The Stewart-Hamilton method (1) for measurement of blood volumes from indicator dilution curves has been validated in both theory (2) and practice (3), provided that certain conditions are met.Measurement of the pulmonary blood volume by the Stewart-Hamilton method would entail injection of indicator into the main pulmonary artery and sampling from the left atrium. However, while insertion of a catheter into the left atrium of the dog is technically feasible, an unknown portion of the left atrial volume will be included in the measurement. A more serious objection is the evidence that the process of mixing of dye with blood in the left atrium is incomplete (4, 5), and thus one of the conditions for valid application of the Stewart-Hamilton method is not fulfilled. The interposition of the left ventricle and aortic valve is essential to achieve uniform cross-sectional distribution of indicator at the sampling site (6, 7). Therefore, it is necessary to place the tip of the sampling catheter in the aortic root distal to the aortic valve. The blood volume between the site of injection (main pulmonary artery) and the sampling site thus comprises both lungs and the chambers of the left side of the heart. This is probably the nearest approximation to the pulmonary blood volume that can be measured at present by the method.Newman, Pearce and associates (8, 9), using a different analysis of indicator dilution curves, introduced the concept that the "slope volume" provides a measurement or index of the pulmonary blood volume. The slope volume is a func-