Water soluble phosphates are the most often used phosphorus sources in fertilizers for ease of P release to the soil. However, in tropical soils, much of this P is adsorbed to soil particles and becomes unavailable to plants. Conversely, reactive phosphates may be slow initial releasers of the nutrient, but be released continually to the developing crops, reducing soil P fixation. The aim of this study was to investigate the application of two P sources to triticale and assess the residual effect of fertilization on pearl millet in no tillage. Soil fertility, dry matter and grain yields, phosphorus content and amount in plants and straw were determined. The experiment was conducted in an Oxisol. Three treatments were applied in April of the first year: (1) without application of P 2O 5, (2) application of 80 kg ha -1 P 2O 5 as triple superphosphate and (3) application of 80 kg ha -1 P 2 O 5 of reactive phosphate (Arad). Triticale (* Triticosecale Wittmack) was planted and grown until grain harvest. In September, after harvest of triticale, pearl millet ( Pennisetum glaucum (L.) R. Brown) was planted to increase the amount of straw on the soil surface. At flowering in November, pearl millet was desiccated. Soluble phosphate, which was applied to the soil surface, increased soil P contents down to the 5-10 cm soil layer, and triticale benefited from P fertilization with higher yields. Despite the increase in available phosphorus in the soil provided by soluble phosphate, P application to triticale did not increase dry matter production, P content and amount in the shoots of the pearl millet grown in sequence.
