Early-generation bulk testing in soybean and inheritance of dwarfness from soybean strain T281

Previous research has produced mixed results on the use of early-generation bulks to measure population performance in soybean (Glycine max (L.) Merr.). The objective of our first study was to further examine the use of early-generation bulk or midparent performance to select among populations within a single-seed descent (SSD) soybean line development program.Parents, F$\sb2,$ F$\sb3,$ and F$\sb4$ SSD bulks of 9 soybean crosses were tested at 3 central Illinois locations in 1990 and 1991. Replicated progeny rows of 100 randomly-derived lines per cross were grown at one location in 1991, and the final measure of cross performance comprised 37 randomly-derived F$\sb{4:6}$ lines per cross, replicated at three locations in 1991. There were differences among crosses for all traits in the three experiments, but correlation coefficients for yield were generally small and non-significant. Much larger correlations were produced for lodging, maturity, and plant height.Midparent values were generally as good or better than bulk performance for selecting among populations. Testing parent lines has an additional advantage because seed supplies and alternative sources of parental data are often available.Our second study investigated the genetic control of dwarfness in soybean strain 'T281', which was found as an offtype in Plant Introduction (PI) 232.992 at Urbana, IL in 1955. The objectives of this study were to determine the inheritance of the dwarf trait in T281 and to determine if these gene(s) were allelic to any previously reported dwarf genes.Our observations on reciprocal crosses of T281 with a normal cultivar indicated dwarfness is controlled by recessive gene(s) and is not cytoplasmically controlled. The phenotypic segregation of the F$\sb2$ and F$\sb3$ generations fit a genetic model for two recessive, independent genes where both genes are required to express the dwarf character. The observed F$\sb2$ phenotypic ratios from crosses of T281 with T243(df$\sb2),$ T256(df$\sb4),$ T263(df$\sb5),$ and T286(df$\sb6)$ fit expected theoretical ratios. The F$\sb2$ phenotypes of the cross T281 x T244(df$\sb3)$ did not fit expected ratios, so further studies of the cross T281 x T244 are indicated.U of I OnlyETDs are only available to UIUC Users without author permissio