Yield and Reproductive Growth of Simulated and Field‐Grown Soybean. II. Dry Matter Allocation and Seed Growth Rates1

Abstract

A better understanding of the main plant traits that ultimately lead to seed yield in soybean [Glycine max (L.) Merr.] is needed. Objectives were (i) to examine the association among various seed growth traits and yield, and (ii) to compare the stability across environments of dry matter allocation coefficient (DMAC) and seed growth rate on a land area (SGR) and individual seed (ISGR) basis with that for yield. Materials were 45 simulated genotypes generated with a model of soybean reproductive growth, and field evaluations of 88 random (1982) and 17 selected (1983) F7 determinate genotypes from two crosses plus their parents. Randomized complete blocks were used in three field growth analyses performed at Gainesville, FL. Negative correlations (r= −0.44**, significant at the 0.01 probability level) between DMAC and yield were predicted using simulated genotypes and were also found in field‐grown soybean (r= −0.18 to −0.40). Simulations predicted strong negative correlations of DMAC with effective filling period (EFP) (r= −0.92**) and with reproductive period duration (RPD) (r= −0.89**). In most cases field data supported this hypothesis. Correlations of DMAC were r=0.30 to −0.72** with EFP, r= −0.79** to ‐0.85** with RPD, and r= −0.29* (significant at the 0.05 probability level) to −0.60** with the period R5 to R7. Simulations predicted linear yield increases with respect to SGR (r²=0.71**). In all cases, field results showed close agreement with the simulations (r²=0.64** to 0.95**). Correlations of ISGR with yield were low and positive (r=0.03 to 0.39). In simulated genotypes many pathways led to medium yields, but the highest yield was attained by a genotype with the highest biomass at the beginning of seed growth, highest potential maximum net hexose crop growth rate, and the lowest DMAC, which resulted in a high SGR and the longest seed‐filling duration. Highly significant genotype by environment interactions were observed for DMAC and yield but not for SGR and ISGR. Results suggest that high DMAC strongly limited seed‐filling duration and that DMAC estimates the intensity of self‐senescence during seed filling. Yield was mainly a linear function of SGR, and ISGR was poorly associated with yield.