Dogs were subjected to volume expansion with physiological saline and 5% glucose solution. During the infusion Na excretion was significantly increased only when saline was given. Maintenance of the volume expansion with 5% glucose solution immediately following saline infusion was ineffective in preventing the decline of Na excretion rate after the cessation of saline loading. Although the indices of hemodilution such as plasma total protein concentration and peripheral venous hematocrit were reduced along with natriuresis, these values also correlated with the total amount of Na retained during saline infusion. And this amount of retained Na has also a good correlation with the natriuresis. It is concluded that hemodilution is not the essential cause of natriuresis in the present volume expansion experiments, and that the amount of Na retained during saline loading determines the Na excretion rate in some unknown manner. The implication of regression coefficient between Na retention and Na excretion is discussed as a characteristic value in the Na homeostatic mechanism of the organism.