Rarity, specialization and extinction in primates

Aim To determine and explain biological traits that distinguish rare from common primate taxa. Location Africa, Americas, Asia, Madagascar. Methods We compare the biology of rare primate taxa with the biology of common taxa. Rarity is defined by (1) small size of geographic range; (2) small geographic range plus low local population density; and (3) small geographic range plus low local density plus narrow habitat specificity. After a linear comparison of size of geographic range with various biological traits, globally and by realm, extremes of rarity and commonness per realm are identified, and then combined for a global analysis. Tests are done both with genera treated as independent data points ( n =62), and also with phylogenetic control by use of an independent contrasts test. Extinction risk in vertebrates, including primates, often correlates with high resource requirements, slow population recovery rate, and specialization. The three indices of rarity are therefore compared with these three general traits. Measures of resource use are body mass, local density, annual range size, and group size; of recovery rate, interbirth interval, and maximum intrinsic rate of natural population increase; and of degree of specialization, variety of diet, of habitats, maximum latitude, and morphological variety. All data come from the literature. Because several measures are compared, probabilities are Bonferroni corrected. Results If rarity in primates correlates with any biological attribute, it consistently correlates with only measures of specialization, and not with measures of high resource use, or slow population recovery rate. Without phylogenetic correction, the first two indices of rarity associate significantly with all four measures of specialization, and the third with maximum latitude. With phylogenetic correction, the first index still associates with all four, the second with two (maximum latitude, number of species per genus), and the third shows no significant associations. While the four measures of specialization are strongly interrelated, stepwise regressions on geographic range indicate that maximum latitude has the strongest effect, followed by dietary variety and number of species per genus and, finally, habitat variety. Main conclusions The most commonly demonstrated traits of susceptibility to extinction are those of high resource use, slow recovery rate, and specialization. Yet, while rarity (almost however, it is defined) is an inevitable precursor to extinction, specialization is the only trait found to correlate with rarity in this study. We cannot explain this apparent contradiction.