The identification of the highest yielding cultivar for a specific environment on the basis of both genotype and genotype x environment interaction could be useful for breeders and producers since yield estimation based only on genotype and environment effects are insufficient. The trial was carried out during six growing seasons using eight registered cultivars of rapeseed developed at European breeding stations and 11 experimental lines developed in the Institute of Field and Vegetable Crops, Novi Sad, Serbia. The objective of the study was to identify genotypes with most stable oil yield by using combination of three parameters: ecovalence (wi), regression coefficient (bi) and deviations mean square (s2di) (first model) and AMMI model analysis (second model). Average oil yield per area of experimental genotypes was higher comparing to registered genotypes. According to the first model, seven experimental lines and two registered varieties (Falcon and Banacanka) were estimated as stable and wide adaptable genotypes. A complete positive and highly significant correlation was estimated between wi and s2 di that implies that both of these parameters could be used independently. According to the AMMI models, in the environmental conditions of Northern Serbia, the genotypes UM-8 and UM-11 were the most stable and with high average oil yield. Such genotypes can be implemented in future breeding programs and recommended for growing in South Eastern Europe.

Brassica napus L.; cultivars; stability; AMMI model; oil yield

Published by University of Zagreb, Faculty of Agriculture, Zagreb, CROATIA
Sponsored by Ministry Science and Education of the Republic of Croatia