<span name="style_bold">Evaluation of three legumes as cover crops with maize under subhumid tropic. </span>The objective of this work was to evaluate the potential of three legumes as cover crops to improve maize production systems. Four treatments were evaluated: 1) control, maize in monoculture; 2) maize - short-cycle, white seed lima bean (<span name="style_italic">Phaseolus lunatus</span>); 3) maize - long cycle, white seed lima bean <span name="style_italic">Phaseolus lunatus</span>), and 4) maize - velvet bean (<span name="style_italic">Mucuna </span>sp.), in a randomized complete blocks design with four replications during four growing cycles (1999-2002), in Yucatán, México. We measured: crop biomass, pH, total N, potential anaerobic mineralization of N, CO<span name="style_sub">2</span> evolution, potassium (K), phosphorus (P), and soil organic matter, and biomass and frequency of weeds. Legume stubble residues accumulated on maize crops were: 3436, 3425 and 4018 kg DM/ha for the first, second and third year, respectively. Nitrogen contributed by stubble material was 25.8, 31.9, 52.8, and 43.4 kg/ha in treatments 1, 2, 3, and 4, respectively. Maize grain yield was 841, 843, and 460 kg DM/ha in the first, second and third year, respectively, without significant differences between treatments. During the third year, velvet bean caused a reduction in weed biomass from 126 to 58 g DM/m<span name="style_sup">2</span>. Cover crops had little influence on soil variables and maize grain yield.. Cover crops had little influence on soil variables and maize grain yield.

Weeds; nutrients; organic matter