This study aimed to evaluate the composition of weed communities in the soybean culture under cultivation of distinct commercial crops, grown for one or three subsequent years, without crop rotation. The experiment was installed under field conditions at Western Agriculture Embrapa, Dourados, Brazil. Five fall-winter treatments, planted over one or three subsequent years, were evaluated: corn planted 90 cm between rows; corn planted 45 cm between rows, intercropped corn + Brachiaria ruziziensis, B. ruziziensisalone, and Cowpea. At the end of September 2011, all areas were desiccated with glyphosate, with evaluation being carried out 30 days after the first desiccation. The phyto-sociological characterization of the weed communities was based on abundance, frequency, dominance, and importance index of each species in the community. Areas were also intra-characterized by the diversity coefficients of Simpson and Shannon-Weiner and inter-characterized by the similarity matrix of Jaccard, through cluster analysis by the UPGMA method. Soybean sowing should be followed by sowing a species which allows accumulation of high amounts of residual dry mass on the soil surface, and this dry mass should present high C/N ratio. At this study, the intercropped corn + B. ruziziensis, or even B. ruziziensisalone, planted following soybean, allowed lower infestation levels by weed species after three years of cultivation.
second crop; phyto-sociology; diversity; similarity
