Factorial analysis of self-incompatibility in alfalfa

Self-incompatibility (seeds/pod and seeds/floret tripped) was evaluated in the growth chamber (27°C) using a 6 × 6 factorial arrangement. Parents represented a broad range of self-compatibility and were selected on the basis of stability of autogamous seed production in the greenhouse and growth chamber. Additive genetic variation appeared to be most important, indicating that a mass selection scheme such as recurrent phenotypic selection would be effective in increasing levels of self-incompatibility in at least some alfalfa populations. Narrow-sense heritability estimates for seeds/pod and seeds/floret tripped were 0.23 (SE = 0.09) and 0.28 (SE = 0.06), respectively. There is a direct relationship between autogamous seed production and pollen production, however, results of this experiment indicate that there may be considerable variation in pollen production, germination and vigor among self-incompatible clones. Increased seed set resulting from self-incompatible × self-incompatible crosses could be a function of greater gamete selection in the self-incompatible parents (due to deleterious genes), improved pollen germination, and hybrid vigor in pollen tube growth. Key words: Medicago sativa L., autogamy, allogamy, hybrid