Detecting between-individual differences in hind-foot length in populations of wild mammals

Hind-foot length is a widely used index of skeletal size in population ecology. The accuracy of hind-foot measurements, however, has not been estimated. We quantified measurement error in adult hind-foot length in yellow-bellied marmots (Marmota flaviventris (Audubon and Bachman, 1841)), mountain goats (Oreamnos americanus (de Blainville, 1816)), and bighorn sheep (Ovis canadensis Shaw, 1804) from long-term capture-recapture studies. Fitting a linear mixed effect model for each species separately, we found that hind-foot length was significantly repeatable in the three species, but repeatability was low, ranging from 0.30 to 0.47. Measurement error explained 53%-66% of the variance in foot length. Differences of 6, 13, and 27 mm would be indistinguishable from measurement error for marmots, goats, and sheep, respectively. At least 4-6 measures per individual were needed to detect variation in foot length between individuals of a population using a mixed effect model. Researchers should strive to limit measurement errors because inaccurate measures may obscure important biological patterns.