differences in the definition of acne (that is, whether or not comedones were included in the definition). Bloch, defining acne as the presence of more than one comedone, examined 4191 children aged 6-19 (2076 children aged 12-17) and found,that the prevalence of acne increased from 13% to 97% in girls (aged 6 and 17 respectively) and from 12% to 99% in boys.6 Burton et al, who defined acne in a similar manner, reported a prevalence of 100% in girls aged 14 and boys aged 15.' Fellows et al, who in a longitudinal study of 1500 children defined acne as the presence of papules and pustules but excluded simple comedones, reported an increasing prevalence, from 1-3% to 78% in girls (aged 10 and 15 respectively) and from 0% to 87% in boys (aged 10 and 16 5 respectively).7 The apparent fall in the prevalence of acne in our study compared with that of Burton et al 20 years ago may not be important.' We were careful to include a single comedone as representing acne vulgaris, but we cannot say that a "snapshot" view of acne on one day means that 15% of girls and 7% of boys are never going to develop acne. We agree with Kligman's argument that the social impact of acne is due to its severity and not simply its presence (which he assumed to be 100%).8 Of greater interest is the fall in the severity of acne over the past 50 years. It is difficult to compare the results of previous studies as they all used different scales for grading acne. For comparison with other studies we arranged our results into five groups: no acne, minimal acne, mild acne, moderate acne, and severe acne. In 1931 Bloch reported that 57% of boys and 19% of girls had moderate to severe acne.6 These figures had fallen to 30% and 20% in 1971' and to 35% and 13% in 1981.7 Proportionally there was an even greater shift from the mild to minimal or subclinical groups in this study: we were unable to find any girls with worse than mild acne and only 1 8% of the boys had moderate acne. During the same period the proportion of 12-17 year olds referred for specialist dermatology opinions declined. In 1967, 275 new patients with acne were referred to this department, of whom 98 (36%) were aged 12-17. In 1977 this proportion had fallen to 62 (31%) of 203 referrals and in 1987 to 45 (21%) of 217 referrals. The reason for the decrease in severity of acne is not obvious. It has probably been due to an increase in awareness of appearance, increasing availability and use of over the counter preparations, and better management of acne by general practitioners. The low number of teenagers seeking advice from chemists is almost certainly an underestimate of the value of the pharmacist as a source of advice as many parents consult chemists on behalf of their children. The economic cost of acne is potentially enormous. In 1974 in the United States it was estimated at $314 million dollars (£200 million): $190 million for doctors' fees, $24 million for prescribed treatments, and $100 million for over the counter preparations.910 Similar figures are not available for the United Kingdom, but if dermatologists saw similar numbers of patients -that is, 21 per 1000 population each year-the cost to the hospital budget would be about £58 million if the consultation for each new outpatient is estimated to cost £50. Our results suggest that most cases ofacne are managed well within the community and by general practitioners at a much lower cost. We conclude that adolescent acne seems to be much less troublesome than it was 20 and 10 years ago. This is probably because of more extensive use of over the counter preparations as well as more suitable treatment by better informed general practitioners. The role of the dermatologist in the management of acne in this age group is minimal other than in educating primary health care workers.