Path analyses of the yield formation process for late-planted soybean

Strategies to increase yield of late-planted soybean [Glycine max (L.) Merr .] in the southeastern USA would be improved by an understanding of how yie ld components interact with one another in affecting yield at both the phen otypic and genotypic levels. Our objective was to apply path analyses to da ta collected from previous studies to determine the relative importance of yield components in the overall yield formation process when environmental and genotypic factors were varied. Path analyses were applied within primar y yield components (the predictor variables seed number and seed size, affe cting the response variable yield), secondary yield components (the predict or variables pod number and seed per pod, affecting the response variable s eed number), and tertiary yield components (the predictor variables pods pe r reproductive node, reproductive node number, percent nodes becoming repro ductive, and node number, affecting the response variable pad number), Path analyses were applied to three data sets constructed from field studies co nducted from 1987 to 1993 near Baton Rouge, LA (30 degrees N lat). Phenotyp ic analyses, which involved altered source strength (canopy assimilatory ca pacity as indicated by crop growth rate or light interception) during the r eproductive period (R1 to R7), affected yield through adjustments in pods p er reproductive node and reproductive node number. Pods per reproductive no de was also the most important yield component influencing pod and seed num ber on the genotypic level, although negative compensation between seed num ber and seed size (at the genotypic level only) precluded a positive effect on yield.