Species composition and turnover that have occurred in a series of permanent sample plots established during the 1930s and 1940s in Budongo, a semi-deciduous Ugandan forest, are reported. The plots were established as part of a sequence first used to describe forest succession, five of which have been maintained and which were last measured in 1992-1993. One plot (plot 7) provides 53 y of data from old-growth pristine forest. Plot 15 was established in wooded grassland at the forest edge and is now closed high forest. Evaluation of the remaining three plots is complicated by silvicultural interventions carried out in the 1950s. Forty species have been added since the first evaluations and a total of 188 tree species (over 80% of Budongo's forest tree flora, and including two exotics) has now been recorded from within the plots. The pattern of shade-tolerance in the original plot series conforms to patterns expected for succession with an increasing proportion of shade-tolerant species with development, and large stems appearing to ‘lag behind’ smaller stems in this respect. The time series data are less consistent, and while plot 7 increased in the proportion of shade-tolerant stems through time, the proportion of shade-tolerant species actually declines. Stem-turnover (the mean of mortality and recruitment) slowed with implied successional stage. Most species have a higher recruitment than mortality rate and stem numbers have thus increased in all plots. This is most pronounced in the putatively ‘early successional’ plot. Stem size structure has changed within the plots, with an increased proportion of smaller stems. Species show different rates of turnover and these vary from plot to plot and period to period. In plot 7, the overall mortality rate decreased with initial stem size. Estimates imply that some tree species may easily live longer than 500 y after reaching 10 cm DBH, and that 1000 y is possible. The importance of large trees in determining forest dynamics is illustrated by the finding that death of only seven stems in plot 7 contributed over 60% of net basal area losses recorded over the 53-y observation period. Many of the observed patterns were not predicted and could only have been found by long-term studies.