There are two endemic subspecies of western honey bees (Apis mellifera L.) in the Republic of South Africa (RSA), A.m. capensis and A.m. scutellata. They have traditionally been identified using morphometric characteristics, but geometric morphometric data from honey bee wings are easier to collect, possibly making them a useful alternative for identifying these subspecies. We compared the accuracy of both morphometric and geometric morphometric methods using linear discriminant and classification and regression tree analyses. We found that using geometric wing shape data from both forewings and hindwings resulted in a lower classification accuracy (73.7%) than did using models derived from the full set of standard morphometric data (97% accurate) in cross-validation. Tergite color and average ovariole number were the most important features for discriminating between the two subspecies. Finally, we used Kreiger interpolation to construct maps illustrating probable distributions of A.m. capensis and A.m. scutellata in the RSA.
CITATION STYLE
Bustamante, T., Baiser, B., & Ellis, J. D. (2020). Comparing classical and geometric morphometric methods to discriminate between the South African honey bee subspecies Apis mellifera scutellata and Apis mellifera capensis (Hymenoptera: Apidae). Apidologie, 51(1), 123–136. https://doi.org/10.1007/s13592-019-00651-6
Mendeley helps you to discover research relevant for your work.