Several reports have been published on blood leptin concentrations in feral animals, including members of the Carnivora, using a commercially available multi-species radioimmunoassay (RIA) kit with anti-human leptin antibody. However, we observed weak immunoreactivity between recombinant canine leptin and anti-human leptin antibody, suggesting a limitation in the applicability of the RIA kit for leptin assays in Carnivora species. We tested the applicability of RIA and sandwich enzyme-linked immunosorbent assay (ELISA) with anti-canine leptin antibody to assay blood leptin in the dog (Canis familiaris) and the raccoon (Procyon lotor). When RIA was used for recombinant canine leptin and dog sera, values were much lower than those determined by ELISA at higher concentrations (>10 ng/ml), while rather higher at lower concentrations (<2 ng/ml). A similar discrepancy between the two methods was found for serum leptin concentrations in raccoons. Clear seasonal variations were observed by ELISA, but not by RIA, with high values in autumn (3.46+/-0.45 ng/ml) and low values in spring and summer (0.71+/-0.07 ng/ml). Serum leptin concentrations in raccoons correlated positively with their body weight (r=0.753) and body mass index (r=0.755), corroborating our previous findings of a strong positive correlation between serum leptin concentrations and body fat content in dogs. Thus, the canine leptin ELISA is useful for assays of dog and raccoon leptin, and blood leptin is a good marker of nutritional condition in the species of Carnivora assayed in this study.