Abstract: This research was aimed at understanding the effects of intercropping on biomass accumulation and nutrient uptake of component species in grass-legume mixtures, and providing a theoretical basis for determining optimal planting configuration of different forage grasses in grass-legume mixtures. The design was a randomized block experiment, with species studied being Stylosanthes guianensis, Heteropogon contortus and Bothriochloa pertusa, and their nutrient absorption were studied. The results indicated that intercropping of H. contortus and B. pertusa with S. guianensis enhanced nutrient absorption ability and yields, compared with monoculture, due to the biological nitrogen fixation of S. guianensis. The grass biomass values in intercropping systems were 0.06% and 26.78% higher for H. contortus and B. pertusa, respectively, than for the same species in a monoculture system. Meanwhile, S. guianensis biomass values in the two intercropping systems were 9.70% and 12.83% lower than those in an S. guianensis monoculture system. For H. contortus intercropped with S. guianensis, the nitrogen and phosphorus absorption were 24.26% and 35.18% higher than those in H. contortus monoculture. Corresponding values of nitrogen and phosphorus absorption for B. pertusa intercropped with S. guianensis were 40.64% and 47.50% higher than those in B. pertusa monoculture. Compared with monoculture, S. guianensis intercropped with H. contortus or B. pertusa showed, respectively, a 3.58% and 20.66% decrease in phosphorus absorption, and an 8.31% and 28.94% decrease in nitrogen absorption, while the two intercropping modes had a land equivalent ratio>1. Intercropping of S. guianensis and H. contortus had higher yield and interspecific competitive ability than intercropping of S. guianensis and B. pertusa. For intercropping systems with a row space of 50 cm, there was an inhibitory effect on the growth of S. guianensis because of competition from the grasses. At this row spacing, S. guianensis was less competitive crop in an intercropping system with either of the two grasses. The results show that for grass-legume intercropping systems the appropriate planting density, and species combination improves the utilization of soil nitrogen and phosphorus, and enhances grass yield and forage quality through the inclusion of the legume component.

Key words: Leguminous, Gramineae, intercropping, growth effect of pasture, nutrient absorption