CORN RESPONSE TO DIFFERENT LEVELS OF NITROGEN IN ROTATION WITH CANAVALIA UNDER TWO TYPES OF TILLAGE. RIO HATO, PANAMA. 1993-1994.

Abstract

During two years (1993-1994) was carried out an experiment to evaluate the response of corn to N application in rotation with Canavalia ensiformis. The legume was planted at the beginning of raining season (May), while the corn was planted in September. In the experiment two types of tillage were evaluated. One method consisted in incorporate the weeds and legumes before planting (conventional tillage) and the other consisted in leave it on the top of soil (zero tillage). There were evaluated four systems: a) Plot consisted in cutting the legume and leave the stubble on the top of soil (Can R+F); b) Plot where grew the legume, cut the foliage and remove it (Can R); c) Plot without legume, but the stubble of the previous plot was move it to this (Can F); d) Plot without legume (SLeg). The experimental design was a Completely Randomized Block design in a split-split-plot with three replications. The principal plots were the systems, the sub-plots were the tillage types and the sub-sub-plots were the Nitrogen levels. A highly significant response was founded in the residual effect of legume systems. The Can R and Can R+F were the best systems with yields of 5.39 and 5.28 t/ha, respectively. The system Can F follows to these systems (3.58 t/ha) and the last was SLeg system with 2.50 t/ha. The N application response was highly significant, with mean yield of 2.82, 4.54 and 4.99 t/ha for the levels 0, 75 and 150 kg N/ha, respectively. The levels of N that optimized the yield of systems Can R+F, Can R, Can F and SLeg were 39, 30, 36 and 110 kg N/ha, respectively. The N level x System interaction was highly significant. Furthermore, it was founded when didn't apply in the system, the best yield was obtained in the Can R+F system.