A prospective study for the development of scoring techniques for the diagnosis of ventricular hypertrophy from the three orthogonal lead electrocardiogram was undertaken. A total of 51 hearts was examined at necropsy in a training group on which the scoring techniques were developed and a test group of a further 82 hearts was studied to assess the sensitivity and specificity of the method. The hearts were classified as being normal or having left ventricular hypertrophy, right ventricular hypertrophy, biventricular hypertrophy, or were placed in a borderline category. These classifications were based on the measurement of the weight of the left ventricular free wall, the interventricular septum, and the right ventricular free wall. Classical electrocardiographic variables were selected and assigned a point score on the basis of their diagnostic value. The electrocardiographic diagnosis of left ventricular hypertrophy, right ventricular hypertrophy, and biventricular hypertrophy was then made when the relevant score exceeded four points. After excluding 25 cases which showed conduction defects or were regarded as borderline at necropsy, 57 cases remained in the test group for assessing the technique. The sensitivity and specificity for left ventricular hypertrophy were 65 per cent and 91 per cent, respectively. For right ventricular hypertrophy corresponding results were 53 and 90 per cent. For biventricular hypertrophy the sensitivity was lower at 42 per cent, with a corresponding specificity of 93 per cent. These results represent a considerable improvement over older techniques, with up to a tenfold increase in sensitivity being obtained. The technique can be applied with or without computer assistance.