On northern Vancouver Island in British Columbia it has been hypothesised that the cultivation effect of repeated windthrow is the cause of the higher nutrient availability in windstorm-derived, natural second-growth stands of western hemlock and amabilis fir (the HA type), compared with adjacent old-growth stands of western red cedar with a smaller component of hemlock, and a dense understorey of the ericaceous shrub, salal (the CH type). In 1988 an experiment was established in a clearcut area containing examples of these two forest types. The experiment was designed to simulate the effects of a broadscale windthrow by mixing mineral and organic horizons using a large rake attached to an excavator. In this study, a range of soil physical and chemical properties, and the biomass of understorey plants, were measured 4.5 years after treatment in uncut, clearcut, and clearcut and mixed plots of the two types. In the HA type, clearcutting and soil mixing decreased surface organic matter and moisture content, and increased the rate of decomposition of cellulose. It had little effect on the rate of microbial activity measured using CO₂ evolved in laboratory incubation, or N and P measured after KCl extraction, anaerobic incubation, or using ion-exchange resin bags in situ. In the CH type, the treatment decreased mineralisable N and phosphate-P held on resin bags, and the rates of cellulose decomposition and CO₂ evolution. Mixing markedly decreased the cover of salal compared with both the clearcut only and the uncut treatment. The anticipated benefits of mixing and soil disturbance (increased soil nutrient availability brought about by the mixing of mineral and organic horizons) have not occurred. This was probably because the mixing treatment brought humus material from deeper in the soil profile to the surface. Decomposition of this material is limited more by its poorer quality for decomposers (lower N concentration, and possibly higher tannins) than by soil microclimate.