CAMELS OF THE FOSSIL GENUS CAMELOPS.By Oliver P. Hay,Research Associate of the Carnegie Institution of Washington.One of the most interesting revelations furnished us by the studyof vertebrate paleontology is that our country was inhabited, still afterthe beginning of the Pleistocene, by camels belonging to more thanone genus and to several species. Our knowledge of these specieshas been meager enough, although the number named has not beenso restricted. Most of these species have been founded on suchscanty materials that comparisons among them could hardly be madewith any accuracy or certainty. In 1898* Doctor J. L. Wortmanconsidered the materials then available, and he ended by includingunder the name Camelops Icansanus, given by Leidy in 1854, not onlythe type of this species, but likewise Leidy's species Megalomeryxniolrarensis and liis Californian Auchenia hestem-a, Cope's Holomenis-cus sulcatiis, and the specimens from Oregon and Texas which theauthor just mentioned had described under the name of Holomenis-cm Jiestermis, and Cragin's Auchenia fiuerfanensis, found in Colorado.Camelops Icansanus had itself been based on a fragment of the snout,consisting of portions of the left premaxilla and maxilla, wath theroot of an incisor and a part of the socket of a canine. This specimenhad been found in 1854, or previously, in what was described as"gravel drift," somewhere within the present State of Kansas.Happily, these camels are beginning to emerge from the obscuritywhich has enveloped them. That wonderful deposit of remains ofPleistocene vertebrates, the asphalt beds of Rancho La Brea, nearLos Angeles, California, has furnished to Doctor John C. Merriam afew complete skulls and the greater part of the skeleton of one, pos-sibly of two, species of camels. The skulls are described by him in apaper recently issued. ^ Two complete skulls are figured, of which oneis identified as representing the species which Leidy called Auchenialiesterna, the other as being near this species and probably belongingto it. Merriam accepts Wortman's conclusion that these camels are
1 Bull. Amer. Mus. Nat. Hist., vol. 10, p. 93.2 Univ. California Publ., Geol., vol. 7, pp. 305-323, fig.s. l-ll.Proceedings U. S. National Museum, Vol. 46-No. 2025. 267
268 PROCEEDTYGS OP THE NATIONAL MUSEUM. tol. 46.generically distinct from Auchenia and with him accepts Leidy'sname Camelops.The present writer has recently had occasion to study some of thematerials belonging to Pleistocene camels. In the United StatesNational Museum is Leidy's type of his Megalomeryx niohrarensis, afragment of the left side of the lower jaw containing a much wornsecond premolar; likewise the little worn right lower second molaridentified as belonging to the same species.^ In looking for othermaterials the writer found Cragin's type of his Auchenia huerfanensis.This was discovered in 1884 in beds of volcanic ashes, along a smalltributary of the Pluerfano River, in Huerfano County, Colorado,by the geologist Mr. R. C. Hills, of Denver. It had afterwards beenfor some time deposited in the Colorado Scientific Society, and whilethere had been described by Professor F. W. Cragin. Later it waspresented, together with a considerable quantitj^ of other fossil ma-terials, by Hills to the United States National Museum. Here itnow has the catalogue number 7819. It presents a part of the rightmaxilla with the last premolar and the molars, all in good condition;a part of the left maxilla with the roots of two premolare and of twomolars; the right ramus of the lower jaw, lacking most of the sym-physis and containing the fourth premolar and the first molar; theleft ramus, lacking most of the ascending portion, and of the sym-physis, and furnishing all the molars in fair condition, and the rootof the fourth molar; portions of both premaxillse; a part of theocciput; many small fragments of the brain-case and of the face;the distal ends of the metapodials of one foot; and one proximalphalanx.The writer has at hand likewise some remains of two or three indi-viduals of a camel which evidently belonged to the same genus asthose already mentioned. These remains consist of two symphysesof lower jaws, with the incisors and canines; a part of a right horizon-tal portion of a lower jaw, containing the cheek-teeth; a few otherlower teeth and parts of still others; and one as yet uncut upper lastmolar. These materials were found in 1905, by Mr. F. C. Horn, atMinidoka, Idaho, not far from Shoshone, in a bed of gravel whichwas overlain by a lava flow. In the same gravels were found bonesof a large elephant, a part of a lower jaw of a horse, and two horn-cores of a bison, which the writer identifies as Bison alleni. A part ofthe camel remains bears the United States National Museum cata-logue number 2579; another part, the number 5315.In the National Museum there are three incisors and a part of alarge molar of a camel which were found in 1867, by Mr. E. L.Berthoud, in "loess deposits of the big ravine on north bank of BigThompson River," near Greeley, Colorado. The locality is close
1 Journ. Acad. Nat. Sci. Phila., n. s., vol. 7,.p. ICl, pi. 14, flgs. 12-14.
NO. 2025. CAMELS OF THE FOSSIL QENU8 CAMEL0P8?HAY. 269'to the line between townships 4 and 5 north, range 66 west.The teeth were discovered at a depth of 35 feet from the surface.The catalogue number is 870.The most important of the specimens above mentioned are, ofcourse, those constituting the type of Cragin's Auchenia Jiuerfanensis.It may be as well, first of all, to establish, if possible, the relation-ships of Leidy's Camelops Tcansanus, Leidy's Auchenia Tiesterna (asrepresented b}^ Merriam's specimens), and Cragin's Auchenia Jiuer-fanensis. That all belong to the same genus there seems to be littlereason to doubt. Comparisons between Camelops Icansanus andMerriam's specimens are limited to the anterior half of the premaxilla,the anterior extremity of the maxilla, the last incisor, and a part ofthe socket for the canine. Merriam ^ writes that compared withLeidy's type the anterior end of the rostral region of the RanchoLa Brea skulls shows little to distinguish it, the general proportionsof the elements present and the location of the teeth bemg nearlythe same. However, it seems to the writer that Merriam's drawing,figure 5, shows that the premaxilla has almost exactly the form andproportions of that of the lama. At the point of the alveolar borderwhere the maxillo-premaxillary suture is encountered, the premaxillabegins to narrow as it passes backward. In the type of CamelopsIcansanus the bone continues to widen backward as far as it is pre-served. Leidy's figure appears to show also that the whole alveolarborder in front of the canine was more strongly sigmoid than in thespecimens from Rancho La Brea. Leidy himself stated that thepremaxilla is of very much more robust proportions than in the lamaor the camel. It seems to me that Merriam has done right in notidentifying his specimens as Camelops Icansanus.It is still more certain that the type of Camelops Jiuerfanensis isdifferent from both C. Icansanus and C. Jiesternus. Plate 25, figure 2,represents an exterior view of the left premaxilla of the Huerfanospecimen and Plate 25, figure 3, the inner surface of the same bone;while figure 4 of the same plate presents a view of the maxillaryborder. That part which was m front of the exit of the incisor iswantmg. It is evident that the maxilla extended forward on thealveolar border nearly to the incisor tooth. Just below, behind, andoutside of the bottom of the socket for the incisor there is a greatthickening of the premaxilla. On this thickening, mesiad of the lineof suture, there is a concave surface which is taken to be a part of thewall of the socket for the canine. The presence of the canine herefurnishes the reason for the thickening of the premaxilla at thisplace. If this conclusion is correct, the canine must have emergedimmediately behind the incisor just as it does in the Bactrian camel.
?Univ. California Publ., Geol., vol. 7, p. 318.
270 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 46.In the t3^pe of Camelops Icansanus there is a space of 45 mm. betweenthe two teeth. It is to be noted further that the maxillo-premaxillarysuture in C. Icansanus crosses the alveolar border 20 mm. behind theincisor. What may have been the form of the nasal border of theprem axilla of C. Icansanus behind the part present in the type, wedo not know. As will be seen, that of C. huerfanensis is stronglyconcave, differing thus from C. Jiesternus, the species of Camelus, andthe lama, m all of which this border is sinuous. Unfortunately, theanterior end and the upper border of the left maxillary which cameinto contact with the premaxilla is broken away.The type of C. huerfanensis differs from Merriam's specimens ofC. Jiesternus in having the posterior palatine foramina placed fartherbackward; that is, opposite the first molar, instead of opposite thethird or fourth premolar. In the specimens described by Merriamthe mental foramina are said to be situated immediately below orslightly behind the canine, as in the lama. In C. huerfanensis theyare placed but little in front of the hinder end of the symphysis andprobably well behind the canines. In the camel last mentioned thecoronoid process of the lower jaw is relatively wider than in theC. hesternus, as shown by Merriam's figure 5.It is possible to describe some of the very fragmentary parts ofthe skull of Cragin's type. A part of the occipital region (Plate 25,fig. 1) is present, but it does not extend down to the foramen magnum.There was a strong sagittal crest, but its summit has crumbled away.The width of the occiput, measured along a line passing through thelateral foramina, was close to 110 mm., the same as in the case of thedromedary present. The lambdoidal crest is thin and sharp. Onthe supraoccipital surface there is a median descending ridge, roughand rounded, and separating two deep excavations. Exterior tothese there is on each side another deep excavation, at the bottomof which is placed the lateral foramen. This region resembles thatin Merriam's specimens. The paroccipital process is longer, thicker,and wider than in the Bactrian camel, and at its extremity presentsa hook. Its form is quite like that of the lama. A fragment of theright maxUla has near its upper edge a depression which corresponds,doubtless, to the fossa mentioned by Merriam.Figiu'es 2-4 of Plate 25, as already stated, give views of the leftpremaxilla. The total length of the fragment is 126 mm.; the widthof the upper end is 26 mm.; at the narrowest part, 20 mm.; near theanterior end, 31.5 mm. The thickness, a short distance below theupper, or hinder end, is 7 mm.; just at the bottom of the socket forthe incisor, 17 mm. The incisor socket indicates that the tooth waslarge, its height, close to the place of emergence of the tooth, being22 mm. The socket was at least 40 mm. deep. The surface believed
NO. 2025. CAMELS OF THE FOSSIL GENUS CAMELOPS?HAY. 271to have supported the upper side of the canine indicates that thistooth also was one of considerable size. The left maxilla shows thatthe infraorbital foramen was above the front of the first molar. Theregion in front of this is not so constricted as in the dromedary.Between the fourth premolars the palate was 50 mm. wide; at thefront of the last molars, 87 mm. The palate was therefore narrowerthan in Merriam's specimens. The left posterior palatme foramen issituated somewhat behind that of the opposite side (Plate 26, fig. 1).The right ramus of the lower jaw permits various measurementsto be taken. The symphysis presents only about 22 mm. of its hinderend. From the hinder end of this to the rear of the bone, above theangle, is 385 mm. ; to the rear of the condyle, 415 mm. ; to the extrem-ity of the coronoid process, about 450 mm. The depth of the jaw atthe rear of the symphysis is 48 mm. ; at the front of pm.4, 59 mm. ; atthe front of m.g, 83 mm.; at the rear of m.g, 110 mm. The measure-ments indicate a jaw longer than that of Merriam's specunen 20028,but with the depth about the same.It is not possible to determine accurately what was the length of thesymphysis m the type of C. huerfanensis. After makmg such esti-mates as are possible with the materials at hand the length is takento have been at least 125 mm. On this assumption the length of thejaw, to a point on the hinder border and on a level with the premolars,will be about 505 mm.; to the rear of the condyle, about 530 mm.;to the rear of the coronoid process, about 540 mm.The width of the coronoid process at the middle of its height is 46mm. ; that from the front of this process to the rear of the condjde, is92 mm. ; the former being therefore just one-half of the latter dimen-sion. In the specimens of Camelus and Auchenia at hand the widthof the process is considerably less than half that of the jaw across thecondyle; and the same appears to be true in the case of Merriam'sspecimens.Returning to the symphyses it may be well here to describe thosefrom Minidoka, Idaho. The largest one. No. 2579, is 120 mm. longand was about 50 mm. wide at the narrowest place; 58 mm. wideat the bases of the outer incisors. The mental foramen is placedfour-fifths of the distance from the front to the rear of the symphysisand well behind the canine. In Merriam's specimens the foramenis below or slightly behind the canine, and somewhat further forwardthan in C. huerfanensis. In the other specimen from Minidoka (Plate26, fig. 2), the symphysis has a length of 103 mm. and the foramenis nearer the rear of the union. In these jaws, which probablybelonged to the same species as Cragin's specimen, the canines aresituated nearer the incisors- than they are in those found in California.It may be noted here that in the type of C. huerfanensis the fourth
272 PROCEEDINGS OF THE -NATIQ-^^AL MUSEUM. vol.46.premolar is placed 65 mm. behind the symphysis, while in Merriam'sspecimen, No. 20028, furnishing a shorter jaw, the same tooth seemsto be at a distance of 75 mm. from the symphysis. Merriam hasstated that the symphysis in Camelus is much longer than in Auchenia.This depends, however, on the species. From the specimens at handit is found that in the Bactrian camel the length of the symphysisequals about 26 per cent of the length from the incisive border to therear of the condyle; in Auchenia, 28 per cent; in the dromedary, 35per cent. On the assumption that the symphysis of Cragin's specieswas 125 mm. long, its length would be 23.5 per cent of the length ofthe jaw. It is not improbable that the symphysis was really longerthan 125 mm. Judging from the drawmgs presented, Cope's Texasspecimen^ referred to Jiestemus, had a symphysis equal to 24 per centof the length of the jaw, estimated as in the other cases. In that jawthe position of the canine and that of the mental foramen are as inCragin's type. That jaw was, however, shorter than the latter byabout 100 mm. Furthermore, the fourth premolar appears to bemuch nearer to the symphysis than in the case of Cragin's type,apparently only about 40 mm. distant.The teeth of the various specimens at hand which are supposed tobelong to C. Jiuerfanensis must be described. None furnishes thelast upper mcisor; but the left premaxilla (Plate 25, figs. 2-4) describedabove contains the socket of this tooth. This has already beendescribed. Likewise, the only trace of the upper canine is shown onthat premaxilla, as already noted.In Cragin's type the fourth premolar of the right side and all themolars are present and in excellent condition. On the left side littleis left of the teeth except the roots of the third and fourth premolarsand of the fu-st and second molars. (Plate 26, fig. 1 .) So far as may bedetermined from the two roots of the third premolar, this tooth hadthe size of the corresponding one in Merriam's specimen, referred toC. hestemus. In the table below are given the measurements of theupper cheek-teeth. The height of the crowns is given as an indica-tion of the stage of wear; for as the teeth are worn down, the antero-posterior diameter, here called the length, diminishes (except in thecase of the third premolar and the last molar), while the transversediameter increases. In these measurements the length of the crownis taken along the middle of the width of the grmding surface, whilethe width of the tooth is taken at the base and where greatest. Thelength of the whole series and of the molar series is taken in a straightine, not along the curve.
1 Geol. Surv. Texas, 4th Ann. Rep., 1892, pp. 71, 93, pi. 21, figs. 3, 4.
NO. 2025. CAMELS OF THE FOSSIL GENUS CAMELOPS^HAY. 273Measurements of the premolars and molars of the upper jazv. mm.Length of the premolar-molar series, pm.^?m.^ inch:sive 171Length of the teeth, from front of pm.* to rear of m.^ 152Length of the molar series 129Pm.^, height.length 18. 8width - 11Pm.*, height 38length 25width 25M.i, height 35length 38. 5width 30M.2, height 57length ^8width 30M.^ height 62length on grinding surface -15length at middle of height 56width 28.5These measurements may be compared with those given by Merriamon page 316 of his paper; but some of his measurements appear tohave been taken somewhat differently. If ui the Huerfano specimenwe measure the distance from the front of pm.* to the rear of m.^ alongthe outer curve we shall have 167 mm.; and along the outer curve ofthe molars alone, 138 mm. Merriam mentions the fact that in hisspechnen 20028 the metastyle of the last upper molar is drawn outposteriorly as a wing; but that this wing is not present in the speci-men 20040. In the Huerfano specimen this metastyle is large and isbent strongly inward, as may be seen from figure 1, plate 26. Thewidth of this metastyle, from side to side, is 14 mm. Accompanymgthe materials from Minidoka is a third upper molar which had not yetbeen cut, and whose base had not yet been completed. The metastyleforms a broad sharp border, but shows no tendency to be bent mesiad.The lower incisors of the Huerfano specimen are missing. Theyare present in the two symphyses from Minidoka. Those of No. 5315are shown in figure 2 of Plate 26. The outer incisors had only justbegun to wear. They have a length of 60 mm., a width of 13 mm. atthe middle of the length, and a thickness of 9 mm. The second in-cisors are naturally more worn. They have a width of 18 mm. and athickness of 10 mm. at the middle of the length. The first mcisors areabout 17 mm. wide and 12 mm. thick. AU are flat on the upper sur-face and convex from side to side on the lower, or front, surface.They are relatively more powerful teeth than in the lama or thedromedary.95278??Proc.N.M.vol.46?13 18
274 PROCEEDINGS OF THE NATIONAL MUSEUM.The incisors of No. 2579, from Minidoka, are still more worn thanthose just described, and they belonged to a larger animal. Those ofthe first pair, in their worn condition, are at least 73 mm. long. Allthese incisors were directed forward more strongly than in the lamaand the Bactrian camel, as the latter is represented in the specimen athand. It must be stated further, that the lateral incisors of figure 2,Plate 26, are, relatively to the others, much larger than in Cope'sspecimen from Texas. ^ None of these m-cisors are as strongly curved as they arein the lama.Figures 3 and 4, Plate 26, represent ofthe natural size the two incisors, i.2 and i.gright side, which were found in 1867 byE. L. Berthoud. It is, of course, notcertain that they belonged to Cam.elops,but it is probable that they did. Theseteeth are spatulate in form, not greatlycurved, and are somewhat twisted. Whatis taken to be the second right incisor(Plate 26, fig. 4) is worn very slightly ; theother incisor (fig. 3) not all. I., has alength, in a straight line, of 92 mm.; awidth of 23 mm. near the anterior end;while at the middle of the length, thewidth is 18 mm.; the thickness 10.5 mm.A section of the tooth at this place wouldgreatly resemble that of the correspondingtooth from Minidoka. 1. 3 had not quitecompleted its growth at the base. It is80 mm. long, 26 mm. \vide near the front;21 mm. wide and 10 mm. thick at themiddle of the length.As in other specimens supposed to be-long to Camelops, the lower jaw of theHuerfano skuU has present no traces ofany premolars in front of the fourth. In a part of a jaw (Plate 26,fig. 5) from Minidoka, which belongs possibly with the symphysisnumbered 5315, there is left the base of the crown and the root ofa small third premolar. This has a diameter of 7 mm., fore and aft,and a transverse diameter of 6 mm. In the anterior border of thefourth premolar there is a groove which appears to have been occupiedby the crown of this third premolar.
Fig. 1.?Molars (m') and premolars(pm^) OF LO"WER JAWS OF THE TYPEOF CAMELOPS HUERFANENSIS. X |.
1 Geol. Surv. Texas, 4th Ann. Rept., 1892, pi. 21, fig. 4.
CAMELS OF THE FOSSIL GENUS CAMFLOPS?HAY. 275The following are the measurements of the fourth premolar and themolars of the type of Canielops Imerfanensis (fig. 1) and of teeth ofNo. 5315, from Minidoka. Merriam's measurements of the lowerteeth of his specimen No. 20040, referred to C. hestemus, are added inthe third column ; and in the fourth the measurements given by Copefor his Texas specimen. In C. liuerfanensis the fourth premolar andfirst molar are present in the right ramus, while in the left ramus thepremolar is represented by the socket only.Measurements of lower premolars and molars.
Teeth measured. Huerfanospecimen.
Length of lower teeth, pm.i to m.3, inclusiveLength of molar seriesPm.4, heightlengthwidthM.i, heightlengthwidthM.2, heightlengthwidthM.3, heightlengthwidth
Minidoliaspecimen.
143US25231430?2717454020
No. 20040Univ.,Cal. Cope'sTexasspecimen.
162.2
276 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 46.the Texan specimen belongs with neither C. Tiesternus nor C.Tiuerfanensis.4. It is impossible at present to decide the status of Megalomeryxniohrarensis Leidy. The decision must await new discoveries.5. The writer accepts, therefore, as species of Camelops to becarried on the rolls, until further knowledge is obtained, C. Icansa^nus, C. californicus, C. liesternus, C. vitakerianus, C. niohrarensis, C.macrocephalus, and C. huerfanensis , the latter to include provisionallyC. sulcatus (Cope) and the Texan mandible referred by Cope to C.liesternus.Leidy and Cragin referred to the genus Auclienia the species men-tioned in this paper. Wortman distinguished the two genera on thepresence of a prominent lamina, or style, at the anterior outer angleof the two hinder lower molars of Auclienia, a structure absent fromthe same teeth of the species of Camelops. Merriam accepts thisseparation. It seems to the present writer that there arc variousother characters which are of perhaps genera importance. It isevident that the species of Camelops, so far at least as representedby C. liesternus, had skuUs relatively longer and narrower than thoseof Auchenia. In the latter the width at the rear of the orbits isequal to about 54 per cent of the length from the front of the fora-men magnum; while, according to Merriam's illustrations and meas-urements, the corresponding width in Camelops equals only about45 per cent of the corresponding length. There is an importantdifference in the upper molars. In Auclienia the length of the grmd-ing surface is nearly equal to the width of the tooth measured atthe base; that is, when these molars are well worn down the grind-ing face is nearly square. In Camelops the teeth are relatively longantero-posteriorly. In Camelops the lower incisors are less curvedthan in Auchenia and directed more strongly forward; that is, theyare more procumbent. In Auclienia the nasals are strongly expandedat the hinder end; in Camelops they are narrow posteriorly. InAuclienia the lachrymal vacuity is crowded outward against themner border of the lachrymal, while in Camelops the vacuity hardlyor not at all comes into contact with the lachrymal. In Camelopsthere is fossa in the upper border of the maxilla; in Auclienia thereis none.
NO. 2025. CAMELS OF THE FOSSIL 0ENU8 CAMELOPS?HAY. 277EXPLANATION OF PLATES.Plate 25.Figs. 1-4. Camelops huerfanensis Cragin.1. Rear of the skull, to show its form, the two excavations on each side of themidline, and the intervening ridges. Part of the right side is missing.2-4. Left premaxilla.2. View from the outside.3. View from the inner side, i, Socket for the third incisor.4. View of the border which articulated with the maxilla, r, Surface whichformed a part of the socket for the canine,Plate 26.Fig. 1. Camelops huerfanensis . Palate showing premolars and molars. Type. XJ.2. Camelops huerfanensisf Symphysis showing incisors and canines. No. 5315U. S. Nat. Mus. XJ.3,4. Third and second incisors. No. 870 U. S. Nat. Mus. XI.5. Part of right ramus of lower jaw. Shows minute pm.3. pm.4 m.i and m.2.No. 5315 U. S. Nat. Mus. XJ.

U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 46 PL. 25
Skull and Premaxilla of Camelops huerfanensis.For explanation of plate see page 277.

U. S. NATIONAL MUSEUM PROCEEDINGS, VOL. 46 PL. 26^' tr-'
\pm3
Details of Teeth and Jaw of Camelops huerfanensis.For explanation of plate see page 277.