Do lizard communities partition the trophic niche? A worldwide meta‐analysis using null models

摘要: Lizards have been widely used as model organisms in community ecology over the past 30 years. I reviewed more than 50 studies from literature on of lizards worldwide, with a focus 1990-2007. determined if there is support for hypotheses that many lizard communities 1) are non-randomly organized along trophic niche dimension, and 2) partition available food resource to minimize interspecific competition. data number prey items percentage volume, calculate dietary overlap among species. compared true datasets randomly generate new produced by 3 x 10 4 Monte Carlo permutations using two algorithms (RA2 RA3). The great majority (more 80%) were dimension RA2 or RA3; frequency occurrence detecting non-random structure did not differ significantly between based either volume. Thus, usually do axis. Concerning items, logistic regression models showed presence dataset depend species, method employed obtain data, matrix size, location (continent), but instead depended whether was tropical. also tropical vs non-tropical origin, marginally dependent continent (South America favoured identifying dataset) size. In general, often uncovered RA3. Overall, conclude unlikely resources, does allow accurate identification (at least at genus level), risk obtaining false null results about structure. This because axis genuinely rare inappropriate methods may erroneously lead this conclusion. Therefore, conclusions utilizing stomach dissection source diet (still common literature) should be cautiously considered ethical reasons it difficult find any difference stomach-dissection fecal-pellet when assessing datasets. My minor differences probability item volume lizards. biases, both statistical biological, associated meta-analysis discussed.