Na current fluctuations in nodes of Ranvier were measured under voltage clamp conditions as described in the preceding paper (Conti, Hille, Neumcke, Nonner & Stämpfli, 1976) and analysed in terms of power spectral density calculated for frequencies between 30 Hz and 5 kHz. 2. External (10(-5) g/ml.) Leiurus scorpion venom or Anemonia Toxin II (3 X 10(-5) g/ml.) or internal 20 mM iodate were applied in order to remove or slow down inactivation in part of the Na channels. The treatment increased the steady-state Na current during the noise measurement one-to eight fold over that in normal fibres. 3. Noise spectra were interpreted as the sum of 1/f noise and noise SNa(f) due to all-or-none, open-close transitions of single Na channels. The drug effects on the inactivation could be accounted for either by assuming two populations of channels, one with and one without inactivation, or by postulating a single population with modified inactivation characteristics. 4. Except for an increase in amplitude, the fluctuation spectra SNa(f) were similar to the ones in normal nodes. Again, the time constants taum obtained from the fit of the spectra agreed within a factor of 2 with the values of taum found in the macroscopic Na currents. 5. From the fluctuation spectra, single Na channel conductances gamma of 5-4 +/- 0-4 pS (iodate), 6-7 +/- 0-5 pS (Leiurus) and 7-0 +/- 0-6 pS (Anemonia) were calculated. The value of gamma was not significantly voltage dependent. 6. Our observations indicate that inactivation of Na channels can be modified with at most small effects on the microscopic properties of the activation process and on the conductance of the open channel. They suggest that the h mechanism normally produces all-or-none, open-close changes of conductance.
