Bone loading was quantified, using in vivo strain recordings, in the tibiotarsus of growing chicks at 4, 8, 12, and 17 weeks of age. The animals were exercised on a treadmill at 35% of their maximum running speed for 15 minutes/day. In vivo bone strains were recorded at six sites on the tibiotarsus. Percentages of the bone's length and a percentage of top running speed were used to define functionally equivalent sites on the bone, and a consistent exercise level over the period of growth was studied. The pattern of bone strain defined in terms of strain magnitude, sign, and orientation remained unchanged from 4-17 weeks of age, a period when bone mass and length increased 10-fold and threefold, respectively. Our findings support the hypothesis that bones model (and remodel) during growth to achieve and maintain a similar distribution of dynamic strains at functionally equivalent sites. Because strain magnitude and sign (tensile versus compressive) differed among recording sites, these data also suggest that cellular responses to strain-mediated stimuli differ from site to site within a bone.