IDENTIFICATION OF THE HONEYBEE SUBSPECIES FROM SOME ROMANIAN COUNTIES USING A SEMIAUTOMATIC SYSTEM FOR ANALYZING WINGS

Autor: Maria Rodica Gurau, Elena Buescu, Doina Danes
Rok vydání: 2018
Předmět:
Zdroj: CBU International Conference Proceedings. 6:1124-1128
ISSN: 1805-9961
1805-997X
DOI: 10.12955/cbup.v6.1304
Popis: Morphometric analysis of honeybee wings can be used to discriminate between honeybee subspecies. The classic morphometric methods used to identify honeybee subspecies are time consuming while geometric morphometric analysis are proven to be more efficient and rapid for the identification of honeybee subspecies. Even if the geometric morphometric analysis involves several steps such as: obtaining the image of the honeybee wing, processing the wings image (settlement of the 19 points), classification and validation, in terms of consumed time,it takes less than classic morphometric methodes and it is easier to apply. The aim of this study was to identify honeybee subspecies from the Romanian South-Eastern area, using the geometric morphometric method. To accomplish this, the semiautomated French system ApiClass was used. This system allows us to identify the honeybee subspecies based on the wing's image of the Apis mellifera honeybee worker. The program is using the recent approaches of geometric morphometry and is analyzing the wing considering 19 points corresponding to the intersections of the main veins of the bee forewings. These coordinates are processed by the system before being analyzed after which the system returns the probability of the analyzed honeybee belonging to one of the honeybee subspecies from its reference system. The system has more than 5000 honeybee wing references off the main lines and honeybee subspecies. Using this system we analyzed samples from the counties of Braila, Buzau, Galati and Ilfov. Each sample was composed from wings of 20 honeybees. Most of the analyzed samples were identified as hybrids without being possible to specify the type or the level of the hybridization: only three samples were identified as being Apis mellifera carnica presenting a homology degree over 90%.
Databáze: OpenAIRE
Externí odkaz: