Sixteen soybean [Glycine max (L.) Merr.] cultivars were evaluated for their response to the rice herbicide propanil (3′,4′,-dichloropropionanilide) applied at two growth stages. Application of 3.4 kg/ha resulted in more injury than 0.6 kg/ha. Application at the stage of growth when three nodes on the main stem had completely unrolled leaves (V3) resulted in more injury than when eight nodes had completely unrolled leaves (V8). As determined by seed yield when treated at 3.4 kg/ha propanil at the V3 stage of growth, ‘Pickett 71′, ‘McNair 600′, ‘Tracy’, and ‘Bragg’ were the most tolerant; ‘Hutton’, ‘Hampton 266A’, ‘Bossier’, ‘Ts 74-34′, ‘Semmes', and ‘Forrest’ were moderately sensitive and Dowling’, ‘Davis', ‘Mack’, ‘Dare’, ‘Ranson’, and ‘McNair 800’ were the most sensitive to propanil injury.

Combining ability analysis for three stem borer resistance parameters, i.e. “dead hearts”, leaf injury and stem tunnelling, revealed that both additive and non-additive types of gene effects were involved in the inheritance of these attributes. The specific combining ability (sca)/general combining ability (gca) variance ratios further indicated predominance of additive gene effects for stem tunnelling and non-additive effects for “dead hearts” and leaf injury. The genotypes, IS 2123 and IS 5469 are good general combiners for all the three parameters. Among F1 hybrids, IS 1054 × IS 5469 and IS 18551 × IS 2123 expressed good sca for all the parameters of stem borer resistance. The use of IS 2123 and IS 5469 parental genotypes in the future breeding programmes is suggested.

Twelve cultivares were grown along with released high yielding hybrid CSH-1 (susceptible to stem-borer) to investigate their relative resistance to stem-borer at three locations. Results obtained have established that the 12 cultivars possess moderate levels of resistance to stem-borer. The damage suffered by them was significantly lower than CSH-1. Further, four cultivars viz. E601, E 602, E 603 and E 604 suffered less damage.

The stability of the resistance was worked out by following the model of Eberhart and Russell (1966) [Crop Sci. 6, 36–40]. The means of various cultivars were calculated and regression coefficients (bi) were tested against unity to locate most adaptable cultivar for borer damage. The cultivar × environment interactions for leaf injury and stem tunnelling (two parameters used for assessing stem-borer damage) were significant and in most of the cases were due to linear response of different cultivars. Cultivar E 601 proved to be most promising and stable for resistance to stem-borer.