Computer-assisted Discrimination of Honeybee Subspecies Used for Breeding in Poland

The aim of this study was to develop a method for automated discrimination of three honeybee subspecies used for breeding in Poland: A. mellifera carnica, A. m. caucasica and A. m. mellifera. The method developed exploits characteristic venation of forewings. Samples of bees were collected from 1042 bee colonies of the three subspecies. From each of the colonies 20 right forewings were dissected and scanned with Nikon Coolscan 5000 ED. For every wing image the coordinates of 19 points were determined automatically using DrawWings software. Within each of the colonies the points were aligned using the Procrustes method, and the average configuration of points were calculated. Two methods of subspecie distinction are presented; one based on discriminant functions and the other based on canonical coefficients. Computer software was developed for automated discrimination of the three honeybee subspecies used for breeding in Poland.