五个箭筈豌豆品系基因型与环境互作效应及农艺性状稳定性

Genotype×environment interactions and consistency analysis for agronomic characteristics of five Vicia sativa lines

通过 2 a、4个区试点和 5种基因型及其互作效应研究 ,运用混合线性模型和 MINQU E(1)法 ,对箭豌豆包括株高在内的9个农艺性状可塑性进行评价 ,揭示了年份和区试点的生态环境效应、基因型与生态环境互作效应对各农艺性状的可塑性。结果表明 ,种子产量和千粒重等性状的基因型与生态环境互作效应达到了极显著水平 (p <0 .0 1和 p<0 .0 0 1)。生态环境分量 (年份、区试点、年份×区试点 )对各农艺性状的可塑性贡献较大 ,同时在不同生态环境间各农艺性状间差异达到了显著水平 (p <0 .0 5)。其中区试点分量对各农艺性状的可塑性贡献最大 ,各农艺性状在 4个区试点之间差异达到了显著水平 (p<0 .0 1) ,肃南和天祝的 2个区试点的牧草干重和种子产量等重要农艺性状的平均值显著大于另 2个区试点。各农艺性状在不同年份间差异达到了极显著水平 (p<0 .0 1) ,2 0 0 2年各农艺性状的平均值显著优于 2 0 0 1年。牧草干重和种子产量数量性状与气候因子的相关分析表明 ,5～ 8月份的月均温对牧草干重和种子产量的影响作用较大 ,较高的温度有利于牧草干重和种子产量的提高 ;7月份的降水量与牧草干重和种子产量存在一定程度的正相关关系。品系 2 556和 2 560在进行了基因型与环境互作效应稳定性评价后 ,4个区试点两年间都?

Vetch (Vicia sativa) is an important forage legume in alpine grassland. It is sown with oats (Avena sativa) to increase herbage yields and protein content. It has been frequently reported that the performance of vetch is inconsistent under differing environmental conditions. Genotype×environment interaction assessment is recommended for identification of genotypes with high performance over a wide range of environments. The genotype×environment interactions for vetch have not been reported previously. The objective of this study was to determine the genetic and environmental control of important agronomic characters by looking at five vetch lines over a two-year period in four locations in Gansu Province, China. Four breeding lines developed by the research team of Gansu Grassland Ecological Research Institute (GGERI) and a local Vicia angustifolia commercial cultivar named 333/A were employed in the experiments. Four representative alpine grassland experimental locations including Xiahe, Luqu, Sunan and Tianzhu in Gansu Province were used as trial sites. Seed of the five Vicia lines were sown into field plots at these four locations in both 2001and 2002. The agronomic traits of each line were measured, including plant height, herbage dry matter yield, seed yield, pods per plant, seeds per pod, dry root weight per plant, 1000 seed weight, days to mature, and days to first 20% flowering. Tests of normality of error variance were carried out using the SAS/UNIVARIATE procedures (SAS,1987), and Bartlett's test to test homogeneity of variance. PROC GLM of SAS/STAT was used to conduct the analysis of variance. PROC VARCOMP of SAS/STAT was used to estimate the variance components for agronomic characteristics. Genetic variation and genotype×environment interactions for important agronomic characteristics were estimated. The genotype×environment interaction effects from combined analysis were significant for three factor analysis of variance. Highly significant interactions were found between genotype and location, and between genotype and year for all characteristics measured. Environment showed the greatest contribution to all agronomic characteristics and there was significant difference between each trial site for all characteristics measured (p<0.01 or p<0.001).There were also significant differences between 2001 and 2002 (p<0.01). The correlation analysis of climatic factors with herbage dry matter yield and seed yield of different Vicia lines tested, showed monthly average temperature from May to August had a direct effect on herbage dry matter yield (DMY) and seed yield (SY), and that high DMY and SY can be obtained with high temperature. DMY and SY characters also showed a positive correlations with total rainfall in July. 2556 and 2560 lines showed more consistent performance with evaluation at four locations between 2001 and 2002, so are identified as outstanding genotypes with high performance over a wide range of environments.