Genetic Variability for Desirable Agronomic Traits in Populations Containing Glycine soja Germplasm 1

The wild soybean ( Glycine soja Sieb. et Zucc.) may be a source of useful characters for improvement of the cultivated soybean [ Glycine max (L.) Merr.] species. Successful introgression of genes from G. soja into G. max will require the development of populations with an adequate frequency of segregates that do not possess the undesirable agronomic characters of the wild species. The objective of this study was to determine the genetic variability for desirable agronomic traits of G. max in backcross populations containing different percentages of G. soja germplasm. No selection was conducted during backcrossing to allow the evaluation of segregation patterns that would not be influenced by genetic linkages or physiological associations between desirable and undesirable characteristics in G. soja . The populations were formed by crossing each of two G. soja accessions to a different G. max cultivar, ‘Amsoy 71’ or ‘Century’, and backcrossing to the G. max parent to obtain BC 0 to BC 5 populations. The F 2 plants and their F 3 progeny from each population were evaluated for agronomic score, vining, lodging, plant height, days to stage R7, seed weight, shattering, leaf abscission, and seed coat color. The percentage of segregates with an agronomic rating equal to the G. max parent averaged across the Amsoy 71 and Century populations was zero in BC 0 , and BC 1 , 2% in BC 2 , 22% in BC 3 , 51% in BC 4 , and 65% in BC 5 . The results indicated that, in the introgression of G. soja germplasm, at least two backcrosses are required to recover any segregates that are agronomically acceptable when at least 50 plants are backcrossed each generation. Three backcrosses may be preferred to obtain a relatively high frequency of segregates that may be useful in a cultivar development program.