ADDITIVE, DOMINANT, AND EPISTATIC EFFECTS FOR MAIZE GRAIN-YIELD IN ACID AND NONACID SOILS

Acid soils severely reduce maize (Zea mays L.) yield in the tropics. B reeding for tolerance to soil acidity provides a permanent, environmen tally friendly, and inexpensive solution to the problem. This study wa s carried out to determine the relative importance of additive, domina nt, and epistatic effects on maize grain-yields in different tropical genotypes. Divergent selection in three populations (SA4, SA5, and SA7 ) provided inbred lines tolerant or sensitive to acid soils. The toler ant and sensitive lines from each population were used to obtain the F -1, F-2, F-3, back-crosses, second back-crosses, and selfed back-cross generations. In addition, the tolerant lines from SA4 and SA5 were cr ossed with a sensitive line from the Tuxpeno Sequia population, from w hich the same generations were also derived. All generations from each of the five sets of crosses were evaluated in three acid-soil environ ments and one non-acid-soil environment. A generation-mean analysis wa s performed on each set for yield. The sequential sum of squares assoc iated with additive, dominance, and digenic epistatic effects were use d to estimate the relative importance of each genetic effect. Epistasi s was not important in any set in the non-acid-soil environment, with dominance accounting for 80.76% of the total variation among generatio n means across sets. In acid-soil environments, epistasis was more imp ortant. The relative importance of digenic epistasis was greater in th ose evaluations with large experimental errors. The tolerant line from population SA5 was prone to severe root lodging, suggesting a very po or root system. Apparently, the tolerance to soil acidity in this line is not associated with a large root system.