This research evaluated the potential benefits of winter cover crops on the utilization and cycling of P in Brazilian tropical cropping systems. The effect of P fertilizer [none, rock phosphate (RP), and soluble phosphate (single superphosphate, SSP)] in combination with cover crop residues (common vetch [Vicia sativa L.], white lupin [Lupinus albus L.], forage radish [Raphanus sativus L.], ryegrass [Lolium multiflorum Lam], black oat [Avena strigosa Schreb.], red clover [Trifolium pratense L.], and fallow) were evaluated on maize (Zea mays L.) yield and P use efficiency over three maize cropping seasons under no-tillage, from 2009 to 2012. Cover crop yields and P uptake were higher under phosphate fertilizers than nil-P across all seasons evaluated. The highest amounts of P recycled in cover crops over the period were under white lupin, followed by radish and ryegrass, but without any significant cover crop effect on maize yield. The largest response and greatest P use efficiency (30 kg grain per kg P applied) was obtained in the third year of evaluation, when maize yield was restricted by low rainfall. In this year, RP promoted greater maize yield than SSP and the nil-P. Soil available P at the end of the experiment was higher under RP than SSP. It is concluded that RP solubility is higher than currently predicted (9% P2O5 in citric acid). Cover crops were not able to affect maize yield after 3 yr of leaving the residues on the surface, however they can reduce the soil loss by erosion and runoff.