A Comparison of Intermating and Selfing Following Selection for Heading Date in Two Diverse Winter Wheat Crosses1

Abstract

Response to two cycles of mass selection for heading date involving selfing or intermating was studied in two diverse winter wheat (Triticum aestivum L. em Thell.) crosses (I and II). Selection was bidirectional resulting in early and late‐heading populations for each mating system, cycle and cross. Generation means and variances within crosses were used in a genetic analysis of heading date. Additive effects of genes were an important source of variation in Cross I. Predicted response to selection for heading date was 7.1 and 9.1 days/cycle for 1981 and 1982, respectively. Heading date in Cross II was affected primarily by non‐additive genetic effects. For this cross, progress from selection for heading date was predicted to be 6.9 days/cycle in 1981, and 1.4 days/cycle in 1982. Observed response to bidirectional selection for heading date in Cross I showed gains of −4.8 and 5.5 days/cycle with selfing and −4.2 and 5.1 days/cycle with intermating in 1982. Response to selection for heading date with intermating in Cross II was superior to selfing for earlier heading date (6.0 vs. 5.0 days/cycle) and inferior to selfing for later heading date (0.9 vs. 1.9 days/cycle) in 1982. The latter results were ascribed to an accumulation of minor genes by intermating and rapid fixation of recessive genes due to selfing, for early and late selection, respectively. Recurrent selection with intermating was not more efficient in changing the heading dates of the two populations than selection followed by selfing. The effect of bidirectional selection for heading date under two systems of mating on seven other traits was also studied. Selection for early heading date under both systems of mating in Crosses I and II resulted in shorter plant stature, fewer spikes/plant, and lower biological yield (P = 0.05) at two locations in 1982. Bidirectional selection for heading date under both mating systems had no significant effect on grain yield, 100‐kernel weight, kernels/spike, and harvest index in Crosses I and II.