sign in sign up
<<
Home , Erinaceidae, Hyopsodontidae, Insectivora, Kennalestes, Leptictis, Macrocranion

University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln

USGS Staff -- Published Research US Geological Survey

1985

On the Classification of the Early erT tiary Erinaceomorpha (, Mammalia)

Michael J. Novacek American Museum of Natural History

Thomas M. Bown U.S. Geological Survey

David Schankler Princeton University

Follow this and additional works at: https://digitalcommons.unl.edu/usgsstaffpub

Part of the Earth Sciences Commons

Novacek, Michael J.; Bown, Thomas M.; and Schankler, David, "On the Classification of the Early erT tiary Erinaceomorpha (Insectivora, Mammalia)" (1985). USGS Staff -- Published Research. 197. https://digitalcommons.unl.edu/usgsstaffpub/197

This Article is brought to you for free and open access by the US Geological Survey at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in USGS Staff -- Published Research by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. AMERICAN MUSEUM Novitates PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY CENTRAL PARK WEST AT 79TH STREET, NEW YORK, N.Y. 10024 Number 2813, pp. 1-22, figs. 1-1 1, table 1 April 10, 1985

On the Classification of the Early Tertiary Erinaceomorpha (Insectivora, Mammalia)

MICHAEL J. NOVACEK,I THOMAS M. BOWN,2 AND DAVID SCHANKLER3

ABSTRACT Definitions are provided for three Early Tertiary odont cheek teeth that superficially resemble pri- families of Erinaceomorpha. The family Dorma- mate . A third family, the , aliidae includes Dormaalius, , Sce- includes the Early Tertiary genera Litolestes, Leip- nopagus, Ankylodon, Crypholestes, Sespedectes, sanolestes, Entomolestes, Neomatronella, Eolestes, and Proterixoides. Several ofthese genera are tra- Adapisorex, Cedrocherus, and living and fossil ditionally known collectively as "Adapisorici- members of the Galericinae, Brachyericinae, and dae," but Adapisorex is probably an erinaceid and Erinaceinae. Another member of the Erinaceidae the name Adapisorncidae is therefore unavailable is Dartonius, proposed here as a new designation for the above listed taxa. Dormaaliids are char- for "Leptacodon"jepseni. Several erinaceomorphs acterized by a reduction in size and complexity of are either too generalized in structure, too diver- the anterior premolars, a reduced, premolariform gent, or too poorly represented to allow assign- P4, and several other dental specializations. A sec- ment to any of the above families. These incertae ond family, the Amphilemuridae includes Am- sedis taxa are Diacodon, Adunator (including philemur, Gesneropithex, Alsaticopithecus, and Mckennatherium), Diacocherus, Litocherus, Tal- Pholidocercus. Amphilemurids have inflated, bun- pavus, and Talpavoides. INTRODUCTION The erinaceomorph are a crit- sification of Erinaceomorpha has, however, ical group for understanding eutherian phy- been subject to diverse interpretations. Greg- logeny. Erinaceomorphs have been variously ory (1910, p. 464) named the Section Eri- cited as ancestors or close relatives of tu- naceomorpha as a group including Erinacei- paiids, , , dermopterans, and dae, Leptictidae, and Dimylidae (table 1). several other major eutherian taxa. The clas- Simpson (1945) retained these families and I Chairman and Assistant Curator, Department ofVertebrate Paleontology, American Museum ofNatural History. 2 Geologist, U.S. Geological Survey, Denver, Colorado 80225. 3Research Assistant, Department of Geological and Geophysical Sciences, Princeton University, Princeton, New Jersey 08544.

Copyright X American Museum of Natural History 1985 ISSN 0003-0082 / Price $2.90 2 AMERICAN MUSEUM NOVITATES NO. 2813

TABLE 1 Classifications of Erinaceomorph Insectivorans Gregory, 1910 Subfamily Creotarsinae (incl. Litolestes, Insectivora Xenacodon, Talpavus, Creotarsus, Dor- Suborder maalius, Entomolestes, Scenopagus, Section Zalambdodonta (incl. Families Centeti- Macrocranion, Amphilemur, Sespe- dae, , Solenodontidae, dectes, Proterixoides, Amphidozotheri- Necrolestidae, Chrysochloridae) um, Ictopidium, Tupaiodon) Section unnamed (incl. Family ) Subfamily Nyctitheriinae Section Erinaceomorpha (incl. Families Lepticti- Family Erinaceidae dae, Erinaceidae, Dimylidae) Subfamily Galericinae Section (incl. Families Soricidae, Subfamily Erinaceinae ) Family Talpidae Suborder Unnamed (incl. Family Hyopsodontidae) Superfamily Soricoidea Simpson, 1945 Russell, Louis, and Savage, 1975 Order Insectivora Suborder Erinaceomorpha Superfamily Deltatheridioidea Family Adapisoricidae Superfamily Tenrecoidea Subfamily Adapisoricinae (incl. Adapisorex) Superfamily Chrysochloroidea Subfamily Dormaaliinae (incl. Litolestes, Superfamily Macroscelidoidea Leipsanolestes, Dormaalius, Entomo- Superfamily Soricoidea lestes, Neomatronella, "Leptacodon" Superfamily Pantolestoidea jepseni, Macrocranion, Scenopagus, Superfamily Mixodectoidea ?Proterixoides, ?Sespedectes, Talpavus, Superfamily Erinaceoidea Ankylodon) Family Krishtalka, 1976a Family Dimylidae Family Adapisoricidae (incl. Mckennatheri- Family Leptictidae (incl. , Pro- um, Scenopagus, Ankylodon, Macro- diacodon, Acmeodon, Emperodon, Myr- cranion, Talpavus) mecoboides, Adapisorex, Diacodon, Family Erinaceidae (incl. Litolestes, Leipsano- Parictops, Protictops, Ictops, , lestes, ?Entomolestes [grangeri], and the ?Xenacodon, ?Sespedectes) subfamilies Galericinae and Erinacei- Family Erinaceidae nae) Subfamily Echinosoricinae (incl. Entomo- THIS PAPER: lestes, Proterioxoides, Metacodon, An- Suborder Erinaceomorpha kylodon, Proterix, Brachyerix, Mete- Family Dormaaliidae (incl. Dormaalius, Mac- chinus, Meterix, Neurogymnurus, rocranion, Scenopagus, Ankylodon, Lantanotherium, , Pseudogale- Crypholestes, Sespedectes, Proterix- rix, Echinosorex, Hyolomys, Podogym- oides) nura, Neotetracus) Family Amphilemuridae (incl. Amphilemur, Subfamily Erinaceinae (incl. , Pa- Gesneropithex, Alsaticopithecus, Pholi- laeoerinaceus, Tupaiodon, Palaeoscap- docercus) tor, Parvericius, Aethechinus, , , Family Erinaceidae (incl. Litolestes, Leipsano- , ) lestes, Adapisorex, Entomolestes, Neo- Van Valen, 1967 matronella, Eolestes, Cedrocherus, Dar- Suborder Erinaceota tonius, Proterix, and subfamilies Superfamily Erinaceoidea Galericinae, Erinaceinae, and Brachyer- Family Adapisoricidae icinae) Subfamily Geolabidinae Erinaceomorpha, (incl. Diaco- Subfamily Adapisoricinae (incl. Mcken- don, Adunator [incl. Mckennatherium], natherium, Leptacodon, Adunator, Diacocherus, Litocherus, Talpavus, Tal- Adapisorex, , Messelina) pavoides) added Zalambdalestidae in his Superfamily lambdalestids, and macroscelidids, in addi- Erinaceoidea. Saban (1954), frequently and tion to the families recognized by Gregory. incorrectly cited as the author of the Erina- Several authors in more recent years (Butler, ceomorpha, expanded this category with his 1956a, 1972; McDowell, 1958; McKenna, inclusion ofthe pantolestids, apheliscids, za- 1960, 1975; Russell, 1964; Van Valen, 1967; 1985 NOVACEK, BOWN, SCHANKLER: ERINACEOMORPHA 3

Novacek, 1973, 1976, 1977; Russell, Louis, UCMP, University of California Museum of Pa- and Savage, 1975; Krishtalka, 1976) have ar- leontology, Berkeley gued that Erinaceomorpha seems realistically USGS, United States Geological Survey limited to the Erinaceidae and other early YPM, Yale Peabody Museum taxa generally recognized as "adapisoricids." Several matters remain unresolved. Sige (1977) regards nyctitheriids as members of SYSTEMATICS the Erinaceomorpha contra opinions of McKenna (1975), Krishtalka (1976a, 1976b), MAMMALIA LINNAEUS, 1758 Novacek (1976), Bown and Schankler (1982), ORDER INSECTIVORA CUVIER, 1817 and others. There is also argument as to whether dimylids (see Schmidt-Kittler, 1973) SUBORDER ERINACEOMORPHA GREGORY, 1910 belong to the Erinaceomorpha (following Van DIAGNOSIS: Insectivorans with the follow- Valen, 1967; Butler, 1972; Novacek, 1976) ing combination of dental features that dis- or the Soricomorpha (following Schmidt- tinguish them from other early eutherian Kittler, 1973; McKenna, 1975). "Amphile- . P1 small, single-rooted and P2-3 dom- murids" (Heller, 1935) were transferred from inated by a single cusp. P4 with a short, ba- Erinaceomorpha to Primates by Russell, sined or unbasined talonid; P4 talonid cusps Louis, and Savage (1975) but were returned absent, or, if present, diminutive. M1 3 para- to Erinaceomorpha by Koenigswald and conids compressed, lophid-like, or cresti- Storch (1983). Gingerich (1983) has united form. Molar trigonid cusps lower, less sec- several problematic "adapisoricids" as mem- torial, and more anteriorly canted than in bers ofthe new subfamily Litocherinae. Here , , Procerberus, Asior- we address problems of relationships among yctes, palaeoryctids (sensu stricto), primitive the selected members of the Erinaceomor- leptictids (e.g., Prodiacodon), geolabidids, pha; namely, the Early Tertiary taxa com- early miacids, and creodonts. Talonids on monly recognized as erinaceids, adapisori- MK2 nearly as wide or wider than trigonids. cids, and amphilemurids. Our purpose is to M1 3 with high entoconids and low hypocon- provide a consensus statement on the higher ids often flattened in early stages of wear. P3 level of the Erinaceomorpha that small, triangular in occlusal outline (second- reflects, with some modification, our inde- arily enlarged in some taxa). P4 hypocone pendent studies of this problem (e.g., Bown usually present, metacone weak or absent, and Schankler, 1982; Novacek, 1982). metastylar crest strong. MI-2 semirectangular with narrow stylar shelves, distinct hypo- ACKNOWLEDGMENTS cones, and posterolingual cingula. INCLUDED FAMILIES: Dormaaliidae Quinet, We thank Drs. Kenneth Rose, David 1964; Erinaceidae Fischer de Waldheim, Krause, Malcolm C. McKenna, and Charles 1817; ?Dimylidae Schlosser, 1887; Amphi- Repenning for their comments and review of lemuridae Heller, 1935. the manuscript. This paper was greatly im- DISCUSSION: The above characterization of proved by the efforts of the staff in the De- the Erinaceomorpha is discussed at length by partment of Paleontology Novacek (1982). Polarity assessments oftooth (AMNH). Mr. Otto Simonis skillfully pre- characters in early insectivorans are difficult, pared and repaired specimens and made owing to the often subtle differences observed opaque epoxy casts of the originals. The ex- among these taxa. Only the combination of cellent photographs of these casts and origi- the above cited features serves to distinguish nals were taken by Mr. Chester Tarka. Sup- erinaceomorphs; no single feature is by itself port for this study was provided by the Frick diagnostic. Readers familiar with the prob- Laboratory Endowment Fund (AMNH, No- lem will be aware of other early eutherians vacek) and USGS (Bown). that share at least some of these traits (see also Novacek, 1982). Nevertheless, the di- ABBREVIATIONS agnosis is useful if it is assumed that high- AMNH, American Museum of Natural History cusped, sectorial molars and premolars seen HTM, Institut Royal des Sciences Naturelles de in early leptictids, palaeoryctids, and a va- Belgique riety of eutherians represent a 4 AMERICAN MUSEUM NOVITATES NO. 2813

FIG. 1. Macrocranion nitens, partial cranium andjaws (USGS 3676). A, lateral view ofleft premaxilla, maxilla, I2-3, C, Pl-4, M'-3 (PI is a reversed photograph of PI from right maxilla); B, lateral view of left ramus with root ofII, I2, I3, C, P1, P2, P4 (P3 damaged), M13. Note small, procumbent, and single-rooted PI-2. Specimen from Willwood Formation (lower ), Bighorn Basin, Wyoming. closer approximation of the primitive euth- reduced and procumbent. Paraconid on erian morphotype. M1_3 transversely oriented and crestiform; trigonids anteroposteriorly compressed in FAMILY DORMAALIIDAE QUINET, 1964 occlusal view. M1_3 trigonids less erect, more Figures 1-3 canted, than in early erinaceids. (* indicates DIAGNOSIS (as modified from Russell, characters for teeth not known in all dor- Louis, and Savage, 1975; Krishtalka, 1976a, maaliids.) 1977; Novacek, 1982): Shares with erina- INCLUDED TAXA: ceids the following derived erinaceomorph characters: P4 metacone lost or greatly re- Dormaalius Quinet, 1964. early Eocene, Eu- duced.* Hypocone on P4 present.* Talonid rope. on M1 as wide or wider than trigonids. Small Dormaalius vandebroeki Quinet, 1964. hypoconulid on M1l2. Lower trigonid relief Macrocranion Weitzel, 1949. early-middle on M1I3. M3 reduced. Differs from primitive Eocene, Europe and . erinaceids in having the following derived Macrocranion tupaiodon Weitzel, 1949. features: P4 with notably short talonid (P4 Macrocranion tenerum (Tobien, 1962), talonid somewhat more elongate in Sce- Russell, Louis, and Savage, 1975. nopagus hewettensis). P2-3 reduced, P2 single- Macrocranion nitens (Matthew, 1918) rooted. Anterior lower premolars generally Krishtalka, 1976a. 1985 NOVACEK, BOWN, SCHANKLER: ERINACEOMORPHA 5

B

I.)v

A4Z4 .1

J_

f

IE I _ ..-- 'k jI~t I ,c " '. 7. 4 #t

FIG. 2. Macrocranion nitens, partial cranium and jaws (USGS 3676). Stereophotographs ofA, ventral view of palate and upper (right and left IP, right 12-3, and left PI missing); B, occlusal view of left ramus with I,12; C, P1 , M1.3. Note lack of metacone on P4, well-developed hypocones on M'-2, short heel on P4, and crestiform paraconid on M1 3 (see diagnosis for Dormaaliidae).

Macrocranion robinsoni Krishtalka and middle Eocene-early , North Setoguchi, 1977. America. Macrocranion sp. (in Russell, Louis, and Ankylodon annectens Patterson and Savage, 1975). McGrew, 1937. "?Entomolestes cf. nitens" (in Russell, Ankylodon progressus Galbreath, 1953. Louis, and Savage, 1975). Crypholestes (Novacek, 1976), Novacek, Scenopagus McKenna and Simpson, 1959. 1980. middle Eocene, North America. early-middle Eocene, North America. Crypholestes vaughni (Novacek, 1976), Scenopagus mcgrewi McKenna and Simp- Novacek, 1980. son, 1959. Sespedectes Stock, 1935. middle Eocene, Scenopagus edenensis (McGrew, 1959), North America. Robinson in McKenna, Robinson, and Sespedectes singularis Stock, 1935. Taylor, 1962. Proterixoides Stock, 1935. middle Eocene, Scenopagus priscus (Marsh, 1872), Krish- North America. talka, 1976a. Proterixoides davisi Stock, 1935. Scenopagus hewettensis Bown and Schank- "Erinaceid-like genus and " (UCMP ler, 1982. 101420, see Novacek, 1976), middle Ankylodon Patterson and McGrew, 1937. Eocene, North America. 6 AMERICAN MUSEUM NOVITATES NO. 2813

FIG. 3. Macrocranion sp., left maxillary fragment with Ml-3 (alveoli for P3-4) (AMNH 46897). Lateral view of A, original; B, epoxy cast; C, occlusal view of cast. Specimen from Quarry 88, Arroyo Blanco, San Jose Formation (lower Eocene), San Juan Basin, New Mexico. See also Maier (1979).

DIscussION: The majority oftaxa included soricids" (within which he included "Mcken- within the Dormaaliidae are perhaps more natherium" [see remarks below], Scenopagus, familiar to specialists as adapisoricids. How- Macrocranion, Dormaalius, Talpavus and, ever, we believe that the Adapisoricidae is tentatively, Sespedectes and Proterixoides) an inappropriate designation for any erina- that differentiated this group from erinaceids. ceomorph higher taxon. As noted below, These were: (1) lophid-like transverse para- Adapisorex seems excluded from a group conids on M1-3; (2) M1 with relatively broader comprising most of the forms usually called talonid than trigonid and M2 with relatively adapisoricids; thus the name Adapisoricidae narrower talonids than trigonids; and (3) high is unavailable for this group. Genera cur- entoconids and low, flat hypoconids. rently recognized as "adapisoricids" can be Use of these characters for the purpose of linked with more modem erinaceids by sev- defining "adapisoricids" is, however, prob- eral derived characteristics (see Russell, Louis, lematic. There is notable variation within and Savage, 1975; Krishtalka, 1976a, 1977; "adapisoricids" sensu Krishtalka (1977) in Novacek, 1976, 1982). These features are also the structure of the paraconid (Bown and commonly used to recognize "adapisori- Schankler, 1982; Novacek, 1982). Moreover, cids," but this family cannot be distinguished Entomolestes grangeri-recognized as an er- as a monophyletic group by application of inaceid by Krishtalka (1976a)-has a lin- the same features suggesting monophyly for gually positioned, crestiform paraconid sim- the Erinaceomorpha. Krishtalka (1977) has ilar to that in the putative "adapisoricids" attempted to remedy this situation by sug- "Mckennatherium" ladae and Macrocranion gesting three derived dental traits of "adapi- nitens. 1985 NOVACEK, BOWN, SCHANKLER: ERINACEOMORPHA 7

Variation is also seen in other allegedly some ofthe more primitive erinaceomorphs diagnostic features of "adapisoricids." from the group, including "Mckennatheri- "Mckennatherium" ladae differs from the um" ladae (and its probable close relatives typical adapisoricid condition in having a Adunator and "Diacodon" minutus, see Bown narrower talonid than trigonid on M1 (M1 and Schankler, 1982, and discussion below) trigonids and talonids are ofsubequal widths and thus recognizing a group ofmore limited in Talpavus nitidus). Erinaceids supposedly diversity. These authors raised the rank of differ from "adapisoricids" in having a rel- the Dormaaliinae (ofRussell, Louis, and Sav- atively wider M2 talonid, but this condition age, l-9-7-5) for such a purpose, but this action, is also known in the "adapisoricid" Macro- without re-diagnosis, is not satisfactory. The cranion nitens (see table 4 in Krishtalka, published definition for dormaaliines (see 1976a). Finally, the "adapisoricid" entoco- Russell, Louis, and Savage, 1975, p. 134) does nid/hypoconid relationship suggested as di- not exclude more primitive erinaceomorphs agnostic by Krishtalka (1 976a, 1977) also oc- nor does it aptly describe all sensu stricto dor- curs in the putative erinaceid Entomolestes maaliids. For example, the diagnosis given grangeri. by Russell, Louis, and Savage (1975) states Even if one could arrive at an internally that the talonids are generally wider than the consistent definition of the Adapisoricidae, trigonids in the lower molars ofdormaaliids, the name of the group is an unfortunate but this does not account for the variation choice. The type genus Adapisorex, repre- in lower molar proportions in erinaceids and sented by A. gaudryi Lemoine 1883, is mark- other erinaceomorphs. edly different from other "adapisoricids" (see To complicate matters, Dormaalius is a Clemens, 1973; Russell, Louis, and Savage, problematic type genus for a higher taxon. 1975). Krishtalka (1976a, 1977) suggested Quinet (1964) erected this name for two thatAdapisorexis an erinaceid. Russell, Louis, species, D. vandebroeki and D. simonsi, but and Savage (1975) separated Adapisorex as did not provide differential diagnosis ofeither the sole member of the Adapisoricinae and species, and did not explicitly establish a type applied Dormaaliinae (originally established species for the genus or type specimen for by Quinet, 1964, as a monotypic family) as either species. All these actions violate suf- the subfamily-group name for all other ficiently the International Code ofZoological "adapisoricids." Bown and Schankler (1982) Nomenclature (Stoll et al., eds., revised 1964; argued that Adapisorex is an "erinaceid-like" articles 13, 67, 72) to constitute grounds for , similar to Phenacodaptes or suppression ofthe generic and specific names. Apheliscus (Gazin, 1959). There is a striking Finally, as Bown and Schankler (1982) ob- similarity, however, between Adapisorex and serve, Dormaalius is poorly represented and erinaceids and reference to that family seems the meager available material suggests very justified (see comments below). close similarity to Macrocranion. Similar problems involve Creotarsus. This Some of these problems have been re- genus serves as the type of the Creotarsinae, solved by recent work. Russell, Louis, and a subfamily recognized by Van Valen (1967) Savage (1975) clarify most ofthe ambiguities to be "adapisoricid" and to include many of of Quinet's (1964) original designation of the genera considered here. Creotarsus, rep- Dormaalius. The type specimen of D. van- resented solely by the type specimen, has some debroeki is validly designated as HTM (In- rather strong hyopsodontid-like characters in stitut Royal des Sciences Naturelles de Bel- the lower premolars and molars, and we do gique) 66, a lower jaw with P4 and alveoli for not recognize the Creotarsinae as a valid er- other teeth. Dormaalius vandebroeki was inaceomorph group. originally referred by Teilhard de Chardin IfAdapisorex and Creotarsus do not belong (1927) to his Omomys belgicus. The type in the group generally termed Adapisorici- mandible (HTM 66) was subsequently fig- dae, a group ofearly non-erinaceid erinaceo- ured by Quinet (1964) and given the name morphs requires diagnosis and an appropri- D. vandebroeki. Russell, Louis, and Savage ate name. Bown and Schankler (1982) argued (1975) argue convincingly that D. simonsi is that clearer definition is attained by excluding a junior synonym of D. vandebroeki. Dor- 8 AMERICAN MUSEUM NOVITATES NO. 2813 maalius is clearly distinct from the type idae. Ictopidium is represented by a single species of Macrocranion (M. tupaiodon) in species, I. lechei Zdansky (1930) from the having a relatively larger P4 compared to M1. Eocene ofChina. Zdansky misinterpreted the For these reasons, Novacek (1982) en- structure of P4 in this species as "molari- dorsed Bown and Schankler's (1982) recog- form," and he accordingly assigned Ictopi- nition of the family Dormaaliidae. This ac- dium to the Leptictidae. The genus was trans- tion does not solve all the problems with ferred to the Erinaceidae by Butler (1956b), Dormaalius. As Krishtalka (1977) noted, and subsequently to the Creotarsinae by Van Dormaalius is very similar to Macrocranion Valen (1967). Ictopidium shows the reduced nitens (figs. 1-3) and several other species P4 talonid characteristic of dormaaliids. referred to this genus (see above listing). However, the genus is poorly represented by Moreover, Dormaalius is so poorly repre- its holotype, a partial lowerjaw fragment with sented that its generic distinction from Mac- P3-4, M2 and the trigonid of M3. Reference rocranion (as currently defined) is open to to the Erinaceidae as suggested by Butler question. Dormaalius differs, however, from (1956b) seems doubtful because the alveolar M. nitens in being significantly smaller and space occupied by M1 does not suggest that in having a relatively anteroposteriorly short- this tooth was any larger than M2. Hence, the er, less labially expanded P4 talonid; a small, progressive reduction in size of M1 through shallow basin lingual to the crest connecting M3 that is diagnostic of erinaceids is not ap- the talonid cusp to the posterior wall of the parent in Ictopidium. P3, P4 morphology of trigonid of P4; a less transverse, labially ex- Ictopidium is also unlike the typical condi- panded M2 talonid with a weak hypoconulid tion in other erinaceomorphs. These teeth represented by a slight rise on the crest join- have high, piercing trigonid cusps more rem- ing the hypoconid and entoconid (M2 hy- iniscent ofpalaeoryctoids. Moreover, the hy- poconulid is distinct and somewhat "swol- poconulid of M2 is very small and placed len" in Macrocranion nitens). These dif- lingually and adjacent to the entoconid, a ferences are slight but adequate for separating condition that contrasts strongly with that in Dormaalius from the member ofMacrocran- erinaceomorphs. Ictopidium is Insectivora ion it most resembles. Moreover, as is stated incertae sedis with possible affinities to pa- in article 40 ofthe International Code, 1964, laeoryctids or soricomorphs. the recognition of Dormaalius as a junior Ankylodon represents perhaps the most synonym of Macrocranion would not justify controversial allocation within the Dorma- the rejection of the family-group name Dor- aliidae. Fox (1983) claimed recently that, maaliidae. In the interest of consistency, we contrary to widespread opinion (Butler, 1972; retain the Dormaaliidae to designate the group Krishtalka, 1976a; Lillegraven, McKenna, including Dormaalius, Macrocranion, and and Krishtalka, 1981; Novacek, 1982), this several other genera listed above. genus is a soricomorph rather than an eri- Macrocranion was allocated to the Am- naceomorph. His evidence (ibid.) for this philemuridae of Heller (1935; see also claim includes the presence in Ankylodon of McKenna, 1960) by Tobien (1962), although (1) an enlarged I1; (2) a raised anterior and Van Valen (1967) recognized the genus as a ventral rim of the orbit; (3) anteriorly elon- member of the Creotarsinae. It is now evi- gated palatines; (4) the opening of the lacri- dent, based on excellent new skeletal mate- mal duct within the orbit; (5) a reduced (in- rial, that amphilemurids are referable to the complete?) zygomatic arch; and (6) inferred Erinaceomorpha, but separable from erina- origin of snout muscles on the maxillary root ceids and Macrocranion and other "adapi- of the zygomatic arch. Of these, only char- soricid" or "dormaaliid" genera considered acter 1 seems to support Fox's argument. here (Koenigswald and Storch, 1983; and Character 2 (cf. a diversity of , in- comments below). cluding tupaiids, macroscelidids, dermopter- Because the matter is not discussed else- ans, and some erinaceids-but note lack in where, we briefly mention our reasons for most tenrecids) and character 3 (cf. leptictids, excluding Ictopidium from the Dormaali- macroscelidids) are widely distributed and 1985 NOVACEK, BOWN, SCHANKLER: ERINACEOMORPHA 9 possibly primitive eutherian traits. Character from MI to M3. Upper molars with meso- 4 may be derived for but it is also style, very narrow stylar shelves, low meta- present in leptictids, dermopterans, macro- crista, well-developed paraconules and meta- scelidids, and, notably, in some galericine er- conules. M'-2 somewhat quadrate in occlusal inaceids. Character 5 is ambiguous because outline with strong hypocones. Lower molars the material Fox described is damaged in the with transverse lophid-like paraconid, and zygomatic region. Moreover, the zygomatic showing only slight difference in height be- is slender in certain erinaceids (cf. Hyolomys, tween trigonid and talonid. Hypoconid as high Neotetracus). It is the marked reduction--or or higher-than entoconid. Talonid wider than loss ofthe jugal elements, not simply the nar- trigonid. M1_2 semirectangular in outline (for row zygomatic arch, that is a significant char- expanded diagnosis that includes general er- acteristic of soricomorphs. Character 6 is inaceomorph characters, see Koenigswald and subject to considerable variation in insectiv- Storch, 1983, p. 451). orans (as noted by Fox, 1983), and the ar- INCLUDED TAXA: rangement of the snout muscles and their at- tachments in the fossil Ankylodon are open Amphilemur Heller, 1935. middle Eocene, to various interpretations. This problemati- Europe. cal evidence hardly challenges the association Amphilemur eocaenicus Heller, 1935. ofAnkylodon with erinaceomorphs based on Gesneropithex Hiirzeler, 1946. late Eocene, shared-derived features ofthe dentition (But- Europe. ler, 1972; Krishtalka, 1976a; Novacek, 1982; Gesneropithex peyeri Huirzeler, 1946. and McKenna and Lillegraven, in prep.). We Alsaticopithecus Huirzeler, 1947. middle thus retain Ankylodon within Erinaceomor- Eocene, Europe. pha and, more specifically, we ally this genus Alsaticopithecus leemanni Hiirzeler, 1947. with Scenopagus and an unnamed dorma- Pholidocercus Koenigswald and Storch, 1983. aliid from the middle Eocene of San Diego middle Eocene, Europe. (UCMP 101420, see Novacek, 1976, 1982). Pholidocercus hassiacus Koenigswald and No attempt is made here to provide a clas- Storch, 1983. sification of higher resolution for the Dor- maaliidae. Novacek (1982) has suggested a DISCUSSION: Amphilemurids are clearly possible cladistic pattern of relationships for more allied with erinaceomorphs (McKenna, some of the taxa listed here. Any formal- 1960; Koenigswald and Storch, 1983) than ization ofthis scheme first requires more de- with primates (Russell, Louis, and Savage, tailed published comparisons of Sespedectes 1975). This group, however, shows traits that and Proterixoides with "amphilemurids" and suggest a number ofalternative relationships other erinaceomorphs. The tortuous history within Erinaceomorpha. The quadrate out- of allocations for dormaaliids and related line of the molars, the well-developed hy- forms is summarized in table 1. pocones, and the slight progressive reduction in size from Ml13 suggest affinity with eri- naceids. The bunodont crown patterns ofthe FAMILY AMPHILEMURIDAE HELLER, 1935 molars are reminiscent offeatures in the dor- DIAGNOSIS (from Koenigswald and Storch, maaliids Sespedectes, Proterixoides, and Cry- 1983): 3 1 4 3and fully functional milk teeth. pholestes. The reduction and procumbent Relatively little differentiation between an- form of P1 3 strongly resemble conditions in temolar teeth. P1I3 single-rooted, crowded, Macrocranion. Detailed dental comparisons and progressively procumbent anteriorly. with some, but not all, of the relevant taxa Marked size difference between P4 and P3. C, led Koenigswald and Storch (1983, p. 477) small, premolariform, or incisiform. 11-3 rel- to conclude that amphilemurids were not atively small, spatulate. P4 premolariform closely related to the groups they recognized with unicuspid talonid and distinct crista ob- as Adapisoricidae (essentially equivalent to liqua and talonid basin. Molars bunodont the Dormaaliidae of this paper) and Erina- with low, inflated crowns. Reduction in size ceidae. The above comparisons suggest, 10 AMERICAN MUSEUM NOVITATES NO. 2813

FIG. 4. Galerix socialis, right ramus with P3, P4, M1_3 and alveoli for I3, C, P1 2 (AMNH 10516D). A, occlusal; B, lateral views of epoxy cast; C, lateral view of original. Note projecting prevallid shearing surface on M1, reduction in size from M1 to M3 (see diagnosis for Erinaceidae). Specimen from La Grive (), St. Albans, Isere, France. nonetheless, the potential for documentation V-shaped valley by flat internal walls of hy- of a close relationship between amphilemu- poconid and entoconid. Ml paraconid salient rids and some other erinaceomorph subgroup. and anteriorly projecting, elongating preval- Pending such a study, the Amphilemuridae lid shearing wall. Hypoconulids on M12 are recognized here as a separate family of markedly reduced and positioned at or just erinaceomorph insectivorans. lingual to midline of the crown on posterior wall. M1-2 (where known) are semirectangular FAMILY ERINACEIDAE or quadrate in outline. Hypocones are better (FISCHER DE WALDHEIM), 1817 developed on Ml-2 than in dormaaliids. M3 Figures 4-10 is markedly reduced and usually oval in out- DIAGNOSIS (after Krishtalka, 1 976a, 1977; line. In most erinaceids, hypocones are con- Butler, 1948): Erinaceids share the following nected via a crest to postprotocrLsta. defining features: Lower molars show pro- INCLUDED TAXA: The following species are gressive reduction in size from M1 to M3. recognized as Erinaceidae but are excluded Lower molars semirectangular in occlusal from the subfamilies Galericinae, Brachyer- view, with some degree of exodaenodonty icinae, Erinaceinae, and Protericinae: (i.e., bases oflabial trigonid and talonid cusps LitolestesJepsen, 1930. late , North are swollen). Talonid basin formed as a America. 1985 NOVACEK, BOWN, SCHANKLER: ERINACEOMORPHA I1I

3 mm

s

I A '. k,. AF, A

I - C

FIG. 5. Galerix socialis, left ramus with Pi (damaged), P24, M1_2, and alveoli for I113, C, M3 (AMNH 10499). A, occlusal; B, lateral views of epoxy cast; C, lateral view of original. For locality information see figure 4.

Litolestes ignotus Jepsen, 1930 (but ex- Dartonius jepseni (McKenna, 1960), new cluding L. notissimus and L. lacunatus, species. seem comments in Krishtalka, 1976a, pp. Adapisorex Lemoine, 1883. middle-late Pa- 29-30; Gingerich, 1983). leocene, Europe. Leipsanolestes Simpson, 1928. late Paleo- Adapisorex gaudryi Lemoine, 1883. cene, early Eocene, North America. Adapisorex abundans Russell, 1964. Leipsanolestes siegfriedti Simpson, 1928. Cedrocherus Gingerich, 1983. late Paleocene, EntomolestesMatthew, 1909. middle Eocene, North America. North America. Cedrocherus ryani Gingerich, 1983. Entomolestes grangeri Matthew, 1909. Neomatronella Russell, Louis, and Savage, The following species are provisionally re- 1975. early Eocene, Europe. (Described ferred to the Galericinae: as Matronella in Russell, Louis, and Erinaceid sp. from type specimen of Tepee Savage, 1975; see note, p. 177, op. cit.) Trail Formation (AMNH 88288), middle Neomatronella luciannae Russell, Louis, Eocene, North America (in McKenna and and Savage, 1975. Krishtalka, in prep.; Krishtalka, 1976a). Eolestes (Bown, 1979), Bown and Schankler, Erinaceid sp. in Krishtalka and Setoguchi, 1982. early Eocene, North America. middle North America. Eolestes simpsoni (Bown, 1979), Bown and 1977. Eocene, Schankler, 1982. For taxa referred to the erinaceid subfam- Dartonius, new genus. early Eocene, North ilies Brachyericinae, Protericinae, Galerici- America. nae, and Erinaceinae, see Butler (1948), Van 12 AMERICAN MUSEUM NOVITATES NO. 2813

A

3mm

t - - i I I 1.ik, I Altd;l , ,. I& 'IT;. 1. p / v ", INIII".)U .1 T...- -

FIG. 6. Galerix socialis, maxillary fragment with P3-4, M1-2, and alveoli for P2, P3, M3 (AMNH 10499). A, lateral view of original; B, lateral; C, occlusal views of epoxy cast. For locality information see figure 4.

Valen (1967), Rich and Rich (1971), and Rich by the dental variation in these early taxa (see (1981). Novacek, 1982). Moreover, assignment of DISCUSSION: The above diagnosis for the these species is difficult because the most Erinaceidae is limited to dental characters, as primitive erinaceid subfamily, the Galerici- pre-Oligocene erinaceids are virtually un- nae, is poorly defined, as this group is rec- known from other parts of the skeleton. It ognized primarily by the lack of erinaceine should be noted, however, that known skulls dental specializations. A reasonable ap- of both fossil and Recent erinaceids show a proach to the problem is to look once again suite of distinctive characteristics (Butler, at the evidence for monophyly of the Gal- 1948; Rich, 1981). ericinae represented by living taxa and more There are four recognized erinaceid completely preserved fossils. subfamilies-the Galericinae (figs. 4-6) Krishtalka (1976) clarified considerably the (equals the Echinosoricinae ofearlier papers), position of certain Early Tertiary erinaceo- the Erinaceinae, the Protericinae, and the morphs that shared derived dental features Brachyericinae-but we do not assign any of with established erinaceids. Litolestes (ig- the named Paleocene or Eocene erinaceids to notus) (fig. 7), Leipsanolestes (figs. 8, 9), and these groups. Contrary to statements by Entomolestes (fig. 10) all show the progres- Krishtalka (1976a, 1977), we do not see a sive reduction of M1-3, and the more quad- clear galericine-erinaceine split exemplified rate dimensions of the lower molars noted in 1985 NOVACEK, BOWN, SCHANKLER: ERINACEOMORPHA 13

3 mm

FIG. 7. Litolestes ignotus, left ramus with P2-4, Ml3 and root of P1 (PU 19387). A, occlusal; B, lateral views of epoxy cast; C, lateral view of original. Note progressive reduction in size from Ml to M3. Specimen from Schaff Quarry southwest, Upper Polecat Bench Formation (upper Paleocene), Park County, Wyoming. the above diagnosis. Leipsanolestes differs with one or more cusps and a shallow basin. from more specialized erinaceids and from P4 in the type ofAdapisorex abundans shows dormaaliids in having an elongate P4 with a strong molariform development, but the tooth well-developed talonid (fig. 9), a primitive may be a deciduous premolar (see Russell, erinaceomorph trait (Novacek, 1982). The 1964 for description). shortening of the talonid of P4 was thus de- Cedrocherus, a monotypic genus from the veloped independently in erinaceids and dor- late Paleocene of northwestern Wyoming, maaliids. was described by Gingerich (1983) as a mem- As noted above, Adapisorex is a likely ber of his proposed adapisoricid subfamily member of the Erinaceidae (fide Krishtalka, Litocherinae. Here we regard Cedrocherus as 1976a, p. 7; Koenigswald and Storch, 1983). closely related to, and possibly synonymous This genus has the quadrate upper molars with, Litolestes. Although Gingerich (1983) with well-developed hypocones, the marked included Litolestes and Leipsanolestes along reduction in size ofM3 relative to M2 and the with Cedrocherus in the Litocherinae, these reduced oval-shaped M3 characteristic of er- taxa in their erinaceid specializations differ inaceids. Other features indicate a more con- distinctly from other "litocherines" (see servative morphology. P4 in Adapisorex has comments below). Cedrocherus was distin- a small, but distinct, metacone, the P4 hy- guished by the shape of the entoconid on M1 pocone is either weak or absent, and, as in and M2 and the very marked gradient in de- Leipsanolestes, P4 talonid is well developed, creasing size from MK3 (Gingerich, ibid., p. 14 AMERICAN MUSEUM NOVITATES NO. 2813

3mm 3mm C

FIG. 8. Leipsanolestes siegfriedti, right man- FIG. 9. Leipsanolestes siegfriedti, right man- dible fragment with M2-3 and roots of P4, M1 dible fragment with P4 and alveoli for P2, P3 (AMNH 22157, Type). A, occlusal; B, lateral views (AMNH 22231). Posterior fragment oframus not of epoxy cast; C, lateral view of original. Note shown. A, occlusal; B, lateral views of epoxy cast; reduced M3. Specimen from Bear Creek, Fort C, lateral view of original. Note distinct talonid Union Formation (Clarkforkian, upper Paleo- basin and talonid cusps. For locality information cene-lower Eocene), Carbon County, Montana. see figure 8.

237). It is debatable whether these characters view. Entomolestes also has a two-rooted P2 justify recognition ofa new genus, rather than (Krishtalka and West, 1977, fig. 2), whereas simply a new species of Litolestes. Cedroch- known P2s of dormaaliids are single-rooted. erus is presently only represented by a single This condition in Entomolestes is simply species (C. ryani) and a single jaw fragment primitive; some early Galericines (e.g., Li- with M1.3. Additional material is required to tolestes, Galerix, Neurogymnura) have two- substantiate its generic status. rooted P2s (fig. 5) but in most erinaceids this Of these early Tertiary erinaceids the as- tooth is single-rooted. Loss of roots and size signment of Entomolestes seems the most reduction of anterior premolars seems sub- problematic. This genus shows a mosaic of ject to much convergence among early Ter- erinaceid and dormaaliid traits (fig. 10). Like tiary insectivorans. the latter, Entomolestes has a P4 with a very In light of these comparisons, we retain short talonid, strong crestiform paraconid on Entomolestes within erinaceids. We also con- M1I3, reduced, somewhat procumbent, P2, P3 cur with Krishtalka's (1976a, 1977) recog- (described as P3, P4 by Krishtalka and West, and 1977), and a hypoconid that is much lower nition of Leipsanolestes, Litolestes, Neo- than the entoconid in M1_3. Like erinaceids, matronella as early erinaceids. To this list we Entomolestes shows a progressive reduction add Eolestes simpsoni, described by Bown in dimension of M1 to M3 (although this re- and Schankler (1982), and an early Eocene duction is not so marked in M2 as in other form described as "Leptacodon" jepseni by erinaceids), and the labial margins ofthe low- McKenna (1960). A new designation and er molar cusps are expanded, giving the comparative diagnosis for "L." jepseni fol- crowns a more quadrate outline in occlusal low. 1985 NOVACEK, BOWN, SCHANKLER: ERINACEOMORPHA 15

r-tl.I O'I f.I -, It -9, .4 *l, 3 mm

C

FIG. 10. Entomolestes grangeri, left ramus with P3, P4, M13 and alveoli for C, P1, P2 (AMNH 11485, Type). A, occlusal; B, lateral views of epoxy cast; C, lateral view of original. Specimen from Grizzly Buttes West, Bridger B, (middle Eocene), Bridger Basin, Wyoming.

DARTONIUS, NEW GENUS Dartonius jepseni (McKenna, 1960) TYPE AND ONLY SPECIES: Dartoniusjepseni Leptacodon jepseni McKenna, 1960, p. 51. (McKenna, 1960, p. 51). DIAGNOSIS: Erinaceid genus that differs MATERIAL: Holotype UCMP 45949, frag- from Entomolestes grangeri in smaller size, ment of right ramus with P3-M2 (formerly in having a smaller P3/P4 ratio, in having a preserving P2; McKenna, 1960, fig. 24). shorter, more curved prevallid, in having less UCMP 47023, 47155 (McKenna, 1960, p. exodaenodont labial margins of the molars, 51); YPM 30559 (Bown and Schankler, 1982, and in having smaller and less linguolabially fig. 24E); possibly AMNH 56313 (Delson, compressed molar entocristids. From the ap- 1971, p. 328). proximately coeval erinaceid Eolestes simp- OCCURRENCE: early Wasatchian (early soni, Dartonius differs in having a smaller P4/ Eocene) , northwest Col- M1 ratio, in having a better developed P4 orado and northeast Wyoming, and Will- talonid basin, in having a shorter, less atten- wood Formation (upper Haplomylus-Ecto- uated entocristid, in lacking a connection of cion Range Zone of Schankler, 1980, the cristid obliqua with the metaconid, and northwest Wyoming). in having a more acute paracristid notch on DIAGNOSIS: Same as for genus. the molars. Differs from Talpavus and dor- DISCUSSION: In 1960, McKenna named maaliids in features that define erinaceids (see Leptacodon jepseni for three specimens of a comments below). small from Alheit Pocket in the ETYMOLOGY: For Nelson Horatio Darton, Four Mile Creek area ofnorthwest Colorado. pioneer western North American geologist. McKenna rightly believed Leptacodon to be 16 AMERICAN MUSEUM NOVITATES NO. 2813 an artificial genus composed of structurally confused (e.g., Talpavus and Scenopagus; very diverse species, and he considered "L." Russell, Louis, and Savage, 1975; and Lep- jepseni and "Leptacodon minutus" (Jepsen's tacodon, McKenna, 1960; Delson, 1971) by 1930 Diacodon minutus) to be primitive En- those characters which serve to distinguish tomolestes-like erinaceoids, perhaps related the Erinaceidae from the Dormaaliidae and to the ancestry ofGeolabididae (equals Meta- (Soricomorpha), respectively. codontidae ofearlier authors, see Lillegraven, Dartonius in no ambiguous way resembles McKenna, and Krishtalka, 1981). nyctitheriids (Bown and Schankler, 1982), but Van Valen (1967) placed Leptocodon (in- could still potentially be confused with Sce- cluding "L." jepseni) in the Adapisoricinae, nopagus and Talpavus because all are eri- though his construct of that group differs naceomorphs. greatly from that ofnearly every other author. From Scenopagus, Dartonius differs in all Delson (1971) referred four new specimens its erinaceid features, these including pro- from the early Eocene Powder River local gressive molar size reduction from MI through fauna to "Leptacodon" jepseni. Delson also M3, greater molar exodaenodonty, the flat in- questioned the generic status of Leptacodon ternal walls of the hypoconulid and entoco- and observed that McKenna's "L." jepseni nid, the anteriorly projecting, salient para- is morphologically more similar to the conid of MI, and the less medial molar hy- Bridgerian erinaceid Entomolestes grangeri poconulids. In addition, Dartonius shows an Matthew (1909), a view that was advocated anteriorly projecting P4 paraconid (also dis- by Bown and Schankler (1982) and is en- tinctive in its close relative Entomolestes and dorsed here. erinaceines) that forms a characteristic para- Russell, Louis, and Savage (1975) placed conid notch in labial view. The internal wall "L." jepseni in the Adapisoricidae, but be- ofthe P4 paraconid is developed as a flat shelf. lieved that it probably constitutes a new ge- The talonid basin of P4 is relatively broader nus closest to Talpavus or Scenopagus. Those than in Scenopagus, and the entocristid notch authors discounted the idea ofDelson (1971) is partly filled, forming a sharp mure in labial that "L." jepseni shares any close relation- and lingual views. The molar paraconids and ship with Entomolestes. Krishtalka (1976a, metaconids in Dartonius are exactly opposite 1977) believed that "Leptacodon" jepseni was each other linguolabially, whereas in Scen- too poorly known to confidently demonstrate opagus the protoconid is positioned some- its generic status, but agreed with Delson what more posteriorly. Dartonius differs from (1971) that the probably does not be- Talpavus in its diagnostic erinaceid features, long in Leptacodon. Krishtalka compared but also in the development ofthe anteriorly "L." jepseni with early nyctitheriids and con- projecting P4 paraconid, larger molar para- cluded that its affinities probably are closer conid, broader P4 talonid basin, and rela- to the Erinaceomorpha than to the Sorico- tively smaller P3 with respect to P4 size. All morpha. Talpavus and Scenopagus species were much Bown and Schankler (1982) recognized a larger than Dartonius jepseni. form close to McKenna's "Leptacodon" jep- We believe that Dartonius is closest in seni in collections offossil lipotyphlan insec- morphology (and possibly ancestral, accord- tivorans from the lower Eocene Willwood ing to T.M.B. and D.S.) to the younger En- Formation ofthe Bighorn Basin ofWyoming tomolestes grangeri. (Yale Peabody Museum 30559). These au- thors removed three ofthe four Powder River Basin specimens from "L." jepseni, but en- ERINACEOMORPHA, INCERTAE SEDIS dorsed Delson's (1971) suggestion that the taxon is very close to the Bridgerian (middle DIscuSSION: Several Paleocene and Eocene Eocene) erinaceid Entomolestes grangeri. species have been recognized as erinaceo- Dartonius (as "Leptacodon" jepseni) has morphs, but their primitive morphology, di- thus been the source of some confusion since vergent features, or poor representation pre- it was first described. Clearly, it differs from clude their assignment to the Dormaaliidae the forms with which it has most often been or the Erinaceidae as defined above. The taxa 1985 NOVACEK, BOWN, SCHANKLER: ERINACEOMORPHA 17 here regarded as Erinaceomorpha incertae this genus. Diacodon has served as a taxon sedis are Diacodon (alticuspis) Cope (1875), for as many as twelve species, but Novacek Talpavus Marsh (1872), Talpavoides Bown (1982) concluded that only the type species, and Schankler (1982), Litocherus Gingerich D. alticuspis, is referable to this genus. Al- (1983), Adunator Russell (1964), Diacoch- though Diacodon was traditionally regarded erus Gingerich (1983) [includes "Diacodon" as a leptictid, this relationship is contradicted minutus of Jepsen (1930)] and Mckennathe- by features of the posterior premolars and rium Van Valen (1965). molars that suggest affinities with erinaceo- The interrelationships of Adunator, morphs. Nonetheless, this genus shows a more Mckennatherium, and "Diacodon" minutus primitive overall morphology than other er- and the higher level affinities of these taxa inaceomorphs (with the possible exception of are matters of current debate. Krishtalka Adunator) and it is excluded from both the (1976a) noted that Adunator and "Diacodon" Dormaaliidae and the Erinaceidae. minutus were probably congeneric but dis- Gingerich (1983), following Krishtalka tinguished Mckennatherium from the former (1976a), justifiably separated Litolestes no- two species on the basis of subtle differences tissimus (fig. 11) and L. lacunatus from the in molar structure. Bown and Schankler type species of that genus, L. ignotus. He es- (1982) argued that these differences are very tablished a new genus, Litocherus, to include slight or virtually unrecognizable and that they L. notissimus, L. lacunatus and a new species, do not justify generic separation of the taxa. L. zygeus. Litocherus was designated by Gin- Following their conclusions, Mckennathe- gerich as the type genus of the subfamily Li- rium would be recognized as a junior syn- tocherinae, to which he referred Leipsano- onym of Adunator. This synonymy is of lestes, Litolestes, Mckennatherium, and the broader interest because Adunator has been new genera Cedrocherus and Diacocherus. We suggested as showing a strong resemblance to endorse Gingerich's (1983) recognition ofLi- Haplaletes and other early hyopsodontid tocherus, although we note that his diagnoses (Krishtalka, 1976a; Bown and distinguishing species of this genus were es- Schankler, 1982). From such considerations, sentially limited to comparisons of size. Not it is clear that only slight departures from the mentioned were the more robust proportions typical erinaceomorph condition can lend a of the protoconid and elongation of the tal- "hyopsodontid appearance" to the lower onid in P4, and the reduced molar paraconids dentition. A pertinent problem is the lack of of L. zygeus that separate this species from an explicit diagnosis of the Hyopsodontidae L. notissimus. that accounts for both dentally primitive and Litocherus is another taxon potentially dentally derived members ofthis group. Un- confused with hyopsodontid condylarths. til such a study is available, Adunator and a Nevertheless, Gingerich (1983, p. 235) cor- variety of other species will remain in the rectly noted that Litocherus is separable from limbo between primitive insectivorans and Haplaletes and other primitive hyopsodon- primitive condylarths. Here, we do not ex- tids by its sharper-cusped cheek teeth, the clude the possibility that Adunator is a very sweeping postmetacrista (lacking in upper primitive hyopsodontid, but Novacek rec- molars of Haplaletes) and the lack ofdistinct ognizes its erinaceomorph affinities to be labial cingula on the lower molars. Further- equally likely. Krishtalka (1976a) has pointed more, at least unworn lower molars of Li- out that although Adunator lehmanni shows tocherus have a crestiform paraconid rather a swelling of the molar metaconids reminis- than the lophid, or shelflike paraconid char- cent of hyopsodontids, this genus does not acteristic of hyopsodontids. approach the low, bulbous, and almost buno- Despite the validity ofLitocherus, its status dont condition of the molar cusps in Hap- as the type genus for the Litocherinae sensu laletes and other hyopsodontids. Gingerich (1983) is highly questionable. The The relationships of Diacodon have been subfamily was defined primarily on the lack considered at length by Novacek (1982), who of the specializations seen in "adapisori- described for the first time much more com- cines" (=Adapisorex) and "dormaaliines"- plete material than was originally known for namely the presence of less reduced anterior 18 AMERICAN MUSEUM NOVITATES NO. 2813

AL j

C

FIG. 11. Litocherus notissimus, right ramus with P2 (damaged), P3_4, MI-3 and alveoli for C, P1 (AMNH 33938). A, occlusal; B, lateral views of epoxy cast; C, lateral view of original. Specimen from Scarrit Quarry, Fort Union Formation (upper Paleocene), Crazy Mountain Field, Montana.

premolars, more transverse upper molars, and contrast strongly with the more general con- smaller hypocones on upper molars. Some ditions in Mckennatherium, Litocherus, and putative diagnostic characters do, as Ginger- Diacocherus (see also Gingerich, 1983, table ich maintained, distinguish "litocherines" 3). Moreover, Diacocherus (including "Dia- from "adapisoricines" but these are shared codon" minutus) and Mckennatherium with typical dormaaliids (e.g., metacone on strongly differ from the more bunodont mo- P4 small or absent). Other "diagnostic" char- lar condition and swollen P4 protoconids of acters allow great latitude in structural vari- Litocherus. We conclude that three "lito- ation (e.g., P4 talonid small, flat, or shallowly cherines" (Leipsanolestes, Litolestes, and basined, lower molars with trigonids ofmod- Cedrocherus) are better recognized as mem- erate size, and basined talonids ofmoderately bers of the Erinaceidae and that a special as- large size). Some traits do not really charac- sociation between Litocherus and Diacoch- terize all litocherines in a way that collec- erus is unwarranted. Litocherus is a distinct tively distinguish them from certain other er- taxon but its relationship with other erina- inaceomorphs (e.g., "reduced" paraconids on ceomorphs is uncertain. lower molars). Gingerich's (1983, p. 238) recognition of Thus, the definition of Litocherinae does Diacocherus illustrates the problem ofuntan- not successfully account for its included taxa gling the "Diacodon" minutus-Mckennathe- nor does it effectively eliminate more plau- rium-Adunator complex. Subtle differences sible alternative relationships. Leipsano- in molar proportions distinguish Diacocherus lestes, Cedrocherus and, particularly, Lito- minutus from Mckennatherium (Krishtalka, lestes show erinaceid specializations that 1976a; Gingerich, 1983) but no characters 1985 NOVACEK, BOWN, SCHANKLER: ERINACEOMORPHA 19 were specified to distinguish Diacocherus mi- Africa. Fossil Mammals of Africa, vol. nutus and Gingerich's new species Diacoch- 11, pp. 1-75. erus meizon from Adunator lehmanni. Per- 1972. The problem of insectivore classifica- haps Adunator best serves as the senior tion. In Joysey, K. A., and T. S. Kemp (eds.), Studies in vertebrate evolution, synonym of Mckennatherium and Diacoch- pp. 253-265. New York, Winchester erus. Regardless of the nomenclatural solu- Press. tion, these species undoubtedly represent a Clemens, W. A. close grouping ofvery small erinaceomorphs 1973. Fossil mammals ofthe type Lance For- with sectorial, primitive dentitions. mation, Wyoming. Part III. Eutheria and Talpavus and Talpavoides show a basic summary. Univ. Calif. Publ. Geol. Sci., similarity with dormaaliids. Yet these taxa- vol. 94, pp. 1-102. at least based on the meager available ma- Cope, E. D. terial-lack the important specializations that 1875. Systematic catalogue of the Vertebrata define the Dormaaliidae. In this regard they of the Eocene of New Mexico. Jour. Acad. Nat. Sci., Philadelphia, 1875, pp. are somewhat "intermediate" in dental struc- 1-36. ture between the Adunator-Diacocherus Cuvier, G. complex and dormaaliids like Scenopagus. 1817. La Regne Animal. Paris, Deterville, vol. Talpavus and Talpavoides, for example, have 1, xxxvii + 504 pp. an elongate P4 with a well-developed talonid Delson, E. that is primitively uncharacteristic of dor- 1971. Fossil mammals of the early Wasatch- maaliids. Talpavus also lacks the typical er- ian Powder River local fauna, Eocene inaceid specializations seen in progressive size ofnortheast Wyoming. Bull. Amer. Mus. decrease M1 to M3 and the salient paraconids Nat. Hist., vol. 146, pp. 305-364. on M1.2. Although we have earlier included Fischer de Waldheim, G. F. Talpavus and Talpavoides with the Dorma- 1817. Adversia Zoologica. Part 2. De syste- mata mammalium. Mem. Soc. Impe- aliidae, it seems more realistic to regard these riale Naturalistes, Mosquae, vol. 5, pp. taxa as incertae sedis, until better material 368-391. may show otherwise. Fox, R. C. 1983. New evidence on the relationships of the Tertiary insectivoran Ankylodon LITERATURE CITED (Mammalia). Canadian Jour. Earth Sci., vol. 20, no. 6, pp. 968-977. Bown, T. M. Galbreath, E. C. 1979. Geology and mammalian paleontology 1953. A contribution to the Tertiary geology of the San Creek Facies, Lower Will- and paleontology of northeastern Col- wood Formation (lower Eocene) Wash- orado. Univ. Kansas Paleo. Contr., vol. akie County, Wyoming. Wyoming Geol. 4,pp. 1-119. Surv. Mem., vol. 2, pp. 1-151. Gazin, C. L. Bown, T. M., and D. Schankler 1959. Early Tertiary Apheliscus and Phena- 1982. A review ofthe Proteutheria and Insec- codaptes as pantolestid insectivores. tivora ofthe Willwood Formation (low- Smithsonian Miscel. Coll., no. 139, pp. er Eocene) Bighorn Basin, Wyoming. 1-7. U.S. Geol. Surv. Bull., no. 1523, pp. 1- Gingerich, P. D. 79. 1983. New Adapisoricidae, Pentacodontidae, Butler, P. M. and Hyopsodontidae (Mammalia, In- 1948. On the evolution of the skull and teeth sectivora and Condylarthra) from the in the Erinaceidae, with special refer- late Paleocene of Wyoming and Colo- ence to the material in the British Mu- rado. Contrib. Mus. Paleo., Univ. seum. Proc. Zool. Soc., London, vol. Michigan, vol. 26, no. 11, pp. 227-255. 118, pp. 446-500. Gregory, W. K. 1956a. The skull ofIctops and the classification 1910. The orders of mammals. Bull. Amer. of the Insectivora. Proc. Zool. Soc., Mus. Nat. Hist., vol. 27, pp. 1-524. London, vol. 126, pp. 453-481. Heller, F. 1956b. Erinaceidae from the Miocene of East 1935. Amphilemur eocaenicus n. g. et n. sp., 20 AMERICAN MUSEUM NOVITATES NO. 2813

ein primitiver aus dem Mit- Bull. Soc. Geol. France, 3rd ser., vol. teleozain des Geiseltales bei Halle a. S. 11, pp. 249-271. Nova Acta Leopoldina, N.F., vol.2, no. Lillegraven, J. A., M. C. McKenna, and L. Krish- 3, pp. 293-300. talka Htirzeler, J. 1981. Evolutionary relationships of middle 1946. Gesneropithex peyeri nov. gen. nov. Eocene and younger species of Cente- spec., ein neuer Primate aus dem Lu- todon (Mammalia, Insectivora, Geola- dien von G6sgen (Solothurn). Ecl. Geol. bididae) with a description of the den- Helv., vol. 39, pp. 354-361. tition of Ankylodon (Adapisoricidae). 1947. Alsaticopithecus leemanni nov. gen. nov. Univ. Wyoming Publ., vol. 45, pp. 1- spec., ein neuer Primate aus dem un- 115. teren Lutetien von Buchsweiler im Un- Linnaeus, C. terelsass. Ecl. Geol. Helv., vol. 40, pp. 1758. Systema Naturae per Regna Tri Natu- 343-356. rae. Editio decima, reformata. Stock- International Code of Zoological Nomenclature holm, Laurentii Salvii, 824 pp. 1964. Adopted by the XV International Con- McDowell, S. B. gress of Zoology (Stoll et al., eds.). In- 1958. TheGreaterAntilleaninsectivores.Bull. ternational Trust for Zoological No- Amer. Mus. Nat. Hist., vol. 115, pp. menclature, London. 113-124. Jepsen, G. L. McGrew, P. 1930. Stratigraphy and paleontology of the 1959. Geology and paleontology of the Elk Paleocene ofnortheastern Park County, Mountain and Tabernacle Butte area, Wyoming. Proc. Amer. Phil. Soc., vol. Wyoming. Bull. Amer. Mus. Nat. Hist., 69, pp. 463-528. vol. 117, pp. 121-176. Koenigswald, W. von, and G. Storch McKenna, M. C. 1983. Pholidocercus hassiacus, ein Amphile- 1960. Fossil Mammalia ofthe early Wasatch- muride aus dem Eozan der "Grube ian Four Mile fauna, Eocene of north- Messel" bei Darmstadt (Mammalia, Li- west Colorado. Univ. Calif. Publ. Geol. potyphla). Senckenbergiana Lethaea, Sci., vol. 37, pp. 1-130. vol. 64, pp. 447-495. 1975. Toward a phylogenetic classification of Krishtalka, L. the Mammalia. In Luckett, W. P., and 1976a. Early Tertiary Adapisoricidae and Eri- F. S. Szalay (eds.), Phylogeny ofthe pri- naceidae (Mammalia: Insectivora) of mates: a multidisciplinary approach, pp. North America. Bull. Carnegie Mus. 21-46. New York, London, Plenum Nat. Hist., vol. 1, pp. 1-40. Press. 1976b. North American Nyctitheriidae (Mam- McKenna, M. C., P. Robinson, and D. W. Taylor malia: Insectivora). Ann. Carnegie Mus. 1962. Notes on Eocene Mammalia and Mol- Nat. Hist., vol. 46, pp. 7-28. lusca from Tabernacle Butte, Wyoming. 1977. Early Eocene Euramerican Insectivora. Amer. Mus. Novitates, no. 2102, pp. 1- GeoBios, Mem. Spec., vol. 1, pp. 135- 33. 139. McKenna, M. C., and G. G. Simpson Krishtalka, L., and T. Setoguchi 1959. A new insectivore from the middle 1977. Paleontology and geology of the Bad- Eocene of Tabernacle Butte, Wyoming. water Creek area, central Wyoming. Part Amer. Mus. Novitates, no. 1952, pp. 1- 13. The late Eocene Insectivora and 12. Dermoptera. Ann. Carnegie Mus. Nat. Maier, W. Hist., vol. 46, pp. 71-99. 1979. Macrocranion tupaiodon, an adapisori- Krishtalka, L., and R. M. West cid (?) insectivore from the Eocene of 1977. The paleontology and geology of the the "Grube Messel" (Western Ger- Bridger Formation southern Green Riv- many). Palaont. Zeitschr., vol. 53, pp. er Basin, southwestern Wyoming. Part 38-62. 2. The Bridgerian insectivore Entomo- Marsh, 0. C. lestes grangeri. Contr. Biol. Geol., Mil- 1872. Preliminary description ofnew Tertiary waukee Publ. Mus., vol. 14, pp. 1-11. mammals. Amer. Jour. Sci., vol. 4, pp. Lemoine, V. 122-128, 202-224. 1883. Etude sur le Neoplagiaulax de la faune Matthew, W. D. eocene inferieure des environs de Reims. 1909. The Carnivora and Insectivora of the 1985 NOVACEK, BOWN, SCHANKLER: ERINACEOMORPHA 21

Bridger Basin. Mem. Amer. Mus. Nat. considerees affines. Bull. Mus. Natl. Hist., vol. 9, pp. 289-567. Hist. Nat., vol. 327, pp. 129-193. 1918. A revision ofthe lower Eocene Wasatch Saban, R. and Wind River Faunas. In Article XVI 1954. Phylogenie des insectivores. Bull. Mus. (by W. D. Matthew and W. Granger); Natl. Hist. Nat., ser. 2, vol. 26, pp. 419- Part V. Insectivora (cont'd.), Glires, Ed- 432. entata. Bull. Amer. Mus. Nat. Hist., vol. Schankler, D. M. 38, pp. 565-657. 1980. Faunal zonation of the Willwood For- Novacek, M. J. mation in the central Bighorn Basin, 1973. Insectivora of the late Eocene (Uintan) Wyoming. In Gingerich, P. D. (ed.), of San Diego County, California. Mas- Early Cenozoic paleontology and stra- ter's Thesis, San Diego State Univ., 301 tigraphy of the Bighorn Basin, Wyo- PP. ming, pp. 99-118. Univ. Michigan, Pa- 1976. Insectivora and Proteutheria ofthe later pers in Paleo., no. 24. Eocene (Uintan) of San Diego County, Schlosser, M. California. Contr. Sci., L.A. County Nat. 1887. Die Affen, Lemuren, Chiropteren, In- Hist. Mus., no. 283, pp. 1-52. sectivora, Marsupialier, Creodonten und 1977. A review ofPaleocene and Eocene Lep- Carnivoren des europaischen Tertiars. tictidae (Eutheria: Mammalia) from Teil 1. Beitrage zur Palaontologie Oes- North America. PaleoBios, vol. 24, pp. terreich-Ungarns. 1-42. Schmidt-Kittler, N. 1980. Crypholestes, a new name for the Early 1973. Dimyloides-Neufunde aus der ober- Tertiary insectivore (Mammalia) Cryp- Oligozanen Spaltenfullung "Ehrenstein tolestes Novacek 1976. Jour. Paleo., vol. 4" (Suddeutschland) und die systema- 54,p. 1135. tische Stellung der Dimyliden (Insectiv- 1982. Diacodon alticuspis, an erinaceomorph ora, Mammalia). Mitt. Bayer. Staatsam. insectivore from the early Eocene of Palaont. u. hist. Geol., vol. 13, pp. 115- northern New Mexico. Univ. Wyo- 139. ming, Contrib. Geol. 20, vol.2, pp. 135- Sige, B. 149. 1977. Insectivores primitifs de l'Eocene su- Patterson, B., and P. 0. McGrew perieur et Oligocene inferieur d'Europe 1937. A soricid and two erinaceids from the occidentale. Nyctitheriides. Mem. Mus. White River Oligocene. Field Mus. Nat. Natl. Hist. Nat., ser. C, vol. 4, pp. 1- Hist., Geol. Ser., vol. 6, pp. 245-272. 140. Quinet, G. E. Simpson, G. G. 1964. Morphologie dentaire des mammiferes 1928. A new mammalian fauna from the Fort eocene de Dormaal. Bull. Groupement Union of southern Montana. Amer. Internatl. Rech. Sci. Stomatologie, vol. Mus. Novitates, no. 297, pp. 1-15. 7, pp. 272-294. 1935. New Paleocene mammals from the Fort Rich, T. H. Union of Montana. Proc. U.S. Natl. 1981. Origin and history of the Erinaceinae Mus., vol. 83, pp. 221-244. and Brachyericinae (Mammalia, Insec- 1945. The principles of classification and a tivora) in North America. Bull. Amer. classification of the mammals. Bull. Mus. Nat. Hist., vol. 171, pp. 1-116. Amer. Mus. Nat. Hist., vol. 85, pp. 1- Rich, T. H., and P. V. Rich 350. 1971. Brachyerix, a Miocene from Stock, C. western North America, with a descrip- 1935. Insectivora from the Sespe Uppermost tion of the tympanic regions of Parae- Eocene, California. Proc. Natl. Acad. chinus and . Amer. Mus. Novitates, no. 2477, pp. 1-58. Sci., vol. 20, pp. 349-354. Russell, D. E. Teilhard de Chardin, P. 1964. Les mammiferes Paleocene d'Europe. 1927. Les mammiferes de l'Eocene inferieur Mem. Mus. Natl. Hist. Nat., ser. C, vol. de la Belgique. Mem. Mus. Hist. Nat. 13, pp. 1-324. Belgique, vol. 36, pp. 1-33. Russell, D. E., P. Louis, and D. E. Savage Tobien, H. 1975. Les Adapisoricidae de l'Eocene infer- 1962. Insectivoran (Mamm.) aus dem Mittel- ieur de France: reevaluation des formes eozan (Lutetium) von Messel bei 22 AMERICAN MUSEUM NOVITATES NO. 2813

Darmstadt. Notizbl. Hessischen Lan- ora, Testudinata) aus dem Mitteleozan desamt Bodenforsch., vol. 90, pp. 7-47. von Messel bei Darmstadt. Abhandl. Van Valen, L. Senckenberg Naturforsch. Gesellsch., 1965. A middle Paleocene primate. Nature, no. 480, pp. 1-24. vol. 207, no. 4995, pp. 435-436. Zdansky, 0. 1967. New Paleocene insectivores and insec- 1930. Die alttertiaren Saugetiere Chinas nebst tivore classification. Bull. Amer. Mus. stratigraphischen Bemerkungen. Pa- Nat. Hist., vol. 132, pp. 1-126. laeont. Sinica, ser. C, vol. 6, no. 2, pp. Weitzel, K. 1-87. 1949. Neue Wirbeltiere (Rodentia, Insectiv-