Evaluation of the Potentialities of Wheat Crosses Based on Mean Parental and Early Generation Values

With a view to determining if early testing in wheat could be used as a basis for eliminating potentially poor yielding crosses, twelve intervarietal crosses of Triticum aestivum L. with their parents, F₁, F₂ and F₃ (20 families in each cross) generations were evaluated for grain yield, ear number, grain number, 100-grain weight, number of spikelets per ear, ear length, plant height and heading date.

The results appear to justify the conclusion that wheat varieties differ widely in combining ability for grain yield and other characters. The proportion of high yielding genotypes in the low yielding crosses will be less than in crosses with a higher average yield. The replicated yield trials of bulk crosses in the F₂ supplemented with the F₃ could be used for discarding certain crosses. Based on these considerations, of all the varieties used in the study, Rio Negro, Frontiera, Kenya Gular Pilot x K25—New Thatch, Supremo X Mentana, N.P. 790 and H.D.(52)-30 seemed to be better varieties for hybridization with N.P. 718 than Kenya C. 10854, Ridley, Timstein derivative 1495-A.I. 31-2-1, Kenya 338.Ac.2E.2.1:49:89, N.P. 723 and Pb.C. 591.

The use of F₁ generation as a means of determining the average yields in subsequent generations appeared to be of limited value.

The parental performance for characters such as ear number, grain number per ear, ear length, spikelet number per ear, plant height and heading date could be used with considerable advantage for predicting the performance of later generations. It is, therefore, suggested that careful choice of the parents, to be used in the crossing programme, should be made for these characters for obtaining desirable segregates in later generations.