ABSTRACT
Effective population size (N(e)) is the key parameter for predicting genetic drift associated with germplasm regeneration. A major factor reducing N(e) below the census population size (N(c)) is variation in seed production among plants in a given population. The objectives of this study were to estimate N(e)/N(c) associated with variation in seed production in three model wind pollinated, perennial grass species [Lolium perenne L., Festuca pratensis Huds., and Pseudoroegneria spicata (Pursh) Á. Löve] and to recommend cost effective sampling methodology to maximize N(e)/N(c) during seed regeneration. Three accessions of each species were grown at two field locations and variation in seed number among plants and mean seed production per plant used to estimate N(e)/N(c). Mean seeds per whole plant, standard deviations, and N(e)/N(c) differed among species, and accessions within species (P < 0.05). For whole plant samples, average N(e)/N(c) for each species differed with values of 0.42, 0.51, and 0.63 for L. perenne, F. pratensis, and P. spicata, respectively. However, average N(e)/N(c) based on two inflorescences per plant was 0.69, 0.88, and 0.86 for L. perenne, F. pratensis, and P. spicata, respectively, which was higher than that of whole plant samples. This higher N(e)/N(c) resulted from the elimination of the variation in inflorescence number per plant, a major source of variation in seed number among plants. The results showed the high potential for genetic drift in small regeneration populations. Increased plant populations and harvesting a constant number of inflorescences per plant are recommended as cost-effective methods to minimize genetic drift during regeneration of outcrossing grass germplasm.
