Common logging practices in tropical rainforest are often not sustainable and can degrade the forest. The Palcazú Forest Management system (strip clear-cutting) has been proposed as one sustainable alternative. However, low growth rates of commercial tree species during early regeneration indicate that strip-cutting may not be sustainable. To test whether a silvicultural treatment enhances the growth, survival, recruitment or richness of regenerating trees, we carried out an experimental thinning in 1996 within two strips cleared in 1989 at Jenaro Herrera, Peru, then censussed and remeasured trees on thinned and control plots in 2000. In addition, for strip 2, we compared tree regeneration on two felling treatments, clear-cutting and deferment-cutting. Thinning significantly enhanced annual growth increment (AGI) for stems of all regeneration categories (recruits, stump sprouts, and advance regeneration) of commercial species in strip 1, and for all categories of recruits and stump sprouts in strip 2. In most cases, mean increment in the thinned treatment was approximately twice that of the control. Recruitment, survival and richness, however, demonstrated little response to thinning. Although growth tended to be higher in clear-cut portions, significant differences in increment for felling treatment were detected for only two of seven categories on strip 2, and there was no effect of felling treatment on survival, recruitment or richness. Thinning is likely to raise the value of the second harvest on the strips by enhancing the growth rates of regenerating trees, so thinning should be included in the strip clear-cutting system. This system should not, however, be considered sustainable unless growth rates are sufficient to ensure the value of successive harvests. Crown Copyright © 2003 Published by Elsevier Science B.V. All rights reserved.