PATTERNS OF EVOLUTION AMONG THE ARTIODACTYLA AND PERISSODACTYLA (MAMMALIA).

Abstract

Diversity changes through time of the mammalian ungulate orders Artiodactyla and Perissodactyla have been touted long and widely as exemplifying ordinal level taxonomic (and presumably ecologic) displacement (Simpson, 1953; Stanley, 1974) since, as Van Valen (1971) has pointed out, they have a nearly identical way of life (=adaptive zone, Simpson, 1953; plus competitive interactions, Van Valen, 1971). The controversial assumption that herbivores are food limited (cf. Hairston et al., 1960; Murdoch, 1966; Ehrlich and Birch, 1967; Slobodkin et al., 1967) is implicit to this generalization. Both orders presumably appeared in the late Paleocene, as they are well differentiated by the early Eocene (Figs. 1 and 2). While perissodactyls seem to have undergone taxonomic diversification at both generic and familial levels somewhat earlier than artiodactyls, the difference is much less than most authors suggest. Perissodactyls declined in importance after the middle Eocene, while the artiodactyls maintained a stable dominance from the early Oligocene until their apparently enormous Plio-Pleistocene radiation. Various deterministic hypotheses have been forwarded in explanation of this pattern. Such hypotheses usually invoke the acquisition of one or more characters (double trochleated astragalus, selenodont molars, ruminant digestion) by the artiodactyls as giving that group a relative adaptive advantage. Here I review the evolutionary histories of these two orders in search of evidence supporting these hypotheses. Further, I analyze overall patterns to detect possible correspondence to or support for the various empirical (Valentine, 1973) and equilibrium (Mark and Flessa, 1977) models of Phanerozoic diversity. All data have been compiled from Romer (1966). Current revisions might alter the data somewhat, but it is doubtful that the differences would be systematic or great. Question-marked taxa or those not designated chronostratigraphically to subepoch were ignored. The time scale used is from Harland et al. (1964).