Recurrent selection procedures are becoming more popular as a means of population improvement in self-pollinated crops. Four cycles of recurrent selection for yield in a broad-based peanut (Arachis hypogaea L.) population have been completed. The objectives of this study were to determine (a) progress made after four cycles of recurrent selection for fruit yield and (b) if sufficient variability remained in the population to permit further improvements in yield from additional cycles of selection. The 40 highest yielding lines from 100 random-paired matings, were intercrossed to produce each successive cycle, resulting in a selection intensity of 40%. Bulked seed of the parental lines from each cycle were evaluated for yield to determine progress made after four cycles of selection. The observed variation in yield among entries was mainly due to differences among cycle means. Differences among cycle means were also detected for pod length, but not for any of the other fruit traits measured. The response to selection for yield was linear with a significant (p = 0.01) and positive regression coefficient (b = 190.7 kg ha-1). Selection based on fruit yield did not significantly change the means for any of the fruit traits measured. Significant estimates of genetic variability among entries within the cycle 4 population for yield and all fruit traits measured suggest that continued progress from additional cycles of selection should be expected in this population. Greater progress from selection may have resulted had a higher selection intensity been applied. However, estimates of genetic variability would be expected to decrease with each cycle of selection as the selection intensity increases and as the genetic base of the original population narrows. In the use of a broad-based population and a relatively low selection intensity, we chose to accept a lower rate of progress/cycle for yield in order to maintain a greater level of genetic variability within the recurrent selection population.

This content is only available as a PDF.

Author notes

1This research was partially funded by the North Carolina Agricultural Research Service