A 3-year field experiment was carried out to determine effective methods of applying phosphate rock (PR) at a rate of P 120 kg hm^-2a^-1 on an acid paddy soil derived from quaternary red clay in a rapeseed-ric-erice cropping system.The yield response of rapeseed to application of phosphorus fertilizer(PF) was much greater than that of rice.SSP was applied separately to rapeseed,early rice,and late rice,while PR was applied only to rapeseed.Kunyang phosphate rock(KPR) was found to be significantly less effective than Gafsa phosphate rock(GPR) on rapeseed,but quite similar on rice.Relative agronomic effectiveness of GPR and KPR in rapeseed,early rice,and later rice seasons were 108.7%,105.0%,99.6%and 89.4%,104.8%,97.1%,respectively.Increasing the application rate of GPR from P 120 kg hm^-2a^-1 to P 240 kg hm^-2a^-1 or mixing RP with SSP and applying the mixture to separately to three crops did not have any extra effect on yields of rapeseed or rice.Available P in the cultivated layer of soil after the rapeseed season was greater than that after the rice seasons.Phosphorus absorbed by rapeseed was mainly from the PF,whereas P absorbed by rice was mainly from soil.