We have used the isolated perfused tubule technique, measurements of adenosine 3',5'-cyclic monophosphate (cAMP) content in single tubules, and freeze-fracture electron microscopy to study the basis of high vasopressin-independent (basal) osmotic water permeability (Pf) in the terminal inner medullary collecting duct (IMCD) of the rat. The results confirmed the observation that the basal Pf of the terminal IMCD is considerably higher than that of the initial IMCD. They also showed that the basal Pf of the terminal IMCD is regulated by in vivo factors related to water intake, such that a very high vasopressin-independent Pf can be induced in isolated tubules by prior in vivo thirsting. Tubules from thirsted rats did not display elevated urea permeabilities, nor did they exhibit measurable cAMP levels in the absence of exogenous vasopressin, indicating that the high basal Pf was not due to residual binding of vasopressin to its receptors. Freeze-fracture studies in thirsted rats demonstrated the presence of intramembrane particle (IMP) clusters in both initial and terminal IMCD, with more in the latter. Water loading of the rats suppressed the incidence of clusters almost entirely but did not fully suppress the basal Pf in the terminal IMCD, raising the possibility that a component of transepithelial water transport may occur independently of the vasopressin-regulated IMP clusters. On the basis of these results, we conclude that the vasopressin-independent Pf in the terminal IMCD can be stably elevated to very high levels in response to in vivo thirsting. This elevation appears to be due to a chronic conditioning effect mediated by unknown in vivo factors and is not due to the short-term cAMP-mediated regulatory effect of vasopressin.