[en] The regulation of the CSF [Cl^-] plays a key role in CNS acid-base homeostasis. The authors have shown in previous studies that chloride influx from blood to CSF is largely dependent upon sodium-coupled carrier mediated movement. Therefore, the mechanism of chloride outflux from CSF to brain was evaluated in anesthetized dogs using ventricular-cisternal perfusion (VCP) with the short-lived isotope ³⁸Cl^- and dextran. The outflux of ³⁸Cl^- from CSF was determined from the different movements between ³⁸Cl^- and dextran using a one compartment model. VCP was performed at a rate of 1.4 ml/min for 14 min, and then slowed to 0.28 ml/min. The ³⁸Cl^- activity decreased to a steady state level about 12% lower than that of dextran within 40-50 minutes. Under control conditions (19 runs in 7 dogs), the rate of chloride outflux was 0.059 +/- 0.004 min^-1 (mean +/- SE). It was not significantly changed after the inclusion of bumetanide (10^-5 molar) in the VCP fluid (n=6), which inhibits sodium-coupled Cl^- transport, or with acetazolamide 4.5 x 10^-3 molar (n=4) which inhibits carbonic anhydrase. The authors conclude that chloride outflux from CSF is not dependent upon sodium-coupled carrier mediated movement, which is in contrast with chloride influx from blood to CSF, nor is it dependent upon carbonic anhydrase activity