Little is known about the agronomic effectiveness of phosphate rocks in high-yielding crop rotation systems under no-till in the Center-South of the state of Parana, Brazil. This field study was undertaken to compare the effectiveness of rock phosphates and soluble P fertilizers to increase the yield of several annual crops grown in two consecutive triennial crop-rotation cycles under no-tillage. The experiment was carried out in an Oxisol, in Guarapuava, PR, from 2000 until 2006. Two phosphate rocks (PR - Gafsa and Arad) and a soluble P fertilizer (TSP - triple superphosphate) were broadcast over the soil surface at rates of 0, 40, 80 and 160 kg ha -1 P 2O 5, at the beginning of the first and the second rotation cycle of the triennial crops (oat, maize, wheat, soybean, barley and soybean). With exception of barley, crop yields were not affected in the first rotation cycle by any phosphate fertilizer regardless of type and rate, probably due to the high P contents available in the soil (8.7 mg dm -3 in the 0-10 cm layer). In the second rotation cycle, when soil available P was 4.1 mg dm -3; P application at oat sowing increased the yield of summer crops by 11% and of winter crops by 20%. The yield increment was lowest for maize (8%) and highest for barley (44%). In this cycle, TSP resulted in higher yield than PR, though only for winter crops (oat, wheat and barley), in a mean of 11%, with no difference between Gafsa and Arad. Water soluble fertilizer (TSP) was more efficient than phosphate rocks in soils under no-till system both in the short and long term. Nevertheless, it is not necessary to apply phosphates for high yields in soils with high available P contents under no-till, except for species with high P requirement, e.g., barley.
