ABSTRACTThe low effectiveness of nitrogen fertilizer (N) is a substantial concern that threatens global sugarcane production. The aim of the research reported in this paper was to assess the residual effect of N-fertilizer applied at sugarcane planting over four crop seasons in relation to sugarcane crop yield. Toward this end three field experiments were established in the state of São Paulo, Brazil, during February of 2005 and July of 2009, in a randomized block design with four treatments: 0, 40, 80 and 120 kg ha−1 of N applied as urea during sugarcane planting. Within each plot, a microplot was established to which 15N-labeled urea was applied. The application of N at planting increased plant cane yield in two of the three sites and sucrose content at the other, whereas the only residual effect was higher sucrose content in one of the following ratoons. The combined effect was an increase in sugar yield for three of the 11 crop seasons evaluated. Over the crop cycle of a plant cane and three ratoon crops, only 35 % of the applied N was recovered, split 75, 13, 7 and 5 % in the plant cane, first, second and third ratoons, respectively. These findings document the low efficiency of N recovery by sugarcane, which increases the risk that excessive N fertilization will reduce profitability and have an adverse effect on the environment.