Abstract A phonocardiographic analysis employing microphones of different frequency filtering and transmission range was made of the heart sounds in seventy-eight cases of acute coronary occlusion and in 100 normal control subjects. The results were correlated with the clinical findings. The first heart sound was absolutely diminished in amplitude in 24 per cent and relatively to the second sound in 54 per cent of the patients with acute coronary occlusion. This diminution in amplitude affected the central group of high-frequency vibrations and was attributed to the change in the physical character of the infarcted left ventricle and possibly, in the first few days of illness, to the lowered intraventricular pressure following acute myocardial infarction. Occasionally the second sound at the apex is increased to an absolute as well as a relative value. An auricular sound was present in 83 per cent of cases of coronary occlusion compared to 38 per cent in normal subjects. In one-third of the cases of coronary occlusion the auricular sound was accentuated and formed presystolic gallop rhythm. This never occurred in normal subjects. Accentuation of the auricular sound was probably the result of the increased intra-auricular pressure following ventricular infarction. It was practically always associated with heart failure. A third sound occurred in 47 per cent of the cases of coronary occlusion as compared to 12 per cent in normal subjects. The high incidence in the former was attributed to the decreased tonus of the infarcted ventricular muscle. In 9 per cent of the cases the third sound appeared accentuated and produced protodiastolic gallop rhythm. Heart failure was invariably associated with it. Superimposition of the auricular and third sounds of normal or accentuated amplitude occurred in 6 per cent of cases, forming summation gallop. This type of gallop rhythm was also associated with heart failure. Clinical heart failure was present in 63 per cent of the cases of coronary occlusion. It occurred predominantly in those who presented a first sound of diminished amplitude (88 per cent) and gallop rhythm (95 per cent). It was much less common in those with an unimpaired first sound (33 per cent). This emphasizes not only the close relationship between impaired heart sounds and heart failure but also the serious import of a diminished first heart sound and gallop rhythm. Gallop rhythm may be present before signs of heart failure are apparent. The impairment of the first heart sound following coronary occlusion is often permanent and may be the only persistent sign following recovery. It thus may be of diagnostic significance.