2. Type of gene action for grain yield.

Yield records were obtained on 45 single crosses in 1949 which represented five S₁ lines selected on the basis of high yield perfomance of test crosses and five S₁ lines selected for their low perfomance on test crosses. These two groups of lines had been selected in the same manner through two cycles of recurrent selection (after Hull, Amer. Soc. Agron. 37. 1945). The ten single crosses of high x high (having two high performing parental lines) averaged 67.4 bushels per acre; the ten single crosses of low x low averaged 46.0 bushels per acre; and the 25 single crosses of low x high averaged 55.5 bushels per acre. Four of the five highest performing single crosses were of the high x high parental combination. These data seem to be more nearly in agreement with the expectations on the basis of additive effects of dominant or partially dominant gene action, and do not support the hypothesis of over‑dominance of factors governing grain yield in corn (if the gene frequencies were at or near the equilibrium point in the starting population).

Paul H. Harvey