More on the validation of fecal samples for the large-scale study of dietary differences between two sympatric carnivores

Prey occurrence from stomach and fecal samples were compared in American marten Martes americana Turton, 1806, and fisher M. pennanti Erxleben, 1777. Diets were analyzed from stomach and fecal contents in four sex-age marten groups and four fisher groups, and dietary differences tested between species, sex, age and sample sources. Prey richness was compared between the two sample sources. Relative occurrence of dominant or co-dominant prey items did not differ between stomach and fecal samples in any of the groups. However, both series revealed inter-specific contrasts. Binomial correlations of prey occurrence between sample sources were significantly negative for most prey items. Correlations of prey richness between samples sources were not significant except in larger marten sample. Our results failed to reveal a bias in using fecal sample source. Thus we suggest that diet analyses using either sample source are valid when diets are assessed in a relatively large number of animals.