PALEOGENE BUNODONT UNGULATES FROM SOUTH AMERICA: ALTERNATIVE HYPOTHESES

GELFO, J.N.; CHORNOGUBSKY, L.

Bristol, Reino Unido

Jornada; 69th Society of Vertebrate Paleontology Annual Meeting; 2009

Universidad de Bristol- SVP

The earliest known therian fossil records from South America (e.g. Paleocene from Tiupampa, Bolivia and Punta Peligro, in Patagonia, Argentina) show a predominance of bunodont ungulates. In the traditional view, these comprised ‘archaic ungulates’ such as the Mioclaenidae Kollpaniinae and Didolodontidae, plus the litoptern Protolipternidae. Since the Kollpaniinae were related to a North American family, the last two were the only strictly bunodont and low-crowned, endemic ‘ungulates’ from the Paleogene of South America. Didolodontids and protolipterns were recognized as different groups by the reassociation of postcranial remains, with ‘condylarth-like’ and ‘litoptern-like’ morphologies respectively. But, these reassociations were questioned in the case of the didolodontids, particularly because no complete skeleton or associated postcranium is yet known for any of them. Consequently, the phylogenetic position of didolodontids is still controversial. A data matrix of 89 dental and tarsal characters for 26 taxa (Protoungulatum donnae as an out-group + Mioclaenidae Kollpaniinae + Didolodontidae + Protoliptenidae) was built and analyzed under parsimony using the TNT software. The matrix was evaluated firstly considering the didolodontid tarsal reassociation as valid, and secondly codifying these conflictive characters as missing data. Both analyses were carried out, with and without implied weighting. As a result, the Kollpaniinae were recovered as monophyletic and basal to the remaining groups in all the analyses. The analysis performed with the complete tarsal information recovered the Didolodontidae as a natural group only when implied weighting was applied. In contrast, for this matrix, Protolipternidae were always recovered as a natural group, even when the two species of Asmithwoodwardia seem not to be related to this node. When the conflictive tarsal characters for the didolodontids Lamegoia and Paulacoutoia were not included in the analysis, the Didolodontidae resulted monophyletic in contrast to the Protolipternidae. In conclusion, these analyses emphasize two different and mutually exclusive hypotheses. The postcranium of didolodontids could be ‘condylarth-like’, but their monophyly needs to be checked in a wider context including the rest of the native South American ‘ungulates’. Or, the didolodontids may have had a ‘litoptern-like’ postcranial anatomy, and the Protolipternidae are in consequence an artificial group.