Radiological examination of the heart has been practised from the earliest introduction of X-rays and gradually the importance of radioscopy, including the right and left oblique views, has gained ground as a valuable method of distinguishing the different chambers of the heart. In this country the method has been developed and applied largely by the teaching of Parkinson (1933 and.As early as 1931 Forssmann thought of the additional help that could be obtained in outlining the cavities of the heart after injecting opaque media by catheter. Laubry et al. (1935) injected hearts post-mortem and published some beautiful illustrations of the normal anatomy that found a useful basis for comparison with subsequent angiocardiograms. They showed the U-shaped right side and the window between the descending limb, formed by the right auricle, and the ascending limb, formed by the outflow tract of the right ventricle, with the right ventricle itself forming the horizontal base, and the right branch of the pulmonary artery an upper cross-bar.In 1938 Castellanos, Pereiras, and Garcia showed some good angiocardiograms of children with congenital heart disease. They demonstrated the normal U-shaped curve of the right side, the window between the right, auricle and the outflow tract of the right ventricle, and the post-stenotic dilatation of the infundibulum. Chavez et al. (1947) have given a summary of the gradual progress towards satisfactory angiocardiograms, referring particularly to the work of Ara, Cossio, Moniz, Carvalho and Saldanha, and Castellanos and Pereiras.Angiocardiography was developed on a practical scale by Robb and Steinberg (1939) who published many illustrations of the additional information they obtained by this method. They found that the right auricle was shown at 1-5 sec. after the injection, the right ventricle and pulmonary artery at 2-3 sec., the pulmonary veins and left auricle at from 6 to 8 sec., and the left ventricle and aorta at from 8 to 10 sec. This work has been extended on a large scale in several centres in America.Steinberg, Grishman, and Sussman have published a series of papers on angiocardiography in congenital heart disease, dealing particularly with a case of Fallot's tetralogy (1941), with dextrocardia (1942), with intracardiac shunts (1943a) with patent ductus arteriosus (1943b), and with the radiology of congenital heart disease (1943c). The paper on shunts is particularly pertinent to our present subject; 18 of the 65 congenital cases they had examined were thought to have shunts; 10 of these being through auricular defects and 2 through ventricular defects, and 6 due to an overriding aorta or one with some degree of dextro-position, 4 of these being regarded as Fallot's tetralogy and 2 as Eisenmenger's complex. Only two had post-mortem confirmation of the diagnosis.They found the time of filling of the pulmonary vein and left auricle (3*5-4-5 sec.) and of the left ventricle and aorta (5 0-80 sec.) rather quicker than Robb and Steinberg, possibly because more children were exam...