Two techniques for determining the proportions of different plant species eaten by the brushtailed opossum (Trichosurus vulpecula) were compared. Both techniques-bacterial degradation and chemical maceration of feces-produced similar results, and confirmed that a correction factor is needed to allow for the differential loss of cuticle of some plant foods during digestion or preparation of feces for microscopic examination. J. WILDL. MANAGE. 43(2):468-473 A reliable method for measuring the species composition and proportions of foods in the diet of herbivores is required; this is especially important where the management of animals or vegetation is contemplated. All the techniques used have drawbacks. Directly observing the feeding animals or the resulting browsed or grazed vegetation to assess species and amounts eaten does not always give accurate results (Wallmo et al. 1973). Examining the macroor microscopic contents of stomachs or rumens is not practicable on live animals, and the results may be unreliable (Westoby et al. 1976). Examining the microscopic fragments in feces is potentially 1 of the best ways, but this can be difficult, and some workers doubt the accuracy of the results obtained (Slater and Jones 1971). However, an advantage of this method is the ease of collecting and storing material. Fresh fecal samples can be kept refrigerated in air-tight containers, or stored in a simple preservative such as 70% ethanol. One of the problems associated with fecal analysis is that of identifying the plant fragments seen on the microscope slide. For predominantly leaf-eating animals, including most ungulates and the brush-tailed opossum, this is simplified by the characteristic shape and arrangement of the epidermal cells of the leaves of each plant species. These features are molded on the overlying cutin, which is indigestible and passes through the gut of herbivores. Some cuticles, however, may break down physically to such an extent that they cannot be recognized. A number of herbs, grasses, and more fragile-leaved tree species have long cells of the epidermis that are too poorly defined to produce a diagnostic cuticle, but they almost always have characteristic stomata, cell inclusions, glands, or hairs, particularly on the abaxial surface of the leaf. Thus, in most cases identification is possible with the aid of a reference collection of representative leaf cuticles. Identifying fragments of food items other than leaves may not be so straightforward. Seeds may be identified either macroor microscopically, and fragments of flowers, petioles, stems, fruit, and other plant parts may sometimes be identified to species. However, even when the available or likely food plants are known, the preparation of reference material to include all these plant parts is very time consuming and often impracticable. 468 J. Wildl. Manage. 43(2):1979 This content downloaded from 157.55.39.181 on Thu, 29 Sep 2016 06:16:05 UTC All use subject to http://about.jstor.org/terms FECAL ANALYSIS OF HERBIVORE DIET Fitzgerald and Waddington 469 A more intractable problem with food studies using fecal analysis is determining whether the proportions in which various species are represented on the microscope slides are similar to the proportions in which they were eaten, or whether the digestive processes of the animal and the methods of preparing material have altered the proportions. Various techniques have been used to prepare fecal material for microscopic examination. For example, Dusi (1949) and Voth and Black (1973) used chloral hydrate solution to clear the cuticle fragments. Martin (1955) and Stewart (1967) used nitric acid to macerate the fecal samples, removing mesophyll while leaving cuticles identifiable, while Dunnet et al. (1973) used a mixture of nitric acid and chromic acid. Williams (1969), after initial mechanical preparation of fecal samples, added boiling water and finally sodium hypochlorite (domestic bleach). Cuticles can also be separated by macerating the plant material with bacteria. This method has been used by Skoss (1955) to study the structure and composition of cuticles, by Luow et al. (1949) on rumen samples, and by Rogerson et al. (1976) on feces from sheep. Rogerson et al. claimed that this is a simpler and more effective technique than chemical preparation, and that it gave repeatable results in terms of botanical composition. To test this claim, we attempted to compare the bacterial technique with the chemical one used by Dunnet et al. (1973) and Fitzgerald (1976) in studies of the diet of the brush-tailed opossum. We also took the opportunity to validate, for each plant species eaten, the use of predetermined correction factors which take into account both the destructive effect of nitric-chromic acid and the digestive processes of the animals. We are grateful to R. E. Brockie, D. G. Dawson, M. R. Rudge, and B. M. Fitzgerald for help with this paper.
Read full abstract