In the early period of cardiac catheterization there was little inclination to pass the catheter tip beyond the right atrium, and therefore only right atrial pressures and cardiac output measurements were available (McMichael and Sharpey-Schafer, 1944a;Stead et al, 1948). The essential findings in this and Stead's laboratory were that digoxin or lanatoside C reduced the right atrial pressure early, and usually increased the cardiac output. Both groups, however, observed a few cases in which the right atrial pressure fell without any significant change in cardiac output. This pressurereducing action was frequent enough to suggest that it might play a primary role; and this view was supported by (1) simultaneous observations that venesection might cause an increase in cardiac output of the same order as that produced by digoxin, and by (2) published work suggesting that digitalis reduced venous pressure in normal animals (Dock and Tainter, 1930) and in man (Rytand, 1933). This conception had the additional attraction of offering a possible explanation for some of the failures of digitalis therapy in certain instances of " high output heart failure " in which increased venous pressure was perhaps compensatory in nature.Further observations, however, have compelled revision of this view. On making detailed comparisons it was realized that the heart did more work after digoxin than after venesection (Howarth, McMichael and Sharpey-Schafer, 1946). Bloomfield et al. (1948) showed that in a wide variety of patients with heart failure ouabain had a direct stimulating action on the myocardium-a result that has been confirmed by Ahmed et al. (1950). In selected groups of cases it has been shown on the one hand that digoxin can increase the cardiac output in left ventricular failure without there being necessarily a concomitant reduction in venous pressure (Harvey et al., 1949), and on the other hand that the venous pressure may fall without there being any measurable increase in output (McMichael and Sharpey-Schafer, 1944a). Deductions from the results of earlier observations were based on Wiggers' interpretation of Starling's law that small changes in filling pressure of the heart were of paramount importance in determining the strength of the cardiac contraction. Such an assumption cannot be accepted unreservedly. Warren et al. (1948) have shown that the output of the normal human heart is subject to controls more complex than simple government by Starling's law. Thus the significance of small changes in filling pressure of the order of 1-2 cm. saline becomes more difficult to assess. Experience has also made us aware of the random error in mneasurement of cardiac output by the technique of catheterization, the great variety of hemodynamic changes in different types and grades of heart'failure, and the varied responses to digitalization.Unanswered problems and conflicting observations still abound. Harvey et al. (1949) have found in left ventricular failure without systemic venous congestion that digoxin produced a fall