Estacas para propagação de capim-elefante (pennisetum purpureum schum) var. guaçú, em plantio tardio

The experiment was carried out on the experimental field of FEAGRI-UNICAMP, at Campinas. SP, and their objectives were to determine the part of the stem and number of buds per cutting to provide more efficiency in the dissemination of the elephant grass (Pennisetum purpureum Schum, var. Guacu) in a late planting by machine. Treatmens were arranged in a 3 by 3 factorial scheme (position of the cutting in the stem X number of buds per cutting) and were distributed in a complete randomized design, with four replications. Each plot was defined by a row of 12m long and 36 buds/row, despite of cuttings size, Lo simulate a machine planting. Cuttings were prepared by circular saw and plots were fertilized using 200kg of single super phosphate per hectare. It was measured the buds development at the 13th and the 24th day and the number of tillers at 56 days. Results of statistical analyses showed that it is not useful to use cuttings from apical part of the stem for propagation and the number of tillers demonstrated that the cuttings from the lower part of stem were better than from other parts with no differences for two or three buds per cutting. Cuttings with two or three buds are adequate for machine planting.