An analysis of data from 212 permanent sample plots provided no evidence of any decline in rainforest productivity after three cycles of selection logging in the tropical rainforests of north Queensland. Relative productivity was determined as the difference between observed diameter increments and increments predicted from a diameter increment function which incorporated tree size, stand density and site quality. Analyses of variance and regression analyses revealed no significant decline in productivity after repeated harvesting. There is evidence to support the assertion that if any permanent productivity decline exists, it does not exceed six per cent per harvest.