The effects of methyl butyl ketone, methyl heptyl ketone and methyl pentanoate on the sodium current of the squid giant axon have been examined. The peak inward current in intact axons was reduced reversibly by each substance. Sodium currents were recorded in intracellularly perfused axons before and during exposure to the test substances and the records were fitted with equations similar to those proposed by Hodgkin & Huxley (1952). Shifts in the voltage dependence of the steady-state activation and inactivation parameters (m infinity and h infinity), reductions in the peak heights of the activation and inactivation time constants (tau m and tau h) and changes in the maximum sodium conductance (gNa) caused by these substances have been tabulated and compared with the effects of methyl octanoate (Haydon & Urban, 1983b). Each compound shifted the voltage dependence of the steady-state inactivation parameter in the hyperpolarizing direction and that of the steady-state activation parameter in the depolarizing direction. The shifts produced by the ketones are compared with those produced by methyl pentanoate and by methyl octanoate. The possible role of an interaction between the carbonyl oxygen of the test substance and the sodium channel protein in producing the h infinity shift is discussed. The peak time constants are reduced and the voltage dependences of tau m and tau h are shifted in a direction commensurate with the shifts in steady-state properties. The maximum sodium conductance is not much affected either by the ketones or by methyl pentanoate. Large reductions in peak inward current coupled with little effect on gNa have been reported for the n-alkanols and other surface-active compounds (Haydon & Urban, 1983b). This lack of a large effect on gNa indicates that whatever direct interaction does take place between the test substance and the channel protein, it does not result in a blockage of the channel.