ONTOGENY AND PHYLOGENY IN HORSE SKULL EVOLUTION.

Abstract

Horses (Family Equidae) have one of the longest and richest fossil records of any group of mammals, and it documents major changes in morphology. For these reasons, fossil horses have been an important source of examples of evolutionary change and evolutionary phenomena (e.g., Simpson, 1944, 1953). There have been several studies of the evolutionary transformation of the limbs of horses (e.g., Sondaar, 1968; Forsten, 1975a; Hussain, 1975), and of differences in skull morphology among closely related species or genera (e.g., Gromova, 1949; Bennett, 1980; Eisenmann, 1980; Groves and Willoughby, 1981). However, for the past 40 years our knowledge of the evolutionary transformation of horse skulls has rested primarily on two studies, those of Robb (1935a, 1935b) and Reeve and Murray (1942), which came to contradictory conclusions. Robb analyzed phylogenetic changes in horse skull proportions by comparing preorbital length to total skull length in 13 species of fossil horses. He concluded that the relatively long muzzles of modern horses resulted entirely from allometry -size related changes in proportions that occurred during evolution from small ancestral horses to the large modem ones. Robb also examined preorbital length vs. skull length in an ontogenetic series of modern horses and in a sample of adult modern horses of different sizes, and concluded that differences in proportions seen in ontogeny and among differentsized adults correspond exactly to changes in proportions seen in the phylogenetic series. Seven years later, Reeve and Murray (1942) re-analyzed Robb's data and added additional specimens of modern horse skulls. They compared preorbital length to braincase length (in their study defined as skull length minus preorbital length) and concluded that ontogeny did not repeat phylogeny in horse skull evolution, and that allometry alone did not account for the proportional changes seen in the phylogenetic series of fossil horses. Reeve and Murray suggested that a reorganization in skull proportions occurred with the evolution of high crowned teeth, based on an increase in size of the facial rudiment at a very early stage of growth. Later workers have followed either Robb, and stated that scaling effects account for modern horse skull proportions, or Reeve and Murray, and concluded that horse skull proportions were reorganized to accommodate enlarged grinding teeth. For example, the first figure in Simpson's (1944) classic Tempo and Mode in Evolution is a replot of Robb's data showing phylogenetic, ontogenetic and intraspecific adult curves almost completely overlapping. On the other hand, DeBeer ( 1958 p. 59) repeated Reeve and Murray's conclusion that horse