小麦收获指数和其它几个农艺性状的基因效应分析

AN ANALYSIS OF GENETIC EFFECTS ON HARVEST INDEX AND SEVERAL OTHER AGRONOMIC CHARACTERISTICS OF WHEAT

本文利用Ｈａｙｍａｎ的双列杂交法对９个产量性状进行了遗传分析。结果表明：籽粒产量、生物学产量、每穗小穗数和穗长存在上位性效应，收获指数、每穗粒数、株高、千粒重和每穗不孕小穗数符合加显遗传模型。株高以加性效应为主，表现为部分显性；收获指数、每穗粒数、千粒重和每穗不孕小穗数以显性效应为主，表现为超显性。多数产量性状的显性基因为增效基因。株高的狭义遗传率最高，其次为收获指数和千粒重。株高在早代选择就有效，收获指数和千粒重次之。９２４２５是改良收获指数和千粒重的较好亲本，苏８５２７是改良每穗粒数的较好亲本，９２０４１是提高千粒重的较好亲本。

The genetic trait of yield components in wheat was studied in a 6×6 diallel cross set involving six wheat varieties. Estimates of genetic parameters following Hayman's method showed that A D model was fitted for harvest index, plant height, kernel number per spike, 1000 kernel weight and spikelets of nonfertilization per spike, the epistatic existed for grain yield per plant, biomass, spikelets per spike and spike length. The additive gene action was more important than dominant gene action for plant height, which was inherited as a partialdominant character conditioned by one or two blocks of genes. HI, kernel number per spike, 1000 kernel weight and spikelets of nonfertilization per spike were inherited as an overdominant character conditioned by one or more blocks of gene. The correlaton analysis between Wr+Vr and Yr indicated that high value was dominant for most yield components. The estimates of narrow sense heritability for plant height was the highest one in five traits, HI and 1000 kernel weight were secondary one. Early selection was more efficient for improving plant height than for HI and 1000 kernel weight. Kernel number per spike and spikelets of nonfertilization may be selected in early generation.