5. Indeterminate. Sujierior aspect of right M,. No. 1510. Upper John Day, Turtle Cove, John Day Valley, Oregon. X 1. Fig. 0. Outer side of M, shown in fig. 5. XU Another peculiar specimen (No. 1516), text-figures 5 and 6, representing a portion of an inferior sectorial with a well-pre- served heel, was fotind in the Upper John Day beds at the lower end of Turtle Cove. It shows a hypoconid and an entoconid of almost exactly eriual size. It is distinguished from the sectorial of the jaw tentatively referred to M. josephi by the presence of a distinctly marked basal band on the outer side of the talonid. This tooth differs from all of the John Day canids with which 1 am acquainted. It may represent a Mesocyon or po.ssibly one of the short-jawed forms. TEMNOCYON ALTIGENIS Cope. PI. 3, Fig. 2, and Text-figures 7, 8, 9, 10, and 11. Temnocyon altigenis Cope, Palaeont. Bull. No. 30, p. (5, 1878; Kep. U. S. Geol. Surv. Terts., Vol. 3, p. 903, PI 68, Fig. 9, 9a, 9b, PI. 70, Fig. 11. This species, constituting the type of the genus Temnocyon, has been only partially known, and was considered by Cope as 22 University of (UiUfornia J‘ti Mica lions. 1 0?;()L(KiY Olio of Die niror .loliti I);iy oni'iiivoros. It is rojirosontod in llio Univorsity oolloolioiis hy two siiooiniens, No. 9999, ji youn<; Jidiilt witli unworn loolli, niul 1549, an old individual witli toi'tli oon- sidei-ably worn. No. 9999 sliows the entire skull witli the eoni- [ilete njiiier and lower dentition, and the jireater [lart of the skeleton exeeiitiiifi: the iiianns and the distal half of the jies. 'I'lie other s])eeiiiien is a nearly eonijilete eraninin with all the ele- ments of the npjier dentition. The dentition of the inandihle in the first sjieeinien e()rres])onds very elosely to that of Cope’s tyjie. 'file n]i|)er dentition is nearer in form to that of Cope’s type of oJiigcnis as fisnired than to any otlier species, with tlie exeefition of the denteroeone of the carnas.sial. Tn this tooth onr specimens exhihit the heavy form of denteroeone seen in the type of T. fcrox. Tn Cope’s specimen showing the upper jaw', the dentero- eone of P*, thonpth prominent, is represented as small, low, nar- row, and not very sharply separated from the protocone. Al- thongh this cranium was not associated with the tyjie, the molars show so dose a resemhlance to those of specimen No. 9999 that there seemed to lie pood reason for helievinp that the denteroeone of P^ had not been correctly drawn. Dr. W. 1). Matthew has very kindly examined Cope’s type of the cranium for me re- cently to ascertain the true character of this tooth, and finds that “the njiper fourth preniolar has a large, well separated dentero- cene. ’’ The molars of this specimen were found by Dr. Matthew to he correctly represented. With this correction of the original description of Temnocyon altigenis, specimens No. 1549 and No. 9999 approach this species very closely. The ])rincipal difference being found in minor variations of the measurements as seen in the table of measurements on page 29. Doth of the specimens used as tyjies liy Cope are slightly larger than the specimens in onr collection. The differences separating this species from the other mem- bers of the genus Temnocyon may he expressed as follows: — T. fcrox ITyerman. Largest known species. Lcmgth of skull in type specimen 266 mm. Sagittal crest high. Denteroeone of P‘ large and heavy, .shar])ly .sej^arated from protocone, with an incipient Inhercle on the anterior side. P® with posterior basal tubercle. AP with V-.shaped protocone. M- with greatly reduced metacone. P,, practically ecpialing Mj in length, with a distinct VoL. 5] Mcrria))i. — John Day Carnivora. 23 j)Osteri()i- basal tubercle in addition to the prominent cusp arising from tlie postei-ior side of the protoconid. T. altiyenis Cope. Species of medium size. Length of skull 178 mm. in No. 9999, somewhat greater in No. 1549. heavy and wide anteriorly; deuterocone very large and heavy, separated from ])i'otoeone by deep valley, without anterior eonule. P® without po.sterior ha.sal tidiercle. Protocone of nearly conical, regularly rounded posteriorly hut with a short antero-external ridge. M- with metacone smaller than paracone, hut consider- ably larger than in the other species. P^ considerably shorter antero-posteriorly than with i)rominent cusp on posterior side of protoconid and with incipient posterior basal tubercle. T. wallovianus Cope. Species of medium size. Deuterocone of P^ apparently not greatly enlarged. P® without posterior basal tubercle. 1\I- small, with reduced metacone. This species should possibly be included in T. altiyenis. The principal features of the skeleton of T. altiyenis as shown in specimens 9999 and 1549 mainly confirm the generic charac- ters represented in the specimen of 1\ ferox described by Eyer- man. In some cases, however, the specific differences are great enough to modify slightly our estimate of the generic characters based on the description of ferox. Skull. — The skull of 2\ altiyenis differs considerably from that of ferox in general oiitlines. As represented by Eyerman,^- the cranium of ferox is relatively high and narrow compared with the undistorted specimen of altiyenis seen in No. 1549. A part of this difference may be due to lateral crushing in the ferox specimen, which would produce the deviation in height and width at the same time. The palate of altiyenis is decidedly narrow, but that of ferox is extreme in this respect. In ferox the sagittal crest is high and narrow over the greater part of the brain ease ; in specimens 1549 and 9999 the sagittal ridge is very low except- ing over the most posterior portion of the brain case, where it attains a considerable height. In specimen 9999 the ridge is divided almost to the posterior third of its length by a sharp but very narrow' groove. The lambdoidal crest is strong and high. The broad frontal region is slightly concave in front of the J. Eyerman. Amer. GeoL, Vol. 17, p. 267. 24 (Jnivo'siijj af (Jalifoniia I’nhlicalioiis. I (‘iK()l,()(;Y ])()st()rl)it;il i)i‘()('ess(‘s, liiil is convox imincdiiilcly licliiiid llicsc ])<)iii1s ill .speciiiieii 154!). 1'lic pfistorhilid processes ;ire very sliorl and hliinl. 'I’lie nasal region is eliaraeteri/.ed hy tlie stiorl- noss of llie nasal bones, wliieli do not roaidi baeU to a line drawn between tlie anterior liorders of the orliits. 'I'Ikw are consider- ably shorter than in Daplntoiiis or in Mrsocijoii. 'flu* zyoyainatie areli is not (‘S|)eeially robust. 1'lie post^de- noid jiroeess is Ion”: and acute, 'fhe paroeeipital jiroeess is sliorter and less acute tlian in /Vro.r, and is not directed down- ward as innch as in that species. Tlie anterior and jiosterior iialatine foramina are small. The anterior border of the jiosterior nares is situated considerably behind the last molars. The nasal chamber is narrow and the inferior borders of the lateral walls are arched inward, giving; it a more nearly tubular form than is seen in C