Recent work on the niche-variation model has yielded results which might lead to doubts concerning the value of the model. The analyses presented here, however, appear to provide more rigorous tests of the model, and two distinct approaches provide support for it. The first test compares the variability of species that are in the same community but have differing abundances. Common species generally have wider niches than less common ones, and therefore measures of variation should be roughly directly proportional to measures of abundance. Abundance and the coefficients of variation for bill length show significant positive correlations for 15 passerine species breeding on the Tres Marias Islands. A second test deals with the expectation that niche width is roughly inversely proportional to the amount of sympatry and hence actual and potential competition a population experiences with closely related forms. The amount of sympatry is quantified with an "index of sympatry" computed by assigning certain poin...