ABSTRACT

This work was conducted at the eastern farm of Michael Okpara University of Agriculture, Umudike, Abia State. The land was cleared of vegetation. ploughed, harrowed and ridged. Twenty blocks were made and the area of each block was 3m x 2m (that is 6m2) with furrows of in between the blocks. The experiment was laid out in a Randomized Complete Block Design (RCBD), with five ii aliments and four replications. The treatments consisted of lime and were applied at two weeks before the crops were sown at the rates 0, 0.5, 1.0, 1.5, and 2.0t1ha. Viable seeds of soybean of the cultivar TGX1844-4E, were sown iii April at a spacing of 50 x 10cm. This gave a total of 240stands/plot, which is euiva1ent to 400,000 stands/hectare. Data were collected on plant height, leaf number, number of nodules and nodule dry weight, at the second, fourth, sixth, eight, and tenth weeks and grain yield at harvest. All the data generated were subjected to analysis of variance (ANOVA) and Fisher's Least Significant Difference (LSD) was used to separate the significant means. The soils analysis showed that sand dominates the area being 96.8%. From the field experiment, the treatments applied at the different levels, had no significant effects (p>O.05) on the number of nodules, nodule dry weight, number of seeds and grain yield. But there was significant effects (p<O.05) on both pod weight and od number. It was observed that the plants with treatment, 1.5L'ha lime had the highest pod weight (26.47g) compared to the plant at control, which had the lowest Pod weight (8.0425g). Also in the pod number, it was observed that plants with treatment O.5tiha had the highest pod number when compared with the plants at the control Ot/ha. From the experiment I . tons per hectare lime is recommended for soybean production in these soils.