Convergence of macroscopic tongue anatomy in ruminants and scaling relationships with body mass or tongue length

Various morphological measures demonstrate convergent evolution in ruminants with their natural diet, in particular with respect to the browser/grazer dichotomy. Here, we report quantitative macroanatomical measures of the tongue (length and width of specific parts) of 65 ruminant species and relate them to either body mass (BM) or total tongue length, and to the percentage of grass in the natural diet (%grass). Models without and with accounting for the phylogenetic structures of the dataset were used, and models were ranked using Akaike's Information Criterion. Scaling relationships followed geometric principles, that is, length measures scaled with BM to the power of 0.33. Models that used tongue length rather than BM as a body size proxy were consistently ranked better, indicating that using size proxies that are less susceptible to a wider variety of factors (such as BM that fluctuates with body condition) should be attempted whenever possible. The proportion of the freely mobile tongue tip of the total tongue (and hence also the corpus length) was negatively correlated to %grass, in accordance with concepts that the feeding mechanism of browsers requires more mobile tongues. It should be noted that some nonbrowsers, such as cattle, use a peculiar mechanism for grazing that also requires long, mobile tongues, but they appear to be exceptions. A larger corpus width with increasing %grass corresponds to differences in snout shape with broader snouts in grazers. The Torus linguae is longer with increasing %grass, a finding that still warrants functional interpretation. This study shows that tongue measures covary with diet in ruminants. In contrast, the shape of the tongue (straight or "hourglass-shaped" as measured by the ratio of the widest and smallest corpus width) is unrelated to diet and is influenced strongly by phylogeny.