运用多元相关、逐步回归和通径分析法对一些大穗小麦的产量构成因素等农艺性状进行研究。结果表明：①单位面积冬季总茎数、穗数和群体平均单穗粒重与产量呈显著正相关，相关系数分别为０．５９２＊，０．４６４＊和０．７３０＊，３个性状均被选入最优方程；②各产量构成因素对产量的直接贡献是：单位面积穗数（Ｐｙ＝１．３５６）＞群体平均单穗粒重（Ｐｙ＝１．２１８）＞每穗粒数（Ｐｙ＝０．９０７）＞千粒重（Ｐｙ＝０．３９５）＞每穗小穗数（Ｐｙ＝－０．３７７）；③提出将群体平均单穗粒重作为衡量大穗小麦的综合量化指标，将“穗足、穗大、穗匀”作为大穗小麦产量结构选育目标 The agronomic traits such as yield components of some large-spike wheat were studied by multivariate correlation, stepwise regression and path analysis. The results showed that: ① The total number of stems per plant, the number of spike per panicle and the average grain weight per spike were positively correlated with the yield, the correlation coefficients were 0.592 *, 0.464 * and 0.730 *, respectively Was selected into the optimal equation. (2) The direct contribution of yield components to yield was: panicles per unit area (Py = 1.356)> average grain weight per panicle (Py = 1.218)> grains per panicle Py = 0.907)> 1000-grain weight (Py = 0.395)> spikelet number per panicle (Py = -0.377). ③The average grain weight per spike was proposed as a comprehensive quantitative index for evaluating large- “Spike feet, spike large, spike uniform” as a large-spike wheat yield structure breeding target