1. The nephron sites involved in the blunted natriuretic response to frusemide during sodium depletion were investigated using micropuncture techniques in anaesthetized rats. 2. Glomerular filtration rate was lower, and fractional sodium reabsorption in the proximal convoluted tubule higher, in sodium-depleted than in sodium-replete rats. Consequently, sodium delivery to the loop of Henle was reduced (by approximately 35%) in the sodium-depleted animals. Intravenous frusemide (2.5 mg h-1 kg-1; urinary water and electrolyte losses replaced) had no effect on glomerular filtration rate or proximal tubular sodium reabsorption in either group. 3. The inhibitory effect of intravenous frusemide on fractional sodium reabsorption in the nephron segments constituting the loop of Henle (measured by free-flow micropuncture) was attenuated during sodium depletion. However, when loops of Henle were microperfused at identical rates with artificial late proximal tubular fluid, no difference in the responses of sodium-depleted and sodium-replete rats to intraluminal frusemide (10(-5) mol/l) could be detected. 4. In sodium-replete animals, the increased load of sodium delivered from the loop of Henle during frusemide administration resulted in a lowering of fractional sodium reabsorption in the distal tubule. In contrast, in sodium-depleted rats given frusemide, fractional distal sodium reabsorption tended to increase, so that values in the two groups of frusemide-treated animals were markedly different (0.30 +/- 0.04 versus 0.51 +/- 0.03). 5. It is concluded that the blunted natriuretic response to frusemide during sodium depletion results from at least three factors: a reduced sodium delivery to the loop of Henle; a reduced inhibitory effect of frusemide on fractional sodium reabsorption in the loop of Henle, which may be a consequence of the reduced sodium load; and enhanced fractional reabsorption of sodium in the distal tubule, which partially buffers the diuretic-induced increase in sodium delivery from the loop.