Toshiro Kinoshita and Takeo Kumagai
Plant Breeding Institute, Faculty of Agriculture, Hokkaido University, Sapporo, 060 Japan
Varietal differences in the tolerance to herbicides, benthiocarb (satan) and molinate have been reported in rice (Unno et al. 1977; Richard and Baker 1979). Quantitative genetic analysis of the reactions to these herbicides is reported in this paper.
Five strains, N-61, Surjamukhi, Modan, Bologona and Taichung 65, were used as the parents for 5x5 diallel cross. They are arranged above in the increasing order of tolerance. For treatments with herbicides, two concentration levels, namely, 0.25 and 0.5 ppm for benthiocarb and 1 and 2 ppm for molinate, were applied throughout the experiment. Bioassay of germinating F2 seeds was carried out in a greenhouse by using culture solutions containing a given concentration of the herbicides. Percent survival data recorded two weeks after treatment were transformed into sin-1√% and were subjected to diallel analysis.
Variance analysis of the survival percentages at the two concentration levels of each herbicide indicated that additive and dominance effects were both significant in all cases. Maternal effect was detected at 0.25 ppm of benthiocarb and at both concentrations of molinate. There was a tendency for the F2 plants from the crosses involving tolerant cultivars, Bologona and Taichung 65, as female parents to have a higher tolerance than the same F2 plants from the reciprocal crosses.
For further analysis to satisfy the assumptions made by Hayman (1954), only the data for 0.25 ppm of benthiocarb were used. The relationship of parent- offspring covariance (Wr) with array variance (Vr) suggested that the dominance effects were more prevalent than the additive effects, but there was little epistatic effect. Two strains, Bologona and Taichung 65 possessed more dominant genes for higher tolerance than other strains (Fig. 1). The estimates of heritabilities in the narrow and broad senses were 29.0% and 45.6% respectively. The number of effective factors were estimated to be 3.05 and 2.52 by two different ways of calculation.
The tolerance to benthiocarb and molinate were significantly correlated, suggesting that the same genes were concerned with both. The F2 and F3 lines from the reciprocal crosses of A-133 (Norin 9, tolerant) and I-32 (Karalath, susceptible) were used for further genic analaysis. The data on an individual basis contained a large amount of non-heritable variation in parental strains, and the F2 plants could not be classified into tolerant and susceptible classes. Therefore, F3 lines raised from non-treated F2 plants were tested for tolerance. The F3 lines of A-133 x I-32 treated with 0.25 ppm bentiocarb could be divided into tolerant and susceptible classes, and showed a good fit to a 3 : 1 ratio. This suggests that a dominant gene for tolerance is involved. In other cases, however, the F3 lines showed a continous variation. It seems that the tolerance to benthiocarb is controlled by a dominant gene and several other genes with smaller effects. Transgressive variations were observed in all F3 line-groups, suggesting the possibility of selection for higher herbicide tolerance.
Fig. 1 Regressions of Wr on Vr for survival percentages at 0.25 ppm benthiocarb treatment.
Hayman, B.I., 1954. The theory and analysis of diallel crosses. Genetics 39:789-804.
Richard, E.P. Jr. and J.B. Baker, 1979. Response of selected rice (Oryza sativa) lines to molinate. Weed Science 27: 219-223.
Unno, Y., T. Kinoshita and M. Takahashi, 1977. Differential responses to herbicides found between Japonica and Indica varieties of rice. Jpn. J. Breed. 27: 226.