Recurrent Selection in Soybeans. II. Selection for Increased Percent Protein in Seeds1
- C. A. Brim and
- J. W. Burton2
Recurrent selection for increased percent protein in seeds was conducted in two soybean (Glycine max (L.) Merrill) populations to test the utility of the selection procedure. Population I was derived from a cross of two highly adapted experimental lines. Population II was generated from backcrosses of nine unadapted plant introductions with high percent seed protein to a highly adapted experimental line. Each population was subdivided into an A and a B population. The effective population size of populations IA and IIA was about three times larger than that of IB and IIB. A cycle of selection involved intermating selected lines, testing S~ lines from the matings, and selecting those with highest percent seed protein as parents for the next cycle. Selection progress was evaluated in replicated field trials using composite popula. tions of the parents chosen from the testing generation in each cycle as entries.
Percent seed protein increased significantly and linearly in all populations. Seed protein increased from 46~$ to 48.4% with six cycles of selection in population IA and from 46.4 to 47.6% with four cycles in population IB. In population IIA, protein increased from 42.8 to 46.1% with five cycles and from 43.2 to 45.9% with four cycles of selection in population IIB.
Yield of the last cyde in both populations IIA and IIB .was significantly lower than that for C0, yet the change in total protein produced per unit area was nonsignificant. In population IA, however, both yield and total protein produced increased significantly from C0 to C8. There was no significant change from C0 to C4 for yield or total protein in population IB. Despite the significant linear decrease in percent seed oil, the total oil produced per unit area decreased significantly in populations IIA and IIB only.
The results show that significant progress can be obtained from recurrent selection for protein in soybeans. Reduced effective population size and numbers of lines tested per cycle had little effect on progress.Please view the pdf by using the Full Text (PDF) link under 'View' to the left.
Copyright © . .