Resumo. The basic density is a key property of wood quality for pulp production, but its determination is very time consuming and costly. The development of non-destructive sampling tools, efficient and low cost is important so the Resistograph can be an alternative. This study aimed to use drilling resistance amplitude data collected by the Resistograph, associated to dendrometrical variables, to develop models to estimate basic density in Eucalyptus clonal stands. It was used four hybrid clones of Eucalyptus grandis x Eucalyptus urophylla distributed in commercial plantations in Paraíba valley, with aging from 2 to 7 years. These materials were sampled by the Resistograph at 1.30 m above ground (DBH) and the removal of a disk at the same height to determine the density in the laboratory. Single and multiple linear equations were selected starting with data usage by the Resistograph (amplitude) as a predictor variable, and then dendrometric variables such as DBH, age, height, average height of dominant trees and quadratic mean diameter. The inclusion of these variables increased the accuracy of estimates. The best equation selected for all four clones, adjusted using the average amplitude (the Resistograph), associated with the average height of dominant trees, the tree age and quadratic mean diameter, presented R 2 aj equal to 68.80%, with error residual standard 0.0201 g/cm 3 or 4.31%. For each clone alone, the more accurate equations also were those involving the Resistograph variables of the individual and the population. Equations adjusted for only dendrometric variables showed precision measurements higher than the ones adjusted for only the Resistograph, in the case of all four clones, and superior to the Resistograph associated with the individual variables in the case of each clone alone. The Resistograph proved to be an efficient tool for the prediction of density. However, equally or more accurate estimates can be obtained without use of the equipment when you have individual variables and population.