Kamilar Jason M, Paciulli Lisa M
Department of Anthropology, Washington University, St. Louis, Missouri 63130-4899, USA.
Am J Primatol. 2008 Sep;70(9):816-27. doi: 10.1002/ajp.20553.
Species extinctions are nonrandom with some taxa appearing to possess traits that increase their extinction risk. In this study, eight predictors of extinction risk were used as independent variables to predict the IUCN category of a subfamily of specialized folivorous primates, the Colobinae. All data were transformed into phylogenetically independent contrasts and were analyzed using bivariate regressions, multiple regression, and a maximum likelihood approach using Akaike's Information Criterion to assess model performance. Once an outlier was removed from the data set, species that devote a smaller proportion of their diet to mature leaf consumption appear to be at a greater risk of extinction. Also, as female body mass increases, so does extinction risk. In contrast, as maximum latitude and the number of habitat types increase, extinction risk appears to decrease. These findings emphasize the importance of examining detailed dietary variation for predicting extinction risk at a relatively fine taxonomic scale and, consequently, may help improve conservation management.
物种灭绝并非随机发生,有些分类群似乎具有增加其灭绝风险的特征。在本研究中,八个灭绝风险预测因子被用作自变量,以预测一类特化的食叶灵长类动物——疣猴亚科的世界自然保护联盟(IUCN)分类等级。所有数据都被转换为系统发育独立对比,并使用双变量回归、多元回归以及使用赤池信息准则的最大似然法进行分析,以评估模型性能。一旦从数据集中去除一个异常值,饮食中用于食用成熟叶片的比例较小的物种似乎面临更大的灭绝风险。此外,随着雌性体重增加,灭绝风险也会增加。相反,随着最大纬度和栖息地类型数量的增加,灭绝风险似乎会降低。这些发现强调了在相对精细的分类尺度上检查详细的饮食差异对于预测灭绝风险的重要性,因此可能有助于改善保护管理。