Eucalyptus grandis and Eucalyptus dunnii are the most planted tree species in Uruguay. Anticipating information about the quantity and quality of wood is important for managing intensive forest plantation. The estimate of merchantable and total wood volume is an essential tool in forest planning and management. The aim of this study was to evaluate four systems of taper and merchantable volume that consisted in a taper, a merchantable volume and a total tree volume function. A modified second-order continuous autoregressive error structure corrected the inherent serial autocorrelation of different observations in one tree. Taper and volume equations were fitted simultaneously after autocorrelation correction by full information maximum likelihood method. The segmented system proposed by Fang et al. (2000) produced the best fit as it explained more than 98% of the taper, merchantable volume and total volume variability for both species. In addition, precision of the segmented system was compared with and without incorporating stand density as a variable. Results of this analysis showed that for E. grandis, the predictive accuracy of the model was improved by including the stand density variable, whereas for E. dunnii this variable was not statistically significant. This modelling framework provides an improvement in taper and tree volume predictions for E. dunnii and E. grandis in Uruguay. The possibilities offered by this methodology could be of interest for its application in countries where fast growing plantations are managed.