Population improvement in alfalfa: fertility and S1 forage yield performance in original and improved populations

Inbreeding in alfalfa (Medicago sativa L.) may reduce heterozygosity and permit population improvement through selection of favorable alleles and elimination of deleterious ones. Two-allele autotetraploid populations (TAPs) are suitable for studying the effectiveness of selection because they are derived from chromosome doubling of heteroxygous diploid plants and therefore have a maximum of two alleles and a single allelic interaction per locus. The effectiveness of inbreeding and selection was tested by comparing TAPs that had undergone inbreeding and selection with their original counterparts. The original (designated OGDC) and improved (designated AGDC) populations were produced by intercrossing two single-cross lines from the original and improved TAPs, respectively. Self fertility, female cross fertility, and S1 forage yield performance were compared in original and improved populations. Self fertility increased by 57% in the AGDC compared with the OGDC, whereas female cross fertility was reduced by 12% in this population. A higher proportion of plants in the AGDC population produced sufficient S1 seeds (82%) for field evaluation than did plants in the OGDC population (57%). The mean forage yield of the AGDC S1 families was significantly higher (13.3%) than their comparative OGDC S1 families. Improvement in self fertility and S1 forage yield by inbreeding and selection was consistent with population improvement due to elimination of deleterious alleles and a corresponding increase in the frequency of favorable ones. However, the concurrent decrease in female cross fertility was inconsistent with this hypothesis. Therefore, selection for cross fertility may be necessary during inbreeding in alfalfa