PHOSPHORUS AND POTASSIUM FOR GUINEA GRASS (PANICUM-MAXIMUM JACQ) CV LIKONI SEED PRODUCTION

A random block design in a factorial arrangement with 5 replications w as used to study the effect of 3 P2O5 (0, 100 and 200 kg/ha/year) and 3 K2O (0, 100 and 200 kg/ha/year) dosages on guinea grass (P. maximum) cv. likoni seed production. A basal nitrogen fertilization of 100 kg/ ha/year was applied. Panicum maximum was sown 0.50 m apart, using vege tative parts at 1 m between rows in 30 m(2) plots. The experiment last ed 4 years. Panicle length was not affected by treatments and its valu e was never higher than 35.4 cm. The same occurred for the number of r aquilla/panicle which reached 33.1. A similar performance was observed in the number of panicles/ha, except for the first experimental year, when there were differences (P < 0.05) between the lowest phosphorus level and the application of this element. With respect to total seed production only in the 4th. year was there a beneficial effect on usin g phosphorus fertilization with a mean value of 62.7 kg/ha. For pure s eed in that same year the highest results (P < 0.001) are found on app lying phosphorus (32.4, 36.7 and 41.4 for 0, 100 and 200 kg/ha/year, r espectively). The comparison between the first and the last year showe d differences (P < 0.01 and P < 0.001) for all indicators in favour of the first year except in the number of panicles where this interactio n (P < 0.05) was found. It is recommended that in soils with similar f ertility to that reported here, annual applications of P and K are nec essary when moderate nitrogen dosages of approximately 100 kg/ha/year, are present.