The objectives of this work were to evaluate the floristic composition and dry biomass of weeds under the canopy of seven perennial species adapted to the Semi-Arid region of Brazil, and correlate these characteristics with growth traits of the perennial species. The following perennial species were evaluated in two experiments (E1 and E2): mesquite (Prosopis juliflora), jucá (Caesalpinia ferrea), white popinac (Leucaena leucocephala), mofumbo (Combretum leprosum), neem (Azadirachata indica), sabiá (Mimosa caesalpiniaefolia) and tamarind (Tamarindus indica). In E1, the seven species were evaluated in a random block design with four replicates and nine plants per plot. In E2, evaluation comprised four species (mesquite, jucá, white popinac, and tamarind) in a random block design with eight replicates and nine plants per plot. A circle with an area of 1.77 m² was established around the trunk of each plant, two years after they were transplanted to the permanent location. The weeds collected within this circle were cut even with the ground, classified and weighed. At this time, plant height, and crown and stem diameters were evaluated in all trees of each plot. In E1 there were no differences between tree species as to weed frequency under their canopies; however, weed growth was smaller under the canopy of sabiá trees. Mesquite and sabiá had the greatest plant height and crown diameter means, but only sabiá had the greatest stem diameter. In E2, the perennial species were not different with regard to weed frequency and growth under their canopies, but mesquite had the greatest growth, as measured by plant height (with significant results for jucá as well) and crown and stem diameter.

Herbicides have simplified weed control, but the use of herbicides, besides being costly, resulted in the selection of herbicide-resistant weed biotypes and has become an environmental contamination factor. Herbicide use reduction is one of the goals of modern agriculture, with several alternatives being investigated, including intercropping. The objective of this study was to evaluate the effects of cowpea and corn cultivar intercropping on weed control and corn green-ear (immature ears with 80% humidity grains) and grain yield. A completely randomized block design with split-plots and four replications was used. AG 1051, AG 2060 and PL 6880 corn cultivars (assigned to plots) were submitted to the four treatments: no weeding, two hoe-weeding (22 and 41 days after planting), and intercropping with cowpea (BR 14 and IPA 206 cultivars, with indeterminate growth). The cowpea was planted (with corn planting) between the corn rows, in pits 1.0 m apart, with two plants per pit. The corn cultivars did not differ from each other as to weed density (WD), fresh above-ground weed biomass (WB), green-ear yield and grain yields. Higher WD and WB mean values were found in no weeding subplots; lower mean values in two hoe-weeding subplots; and intermediate mean values in intercropped subplots, indicating that cowpea plants had, to a certain extent, control over weeds. The no-weeded plots and the intercropped plots had lower green-ear and grain yields. Although the cowpea cultivars had a certain control over weeds (mean reductions of 22.5 and 18.3%, in terms of green matter density and weight of the above-ground part of weeds, respectively), they also competed against the corn plants, leading to yield reduction (mean reductions of 17.0 and 32% in green ear and grain yield, respectively). The cowpea cultivars did not produce grain, certainly due to the strong competition exerted by the corn and weeds on cowpea plants.