Abstract: A F1 derived doubled haploid （DH） population from a cross between the German cuhivar Sollux and the Chinese cultivar Gaoyou was used in this study. In total,282 DH lines were evaluated in replicated field experiments in two locations,Xi ＇an and Hangzhou,respectively. Ten agronomic and seed traits including yield components, developmental stages, oil and protein content ,were analysed by classical genetic statistics and compared with the QTL mapping approach. The results indicated that all the traits showed typical quantitative characteristics in nature. There was no distinct difference between the QTL number and estimated number of genes in plant height, number of primary branches, silique length, days from sowing to flower and days from sowing to maturity, while the number of QTL identified for thousand seeds weight and number of seeds per silique were much less than that obtained from classical genetic analysis. According to estimated coefficient of skewness and kurtosis,duplicate interaction was deduced from plant height, days from sowing to flower and days from sowing to the end of flower. Partial correlations among the 10 traits revealed significantly positive correlation between plant height and number of seeds per silique,and between plant height and 1000 seeds weight as well;Significantly negative correlations among the three developmental stages and between oil content and seed protein content were in agreement with previous researches.

Key words: Brassica napus L., quantitative traits, DH population, genetic analysis, QTL mapping