MamnralRev. zyxwvutsrq 1991, zyxwvutsrqp Volume 21, No. 3,97-122. Printed in Great Brirain zyxwv Comparison and quantification of carnivore diet by faecal analysis: a critique, with recommendations, based on a study of the zyx Fox Vulpes vulpes JONATHAN C. REYNOLDS and NICHOLAS J. AEBISCHER The Game Conservancy, zyxwvu Fordingbridge, Hants. SP6 IEF, zyxw U.K. ABSTRACT Failure to address many of the difficulties inherent in the analysis of carnivore faeces has hitherto limited the value of this technique in comparing carnivore diets or quantifying carnivore food consumption. In this review, major problems are discussed, and improved procedures designed to overcome them are recommended. The principal improvements recommended for laboratory procedures address the microscopic fraction of faeces, which cannot be analysed visually. This fraction may be derived from a variety of sources, leading to bias in previous techniques, particularly with regard to the importance of earthworms and birds. Attention to the microscopic fraction is necessary for correct comparison or quantitative determination of diet. Particular attention is given to statistical procedures which quantify the errors that arise in collection of samples, or during analysis. Clear statement of such errors will lead to greater comparability between studies. The procedures recommended here define the limits to accuracy in relation to sample size, total faecal production, and laboratory methods. In conjunction with adequate pilot studies, these techniques will allow more effective design of ecological studies relating to carnivore diet and food consumption. CONTENTS Introduction. zyxwvutsrq . . . . . . . . . . . . Background . . . . . . . . . . . . Illustrative material. . . . . . . . . . Critique of procedures used by previous authors . Methods . . . . . . . . . . . . . Associated problems . . . . . . . . . Recommended procedures . . . . . . . . Collecting faecal samples . . . . . . . . Hygiene and safety . . . . . . . . . . Laboratory procedures . . . . . . . . Transformationsofthe raw data . . . . . Detecting differences in diet . . . . . . . Required sample size . . . . . . . . . Estimating sampling errors in quantitative data Presentation of results . . . . . . . . . Discussion . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . References . . . . . . . . . . . . . Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 . . . . . . . . . . . . . . . . . 97 . . . . . . . . . . . . . . . . . 98 . . . . . . . . . . . . . . . . . 98 . . . . . . . . . . . . . . . . . 98 . . . . . . . . . . . . . . . . . 99 . . . . . . . . . . . . . . . . . 108 . . . . . . . . . . . . . . . . . 108 . . . . . . . . . . . . . . . . . 108 . . . . . . . . . . . . . . . . . 109 . . . . . . . . . . . . . . . . . 110 . . . . . . . . . . . . . . . . . 113 . . . . . . . . . . . . . . . . . 113 . . . . . . . . . . . . . . . . . 116 . . . . . . . . . . . . . . . . . 117 . . . . . . . . . . . . . . . . . 117 . . . . . . . . . . . . . . . . . 118 . . . . . . . . . . . . . . . . . 118 . . . . . . . . . . . . . . . . . 119 INTRODUCTION Background Estimating the consumption of any particular prey type by canids and other carnivores depends upon reliable analysis of diet. The analysis of either stomach contents or faeces