Arginine vasopressin (AVP) has been shown to stimulate active Cl transport across the medullary thick ascending limb of Henle's loop (MAL) in association with an increase in adenylate cyclase activity. To determine whether the failure to demonstrate active Cl transport across the thin ascending limb of Henle's loop (TAL) in previous in vitro perfusion studies was due to the absence of AVP in the preparation, we examined the effect of AVP on adenylate cyclase activity and Cl transport in the hamsters TAL. AVP (1 mU/ml) increased adenylate cyclase activity in the hamster TAL (20.7 +/- 5.2 control vs. 46.2 +/- 10.1 fmol . mm-1 . 30 min-1, n = 6, P less than 0.05) but not in the descending limb (27.8 +/- 7.0 control vs. 20.4 +/- 2.7, n = 4, P less than 0.05). When both MAL and TAL were perfused, a lumen-positive transepithelial voltage (Vt) was observed. The Vt was increased by adding 1 or 10 mU/ml AVP to the bath. When only the TAL was perfused, the Vt was not different from zero. Similar results were obtained in mouse renal tubules. In other experiments, AVP did not affect the diffusion potential generated when a transepithelial NaCl gradient was present. AVP or dibutyryl cAMP caused little or no change in efflux of radioactive chloride across the hamster TAL. These findings suggest that electrogenic chloride transport is not demonstrable in the TAL even in the presence of AVP. The physiologic role of AVP-sensitive adenylate cyclase in the TAL remains to be established.