Morphological and molecular evidence for the phylogeny of Cetacea and Artiodactyla: Explaining incongruence between types of data

Possible resolutions of the conflict between morphological and molecular evidence for the phylogeny of Odontoceti and Mysticeti and for the phylogeny of Artiodactyla and Cetacea were achieved by compiling extensive anatomical observations and then using these data, in conjunction with previously published nucleotide sequences, to develop and test five hypotheses that explain the incongruence between both types of data. Anatomical observations for 68 cetaceans, artiodactyls, and related taxa were tabulated in a matrix of 186 morphological characters. Cladistic analyses of the data corroborated monophyly of Artiodactyla (even-toed ungulates), Suiformes (hippos, pigs, peccaries), and Neoselenodontia (camels, deer, cows). These results contradict molecule-based studies that support a paraphyletic Artiodactyla, Neoselenodontia, and Suiformes. Instead of these taxa, molecular data supports Whippomorpha (Cetacea and Hippopotamidae) and Cetruminantia (Whippomorpha and ruminants).; To investigate phylogenetic relationships among whales, I developed a matrix of 304 morphological characters for 52 odontocete and mysticete taxa. All most parsimonious trees for this data include a monophyletic Odontoceti (extant toothed whales) and Mysticeti (baleen-bearing whales). Contrary to previous morphological studies, Squalodontidae and Squalodelphidae are not closely related to Platanistidae. Instead Platanista and other extant river-dwelling odontocetes form a monophyletic Platanistoidea. The most parsimonious trees for morphology contradict previous molecular studies, which claimed that Odontoceti is paraphyletic because Physeteridae (sperm whales) is more closely related to Mysticeti than to other odontocetes.; I tested five hypotheses for the phylogenetic incongruence between morphological and molecular data: (1) poorly aligned molecular sequences, (2) randomly distributed observational errors, (3) convergence of molecular characters, (4) interdependence of morphological characters, and (5) convergence of morphological characters. Conflict between evidence for the phylogeny of Odontoceti and Mysticeti is best explained as incorrect alignment of molecular sequences. This hypothesis is supported by the discovery of alternative alignments of cetacean genes that are remarkably congruent with morphological data. Conflict between evidence for the phylogeny of Artiodactyla and Cetacea can be attributed to interdependent morphological characters. Characters within the molar and hindlimb subpartitions are highly compatible with each other. Removal of these subpartitions reduced the incongruence by 22% to 55%. When molecular characters were analyzed with all morphological characters except the molar partition, limited morphological support for Whippomorpha and Cetruminantia appeared in the temporal, orbit edge, and face subpartitions.