A study of the floristic structure of the bracken community indicated a continuum from high frond densities with a species-poor sward, growing on an organic and acidic soil, to lower frond densities with a relatively species-rich sward on a more mineralised, less acid soil. An analysis of the biomass structure of the community showed that an increase in frond density, litter deposition and long shoot rhizome length, together with a decrease in soil pH, were correlated with a decrease in turf biomass. Excluding livestock appeared to be associated with a decrease in bracken frond mortality but also significantly fewer frond buds and unemerged fronds were found below ground. This suggests that frond damage caused by livestock stimulates the production of frond buds and, paradoxically, may lead to higher frond densities. Asulam reduced frond production, in the year after spraying, by approximately 80 per cent. However, the rates of recovery were very variable such that full recovery or no recovery were observed after three years. The percentage recovery was most rapid in areas with the highest frond densities prior to treatment and may be due to proportionally more rhizome remaining alive after spraying. No significant effect of bracken control was seen on the productivity of the underlying native sward. Furthermore the application of lime and phosphate and the restriction of grazing failed to significantly increase sward cover. The application of agricultural seed mixtures indicated that those containing Agrostis tenuis appeared to have the most potential for rapid sward development, with the removal of bracken litter and restriction of grazing being unnecessary. A model was constructed to suggest two appropriate management strategies for long-term bracken control after spraying. For sparse turf, reseeding and the application of lime and phosphate are necessary but for reasonably well-developed turf no further management is required.