A low-intensity selection strategy was recommended for timber trees in the Peruvian Amazon to maintain genetic variation on farms and produce modest gains in tree growth. The effectiveness of this strategy was evaluated using Calycophyllum spruceanum. Farmers selected 66 mother trees of different ages on farms in seven locations (~ 20% of all trees in the locations) in one watershed, based on a visual assessment of growth, form and external disease symptoms. Another 66 mother trees were chosen at random. Tree height, stem diameter, stem bifurcations and mortality of progeny of the selected and random groups of mother trees were evaluated at 15, 26 and 38 months in a provenance/progeny test located on farms in the same watershed. Height was significantly greater (10%) in the selected group at 15 months, but it did not differ significantly between the selected and random groups at 26 and 38 months. There were no significant differences in diameter, bifurcations and mortality between the groups. There was significant variation in height and diameter at all measurement ages due to families, and results suggested that variation in bifurcations and mortality due to families was also significant. Based on approximate 95% confidence