O estudo desenvolvido em casa de vegetação e no campo, teve por objetivos determinar os níveis críticos de P no solo, pelos extratores de Mehlich I e Resina, e os níveis críticos de P na planta, para o estabelecimento das gramíneas Brachiaria decumbens Stapf., Brachiaria brizantha (Hochst.) Stapf. cv. Marandu e Panicum maximum Jacq., em Latossolo Vermelho Amarelo, álico. O Delineamento experimental foi de blocos casualizados, com 4 repetições, no esquema fatorial 7 (doses de P) x 3 (gramíneas). No ensaio em vasos as doses de P corresponderam a 0; 17,5; 35,0; 70,0; 140,0; 280,0 e 560,0 ppm. Realizaram-se dois cortes, o primeiro, 50 dias após a semeadura, e o segundo, com 30 dias de rebrota. No ensaio a campo as doses de P corresponderam a 0, 20, 40, 80, 160, 320 e 640 kg de P₂O₅/ha. Foram realizados dois cortes, o primeiro, 80 dias após a semeadura, e o segundo, com 60 dias de rebrota. Os resultados mostraram que: - No ensaio em vasos os níveis críticos de P no solo, no primeiro corte, variaram de 8,0 a 12,0 ppm (Mehlich I) e de 13 a 18 ppm (Resina), para as três gramíneas. No segundo corte foram de 11,5, 12,0 e 14,0 ppm (Mehlich I) e de 18,0, 22,0 e 24,0 ppm (Resina), respectivamente, para a B. decumbens, B. brizantha e o P. maximum. Os níveis críticos de P na planta foram na mesma ordem, de 0,11, 0,13 e 0,12%, no primeiro corte e 0,09, 0,12 e 0,10% no segundo corte; - No ensaio a campo os níveis críticos de P no solo, no primeiro corte, foram de 4,7, 10,8 e 8,0 ppm (Mehlich I) e de 10,7, 21,8 e 16,0 (Resina), respectivamente, para a B. decumbens, B. brizantha e o P. maximum. No segundo corte foram de 3,0 e 3,5 ppm (Mehlich I) e de 5,8 e 6,5 ppm (Resina), respectivamente, para a B. decumbens e o P. maximum. Os níveis críticos de P na planta foram de 0,08, 0,13 e 0,06%, no primeiro corte e 0,09 e 0,08%, no segundo corte, na mesma ordem acima, respectivamente; - Os níveis críticos de P obtidos no experimento em vasos não foram equivalentes aos obtidos no experimento a campo; - Os extratores de Mehlich I e Resina foram eficientes e similares na avaliação do P disponível para as três gramíneas, sendo que o método da Resina apresentou maior capacidade de extração do P aplicado.

The objectives of the present study, carried out in a greenhouse and in a field, were to determine the critical levels of soil P by the Mehlich l and the Resin extractors and the critical level of P in the plants, for the establishment of Brachiaria decumbens Stapf., Bracharia brizantha (Hochst.) Stapf cv. Marandu and Panicum maximum Jacq., in a Red Yellow Latosol. The experimental design was a randomized complete block with four replications and 21 treatments (7 levels of P X 3 species). In the greenhouse, the levels of P corresponded to 0, 17.5, 35.0, 70.0, 140.0, 280.0 and 560.0 ppm. Two cuts of plant shoots were done: the first, fifty days after sowing and the second, thirty days later in the field, the levels of P corresponded to 0, 20, 40, 80, 160, 320 and 640 kg of P₂O₅/ha. Two Cuts of plant shoots were also done: the first, eighty days after sowing and the second, sixty days latter. The results showed that: - In the greenhouse, the critical level of soil P, in the first cut, varied from 8.0 to 12.0 ppm (Mehlich I) and 13.0 to 18.0 ppm (Resin), for the three grasses. For the second cut, the levels were 11.5, 12.0 and 14.0 ppm (Mehlich I) and 18.0, 22.0 and 24.0 ppm (Resin), for the B. decumbens, B. brizantha and P. maximum, respectively. The criticai leveis of P in the plant were, in the same order, 0.11, 0.13 and 0.12% for the first cut and 0.09, 0.12 and 0.10%, for the second cut; - In the field, the critical levels of soil P were 4.7, 10.8, and 8.0 ppm (Mehlich I) and 10.7, 21.8 and 16.0 ppm (Resin), for the first cut, for the B. decumbens, B. brizantha and P. maximum, respectively. For the second cut, the levels were 3.0 and 3.5 ppm (Mehlich I) and 5.8 and 6.5 ppm (Resin) for the B. decumbens and P. maximum, respectively. The critical levels of P in the plant were 0.08, 0.13 and 0.06%, for the first cut and 0.09 and 0.08% for the second cut, in the same order as above, respectively; - The critical levels of P obtained in the greenhouse were not similar to those obtained in field conditions; - The Mehlich I and Resin extractors were both efficient and similar for the evaluation of the availability of P to the three grasses, out the Resin method had a greater extraction capacity of the P applied.