Nitrogen (N) and phosphorus (P) deficiency is one of the important causes of degradation of cultivated pasture under tropical conditions. The aim of this study was to evaluate phosphate rates and sources, and N rates on the concentration and uptake of N and P, and shoot dry mass (SDM) yield of Megathyrsus maximum grass cv Mombasa in an Ultisol. The trial was carried out in a greenhouse in pots with 4.0 dm −3 of soil. The experiment was arranged in a completely randomized design with four replicates. The 3 × 3 × 3 factorial treatments consisted of phosphorus sources [reactive rock phosphate from Morocco (RPM), reactive rock phosphate from Algeria (RPA) and triple superphosphate (TSP)], three phosphorus rates (0, 150, and 300 mg kg −1), and three N rates (0, 250, and 500 mg kg −1). The SDM and tillering of Mombasa grass were significantly influenced with the TSP, RPM, and RPA application associated with N fertilization. The RPM, RPA, and TSP met the nutritional demands of Mombasa grass. The three P sources showed the same effect on the total N uptake by Mombasa grass. The P use efficiency (PUE) when fertilizer-P sources were added alone by Mombasa grass was <12% of the added P, and PUE decreased as follows: TSP > RPA > RPM. When P and N-fertilizer were added together, the fertilizer-N use efficiency (NUE) was 62%. The reactive phosphate (RPM and RPA) is an efficient P sources for Mombasa grass, but requiring higher rate of application compared to TSP source.