摘要：Abstract. During the spring-summer of 2000, the investigation was carried out in field conditions utilizing a design of complete blocks, with 30 treatments and three repetitions with a factorial arrangement. The characteristics evaluated were height of the first legume (apv), final height (af), number of stems (nt), number of grains per legume (ngv), number of legumes per plant (nvp), weight of 100 seeds p(100)s and grain yield (rend). The analysis of variance (anova) revealed significant effects for genotypes in the variables p(100)s**, rend** and apv*. As for the phosphorus, the af and rend presented significant values to 0.05. The interaction gxf was highly significant for the rend. The best variety according to the dms was the v-32 (3.92 t ha-1). The characteristics that influenced the expression of grain yield were: nt, nvp and p(100)s.
其他摘要：Abstract. During the spring-summer of 2000, the investigation was carried out in field conditions utilizing a design of complete blocks, with 30 treatments and three repetitions with a factorial arrangement. The characteristics evaluated were height of the first legume (apv), final height (af), number of stems (nt), number of grains per legume (ngv), number of legumes per plant (nvp), weight of 100 seeds p(100)s and grain yield (rend). The analysis of variance (anova) revealed significant effects for genotypes in the variables p(100)s**, rend** and apv*. As for the phosphorus, the af and rend presented significant values to 0.05. The interaction gxf was highly significant for the rend. The best variety according to the dms was the v-32 (3.92 t ha-1). The characteristics that influenced the expression of grain yield were: nt, nvp and p(100)s.
