摘要
Este trabalho teve por objetivo estudar a distribui o de formas n o-troc芍veis de pot芍ssio, em fra es texturais de um Latossolo cultivado com bananeira, por dois anos, em dois regimes h赤dricos (parcelas principais), e doses de pot芍ssio (0, 300, 600 e 900 kg ha-1)(subparcelas), seguindo delineamento em blocos casualizados, com parcelas subdivididas. Verificou-se que os maiores teores de pot芍ssio (troc芍vel e n o-troc芍vel) est o contidos nas fra es mais finas: argila e silte. Neste sistema de cultivo e com a aplica o dos adubos na superf赤cie, a movimenta o do nutriente, nas camadas estudadas, esteve associada 角 presen a e movimenta o da 芍gua no perfil do solo. A irriga o desempenhou papel relevante na distribui o das formas do K. Nas parcelas irrigadas, houve mobiliza o do K at谷 60 cm deprofundidade. Houve efeitos significativos das doses de pot芍ssio nas formas n o-troc芍veis do nutriente, evidenciando a migra o do pot芍ssio da solu o do solo para o complexo de troca e deste para formas mais fortemente retidas. A quantidade de pot芍ssio exportada pela planta foi maior que a varia o dos teores troc芍veis, indicando que formas n o-troc芍veis estariam sendo aproveitadas pelas plantas. The aim of this work was to study the distribution of non-exchangeable forms of potassium in different size fractions ofan Oxisol, cultivated with banana for two years in plots with and without irrigation (main plot) and varying potassium levels (0, 300, 600 and 900 kg ha-1 of K2O) (sub-plots), in a randomized split-plot block design. It was verified that the highest K levels (exchangeableand non-exchangeable forms) derived from the finest particle size fractions: clay and silt. The K moving trough soil layers was associated with the presence and action of percolating water in the soil profile. The role of irrigation was significant in the distribution of the different K forms in the soil. Irrigated plots showed the best distribution of K from the surface down to 60 cm of depth. Statistical significant effects of the doses of K on the nonexchangeable K forms could be explained by the translocation of K from soil solution to soil exchange complex, and from that to strongly retained forms. Amounts of K absorbed by the plants were greater than the variation of exchangeable K content, suggesting that nonexchangeable forms are being taken up.