On the relationship between hypsodonty and feeding ecology in ungulate mammals, and its utility in palaeoecology.

High-crowned (hypsodont) teeth are widely found among both extant and extinct mammalian herbivores. Extant grazing ungulates (hoofed mammals) have hypsodont teeth (a derived condition), and so extinct hypsodont forms have usually been presumed to have been grazers. Thus, hypsodonty among ungulates has, over the past 150 years, formed the basis of widespread palaeoecological interpretations, and has figured prominently in the evolutionary study of the spread of grasslands in the mid Cenozoic. However, perceived inconsistencies between levels of hypsodonty and dental wear patterns in both extant and extinct ungulates have caused some workers to reject hypsodonty as a useful predictive tool in palaeobiology, a view that we consider both misguided and premature. Despite the acknowledged association between grazing and hypsodonty, the quantitative relationship of hypsodonty to the known ecology of living ungulate species, critical in making interpretations of the fossil record, was little studied until the past two decades. Also, much of the literature on ungulate ecology relevant to understanding hypsodonty has yet to be fully incorporated into the perspectives of palaeontologists. Here we review the history and current state of our knowledge of the relationship between hypsodonty and ungulate ecology, and reassert the value of hypsodonty for our understanding of ungulate feeding behaviour. We also show how soil consumption, rather than the consumption of grass plants per se, may be the missing piece of the puzzle in understanding the observed correlation between diets, habitats, and hypsodonty in ungulates. Additionally, we show how hypsodonty may impact life-history strategies, and resolve some controversies regarding the relevance of hypsodonty to the prediction of the diets of extinct species. This in turn strengthens the utility of hypsodonty in the determination of past environmental conditions, and we provide a revised view of a traditional example of evolutionary trends in palaeobiology, that of the evolution of hypsodonty in horses and its correlation with the Miocene spread of grasslands in North America.