Comparison of two faecal analysis techniques to assess Formosan pangolin Manis pentadactyla pentadactyla diet

The diet of myrmecophagous mammals is usually studied by investigating the chitinous remains of termites and ants obtained from stomachs, guts, and faeces; however, this process is time and labour intensive. Moreover, faeces of obligate myrmecophagous mammals contain other materials such as soil, sand, rocks, and plant matter, which hinder the accurate and efficient identification of faeces' contents; this makes estimating the biomass of each prey species difficult. This study tested non-filtering and filtering faecal analysis procedures and compared their identification accuracy and efficiency in the analysis of Formosan pangolin Manis pentadactyla pentadactyla faeces. In the non-filtering procedure, 1 g of dried faeces was distributed over an area greater than 360 cm(2) to be able to see most ant capsules and termite mandibles. In the filtering procedure, 0.5 cm(3) of filtered chitinous remains of termites and ants was spread over a 45 cm(2) area to obtain a high resolution of the diet composition. The filtering procedure can sieve out insect remains from the debris, which facilitates further identification of the insect chitin. Using the two proposed faecal analysis procedures, the prey species composition and the number of prey remains could be obtained. Our results provide a quantitative method to evaluate the diversity of wild pangolin's diets for determining their ecological requirements and developing in-situ and ex-situ conservation plans.