What determines the success of the species identification? The identification of 10 deer (Cervidae) species in China based on multiple parameters of hair morphology

Abstract

Hair morphological structure is widely utilized for species identification based on the differentiation of scales and medullar pattern of mammal hairs. To determine what may influence the accuracy of identification using hair morphology, we measured and calculated 11 parameters of hair morphometry of 10 deer species in China. Our results suggested that the morphological parameters of deer hairs have extensive overlap within Cervidae species and we obtained a correct discriminant rate of 90.1% for 10 deer species. For the five sympatric deer species in the northeastern forests of China, 94.2% of hairs can be identified correctly, with a correct discriminant rate of 89.7% and 83.9% when the hair tip or root was absent, respectively. When both hair tip and root were absent, we obtained a correct discriminant rate of 73.6%. In addition, we obtained a correct discriminant rate of 97.9% for five sympatric deer species using a blind test approach to remove observer bias. Hair morphological characteristics are similar within the family or genus because of their close genetic relationships. Furthermore, species with similar living habitat conditions may have similar hair morphological structure. These factors influence discriminant capacity, and we evidently cannot identify them more accurately when using only one morphological parameter of hair. While understanding the above, our quantitative multi-parameter morphometric analyses successfully identified the hairs of deer, and likely have significant applications concerning further mammal species.