Three selection strategies that utilize recurrent selection to increase protein yield in oats
Increases in protein production per unit land area for cereals (protein yield) generally have resulted from selection for high grain yield while maintaining or decreasing protein percentage. However, recent research showing grain yield and protein percentage may be inherited independently suggests that both of these traits can be utilized to enhance protein yield. In this study, three selection strategies were imposed upon different oat (Avena sativa L.) populations (i.e., lines of descent). The strategies were (a) protein yield selection per se (HGP); (b) selection for high protein yield due to high grain yield (HG); and (c) selection for high protein yield due to high levels of both protein content and grain yield (HP). S(,1) line recurrent selection was the breeding method employed, and each cycle took one year to complete;Gains in groat-protein yield were significant and similar in all lines of descent, ranging from 21 kg ha('-1) in HG and HGP to 27 kg ha('-1) in HP. Correlated increases in groat yield were 147 kg ha('-1) in HG, 95 kg ha('-1) in HP, and 112 kg ha('-1) in HGP. Groat-protein content increased significantly in HP (3.1 g kg('-1)), decreased in HG (-1.5 g kg('-1)), and remained constant in HGP. In all lines of descent, heritabilities for groat-protein yield, groat yield, and groat-protein content were moderately high ((GREATERTHEQ)0.41), and genetic variances generally were high and unchanged from C0 to C3, indicating the potential for continued progress from selection for single or multiple traits;Favorable correlated responses occurred for harvest index and bundle weight, whereas response in seed weight was dependent upon line of descent. Mild selection for heading date and plant height caused favorable response or no change for these traits, but reductions occurred for test weight and groat percentages in two lines of descent;These results indicate that groat yield and protein content can be increased concurrently in oats provided the proper germplasm source, selection pressure, and breeding methodology are applied. Further, S(,1) line recurrent selection can be effective in rapidly improving an oat population for the trait under selection without losses in performance for other traits.