The aim of this study was to determine the effect of the main nutrients N, P, K and the N/K ratio on the rice yield index. To this end, four treatments and a blank control were used in a randomised complete block design with three replications to carry out this study in a lowland area with a histosol character in the commune of Songon in the south of Côte d'Ivoire. In this design, the blocks were separated by 1 m and, within each block, the microplots of 15 m2 containing the treatments were separated by 0.5 m. The principle of the experiment is a response curve to increasing doses of N and K combined with a constant dose of P set at 14 kg P ha-1. The rice variety AR051H, a hybrid variety from Senegal evaluated by the GIPS expert group, was transplanted after a 21-day nursery period, with one seedling per poquet at 20 × 20 cm within the microplots. Fertiliser was applied in three doses at transplanting, tillering and bolting. A minimum of 5 cm of water was applied to the microplots 10 days after transplanting and maintained until the rice reached maturity. A selective systemic foliar herbicide (Weedkill 720 SL) was used for weed control and a systemic insecticide (Decis 50 EC) for insect control. The results showed a very significant effect of the treatments on the harvest index, which was 43% with the optimum doses of 105 kg N ha-1 and 75 kg K ha-1.