The presence of Taeniodonta (Mammalia) in the Early Eocene ofEurope

Carmen ESTRAVIS* Donald E. RUSSELL**

* Centro de Estratigrafia e Palcobiologia da UNL, Faculdade de Ciencias e Tecnologia, Quinta da Torre, P-2825 Monte de Caparica, Portugal. ** Institut de Paleontologic (URA 12, CNRS), 8 rue Buffon, 75005 Paris, France.

pp.191-201 Ciencias da Terra (UNL) Lisboa NQ 11 1992 1 pI.

RESUMO

Palavras-chave: Eurodon- Taeniodonta- Eocenico-:-­ Silveirinha - Portugal.

Eurodon silveirinhensis nov. gen., nov. sp., do Eocenico inferior de Silveirinha (portugal) econsiderado o primeiro representante, na Europa, da extinta ordem de mamiferosTaeniodonta.0 enigrnaticogeneroLessnessina Hooker, 1979, de Abbey Wood(lnglaterra), sensive1mente contcrnporaneo de Eurodon, e tambern atribufdo aos teniodontes.

RESUME

Mots-dis: Eurodon - Taeniodonta - Eocene­ Silveirinha - Portugal.

Eurodon silveirinhensis nov. gen., nov. sp., de I'Eocene infericur de Silveirinha (portugal) est interpretee commeIapremiererepresentanteen Europedes Taeniodonta (ordre eteint de mammiferes). L'enigmatique genre Lessnessina Hooker, 1979, de Abbey Wood (Angleterre), apeu pres contemporainde Eurodon,estrapporteegalement aux Taeniodonta.

ABSTRACT

Key-words: Eurodon - Taeniodonta - Eocene ­ Silveirinha - Portugal.

The first representative of the extinct mammalian order Taeniodonta in Europe is described, Eurodon silveirinhensis n. gen., n. sp., from the early Eocene local­ ity of Silveirinha, Portugal. A formerly enigmatic form, Lessnessina Hooker, 1979, from Abbey Wood, England, and approximately contemporary, is also referred to the Taeniodonta.

193

INTRODUCTION Type species - Eurodon silveirinhensis, new species. Study of the early Eocene mammalian fauna Diagnosis - Small mammal, smaller than any from Silveirinha, Portugal (ANTUNES, 1981; described taeniodont. M3 with large hypoconulid, ANTUNES & RUSSELL, 1981), is the subject of a apparently subequal hypoconid and slightly smaller, doctoral dissertation for one ofus (c. E.). One speci­ conicalentoconid; protoconid and metaconidinflated men, SV2-15, a small lower last molar, has long and subequal; paraconid very small, situated on ante­ defied identification. The bulbousness of its cusps rior side ofmetaconid. initially lead to the supposition that it represented Differs from all conoryctidsby lesserhypsodonty some sort of condylarth. We here present a new and by a larger M3 hypoconulid. interpretation, that of affinity to Taeniodonta. Age and distribution - early Eocene ofEurope. Specimens of the order Taeniodonta are rarely Etymology - euro - pertaining to Europe; ­ found. Until now, the few taxa constituting this group odon (Gr. tooth). were known only from North America, ranging in time from early Paleocene to middle Eocene. The rarity oftaeniodonts was discussed by GINGERICH Eurodon silveirinhensis, new species (1989) and attributed to ecologicalconditions, wherein their habitat was inland and upland and far from the Type specimen - SV2-15, isolated, right M3. low floodplain and basin environments that furnish Referred material - SV3-234, right Mi' and, most fossil collections. SCHOCH (1986) deduced more questionably, SV3-208, left M1 or M; both that their alimentary source consisted primarily of specimens lack most oftheir enamel. underground roots and tubers. The early generalized Diagnosis - Same as for genus. form, Onychodectes, in addition to being a Locality and age - Silveirinha(BaixoMondego noncursorial, plantigrade mammal, is presumed to region of west central Portugal); Donnaalian mam­ have been fairly adept at climbing. It was also the mal age. smallest of the Taeniodonta. Etymology - named after the locality in Portu­ gal from which it was collected. SYSTEMATIC PALEONTOLOGY

Order Taeniodonta COPE, 1876 DESCRIPTION Family Conoryctidae WORTMAN, 1896 Subfamily Onychodectinae WINGE, 1917 The trigonid ofM3 is moderately high, as were Onychodectes COPE, 1888 apparentlythe threestrongtalonid cusps (hypoconid, Subfamily Conoryctinae WORTMAN, 1896 hypoconulid and entoconid; now truncatedby wear). Conoryctella GAZIN,1939 Although wear has reduced the height of the Conoryctes COPE,1881 protoconid (morethanthatofthemetaconid)it seems Huerfanodon SCHOCH & LUCAS, 1981 to have been subequal in size to the metaconid; both Subfamily Eurodontinae, new subf. are large, bulbous cusps. A paracristid, also worn, Eurodon, new gen. extendsfrom the anterio-medianside ofthe protoconid Lessnessina HOOKER, 1979 to the very small paraconid situated at the anterior Eurodon, new genus baseofthemetaconid. Theratherweakcristidobliqua

195 linksthehypoconid and the baseofthe posteriorwall the order, Schoch cited thelackofcingulids in lower of the trigonid, contacting the latter far below the molars, which indeed agrees with the condition in protoconid-metaconid notch. Forming a veritable Eurodon (exception madeofan anteriorcingulum in third lobe, the hypoconulid is bulbous, prominent, Onychodectes and Eurodon). He also stated that the and sharply separated from the hypoconid and trigonids and talonids ofall molars were subequal in entoconid.Theentoconid is large and conical and has length and width, but exceptions can be seen in the at its anterior base a small accessory cusp, the figurations of conoryctids in his monograph, where entoconulid. The talonid basin is open lingually. the talonid of M3 can be either shorter than the Despiteconsiderablewearon theposteriorsideofthe trigonid or appreciably longer, as in Eurodon. The metaconid, a postmetacristid of some sort is indi­ hypoconulid is typically notexpanded onconoryctid cated, or a linguo-vertically placed ridge. Enamel is M3 talonids, but this feature is shown to be very smooth and there is no indication of hypsodonty. variable in (at least) Huerfanodon torrejonius Length, 3.7 mm; width, 2.3 mm. (SCHOCH, 1986,Pl. 16,Fig. 5 and 7); the specimen Twootherteethare perhapsreferableto Eurodon illustrated in his Fig. 7 is similar to the Silveirinha silveirinhensis. One of these, SV3-234, aright M3 M3, with an elongate talonid and a prominent like the type specimen, is quite probably of this hypoconulid. Taeniodonts are well known for their taxon. Identification is hampered by the effects of tendency for the cheek teeth to develop hypsodonty, chemicalerosionwhichremoved mostoftheenamel; butin the earliestforms this is not always perceptible only a small amount remains on the anterior and oris onlyslightly manifested. Its absence in Eurodon lingual sides. This specimen provides little informa­ can be interpreted as a retention ofthe atavistic state. tion, beyond the presence of a more strongly devel­ Despite the distinctiveness of the Silveirinha oped anterior cingulum. specimen, similarities in its morphology to various Theuppermolar,SV3-208, also lacksmostofits conoryctids are preponderant. Theinflated cusps, the enamel, and the labial borderofthe tooth, exteriorto large conical entoconid and the presence of an the paracone and metacone, is missing as well. Nev­ entoconulid at its anteriorbase, are all features found ertheless, the specimenis ofa size thatconcords well in Onychodectes and Conoryctella. A very small withthatofthetypespecimen. The protocone is large paraconid anterior on the metaconid of M3 is also and bulbous and situated rather far from the lingual shared with Conoryctella and Conoryctes. It is con­ base of the tooth. Both conules are present, the cluded that a sufficient number of elements charac­ metaconulebeingthelarger. Thepreparaconulecrest teristic of recognized taeniodonts are also found in is highandstronglydeveloped, as is the premetaconule the M3 of Eurodon to render its attribution to this crest; the latter contacts the lingual side of the order and to the family Conoryctidae quite likely. metacone high, near its summit, while the preparaconule crest extends to the region of the parastyle. Apparently the postparaconule crest was weak, although the paracone possessed a lingual, vertical crest that could be regarded as the continua­ The problem of Lessnessina HOOKER, tion of this crest. The postmetaconule crest is ob­ 1979· scured by wear and erosion, but it extended to the metastyle, Anteriorand posteriorcingulaare strongly Schoch, in his monograph, argued against the developed with the latter being the broader. It ap­ presence ofany members ofthe Taeniodontaoutside pears to have supported a small, lingually placed ofNorth America. None ofthe specimens from Asia hypocone. - or Europe that had been proposed as attributable to It is, of course, uncertain whether or not this this order can be unequivocally accepted. However, specimenis referable to Eurodon silveirinhensis. But no one has as yet proposed this relationship for the by its size and morphology, it seems possible. enigmatic genus Lessnessina from the early Eocene locality of Abbey Wood, England. The principal specimen is a maxillawithp3_M3. HOOKER (1979) DISCUSSION was convinced that its dental morphology indicated affinity to the anisonchine Periptychidae; he was Attribution of the M3, SV2-15, to the particulary impressed with an apparent similarity to Taeniodonta was inspired by the inflated aspect of Oxyacodon. the protoconid and metaconid, the small size of the ARCHIBALD et al. (1983), in a revision ofthis paraconid, and the large, conical shape of the genus, denied the existence of any relationship be­ entoconid. tween Oxyacodon and Lessnessina, excluded the Consultation of the most recent revision of latter from the Anisonchinae and doubted it was a taeniodonts (SCHOCH, 1986) revealed that some of condylarth. Without making a definitive assessment the morphologic features that typify Eurodon corre­ of its ordinal affinities, they noted some resem­ spond to traits that are rather marginal within the blances to pseudictopids or other anagalidans (sensu known taxa. In his listofderived character-states for SZALAY & McKENNA, 1971).

196 hypsodonty, a widely separate, round paracone and 33% larger than Lessnessina packmani, making al­ metacone (with a mesostyle present) and a similar lowances for the variability oflast molars. They are occlusal contour in the upper teeth of Lessnessina not referable, therefore, to the samespecies. Generic and those of the Mixodontia (revised by identity would be extremelydifficultto demonstrate, DASHZEVEG & RUSSELL, 1988), the with only the available material. ressemblance does not seem to be significant. The HOOKER (1979), speculating on the possible presence ofthe Asiatic Mixodontia in Europe seems natureofthe unknownlowerdentitionofLessnessina, unlikely in view of the lack of communication be­ remarked thatM3 wouldhavebeenrelativelysmaller tween Asia and Europe at the time and the absence of than Ml-M2 and been characterized by a reduced mixodonts in North America, the only migrational hypoconulid. Furtheron, he emphasized the similar­ path available. Recognition of a diminutive ity in dental morphology to Lessnessina of taeniodont, Eurodon, in Europe, approximately con­ Oxyacodon. Ifthe shapeofthe lastmolarsin thelatter temporaneouswith Lessnessina,encouragedan analy­ is regarded, it can be seen that the upper M3 is sis of the latter's characters from the viewpoint of narrower (antero-posteriorly) and more transversely possible taeniodont relationships. elongate than this tooth in Lessnessina and that the Dental similarities to the earliest taeniodonts, lowerM3 possesses ahypoconulid evenmorepromi­ Onychodectes, Conoryctella, and Huerfanodon, are nent than that in Eurodon. We feel, therefore, thatthe evident in the upperdentition (p3-M3) that is all that morphology ofthe upperM3 in Lessnessina is com­ is known ofLessnessina. Particulary striking in both patible with that of the lower M3 ofEurodon. is thelingual height ofthe protocone and the absence of a strong hypocone giving the upper cheek teeth a triangularaspect in occlusal view. p4 morphology in CONCLUSION the most primitive form, Onychodectes tisonensis, with the paracone and protocone prominent and the Sparse remains from Portugal and England fur­ metacone absent, is shared by Lessnessina. The nish evidence suggesting that the orderTaeniodonta mesostylethatoccurs in the molars ofthe latteris also was not restricted to North America. But in order to found in Conoryctella, Conoryctes and Huerfanodon, include Eurodon and Lessnessina in the Taeniodonta along with the conical character of the principal one must hypothesize that a branch, whose known cusps. members represented a primitive stagesimilarto that Differences in the upper molars ofLessnessina of the Puercan Onychodectes, existed in Europe. principally concern the acquired existence of ante­ That this possibility is not totally unrealistic is sup­ rior and posterior cingula, terminating lingually in a ported by the facts that taeniodonts are known to be small protostyle and an even smaller hypocone. In always rare in any fauna and that the Paleocene view ofthe considerable lapse oftime that separates mammals ofEurope are very incompletely sampled. the earliest taeniodont species (which Lessnessina It may be postulated that increased data from more mostclosely resembles) and the British form, it is not representative collections will substantiate the real­ surprising that some morphological evolution has ity of this European taeniodont lineage. occurred. It is perhaps more surprising that Lessnessina apparently changed so little in other respects from a Puercan early Paleocene pattern. Ifit is postulated that Lessnessina and Eurodon ACKNOWLEDGEMENTS are related, the eventuality oftheirbeing conspecific We wish to express our gratitude to Professor M. T. should be considered. Comparative upper and lower Antunes for his support. We are indebted to "Institute toothdimensions in earlyconoryctidsreveals thatthe Nacional de Investigacao Cientifica (INIC, Portugal)" for lower M3 is about as long as, or shorter than the a fellowship garanted to oneofus (C. E.). Weacknowledge transverse width of the upper M3. Given this rela­ with gratitude the talents of Madame C. Weber for the tionship, Eurodon silveirinhensis would be 25 to electron scanning microphotographs.

197 BIBLIOGRAPHY

ANTUNES. M. T. (1981) -Asareias e argilasde Silveirinha: 1/ Aspectospaleontol6gicos. Mem6rias e Noticias, Mus. Lab. Mineral. oeei., Univ. Coimbra, 91-92: pp. 253-267. ANTUNES. M. T. & RUSSELL. D. E. (1981) -Legisement de Silveirinha (Bas Mondego, Portugal): laplus ancienne faune de vertebres eocenes connue en Europe. C. R. Acad. Sc.• Paris. (II) 293: pp. 1099-1102. ARCHIBALD. J. D.; RIGBY. J. K. JR. & ROBISON. S. F. (1983) - Systematic revision ofOxyacodon (Condylarth, Periptychidae) and a description ofO.ferronensis n. sp. Jour. Palcont.• 57 (1): pp. 53-72. DASHZEVEG. D. & RUSSELL. D. E. (1988) - Palaeocene and Eocene Mixodontia (Mammalia. Glires) ofMongolia and China. Palaeontology. 31 (1): pp. 129-164. GINGERICH. P. D. (1989) - New earliest Wasatchian mammalianfauna from the Eocene ofnorthwestern Wyoming: compositionanddiversityin a rarely sampledhigh-floodplainassemblage. Univ. MichiganPaperson Paleontology. n Q 28: pp. 1-97. HOOKER. J. J. (1979)- Two newcondylarths (Mammalia)from the early Eocene ofsouthernEngland. Bull. Brit. Mus. Nat. Hist. (Geol.). 32 (1): pp. 43-56. . SCHOCH. R. M. (1986) - Systematics.functional morphology andmacroevolution ofthe extinct mammalian order Taeniodonta. Peabody Mus. Nat. Hist. Bull.• 42: pp. 1-307. SZALAY. F. S. & McKENNA. M. C. (1971) -Beginning ofthe age ofmammals in Asia: the late Paleocene Gashato Fauna. Mongolia. Bull. Amer. Mus. Nat. Hist., 144 (4): pp. 269-318.

198 DOCUMENTA~AO FOTOGRAFICA PLATE 1

A and B -s-Lessnessina pockmani HOOKER, 1979; A - M29632, holotypc, left P3_M3, bar=1 mm; B - MC 17, right M', bar=1 mm.

Collections of the British Museum (N.H.). C - Eurodon silveirinhensis n. gen., n. sp., SV2-15, holotypc right M • C , lingual view; 3 1 C , occlusal view; C labial view, bar=1 mm. Collections of the Centro de Estratigrafia e Paleobiologia da Universidade Nova de 2 3 Lisboa. PLATE 1

A ,1mm,

Cz ,1mm

B 1mm