Skip to main content
ARS Home » Midwest Area » West Lafayette, Indiana » Crop Production and Pest Control Research » Research » Publications at this Location » Publication #86813

Title: INCREASING SEED PROTEIN IN SOYBEAN [GLYCINE MAX (L.) MERR.] WITH EIGHT CYCLES OF RECURRENT SELECTION

Author
item Wilcox, James

Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/16/1998
Publication Date: N/A
Citation: N/A

Interpretive Summary: Increasing the amount of protein in soybean seeds would increase the nutritive value of soybean meal as a livestock feed. Recurrent selection, a breeding method that increases the frequency of genes controlling a specific trait in populations, was used to increase protein concentration in soybean seed. Successive cycles of recurrent selection increased seed protein from an average of 43.8 to an average of 48.4 percent in soybean breeding populations. Increases in seed protein were accompanied by decreases in seed oil, the other major component of the seed. In the final cycle of recurrent selection, 53 percent of the plants in the population had seed protein concentration in excess of 48 percent. This demonstrates that recurrent selection is an effective breeding method to increase seed protein in soybean. The high protein plants that were developed will be an important source of genes controlling increased seed protein in soybean. The information and the germplasm will be useful to soybean breeders who want to develop soybean varieties with higher than normal seed protein.

Technical Abstract: Soybean meal, containing high quality protein, is the more valuable component of the seed than the oil. This study was conducted to determine the usefulness of eight cycles of recurrent selection for seed protein on increasing protein and on associated changes in seed oil. Seed of a breeding line with 483 g kg-1 seed protein was blended with seed of two F2 populations segregating for male sterility (ms2ms2). S0 plants from random matings in the population were evaluated in each cycle for plant maturity and for seed protein and oil concentration. The 20% of plants with highest seed protein in cycles 0 through 3 and the 10% of plants with highest seed protein in cycles 4 through 7 were randomly intermated to generate the next cycle. Recurrent selection increased mean seed protein 5.8 g kg-1 and decreased mean seed oil 2.3 g kg-1 per cycle. There was a stronger inverse relationship between seed protein and oil in later than in earlier cycles as indicated by greater slopes of regression lines and by decreased variability around regression lines in later cycles. Fairly consistent values for high seed protein in cycles 5 through 8 suggest that most of the alleles for this trait had been accumulated in the high protein selections by cycle 5. An average of 53% of the plants in cycles 6 through 8 had 480 g kg-1 or greater seed protein. Progenies of these plants would be excellent sources of phenotypes combining high seed protein with good agronomic traits.