AN attempt was made to determine the amount smoked by the pupils of an independent part boarding school in Yorkshire where smoking was prohibited. An anti-smoking demonstration was given to the pupils and its effect on them was examined.Method . Three sets of questionnaires were given to the children. The first-I-in the morning of the day an anti-smoking demonstration was given (at the end of the Christmas term, 1963), the second-II-the morning after the demonstration, and the third-IIIthree months later. The first and third questionnaires were similar, and were designed to find out the smoking habits of the pupils in relation to certain factors, the second questionnaire attempting to find out what in the demonstration had impressed them. From the answers to all three could be found the differences in knowledge of and attitude to the hazards of smoking before and after the demonstration and also three months later.The questionnaires were answered anonymously. Other surveys (I for discussion) have found the questionnaire method the best to use, and by comparing the answers to similar questions in questionnaires I and III of this survey the conclusion that the truth was being told was reached. The author explained the situation to the pupils at their morning assembly, stressing that the children's answers were important and that no teacher would see them. On dispersal to the form rooms each child filled in his answer sheet privately and folded it up. It was then put into a