The selection of an external site for a radioisotope determination of cardiac output is discussed with respect to satisfying the analytical conditions of the Stewart-Hamilton dilution equation. Examples of both single-chamber and multichamber recording are illustrated, along with a discussion of the difficulties of curve calibration and extrapola. tion from multiple sites. The development of a method independent of critical placement or patient variability is described.A LTHOUGH earlier determinations of 14A cardiac output in this laboratory 1-3 by the radioisotope dilution method were performed by isolating the flow from a specific artery, such sampling is not required by the general formulation of the method. Since the obviation of the arterial puncture would considerably simplify this measurement clinically, studies have been extended 4-to determine cardiac output by focusing a gamma ray detector externally over the region of the cardiac chambers or great vessels. We have been concerned since 1951 with the selection of a site that could be observed externally with sufficient reliability for this technic to be accepted as a routine clinical measurement. Many sites of arterial flow may be chosen for quantitation of a dilution curve of cardiac output; therefore, the problem has resolved to a selection of such sites and technics that will not be critically affected by variations in patients.It is the object of the following investigation to examine the choice of the external site From the