GENETIC CONTROL, BREEDING AND PREDICTION OF MATURITY IN WINTER WHEAT

DONALD PAUL AVEY, Purdue University

Abstract

The mode of inheritance of maturity was studied in winter wheat (Triticum aestivum L. em Thell) using both an early and a late generation analysis. The early generation analysis measured the combining ability of hybrids resulting from the diallel mating of nine diverse parents and was carried out in 1977 and 1979. The late generation analysis was a variance component analysis of approximately 400 advanced-generation lines from each of three crosses planted at two locations in 1979. The combining ability analysis showed the GCA to be highly significant over years while the SCA effects were not significant when averaged over years but were highly significant when the 1979 data were considered alone. In general, additive and/or additive X additive effects would seem to be the predominant controls. There is some indication that genotypes may react somewhat differently under different sets of environmental conditions. The advanced generation analysis showed additive effects to be significant for Aobachomugi/Doublecrop, additive and dominance effects to be significant for Aobachomugi/P6879, and additive X additive effects to be significant for Doublecrop/P6879. These data taken together indicate that dominance may be of some consequence in specific crosses, but in general, the fixable components are the main controls. Further examination of the data shows there to be some major as well as minor genes acting in this system. A recurrent selection program for earliness was effected using 12 two-, three-, and four-way hybrids as parents. Three cycles of intercrossing were carried out in the greenhouse with F(,1) plants being used in the crosses. F(,1) progeny from each of the three cycles, F(,2) progeny from plants selected to intercross in each cycle, parent F(,1)'s hybrids, and original parents were planted at two locations in the fall of 1979. The largest gain by selection was seen in the first cycle. Additional progress was made in the following two cycles, but at a slower rate. Realized heritability estimates were highest for the F(,1) data of Cycle 1, followed by the F(,2) data of Cycle 1. The combined heritabilities of Cycle 2 and 3 were lower and comparable for both the F(,1) and F(,2). These data imply that the first cycle operated mainly on major genes which showed some nonadditive gene action while the latter cycles operated mainly on minor genes governed by additive and/or additive X additive effects. Several prediction models for days to heading based on growing degrees were tried and two were found which yielded acceptable results. The first was by accumulating growing degrees after March 1 using a base temperature of 36F and a maximum limit of 70F. The second was by weighting the fall growing degrees, based on soil temperature, by a factor of 0.3 and the spring growing degrees, based on air temperature, by a factor of 0.7. A base temperature of 40F and a maximum limit of 80F produced the best results in this latter method. Using both of these approaches, estimates were obtained which predicted heading within about two or three days of the mean over the three years studied. The latter method further indicates that the spring growing degrees are about 2 1/3 times more effective in eliciting heading than fall growing degrees.

Degree

Ph.D.

Subject Area

Agronomy

COinS