SMITHSONIAN MISCELLANEOUS COLLECTIONSVOLUME 116, NUMBER 2
Cfjarlea B. anb Jfflarp "^aux OTalcottJXeaearcf) Jftmb
REMAINS OF LAND MAMMALS FROMTHE MIOCENE OF THE CHESAPEAKEBAY REGION
ByC. LEWIS GAZINU. S. National MuseumANDR. LEE COLLINSUniversity of Tennessee
(Publication 4019)
CITY OF WASHINGTONPUBLISHED BY THE SMITHSONIAN INSTITUTIONOCTOBER 12, 1950

SMITHSONIAN MISCELLANEOUS COLLECTIONSVOLUME 116, NUMBER 2
Cjjarle* B. anb Jflarp ^aux 2Hakott&e*earcf) Jftmb
REMAINS OF LAND MAMMALS FROMTHE MIOCENE OF THE CHESAPEAKEBAY REGION
ByC. LEWIS GAZINU. S. National MuseumANDR. LEE COLLINSUniversity of Tennessee
(Publication 4019)
CITY OF WASHINGTONPUBLISHED BY THE SMITHSONIAN INSTITUTIONOCTOBER 12, 1950
Zfye JSoro (g>afttmore tyreeeBALTIMORE, MS., U. B. A.
Ctjadeg B. anto JHarp \Xaux flSalcott &egeard) JfunbREMAINS OF LAND MAMMALS FROM THEMIOCENE OF THE CHESAPEAKEBAY REGIONBy C. LEWIS GAZINU. S. National MuseumANDR. LEE COLLINSUniversity of TennesseeThe occasional discovery of the remains of various land mammalsin the Miocene marine deposits, principally along the cliffs of theChesapeake Bay in Maryland, has, in the course of a number ofyears, resulted in a rather significant representation of these forms.Heretofore no systematic attempt has been made to bring togetherthe various previous reports of land mammals from these beds and toplace on record the undescribed remains known to exist. Much ofthe material extant is preserved in the U. S. National Museum. Itis comprised of an accumulation of gifts from private collectors,Johns Hopkins University, U. S. Geological Survey, and U. S. Na-tional Museum personnel and, in particular, represents the resultsof searches by the junior author. The drawings depicting the speci-mens were prepared by William D. Crockett, of the U. S. NationalMuseum.The materials described in this paper are for the most part fromthe Calvert formation, associated with a wealth of invertebrate fossils,and with unquestionably the most abundant representation known,both in numbers and diversity, of Miocene marine mammals. In someinstances the fossils have weathered out and were found loose on thestrand at the foot of the cliffs, and in cases where only Calvert isreported exposed in the immediate section their derivation from thisformation is logically inferred. However, where Choptank formation,also Miocene in age, is exposed in the upper portion of the cliffsection some doubt attaches to the origin, unless the specimen wasuncovered in place.SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 116, NO. 2
2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Il6PREVIOUS REPORTSThe occurrence of remains of Mammalia in the Chesapeake Mio-cene is the basis of some of the very early reports on vertebrate fossilsin this country. The cast of a mastodon tooth that Leidy 1 includedin his description of Mastodon obscarns was considered to have beenmade from the tooth mentioned by Harlan - in 1842 as coming fromthe Miocene of Maryland, and by Lyell in the Proceedings of theGeological Society of London for 1843 t0 x ^45 (P- 3$)- Furtherdiscussion is found in Warren's 3 work on "Mastodon giganteus."Manatee and porpoise were also described from the MarylandMiocene at about this time, 4 with subsequent papers by Leidy andothers on occasional finds. However, the first extensive reports onthe marine mammals of these beds were Cope's studies of 1867- 1869.Included in Cope's 1867 5 paper was the description of Cynorca pro-terva, the second type of land mammal to be described from thesebeds, first thought to be a squalodont but later recognized as a peccary.In the general review of the fauna from the Maryland Miocene inthe Maryland Geological Survey report of 1904, E. C. Case, in histreatment of the mammals, overlooked the records of land types.GEOLOGIC RELATIONSThe Miocene series in Maryland, known as the Chesapeake group,includes three formations, the Calvert, Choptank, and St. Marys, inascending order. These consist essentially of sands, clays, marls, anddiatomaceous sediments, for the most part highly fossiliferous. Thebeds are exposed in southern and eastern Maryland, to the southeastof Baltimore and Washington, where they unconformably overlie thegreensand marls of the Eocene. For the most part they dip gentlyto the southeast, preserving the higher formations of the group inthe more southerly portions of the State. The Chesapeake group is,in turn, unconformably overlain by Pleistocene and possibly Pliocenedetritus over much of its extent, exposure of the Miocene resulting
1 Leidy, Joseph, Journ. Acad. Nat. Sci. Philadelphia, ser. 2, vol. 7, pp. 244,396, pi. 27, fig. 13, 1869.2 Harlan, Richard, Amer. Journ. Sci., vol. 43, p. 143, 1842.3 Warren, John C, The Mastodon giganteus of North America, pp. 78-85,175-180, 1852.4 Harlan, Richard, Medical and physical researches, p. 385, 1835 ; Bull. Proc.Nat. Inst. No. 2, p. 195, 1842. Also, Dekay, J. E., Zoology of New York, pt. 1,pp. 123, 136, 1842.5 Cope E. D., Proc. Acad. Nat. Sci. Philadelphia, 1867, pp. 151-152.
NO. 2 MIOCENE LAND MAMMALS GAZIN AND COLLINS 3largely from dissection of the older surfaces and terraces. Its de-velopment is most impressive in the cliffs along the Chesapeake Bay.The Calvert formation, with the superadjacent portions of theChoptank, is the principal mammalian fossil-bearing formation ofthe group. It is exposed from near Washington to below the mouthof the Wicomico River along the Potomac, associated with Eocenenorthward and with the Choptank southeastward. It extends southinto Virginia, and as a belt northeastward across eastern Marylandand Delaware, and probably has for its equivalent the Kirkwood inNew Jersey. AGE AND FAUNAL RELATIONSHIPSThe age of the Chesapeake series has been regarded by Dall 6 asmost nearly equivalent to the Helvetian of the European sequence,which in turn is currently regarded as about middle Miocene. In thecommittee report 7 on the "Nomenclature and Correlation of theNorth American Continental Tertiary," the Calvert is regarded asmore nearly equivalent to the Tortonian of Europe, a stage higherthan Helvetian, and, hence, approximately upper middle Mioceneor upper Hemingfordian in the North American continental sequence.It is compared to the "lower Sheep Creek" of the Nebraska Miocene,
"or, possibly, a slightly earlier age."From a critical examination of the fauna, the composition of whichwas not fully known to the committee, we are inclined to regard thestage of development as slightly later. This is based to a largeextent on the development reached in the Equidae and to someextent on certain of the other forms. The horizon represented isprobably a little later than the lower Sheep Creek, or than the FloridaHawthorn (as determined by the vertebrate localities in the northernpart of the State), although the extent to which the fauna is ad-vanced over the faunas of those horizons is not great. The mostnearly comparable fauna is to be seen in the Merychippus zone ofthe Temblor formation in California. The stage represented, if notthe latest Hemingfordian, is early Barstovian, earlier than the typicalBarstow fauna, and might be regarded as early Sarmatian.The land mammals thus far recognized from the Calvert formationare as follows
:
Carnivora
:
Tomarctus marylandica Berry
6 Dall, W. H., The Maryland Geological Survey, Miocene, p. cxliii, 1904.7 Wood, H. E., 2d, et al. Bull. Geol. Soc. Amer., vol. 52, pi. 1 and p. 16, 1941.
4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Il6
Proboscidea
:
Gomphotherium calvertcnsis, n. sp.Perissodactyla
:
Tapiravus cf. validiis (Marsh)Aphelops?, sp.Archaeohippus, sp.Merychippus, sp.Artiodactyla
:
Cf. Cynorca proterva (Cope)Hesperhys (Desmathyus) ?,8 sp.In a consideration of the age assignment, the Carnivora andArtiodactyla listed above are not significant, although Hesperhyswould suggest a relatively earlier age than some of the other forms.Its listing, however, cannot be regarded as in any way indicative,inasmuch as Hesperhys has not been identified with any degree ofassurance. The presence of "Desmathyus" in the base of the Kirk-wood is open to various interpretations, but for whatever it is worthattention should be called to the fact that Dall 9 regarded the southernNew Jersey Miocene as containing reworked fossils from oldersediments.Gomphotherium is significant in either of two ways. If the presenceof mastodon is to be regarded as placing an early limit on thepossible age assignment, then a lower Barstovian age would likely beindicated. On the other hand, there is always the possibility that theCalvert represents the first appearance of mastodons in North Amer-ica, a conclusion not out of keeping with its geographic remotenessfrom the western occurrences. Early appearances of mastodons inother regions of North America include their presence in the PawneeCreek, Deep River, Mascall, Skull Spring, Sucker Creek, and Temblorfaunas. So far no mastodon has been recorded from the lower SheepCreek beds of Nebraska, from the Hawthorn of Florida, or fromthe Phillips Ranch locality in California.Among the Perissodactyla, the tapir gives little evidence regardingthe age assignment which may be made, other than to indicate arelationship with the Shiloh marl of the Kirkwood formation inNew Jersey (from which Marsh described the species, Tapiravusvalidus), inasmuch as tapir remains are exceedingly scarce in thisportion of the geologic column. Also, the rhinoceros present is ofno great help because the remains are too fragmentary for certaingeneric reference, and Aphelops has been recorded from Hawthornto Pliocene times.
8 So far recorded only in zones 17 and 19, which are allocated to the Choptank.9 Dall, W. H., ibid., p. cxliv, 1904.
NO. 2 MIOCENE LAND MAMMALS GAZIN AND COLLINS 5The horses, however, although only the lower teeth have been found,give rather good indication of the portion of the Miocene represented.The Merychippus teeth appear advanced over related types from theHawthorn or from the Sheep Creek beds, but, as noted under thesystematic treatment of this portion of the fauna, the stage ofdevelopment reached is nearly comparable to that in Merychippuscalifornicus of the Temblor. The Temblor mammalian fauna is repre-sented, as is the Calvert, by materials derived from a marine sequenceand includes a strikingly similar assemblage, suggesting an ecologicrelationship in addition to one in time.SYSTEMATIC DESCRIPTION OF THE MAMMALIAN REMAINSOrder CARNIVORAFamily CANIDAETOMARCTUS MARYLANDICA BerryA second lower molar and the greater part of the lower carnassial ofa dog (U.S.N.M. No. 15561) were found by Charles T. Berry inzone 10 of the Calvert formation, about 1^ miles south of PlumpointWharf, Calvert County, Md. These were described and given thename Tomarctus marylandica by Berry in 1938. He noted a markedresemblance of his form to that described by Merriam 10 as Tephro-cyon kelloggi from the Miocene Virgin Valley beds of northwesternNevada. The Calvert material appears inadequate for good specificdiagnosis. Nevertheless, the geographic remoteness of its occurrencesuggests a probable distinctness from previously described forms,a matter for future collecting to demonstrate.Order PROBOSCIDEAFamily GOMPHOTHERIIDEAConsiderable confusion has existed, and still exists, in mastodontnomenclature, particularly as applied to the mastodon known to occurin the Maryland Miocene. Details of its complexity in this instancewere not clarified by Osborn in his monograph of 1936. 11 In orderto illustrate the taxonomic procedure that the present authors havefollowed, it is necessary to review briefly the origin of the name
"Mastodon" obscurus Leidy and the history of the use to which ithas been put.10 Merriam, J. C, Univ. California Publ., Bull. Dept. Geol., vol. 6, pp. 235-238,191 1.11 Osborn, H. F., Proboscidea, vol. 1, pp. 285-287, figs. 232 and 233, 1936.
6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Il6
Leidy,12 in his work on the "Extinct Mammalian Fauna of Dakotaand Nebraska," described several mastodon teeth and figured three,which were considered to be from various localities in the easternStates. One of these, the first of the series discussed, was a cast ofa tooth (see fig. 3), the original of which was lost, but demonstratedwith some degree of probability by Leidy to have come from nearGreensboro, Caroline County, Md. As shown by the MarylandGeological Survey 13 this area is underlain by Choptank Miocene. Asecond tooth of confused history included in the discussion was pur-chased in London and purported to be American. Two incompleteteeth from Tarboro, N. C, a fragment of a molar from an unknownlocality referred to as M. chapmani, and a tooth portion from Darien,Ga., were also described.In a later part of the paper Leidy " proposed the name Mastodonobscurus, citing the description beginning on an earlier page and thethree figured teeth. However, he did not designate a type. Beneaththe synonymy the following statement is made:Apparently a species distinct from the preceding (M. mirificus), indicated byspecimens from North Carolina and Georgia. Other specimens from unknownlocalities supposed, however, to be American probably belong to the same. Oneof the latter was referred to a separate species by Dr. Hays under the name ofM. chapmani, but Dr. Hays expresses the opinion to me that this is distinctfrom the former. Under the circumstances I propose to distinguish the speciesrepresented by the undoubted American specimens by the name heading thearticle. The species has been suspected to be of Miocene ageThis discussion, constituting the designation of the typical andreferred materials, does not include Maryland specimens as indicativeof the species, but regards them as of dubious origin ; hence the castof the lost tooth, believed to have come from Caroline County, iseliminated from consideration as the type.In keeping with the above limitations, Merrill 15 in 1907 selectedas type, by lectotype, one of the Tarboro, N. C, specimens, U.S.N.M.No. 426. At this point the record is clarified, but in 1936 Osborn 16stated that Merrill's designation of the type was an error and thatthe cast of the lost tooth is the proper type. He further indicated thatthe Tarboro tooth really represents the species, Ocalientinns (Ser.)12 Leidy, Joseph, Journ. Acad. Nat. Sci. Philadelphia, ser. 2, vol. 7, pp. 244-249, pi. 27, figs. 13, 15-16, 1869.13 Maryland Geological Survey, Miocene, map facing p. xxiii, 1904.14 Leidy, Joseph, ibid., p. 396, 1869.15 Merrill, G. P., U. S. Nat. Mus. Bull. 53, pt. 2, p. 45, 1907.™ Osborn, H. F., ibid., p. 286, 1936.
NO. 2 MIOCENE LAND MAMMALS GAZIN AND COLLINS J
obliquidens, of somewhat later age. The taxonomic arrangementmade by Osborn is untenable, and we are forced to the followingconclusions: The Tarboro specimen is the type of "Mastodon"obscurus, and Serridentinus obliquidens is a synonym of Serridentinusobscurus (Leidy). "M." chapmani Hays must be ignored, as thereis a strong probability that this specimen is of foreign origin. Thevarious mastodon teeth now known from the Miocene of Marylandare without a specific designation. For these we propose the followingname: GOMPHOTHERIUM CALVERTENSIS, new speciesType.—Left upper second molar, U.S.N.M. No. 13764.Horizon and locality.—Zone 13 or 14, Calvert formation; in greensandy clay about 4 feet above beach, 1.1 miles north of GovernorRun Road, Calvert County, Md.Specific characters.—Size of teeth approximating those in thegenotype, G. angustidens (Cuvier), from the Miocene of France.Teeth low-crowned. Bunodont lophs of upper molar arranged trans-versely, with well-developed trefoils. Anterior and posterior lobesof trefoils partially distinct or cuspate. Lophs of lower molars morearcuate and forward sloping, and with posterior lobe of trefoils betterdeveloped than anterior, and with more noticeable separation. Slighttendency to the formation of trefoil form on inner part of lowerteeth and outer part of upper teeth. Teeth noticeably resemble thoseof G. (Serridentinus) obscurum from North Carolina, but decidedlysmaller and lower-crowned.Description.—In addition to the type upper molar (fig. 1), whichwas collected by M. H. White about 1931 and given to the NationalMuseum in 1935, there is a complete unworn last lower molar, alsofrom near Governor Run, collected by A. Hecklinger in 1931 anddeposited in collections of the Maryland Academy of Sciences. Acast of this tooth (fig. 2) is in the collections of the National Museum,U.S.N.M. No. 1 21 34. The controversial lost tooth that Leidy re-garded as coming from Caroline County, Md., and which Osbornconsidered as the type of G. obscurum, is represented also in thenational collections by a duplicate (fig. 3) of the cast in the Academyof Natural Sciences of Philadelphia.The second upper molar of G. calvertensis lacks only the front halfof the anterior loph and the root portions. It is relatively unworn andshows clearly the details of the cusp development. In occlusal view thesides of the tooth are nearly parallel and the lophs almost at right
8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Il6
angles to the long diameter. The crests of the lophs are characterizedin the unworn condition by a series of rounded cusps. The trefoil planis markedly developed with the anterior and posterior lobes, betweenthe lophs, about equally developed and tending to be separated from
Fig. i.—Gomphotherium calvertensis, new species : Left upper second molar(U.S.N.M. No. 13764) type specimen, lateral and occlusal views, X i- CalvertMiocene, Maryland.the lophs at their summits. A slight, or subdued, trefoil arrangementis also seen on the outer half of the lophs on the opposite side ofthe clearly denned median crease or fold extending the length ofthe tooth. In end view the inner and outer portions of the lophs areof about equal height, except for the first loph where wear has
NO. 2 MIOCENE LAND MAMMALS—GAZIN AND COLLINS
somewhat lowered the lingual half. The slopes of the inner and outercolumns are nearly equal, although the inner wall appears to be lesserect. No cingular crest is developed on the buccal wall and onlybetween the lophs lingually. The character of the forward margin ofthe tooth cannot be determined, but the posterior margin exhibits
Fig. 2.
—
Gomphotherium calvertcnsis, new species : Drawing from cast(U.S.N.M. No. 12134) of right last lower molar belonging to MarylandAcademy of Sciences, lateral and occlusal views, X I- Calvert (?) Miocene,Maryland.
a markedly cuspate crest across the midportion and about halfway upthe posterior loph, rising toward the inner extremity.The last lower molar, belonging to the Maryland Academy ofSciences, appears somewhat smaller than would likely have beenassociated with the type upper molar but may well be within therange of this species. It is also a little shorter and narrower thanthe cast of the Caroline County specimen. The tooth is unworn andcomplete except for roots. Resemblance to the corresponding toothof G. angustidens is striking, although the Maryland Academy tooth
10 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Il6
is noticeably smaller than U.S.N.M. No. 150 from the Miocene ofFrance. These teeth, including the cast of the Caroline Countyspecimens, are alike in structural pattern with arcuate lophs in whichthe outer trefoil portion is offset or directed backward, and the poste-rior buttress or lobe of the trefoil is very well developed. This lobeis even more prominent and shows better separation directly behind
Fig. 3.—Cf. Gomphotlierium calvertensis, new species : Drawing from cast(duplicate of Acad. Nat. Sci. Philadelphia No. 13278), lateral and occlusal views,X §. Original purported to be from Choptank Miocene, Maryland.the first loph. It is essentially single in the Maryland Academy speci-men but may be bilobed in G. angustidens. The posterior lobe ofthe trefoil is less distinctly separate from the second loph but thereis a suggestion of bilobing near the base in the Maryland form.Slight basal cuspules or weak and irregularly developed trefoilingis seen inward of the median longitudinal cleft.In critically examining the type and other materials of Serridentinusproductus and Serridentinus floridanus the structural arrangement isscarcely different from that in typical Gomphotherium. Except for
NO. 2 MIOCENE LAND MAMMALS GAZIN AND COLLINS II
size and height of crown, the molar teeth of Pliocene Scrridentinusin this country differ but little from Miocene Gomphotherium molarsat hand. We suspect that but one phyletic group is represented.Measurements {in millimeters) of cheek teeth ofGomphotherium calvertensisTransverse diameter of M 2 , U.S.N.M. No. 13764 (type) 70.5Length of Ms, U.S.N.M. No. 12134 (cast of Maryland Acad. Sci. speci-men) 140.Transverse diameter of M 3 , U.S.N.M. No. 12 134 61.Transverse diameter of Ms, duplicate of cast in Acad. Nat. Sci. Philadelphia. 69.Order PERISSODACTYLAFamily TAPIRIDAETAPIRAVUS cf. VALIDUS (Marsh)A right maxilla, U.S.N.M. No. 18372 (fig. 4), with P 3 and M 1preserved, but exhibiting the root portions or alveoli for the re-maining cheek teeth, was found by J. E. Smedley, of the U. S.Geological Survey, about 2 miles south of the Chesapeake Beachresort. The specimen was discovered weathered-out on the beach, butat this point only the Calvert formation is reported in the adjacent cliffsection so that there seems little doubt of its origin.The teeth resemble to a marked degree those seen in the moderntapir, but are very much smaller and appear to be relatively lower-crowned. Their size is rather close to the dimensions given byMarsh 17 for Tapiravus validus from the "Miocene marl of Cumber-land Co., New Jersey." The Calvert specimen is tentatively referredto the New Jersey species.Schlaikjer,18 in his work on the tapir from the lower Miocene ofWyoming, considered that Marsh's form represented uppermost Mio-cene or lower Pliocene time. However, from a consideration ofTertiary stratigraphy in southern New Jersey it seems probable thatthe type is from the marl of the Kirkwood formation, which isconsidered to be very close in age to the Calvert formation inMaryland. Except for clay lenses in the Cohansey formation, theredoes not appear to be any beds of later date that could properly bereferred to as marl. Since we have some assurance that the Marylandspecimen is from the Calvert we may also reason that Marsh's typemay well have come from the Kirkwood.
17 Marsh, O. C, Proc. Acad. Nat. Sci. Philadelphia, vol. 23. pp. 9-10, 1871.18 Schlaikjer, E. M., Bull. Mus. Comp. Zool., vol. 70, No. 4, p. 249, 1937.
12 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Il6Comparison of the Calvert specimen with modern Tapirus showsthat its maxilla, though smaller, is relatively a little deeper, when thelengths of the cheek tooth series are considered. The infraorbitalforamen opens anteriorly above the posterior part of P2 , much as in re-cent specimens ; also the notch between the zygomatic wing of the max-illa and the maxilla proper is about opposite M 3 , and the anterior mar-gin of the orbit is located above about the middle of P 4 . However, the
Fig. 4.
—
Tapiravus, cf. validus (Marsh) : Right maxilla with P3 and M 1(U.S.N.M. No. 18372), lateral and ventral views, X i- Calvert Miocene,Maryland.
outward face of the lachrymal is less modified, though actually nearlyas long, and the lachrymal processes are not nearly so prominent. Thelachrymal foramen is divided posteriorly, and the two openings areclose together and face posteriorly. In modern Tapirus skulls the lowerof these foramina, when separate, is often observed to face outwardbeneath the posterior lachrymal process. The sulcus on the lateralface of the ascending portion of the maxilla is shallow and confinedto the maxilla and probably the frontal, whereas in Tapirus this sulcusis much more deeply impressed, involves the lachrymal bone, and
NO. 2 MIOCENE LAND MAMMALS—GAZIN AND COLLINS 1
3
may extend as far down as the infraorbital foramen. A small segmentof the frontal bone in the Calvert specimen is seen wedged betweenthe maxilla and lachrymal, much as in the modern tapir, so that wemay infer frontal and nasal modification for a proboscis, althoughprobably less strikingly developed. Modification of this portion ofthe skull was probably not greatly different from that shown bySchlaikjer 19 for the Harrison tapir.
Measurements in millimetersU.S.N.M. No. 18372 T. validus typeMeasurement P 3 M1 P4 or M 1Anteroposterior diameter 13.4 15.0 14.8*Transverse diameter 16.4 16.0 17.5 ** 7 and 8} lines, respectively.Family RHINOCEROTIDAEAPHELOPS?, sp.Three isolated lower molars, representing two individuals of rhi-noceros, were collected by Collins from a locality about 0.8 mile southof Randle Cliff Beach, Calvert County, Md. Two of these, unwornright and left (fig. 5a) molars, U.S.N.M. No. 18427, were found to-gether on the beach, but with matrix attached which suggested theirderivation from a weathered zone high on the cliff, probably zone 14or 15 of the Calvert. The third tooth, partially worn molar in jaw frag-ment, U.S.N.M. No. 18428 (fig. 56), was found in the same generallocality at a later date, and presumably represents a different individ-ual. The worn tooth is considered to be a first molar and the unwornteeth may well be second molars.The teeth are of moderate size, broad, and appear fairly high-crowned, particularly when viewed from the outer side. The toothwhich Marsh 20 described as Rhinoceros matutinus from Squankum,in Monmouth County, N. J., revised by Wood 21 as Diceratheriummatutinum, may not be significantly different in size from the Calvertteeth, as indicated by Marsh's illustrations, but appears appreciablylower-crowned and exhibits a strong external cingulum not seen inthe Maryland material. The only other Miocene rhinoceros material19 Schlaikjer, E. M., ibid., figs. 1 and 2, 1937.20 Marsh, O. C, Proc. Acad. Nat. Sci. Philadelphia, vol. 23, p. 3, 1870. Seealso Amer. Journ. Sci., vol. 46, pp. 411-412, pi. io, fig. 4, 1893.21 Wood, H. E., 2d, Bull. Geol. Soc. Amer., vol. 50, pp. 1968-1969, 1939(abstract).
14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Il6from eastern United States is that described by Colbert 22 from theHawthorn formation in Florida. The material which the FloridaGeological Survey lent to Dr. Colbert for study included five lowerteeth, apparently of one individual, which he referred to Aphelops.Again, these teeth do not appear to be significantly different in sizefrom those of the Calvert form but are much more worn, and aredescribed as having a strong internal cingulum, closing the inner valleyof the premolars. No premolars are represented in the Calvert material
Fig. 5.—Aphelops?, sp. : a, Unworn left M 2 ? (U.S.N.M. No. 18427) ; b, leftMi (U.S.N.M. No. 18428). Lateral and occlusal views, X f • Calvert Miocene,Maryland.but the molars do not have the cingulum developed either internallyor externally.Reference of the Maryland material to Aphelops is highly tentative.The specimens furnish little or no information which would serveto distinguish the form represented from Aphelops, and at the sametime the material is too incomplete to warrant certain generic reference.Measurements in millimetersMx ? (U.S.N.M. No. 18428) Ma ? (U.S.N.M. No. 18427)Anteroposteriordiameter40.0 Transversediameter28.6 Anteroposteriordiameter47.0 Transversediameter28.5
22 Colbert, E. H., Florida State Geol. Surv. Bull. 10, pp. 55-58, fig. 30, 1932.
NO. 2 MIOCENE LAND MAMMALS GAZIN AND COLLINS 15Family EQUIDAEARCHAEOHIPPUS, sp.A slightly worn lower cheek tooth in a jaw fragment, U.S.N.M.No. 1 843 1 (fig. 6a), is the only specimen of the small brachyodonthorse Archacohippus known from these beds. It was found byF. Stearns MacNeil, of the U. S. Geological Survey, a little less thana mile south of Plumpoint, Calvert County, Md. The Calvert forma-tion is the only Tertiary reported in the cliff section in this vicinity.
aFig. 6.
—
a, Archacohippus sp., portion of right ramus of mandible with M2(U.S.N.M. No. 18431) ; b, Merychippus sp., left lower premolar (U.S.N.M.No. 12032) ; c, Merychippus sp., left lower premolar (U.S.N.M. No. 18430).Lateral and occlusal view, X I. Calvert Miocene, Maryland.The tooth of No. 18431, apparently M 2 , is rather simple in con-struction, decidedly brachyodont, without cement, and with verylittle evidence at its present stage of wear for any separation of themetaconid and metastylid. A cingulum is developed only on theanterior and posterior walls, where it rises internally to the parastv-lid and hypoconulid, respectively. Its size is distinctly less than thistooth in the paratype of Archacohippus mourningi (Merriam) 23from the Barstow beds, but only a little less, and it is somewhat moreslender than in A. mourningi from the Temblor 24 formation in Cali-fornia. Size difference is not significant in comparison with isolated23 Merriam, J. C, Univ. California Publ., Bull. Dept. Geol., vol. 7, pp. 427-434,I9I3-24 See Bode, F. D., Carnegie Inst. Washington Publ. 440, pp. 48-50, 1933.
l6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Il6lower molars of Archaeohippus ultimus Cope 25 from the Mascallformation of Oregon. The tooth is longer and narrower than lowercheek teeth in the type of Archaeohippus penultimus Matthew 26from the Sheep Creek horizon but appreciably smaller than inreferred specimens. The Maryland tooth is also, presumably, lower-crowned.The only Archaeohippus species known from the eastern regionis A. nanus Simpson 27 from the Thomas Farm fauna of the Haw-thorn formation in Florida. The Maryland tooth is very close insize to M 2 in A. nanus. However, there is no assurance that the formrepresented is this species, as the specimen is much too incompleteto warrant specific diagnosis.Measurements (in millimeters) of M2, US.N.M. No. 18431Anteroposterior diameter 11.3Transverse diameter 7-6MERYCHIPPUS, sp.Two left lower premolars of the horse Merychippus were foundon the beach between the Chesapeake Beach resort and Randle Cliff.One of these, an unworn tooth, U.S.N.M. No. 12032 (fig. 6b), wasdiscovered by John J. O'Brien, of the Soldiers' Home, and the other,U.S.N.M. No. 18430 (fig. 6c), worn for probably half of its originallength, was collected by Charles Welter, of New York City. Betweenthe two beach resorts only Calvert of the Miocene sequence isrepresented in the cliff section, as shown by the columnar sections inthe Maryland Geological Survey's report on the Miocene.Two phalanges, collected by W. E. Salter, of the U. S. GeologicalSurvey, from the vicinity of Scientists Cliffs, a short distance northof Governor Run, may also represent Merychippus. Scientists Cliffsincludes the lower part of the Choptank formation as well as the upperpart of the Calvert, so that although a Miocene age is assumed forthe toe bones they are not certainly Calvert.Unfortunately, no upper teeth have been found in the Calvert, andlower teeth are less satisfactory for comparative purposes. Never-theless, the developmental stage reached in the lower teeth affordsstrong evidence toward conclusions as to the stage of the Miocenerepresented.
25 See Bode, F. D., ibid., pp. 51-52, pi. 3, figs. 1-3.26 Matthew, W. D., Bull. Amer. Mus. Nat. Hist., vol. 50, pp. 158-159, 1924.27 Simpson, G. G., Florida State Geol. Surv. Bull. 10, pp. 28-31, 1932.
NO. 2 MIOCENE LAND MAMMALS—GAZIN AND COLLINS 1
7
The two lower teeth have very nearly the same cross section andmust surely represent the same species. Together they demonstratethe unworn or total length of the tooth and the enamel pattern atabout one-half the tooth length. Their construction does not representa very early stage of Merychippus, but demonstrates pronouncedhypsodonty and well-developed enamel folding. Both teeth showcement-filled folds on the lingual side and discontinuous or a scatteringof cement deposition on the outer walls and in the median fold. Theteeth are of small size, but distinctly more advanced than Merychippusprimus from the Sheep Creek horizon. They more nearly approachin pattern and proportions those of Merychippus californicus Mer-riam 28 from the Merychippus zone of the Temblor. They are dis-tinctly less advanced than material of this genus from the Barstow.It would seem that the stage of evolution reached in the Calvertteeth is approximately that of Merychippus californicus, but with asimpler pattern than those illustrated. They may not be far removedin time from the Merychippus horizons represented in the Mascalland Virgin Valley beds of Oregon and Nevada but appear advancedover the stage indicated by Merychippus gunteri and Merychippuswestoni from the Hawthorn beds of Florida.
Measurements (in millimeters) of Merychippus sp. cheek teethU.S.N.M. No. 12032 U.S.N.M. No. 18430Antero- Antero-posterior Transverse posterior Transverse21.0 II.O* 19.0 12.0
* About midsection of tooth length.Order ARTIODACTYLAFamily TAYASSUIDAECf. CYNORCA PROTERVA (Cope)The presence of a distinctly small peccary in the Maryland Miocenehas been a matter of record since 1868, but since that time very fewreferences have been made to the occurrence. In the preceding yearCope 29 described what he regarded as a new type of squalodont onthe basis of material found by James T. Thomas near his residencein the eastern part of Charles County, Md., not far from the Patuxent28 Bode, F. D., Carnegie Inst. Washington Publ. 453, pp. 39-63. figs. 1-6,pis. 1-2, 1934.29 Cope, E. D., Proc. Acad. Nat. Sci. Philadelphia, vol. 19, pp. 151-152, 1867.
l8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Il6River—not the Ashley River, S. C, as Palmer 30 indicated. Cope'sdescription was based on characters of an upper canine of a peccary,but included in the material mentioned were two squalodont cheekteeth. In 1868 Cope 31 announced that Leidy had called his attentionto the fact that the canine belonged to a small peccary, whereupon heattempted to retain this name for the other teeth. The correctionwas not noticed by Case, 32 but Kellogg 33 later demonstrated that sincethe distinctive characteristics of Cynorca proterva were based on apeccary canine, obviously the type, the name was not available foruse as a squalodont, and in this he was supported by Hay.34Leidy,35 on the other hand, followed Cope in retaining the namefor a squalodont and referred the peccary canine to his species,Dicotyles lenis, the type of which is from the Ashley River deposits,and also included additional peccary material from Charles County,Md., which Cope 36 regarded as Pleistocene and referred to themodern species under the name Dicotyles torqnatus (Pecari angulatus).We have not examined these specimens so are unable to verify theirage assignment. The Chesapeake Bay collections of the NationalMuseum include peccary material from both the Miocene and Pleisto-cene, the latter apparently being referable to Tayassu lenis (Leidy).The peccary material at hand from various localities in CalvertCounty, along the cliffs adjacent to Chesapeake Bay, indicates anoticeably small form which is distinct from species of Miocene pec-caries known from elsewhere. Since Cope described Cynorca protervaas coming from the Miocene of the same general area, and Leidycharacterized it as distinctly smaller than modern forms, it seemsprobable, although no other canines are at hand, that our materialbelongs to this species.Description.—Among the materials here referred to Cynorca pro-terva are some six isolated upper and lower cheek teeth and a lower jawportion, U.S.N.M. No. 18429 (fig. 7), including M 2 and M 3 . More-over, some incomplete limb and foot bones may belong to the samespecies. The lower jaw was found on the beach 0.7 mile south of30 Palmer, T. S., Index generum mammalium, p. 213, 1904.21 Cope, E. D., ibid., vol. 20, pp. 185-186, 1868.32 Case, E. C, Maryland Geological Survey, Miocene, pp. 7-8, 1904.33 Kellogg, R., Proc. U. S. Nat. Mus., vol. 62, art. 16, pp. 28-29, 1923.34 Hay, O. P., Second bibliography and catalogue of the fossil vertebrates ofNorth America, vol. 2, p. 770, 1930.35 Leidy, Joseph, Journ. Acad. Nat. Sci. Philadelphia, ser. 2, vol. 7, pp. 384-385, 1869.36 Cope, E. D., ibid., vol. 19, p. 155, 1867.
NO. 2 MIOCENE LAND MAMMALS—GAZIN AND COLLINS 19Randle Cliff Beach, where the Maryland Geological Survey indicatesonly Calvert.Two premolars, possibly P 2 and P 3 , but not of the same individual,are simple and smooth with the elevated crest of the trigonid and thelower talonid divided but free of accessory cuspules. The paraconidis weak on the smaller of the two teeth, but prominent, though lower,on the other. These premolars, if properly referred to the samespecies as the molars, are relatively large.
Fig 7—Cf. Cynorca proterva (Cope): Right ramus of mandible with Maand Ms (U.S.N.M. No. 18429), lateral and occlusal views, X I. Calvert Miocene,Maryland.Two last upper molars are strikingly smaller and relatively nar-rower than in the living collared peccary, the cusps are more nearlyconical, the cingulum is much less developed, and there are feweraccessory cuspules.The lower jaw, No. 18429, is relatively shallow and exhibits rathersmooth, narrow molars, particularly M 3 , with cusps, as in M 3 , morenearly conical than in modern peccaries. It should be noted, however,in comparison with modern forms that the large white-lipped peccary,Tayassu pecari, exhibits teeth having less tendency toward the for-mation of transverse crests, lower trigonid portions in the premolars,and more nearly conical cusps in the molars than the smaller Pecari
20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Il6
angulatus. The absence of cingular cusps between the external cuspsof the lower molars, noted by Leidy for T. lenis, is also characteristicof T. pecari, but not P. angulatus.The dental characteristics noted for the form Cynorca are muchmore prophetic of the modern peccaries than are characters of thebetter-known Pliocene species of Prosthennops. Although it is pos-sible that Prosthennops is a synonym of Cynorca it seems morelikely that the two are distinct and that Cynorca is actually in theline of descent for Tayassu and Pecari. As noted by Colbert,37Prosthennops does not fall in this line and Platygonus has become toospecialized in its dental characters. On the other hand, "Desmathyus"and the closely related form, Floridachoerus, are known only fromlarge species, having teeth so modified that it seems unlikely theyin turn could have given rise to Cynorca.
Measurements in millimeters of lower jaw(Cf. Cynorca proterva, US.N.M. No. 18429)Depth of jaw beneath point between M2 and M., 24.0Anteroposterior diameter : transverse diameter, M2 1.21 : 0.95Anteroposterior diameter : transverse diameter, Ms 1.47 : 0.86HESPERHYS (DESMATHYUS) ? sp.A peccary of considerable size is represented in the collections by ahumerus, the distal portion of a radius, several carpals and phalanges,and some vertebrae and ribs, all belonging to one individual. Thesewere found by Dr. W. F. Foshag, of the U. S. National Museum, in1940 in fallen blocks from zone 17 of the Choptank Miocene immedi-ately above the Calvert, at Calvert Beach, Md. Three isolated meta-podials from along the Chesapeake are of a size comparable to theforegoing. Two of these are not documented as to horizon, but thethird was collected by Collins from zone 19 of the Choptank.There is no certainty as to the genus represented, but it is thoughtthat the form might possibly be "Desmathyus/' as a peccary fromthe base of the Miocene Kirkwood formation in New Jersey, whichMarsh 38 described as Perchoerus (Dicotyles) antiquus, has sincebeen referred by Wood 39 to "Desmathyus." Moreover, "Desmathyus"
37 Colbert, E. H., Nebraska State Mus. Bull., vol. 1, No. 44, p. 421, 1935.38 Marsh, O. C, Amer. Journ. Sci., vol. 46, p. 411, 1893.30 Wood, H. E., 2d, Bull. Geol. Soc. Amer., vol. 50, pp. 1968-1969, 1939(abstract).
NO. 2 MIOCENE LAND MAMMALS—GAZIN AND COLLINS 21has been described from the Merychippus primus zone of the SnakeCreek beds as well as from the Upper Rosebud and Harrison.The humerus and other elements of this peccary are as large andmassive as in the Pleistocene form, Platygonus cumberlandensis. Nolimb material of true Hcsperhys was available for direct comparisonand these seem undescribed in the literature.