Abstract
Assessments of feeding habits based upon stomach contents and faecal analyses appear to differ and results of analyses of faecal material may not give an accurate representation of the spectra of prey consumed. Underrepresentation of the remains of prey species with fragile otoliths in faecal samples will lead to biases in the evaluation of the relative importance of different prey species in the diet. These biases may be so large that results of analysis of faecal material may not be reliable in quantitative estimation of prey consumption. Therefore multispecies ecosystem models using input data derived from scat analysis may not produce realistic predictions of predator-prey interactions.