Primate phylogeny: morphological vs. molecular results

Mol Phylogenet Evol. 1996 Feb;5(1):102-54. doi: 10.1006/mpev.1996.0009.


Our comparative study of morphological (our data on selected living primates) and molecular characters (from the literature) confirms that, overall, phylogenetic reconstructions of Primates, and consequently their classifications, are more similar than dissimilar. When data from fossil Primates are incorporated, there may be several possible relationships among living Primates; the difference between most of them hinges mainly on the position of Tarsius. In one hypothesis, tarsiers are closely related to lemurs and lorises, and thus Primates is divided into Prosimii [lorises, lemurs, and tarsiers] and Anthropoidea [Platyrrhini and Catarrhini, i.e., monkeys, apes, and humans]. Two additional alternatives are that Tarsius is a sister group to the clade embracing lorises + lemurs and Anthropoidea and that in which all three lineages (Tarsius, lorises + lemurs, and Anthropoidea) form a polychotomy. In another hypothesis, tarsiers are closely related to anthropoids, giving these two branches: Strepsirhini [lemurs, lorises] and Haplorhini [tarsiers and Anthropoidea (Platyrrhini, the New World monkeys, and Catarrhini, Old World monkeys and Hominoidea)]. The first three alternatives gain some support from the fossil record, and the fourth from morphology of the living Tarsius and molecular data. It is emphasized that the morphological characters employed in this study for Tarsius are based on the only surviving genus of once-diverse tarsiiform primates known from the Eocene, and, although considered a "living fossil," it cannot represent all of them. Furthermore, Tarsius embodies derived features of its own which may affect its systematic position, but not necessarily the position of Tarsiiformes. Although the early Tertiary adapoids might have more nearly resembled anthropoids in their biochemistry and placental developments, this hypothesis is not testable from fossils, and any inferred relationships here must be based on characters of skeletal anatomy. Alternatively, anthropoids may be derived from certain omomyids or from some as yet undiscovered Eocene African taxon. Close relationships among Homo, Pan, and Gorilla have been confirmed during recent decades; Pongo is the sister group to this trichotomy. With increasing molecular data, Homo and Pan appear to be closer to each other than to any other living hominid taxon. Gorilla is a sister group to the Homo-Pan clade and Pongo is a sister group to all of them. Morphologists have given limited evidence for such a dichotomous grouping. In this study, we support the Homo-Pan clade, although with characters not as strong as for other clades.

Publication types

  • Comparative Study
  • Review

MeSH terms

  • Animals
  • Evolution, Molecular
  • Female
  • Fossils
  • Hominidae / anatomy & histology
  • Hominidae / classification
  • Hominidae / genetics
  • Humans
  • Male
  • Phylogeny*
  • Primates / anatomy & histology*
  • Primates / classification
  • Primates / genetics*
  • Software