Impact of increased character sampling on the phylogeny of Cetartiodactyla (mammalia): combined analysis including fossils

The phylogenetic position of Cetacea (whales, dolphins and porpoises) is an important exemplar problem for combined data parsimony analyses because the clade is ancient and includes many well-known and relatively complete fossil species. We combined data for 71 terminal taxa (43 extinct/28 extant) to test where Cetacea fits within Cetartiodactyla, and where various fossil hoofed mammals (e.g., dagger entelodonts, dagger anthracotheriids and dagger mesonychians) are positioned. We scored 635 phenotypic characters (osteology, dentition, soft tissue, behavior), approximately three times the number of characters in the last major analysis of this clade, and combined these with > 40 000 molecular characters, including new data from 10 genes. The analysis supported a topology consistent with the majority of recently published molecular studies. Cetacea was the extant sister taxon of Hippopotamidae, followed successively by Ruminantia, Suina and Camelidae. Several extinct taxa were phylogenetically unstable, upsetting resolution of the strict consensus and limiting branch support, but the positions of several key fossils were consistently resolved. The wholly extinct dagger Mesonychia was more closely related to Cetacea than was any artiodactylan.dagger Anthracotheriids were paraphyletic, and, with the exception of one species, were more closely related to Hippopotamidae than to any other living taxon. The total evidence analysis overturned a highly nested position for Moschus supported by molecular data alone. The character partition that could be scored for the fossil taxa (osteological and dental characters) included more informative characters than most molecular partitions in our analysis, and had the fewest missing data. The osteological-dental data alone, however, did not support inclusion of cetaceans within crown Artiodactyla. Recently discovered ankle bones from fossil whales reinforced the monophyly of Cetartiodactyla but provided no particular evidence of derived similarities between hippopotamids and fossil cetaceans that were not shared with other artiodactylans. (c) The Willi Hennig Society 2007.