RECURRENT SELECTION IN OAT - STABILITY OF YIELD AND CHANGES IN UNSELECTED TRAITS

Recurrent selection, primarily for grain yield in oat (Avena sativa L. ), has been practiced at a single location for four cycles. Previous e valuations of this program had indicated that grain yield had been inc reased and that correlated changes, some undesirable, had also occurre d. In some studies in other species, yield increases have come at the expense of stable productivity. The objectives of this study were to e valuate the stability of grain yield and to assess changes in nine uns elected traits to determine the potential of lines from this program f or cultivar development. Grain yield was measured in 23 trials during a 5-yr period to evaluate gain from selection and compare the average yield stability of the 12 initial parents with that of the 21 fourth c ycle parents (lines selected from the third cycle). Seven agronomic an d seed quality traits (days to heading, plant height, resistance to lo dging, test weight, groat percentage, percentage groat protein, and pe rcentage groat oil) were evaluated in 12 trials to examine correlated response to yield selection. Barley yellow dwarf (BYD) and loose smut reactions were evaluated in inoculated disease nurseries. Gain from fo ur cycles of yield selection ranged from - 2.0 to 44.6% in the individ ual environments, and was usually significantly positive. Also, based on regression stability statistics, selection for grain yield increase d the ability of the selected materials to respond to more favorable e nvironments. Finally, there were significant correlated responses for all but two of the unselected traits, groat percent and BYD reaction; however, even though most of the changes were undesirable they were re latively small and should not detract from the potential of the fourth cycle parents in a cultivar development program.