This paper presents the main findings of a review of the literature on leisure noise (sociacusis) published elsewhere (MRC Institute of Hearing Research, 1985). From this, it has been possible to draw certain limited conclusions, despite the research's generally poor design, methodology and reporting. Given that the present occupational hearing conservation programme in the U.K. is geared to levels of 90 dB(A), non-occupational noise exposure is unlikely to constitute a major source of noise-induced hearing loss in the U.K. population. The Review contains recommendations for research to fill gaps in the literature. Some of the most important questions remain unanswered or unasked even in the most heavily-researched categories such as amplified music. While sound levels have been fairly well described more attention needs to be paid to the often neglected parameters of leisure noise exposure, i.e. the numbers exposed and the patterns and durations of their exposures. The outstanding need in leisure noise research is for a large, random-sample, whole-population survey in the U.K. of the numbers exposed and the patterns of exposure to leisure noise in general and amplified music in particular.