An experiment was carried out in Magé, Rio de Janeiro State, Brazil, to evaluate the performance of taro (Colocasia esculenta) intercropped with sunn hemp (Crotalaria juncea), used as green manure. A randomized blocks design with four replicates was used and the soil of the field plots was classified as Haplic Gleysol. Treatments consisted of taro (1) intercropped with unpruned sunn hemp, (2) intercropped with sunn hemp cut at soil level, (3) intercropped with sunn hemp pruned at its mid-height, and (4) in monoculture. The taro was planted in double rows spaced by 1.0 m. The planting furrows had a space of 0.5 m, with rhizomes placed at every 3.0 m. The experimental plot consisted of four 3.0 m rows and the useful area corresponded to the five central plants in each of the two central rows. The sunn hemp was sown in parallel lines in the spaces between the taro double rows, 120 days after planting the taro. When the sunn hemp was cut at soil level, an accumulation of 211 kg ha-1 of N was observed. Besides this, the sunn hemp's biomass promoted the cycling of an average 17 kg ha-1 of P, 85 kg ha-1 of K, 151 kg ha-1 of Ca, and 27 kg ha-1 of Mg. Worth mentioning, it represented also a deposition of 6.85 Mg ha-1 of dry substance. Even when pruned at its mid-height, the sunn hemp still provided 60% of nutrients relatively to the treatment in which it was cut at soil level. In addition to the nutrient supply, the use of taro intercropped with sunn hemp cut at soil level or pruned at mid-height represented efficient alternatives for controlling the spontaneous vegetation and for reducing taro leaf burning by solar radiation, without compromising the yield of taro. In all intercropping systems, taro plants had a larger leaf area and were taller than plants in monoculture.
Crotalaria juncea; Colocasia esculenta; organic farming; leaf area; leaf burning; weed control