Resumo. The study was undertaken to evaluate the effect of thinning in the end splits of the logs of the Eucalyptus grandis at 18 years of age. There was analyzed a thinning experiment, located in the northern coast of Rio Grande do Sul State, Brazil, installed in randomized blocks with four repetitions. The seven treatments were defined according to the number of thinnings applied and ranged from zero to six interventions. The trees were selected, in each treatment, on the basis of the Assmann’s dominant diameter and the central diameter of tree, being referred to as dominant and central trees, respectively. After cutting down the selected the logs were located between the DBH (diameter at breast height) and 25% positions, 25% and 50%; 50% and 75%; 75% and 100% of the commercial height, for the end split study. The results indicate that the average end splits of the logs in the dominant and central trees was higher at the second log sampled, suggesting that sectioning should be carried in conjunction with techniques that minimize the growth stress, especially when this is performed near the half of the commercial height. The thinning induced variation in the logs end split, both for dominant and central trees; however, there was no clear positive or negative trend, leading to the conclusion that changes in growth rate did not af fect these parameters uniformly.