SUMMARY1. A two-microelectrode voltage clamp was used to determine the effects of n-butanol, n-hexanol, n-octanol, n-decanol and methyl hexanoate on a transient potassium (IA) current in identified Helix aspersa neurones. Experiments were carried out at a temperature of 10-12 'C.2. Each n-alkanol reversibly reduced the amplitude of the IA current. Logarithmic dose-response curves for the current reduction by each homologue were sigmoidal and had slope factors of around four. The concentrations required to reduce the peak (with time) current at -30 mV by 50 % (ED50 ± fitted standard error) were: 57+5 mM (n-butanol); 2-0 + 0-1 mm (n-hexanol); 0-28 + 0-02 mm (n-octanol) and 0-016+ 0.001 mM (n-decanol). Methyl hexanoate also reduced the current amplitude, with an ED50 of 1-2 mm. The Helix IA current thus showed a similar sensitivity to nalkanols to that of squid and rat sodium currents but was rather more sensitive than the squid delayed rectifier potassium current.3. The n-alkanol ED50 concentrations were used to calculate a standard free energy per methylene group for adsorption to a site of action in the cell of -3-1 + 0-2 kJ/mol. This suggested a hydrophobic site or sites of action. The regularity of the change in free energy with chain length was maintained up to, and including, n-decanol. This implied that the site(s) could accommodate a ten-carbon chain as readily as an eight-carbon chain.4. The voltage dependencies ofIA current activation and steady-state inactivation were not consistently altered by treatment with n-alkanols at concentrations around or above their current suppression ED50 concentrations.5. The kinetics of current activation and inactivation were affected, particularly by lower chain length compounds. At 60 mM n-butanol reduced the time constant for development of inactivation of open channels (Tb) by 56 %, while 0-016 mM n-decanol produced only a 13% reduction. n-Butanol (60 mM) also caused a substantial (76%) reduction in the time constant for development of inactivation in channels which were presumed to be closed. The effects of n-alkanols on the current time-to-peak (tj) were complex, showing both increases and decreases, but these actions also declined with chain length. J. P. WINPENNY, J. R. ELLIOTT AND A. A. HARPER conductances. However, even quite superficial analyses of the mechanisms underlying the inhibition of such conductances have revealed a substantial specificity of action of these structurally simple compounds. The evidence for this specificity and the subsequent implications for more general theories of channel perturbation will be discussed.