This paper reports results from a 6-year long-term fertilizer X banana weevil trial for highland banana in Mbarara, Uganda. The objective of the study was to quantify the effect of mineral fertiliser (100 kg N, 50 kg P, and 100 kg K ha -1 yr-1) on crop and economic performance and weevil pest status. Soil and foliar analyses, and visual observations (i.e. yellowing of leaves) revealed that potassium deficiency was the major soil fertility constraint and a function of slope. Topsoil (0-15 cm) exchangeable K content was low down slope (<0.3 mmolc 100g-1 dm), but higher upslope (>0.6 mmolc 100g-1 dm). As a result, annual yields in the lower block (9 t ha-1) were much lower than in the upper block (17 t ha-1). Yield increase due to fertilizer application (4t ha-1 yr-1) was independent of initial soil fertility status. This observation was in line with foliar analysis, which revealed that K concentrations (2.0 %) were still at deficiency level in fertilized plots, suggesting that 100 kg K ha-1 yr-1 is too little to correct for the deficiency. K deficiency did not only reduce bunch weight, but also increased cycle length, resulting in a significant (r2= 0.57) correlation between these two crop parameters. Mean weevil pressure was too low (< 4%) to result in significant yield loss, but maximum bunch weight was less than 50% in plants with large corm damage (>8%). Overall, the applied fertilizer dose was not profitable; i.e. the mean benefit-cost ratio was 0.7. Probably, application of N and P fertilizer did not increase yields, but the mean benefit-cost ratio is still low (1.6) when N and P fertilizer costs are omitted, and drought-related risks are high. We recommend testing the use of mulch to decrease drought risk and enhance fertilizer use efficiency.

Banana; Plantain; Soil Fertility; Plant Diseases; Plant Health; Disease Control

Cosmopolites Sordidus; Fertilizers; Musa

Africa; East Africa

Uganda