Measuring the inefficiency of allowing self-pollinated alfalfa progeny into breeding nurseries

Authors: Riday, Heathcliffe; JOHNSON, DAVID - Retired Non ARS Employee; DARLING, MARK - Retired Non ARS Employee

Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/9/2016
Publication Date: 10/12/2016
Citation: Riday, H., Johnson, D.W., Darling, M.E. 2016. Measuring the inefficiency of allowing self-pollinated alfalfa progeny into breeding nurseries. Crop Science. 57:161-167.

Interpretive Summary: Alfalfa is treated as an outcrossing forage species (cross pollination between plants in the field) when in fact there is a substantial incidence of self-pollination in the species (an observed average of 30%). This unique study identified selfed individuals in an alfalfa breeding nursery and determined their performance relative to outcross plants in the nursery. It was found that the selfed alfalfa plants were clearly inferior and unlikely to be selected by alfalfa breeders using common alfalfa breeding protocols. The estimated maximum loss in genetic gain due to the presence of selfed progeny in breeding nurseries was estimated at a 10% genetic gain reduction. This study raises awareness among alfalfa breeders of hitherto ignored incidence of selfed progeny in their alfalfa breeding programs and the impact this has on breeding efficiency.

Technical Abstract: Although alfalfa is generally considered an outcrossing species the incidence of self-pollination during insect pollination can be substantial with an observed average 30% selfing rate. Despite this, alfalfa breeders utilizing space plant evaluation nurseries make no attempt to discriminate self-pollination derived progeny (selfs) from outcross progeny. Due to alfalfa‘s substantial inbreeding, the selfed individuals have much lower breeder utility during selection. This study phenotyped an alfalfa space plant breeding nursery derived from a single polycross, in which selfed and outcross progeny had been identified using DNA markers for the following phenotypic traits: plant vigor, fall height, fall dormancy, 1st and 2nd winter survival, plant lodging, and profusion of flowering. Trait means, phenotypic and additive variances, and narrow sense heritabilities on a halfsib mean basis were determined for selfed and outcross progeny. For each trait, inbreeding depression was determined as the percent performance of the selfed progeny relative to outcross progeny. For most traits inbreeding depression was severe with approximately 80% reductions in performance. Selfed progeny for most traits had reduced phenotypic variances compared to outcross progeny. No differences in estimated additive variance and heritabilities were observed between the two types of progeny. Based on observed phenotypic distribution at common selection intensities from 1% to 10% few, if any, of the selfed progeny would be selected if the selfing rate was 30%. The estimated maximum loss in genetic gain due to the presence of selfed progeny in breeding nurseries for selection intensities from 1% to 10% was estimated at a 10% genetic gain reduction. We conclude that selfed progeny in breeding nurseries are an impediment to progress and if possible should be minimized or eliminated.