Non-invasive determination of cardiac output by aid of precordial first passage radiocardiography may prove a useful tool in the management of cardiac disorders. The basic concepts of a method together with theoretical assumptions necessary for the clinical application of the method are described. The principle of the technique was tested in a cardiac phantom model with adjustable stroke volume and heart rate, in which background problems were avoided. First passage radiocardiography gave highly accurate cardiac output determinations compared with the 'true' phantom values, r = 0.998 (p less than 0.001).