Nitrogen-mediated interactions in legume-nonlegume intercropping systems.

Abstract

Intercrops of nitrogen-fixing legumes with non-fixing plants are common throughout the tropics. Reports of greater intercrop yields, compared to monocultures of the component species, occur most often where available soil nitrogen (N) limits plant production and inputs of biologically-fixed N from the legume are presumed to be most important. A one-time application of $\sp{15}$N-enriched (NH$\sb4)\sb2$SO$\sb4$ was used to label soil N in a 13 m x 17.5 m field plot in Costa Rica in which three consecutive experiments were planted with the common intercrop of maize and beans. Using a target-neighbor planting design, I measured the intra- and interspecific effects on N accumulation in maize. A net positive effect, after the first season, of bean (compared to maize) on soil N ($\sp{15}$N-enriched) accumulation in maize, shifted to a net negative effect after the third season. There was no evidence of fixed N transfer (unlabeled N) from bean to maize, despite frequent claims of direct transfer of fixed N from legumes to associated non-fixing plants. Since N transfer from soy to maize via vesicular-arbuscular mycorrhizal (VAM) fungi has been demonstrated in greenhouse experiments, I conducted a similar experiment to investigated the role of VAM fungi in the N-mediated interactions of maize and beans. VAM infection did not result in the transfer of fixed N or soil N from bean to maize, despite a VAM-stimulated increase in N fixation in bean. In fact, beans were more competitive for soil N when mycorrhizal. The competitive effect may have resulted from a VAM-mediated shift in carbon allocation in beans (but not maize) from shoots to roots. It is suggested that N fixation in beans may increase N accumulation in the intercrop, but only under conditions where soil N is not severely limiting. Where available soil N falls below some critical level, establishment of effective N fixation in beans can not occur, and beans may become competitive for soil N. Results from the greenhouse experiment (with limiting soil N) suggest that VAM fungi increased the competitive ability of the more mycotrophic beans at the expense of the maize.