In an attempt to investigate the relationships between allometry and locomotory adaptations, we studied the long limb bones of 45 species of insectivores and rodents. Animals ranged from a few grams to about 50 kilograms. Diameter and length of the bones and body mass (when known) were recorded. Regressions of diameter to length, diameter to body mass, and length to body mass were calculated by the least-squares and Model II, or major axis, methods. The results obtained do not agree with the predictions of either the theory of geometric similarity or the theory of elastic similarity. The discrepancies could be due to the fact that animals studied exhibit various modes of locomotion. Moreover, the allometric relationships of the different locomotor patterns are better reflected in insectivores and rodents than in other groups of mammals. The use of a single regression analysis seems to be inadequate when the sample includes a large range of body sizes.

The allometric relations of diameter and length of humerus, ulna, femur, and tibia of 108 specimens, from 63 different breeds of dogs and 12 specimens of wolves, were calculated by means of model II of regression or major axis method. Only for the tibia were the values of wolves included in the cluster formed for dog breeds. Consequently, separate lines of regression were calculated for the other bones. Results agree in general with the exponents predicted by the theory of geometric similarity; however, the slope obtained for femur (0.865) differed significantly from this. Morphology of the long bones of the legs does not differentiate dogs and wolves; this probably reflects secondary convergence among wolves with relatively modern breeds of dogs.