Research on moral development has relied almost exclusively on the projective method of story completion. Although occasional doubts have been expressed, no systematic investigation of the technique has been reported. In the present study, scores on a test of ‘projective facility’ are found to be associated with maturity of moral judgement. Children aged 9–12 are formed into 2 × 2 groups by age and sex and matched on projective facility. Each child completes 12 stories and the number of immanent justice responses is used as a measure of moral maturity. The main independent variable is type of story, half depicting familiar situations and half originating from foreign cultures. Multiple-choice v. open-ended forms and the sex of the central figure in the story are also systematically varied. All three story variables give significantly different measures of moral maturity. Of the subject variables, age is significant but sex shows no difference. It is concluded that the projective method is liable to serious imperfections and more stringent controls are urged for its future use.