RFLP mapping of quantitative trait loci for reproductive, morphological, and seed traits of soybean (Glycine max L.) across environments

Quantitative trait loci (QTL) and QTL x environment interaction were identified for 11 agronomic traits in soybean, when measured at two locations in each of two years in Iowa. The segregating population consisted of 76 F[subscript]2:5/F[subscript]2:6 families derived from the cross between the two soybean (Glycine max (L.) Merr.) cultivars \u27Minsoy\u27 (PI27.890) and \u27Noir 1\u27 (PI 290.136). A genetic linkage map consisting of 81 markers in 25 linkage groups was used to locate QTL by interval mapping. Analyses of variance was performed on 72 linked and unlinked markers to test for associations between markers and QTL and for QTL x environment interaction;QTL were identified for all traits except seed-protein content in at least one environment. For the developmental (R1, R8, reproductive period) and morphological traits (plant height, canopy height, lodging, leaf area) and seed yield, QTL were clustered on two linkage groups, 2 and 16. Three QTL were identified for seed-oil content on linkage groups 3, 9, and 31. A QTL for seed weight was identified on linkage group 1 also. These results were consistent with the genetic correlations among traits;Analyses of variance showed significant genotype x environment interaction for all traits except seed-oil content. Evidence for QTL x environment interaction varied among traits and depended on method of analysis. Interval mapping identified QTL consistently expressed in all environments only for plant height and lodging. QTL x environment interaction occurred for all other traits, in that QTL were detected only in a subset of environments;Single factor analyses of variance generally confirmed results from interval mapping. Several markers were significant in the analyses of variance that were not identified as significant using interval mapping, notably marker R79;QTL x environment interaction was significant in the combined analysis of variance only for canopy height and marker G173 on linkage group 16. These results indicated a lack of evidence for QTL x environment interaction, and were supported by the fact that plots of LOD scores along the linkage groups in interval mapping showed coinciding peaks across environments