Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/15/1996
Publication Date: N/A
Interpretive Summary: Soybean is produced for the oil in the seed and the high-protein meal after the oil is extracted. Soybean varieties with more than about 40 percent protein in the seed have not been developed because of an inverse relationship between seed yield and seed protein. This inverse relationship was overcome by developing determinate soybean progenies that combined good seed yield with higher-than-normal seed protein. Determinate soybean strains were developed that were similar in seed yield to the high-yielding variety, Charleston, and varied from 44.9 to 47.8 percent protein compared to Charleston with 42.0 percent seed protein. In contrast, indeterminate strains showed the typical inverse relationship between seed yield and seed protein. No indeterminate strains were identified that had as favorable a combination of high seed yield and high seed protein as did the determinate strains. Indeterminate soybean varieties are typically grown in the Midwest. Determinate varieties adapte to production in the Midwest are short-statured and lodging resistant. The high-yielding, high-protein determinate strains developed suggest that the determinate growth type may offer better opportunities for overcoming the inverse relationship between seed yield and seed protein than the indeterminate growth type.
Technical Abstract: Development of soybean cultivars with higher-than-normal seed protein has been hampered by the inverse relationship between seed yield & protein concentration in the seed. Determinate & indeterminate isolines of soybean are usually similar in seed protein but differ in growth type. This study was conducted to determine if relationships between seed yield & seed protein differed in populations of the two plant types. (Det) and (Indet) progenies from 2 crosses between high-protein (Indet) breeding lines & (Det) strains w/average seed protein were evaluated for 2 years in 3 replicate tests at W. Lafayette, IN. Mean seed protein concentration was similar for the 2 plant types in each cross. Regression of seed protein on seed yield was not significant among determinate progenies in either cross. Both the slopes of regression lines & R2 values were significant when seed protein was regressed on seed yield for indeterminate progenies in each cross. The 3 highest yielding determinate entries from 1 cross were similar in yield to the determinate cultivar Charleston, & varied from 449 478 g kg-1 seed protein compared to Charleston with 420 g kg-1 protein. In the 2nd cross, the 3 highest yielding determinate progenies were similar in yield to Charleston and 2 of these had higher seed protein than Charleston. The 3 highest yielding indeterminate lines in each cross were similar in yield to the indeterminate check cultivars and varied in seed protein from 418 to 435 g kg-1 compared to 414 g kg-1 protein for the checks. The data demonstrate that in these 2 crosses, determinate progenies were a better source of selections that combined high seed yield with high seed protein than were indeterminate progenies.