1. Plasma from hypervolaemic rats was fractionated on a G-200 Sephadex column. In addition to three different protein peaks, a fourth non-protein fraction was obtained. Each of the four peaks was desalted, freeze-dried, reconsituted and injected into normal anaesthetized rats. Significant natriuretic responses resulted, from injection, of the middle protein peak and of the small-molecular-weight peak. It was concluded that a natriuretic humoral factor was present in the blood of hypervolaemic rats, and that this factor was of low molecular weight but normally occurred bound to plasma protein. 2. The renal response to injection of non-protein fraction, obtained from either hypervolaemic donors or from iso- or hypo-volaemic donors, was compared in two groups of bioassay rats to test whether the natriuretic factor was present only in plasma of the blood-volume-expanded animals. Both types of reconstituted fraction caused diuresis, natriuresis and kaliuresis in bioassay animals. Only the natriuretic response was statistically greater when the fraction obtained from hypervolaemic plasma was used. In addition to non-specific increase in fluid and ion excretion, possibly due to the extraction and/or methodological procedures, these results demonstrate that blood-volume expansion releases a humoral natriuretic factor into plasma. Since there were no increases in filtration rate, the factor specifically inhibited tubular sodium reabsorption. 3. To determine the maximum possible effect of the non-protein factor, the dose given to bioassay rats was tripled. There was no further increase in sodium excretion, indicating that the effect was quantitatively limited and suggesting that the physiological importance of natriuretic hormone lies in long-term regulation of body-fluid balance.