Summary European forestry strategies place emphasis on developing alternative management practices to clearfelling within commercial forests as a means of increasing the non-market benefi ts of sustainable forestry. In the UK, many thousands of hectares of forest are being transformed to continuous cover forestry and a number of minimum intervention natural reserves are being created to encourage the development of old-growth conditions. This paper defi nes the term ‘old growth’ in the context of upland spruce-dominated plantations in Britain and evaluates different options for the location, design and management of old-growth areas to enhance biodiversity. Evidence outside of Britain from semi-natural analogues of upland spruce plantations suggests that old growth can develop 100–200 years after stand initiation in those parts of the landscape not subject to frequent catastrophic disturbance by wind and fi re. Old-growth stands in these forests are characterized by a high proportion of large, old trees, multiple age classes and high volumes of fallen and standing deadwood. Studies of old spruce stands in the British uplands suggest that old-growth features can begin to develop after 80–100 years, conferring substantial benefi ts to species-groups such as hole-nesting birds, mammals (e.g. red squirrel), bryophytes, lichens and fungi. Based on the likelihood of wind damage it is suggested that ~50 per cent of the current land area in upland Britain could support large patches (50–100+ ha) of old growth. To enhance colonization by woodland species, these old-growth patches should be within 2 km of existing semi-natural woodland, managed ideally on a minimum intervention basis. If production of timber was also an objective, old-growth stands could be managed by single tree selection or small group-fell silviculture, provided that over-prolifi c regeneration of shade-tolerant conifers was controlled and some deadwood and veteran trees were maintained. In surrounding areas subject to normal patch-clearfelling, small enclaves of old-growth forest (0.25–2 ha) could be retained to provide temporary habitat for species and facilitate dispersal through the landscape. The planning and design of old-growth areas needs to be considered at the landscape scale to ensure an appropriate balance between old growth and other types of woodland and nonwoodland habitats. An imaginative approach to incentives will be required to encourage positive management for old growth.