Site studies have shown that conifer plantations in northern Britain have increased in General Yield Class (GYC) by 1 m(3) ha(-1) a(-1) per decade or more (20-40 per cent) since the 1930s. Large increases in forest productivity have also occurred in many other regions of Europe. Are these increases due to improved silvicultural practices or to increases in N deposition, CO2 and temperature? Two process-based mathematical models of forest growth were used to simulate the responses of conifer forests growing in the Scottish southern uplands to increases in atmospheric N deposition, CO2 concentration and temperature, during this century and next century. The models differed substantially in the ways in which underlying processes were represented: one simulated a managed plantation, the other a natural forest. Nevertheless, both showed that: (1) increases in N deposition, CO2 and temperature together might account for up to half of the observed increase in GYC this century; (2) increased N deposition and CO2, considered separately, probably increased forest productivity by a modest amount (7-14 per cent), but their combined effect has been approximately additive; (3) increased temperature, even when combined with increasing CO2 concentrations, promoted growth less than expected from site studies relating GYC to temperature; and (4) substantial further increases in productivity, GYC, leaf area index and standing biomass are forecast during the next century as a result of increasing CO2 concentrations and continued N deposition, with or without climatic warming. The predicted increases in GYC could be large enough to have profound effects on the forest industry.