Cr51-labeled red cells and Evans blue dye were used to measure cardiac output, circulating pulmonary and total red cell volumes, plasma volumes and resultant pulmonary and body hematocrits in nembutalized dogs. The pulmonary blood volumes were measured simultaneously by the mean circulation time-output, slope output and arteriovenous equilibration methods and averaged 210, 136 and 151 cc/10 kg, respectively, representing 28.6, 18.7 and 20.4% of the total blood volume. Pulmonary hematocrits averaged 37.6, 36.8 and 33.5%, respectively, compared to mean large vessel hematocrit of 39.7%. Total blood volume averaged 736 cc/10 kg. The mean of the ratios of body to large vessel hematocrit was .92. Cardiac outputs determined from plasma and red cell indicator dilution curves were in good agreement. The mean circulation time of red cells was consistently shorter and the exponential washout slope consistently steeper than of plasma. These differences resulted in a larger calculated total pulmonary volume by the mean circulation time and slope output methods using a plasma compared with a red cell tag; the resultant pulmonary hematocrits were significantly lower than large vessel hematocrits. The pulmonary hematocrits determined by the arteriovenous equilibration method were still lower than those calculated with the other methods. The explanation of this is uncertain. The demonstration by all three methods of a pulmonary hematocrit significantly lower than large vessel hematocrit indicates that the pulmonary bed participates in the lowering by capillary blood of large vessel hematocrit.