Dental microwear is of special interest for two reasons. First, it has been proposed that specific dental microwear patterns are associated with specific diets and therefore that the diets of extinct forms may be deduced by analysis of microwear. Second, it has been suggested that the geometry of wear striations indicates the direction of masticatory movement. We tested these ideas by analyzing microwear of laboratory animals fed different diets. Twelve American opossums (Didelphis marsupialis) were fed soft cat food for 90 days. Two control animals were fed only this base diet, five animals had plant fiber added to their diet, four animals had chitin added to their diet, and one animal had fine ground pumice added to its diet (for the last 30 days of the feeding period). We examined the wear surface below the paracristid on the M3 and M4 of each animal by SEM. No microwear pattern differences were observed on the plant fiber-fed, chitin-fed, or control animal's molars. The pumice-fed opossum had a distinct microwear pattern with many parallel striations, resembling those found on the teeth of grass-eating hyraxes (Walker et al., 1978). These results suggest that (1) exogenous grit (this study) or plant parts containing opaline phytoliths (Walker et al., 1978) produce similar microwear patterns, and (2) the diets of extinct forms cannot always be deduced by the analysis of dental microwear. The absence of fine parallel striations on teeth of Sivapithecus examined by us suggests that grass parts were not an important part of its diet and that it avoided dietary fine grit. Furthermore, we found striations on opossum molars with deep, broad heads and shallow, narrow tails oriented in opposite directions on the the same Phase I wear facet. This suggests that the geometry of striations on Phase I wear facets does not allow one to determine the direction of masticatory movement.