1. The use of very fine-tipped and mechanically strong microelectrodes has allowed reliable recordings of resting and action potentials to be made in cockroach giant axons in sheathed and desheathed nerve cords.
2. When the microelectrode was withdrawn from a giant axon in an intact connective the first positive change in the potential from the resting level, was in most cases followed by a negative deflexion to the original zero level, the ‘sheath potential’. The values of this ‘sheath potential’ together with the resting potential, the action potential, the maximum rate of rise and maximum rate of fall of the action potential have been measured in three different salines.
3. In normal saline, resting potentials were lower in sheathed preparations (58·1 ± 55·4 mV.) than in desheathed ones (67·4 ± 6·2 mV.), whereas action potentials were higher in the former (103±5·9 mV.) than in the latter (85·9±4·6 mV.).
4. Elevation of K+ and Ca2+ concentrations in the saline to the haemolymph level resulted in a decrease of resting and action potentials in desheathed cords, to 57·3±5·3 mV. and 36·5±7·6 mV. respectively. No alterations in the membrane potentials were recorded in intact connectives bathed in this saline, the mean resting potential being 55·6±4·2 mV. and the mean action potential 107·9±6·0 mV. Local desheathing of the nerve cord led only to local disturbance of the resting and action potentials, thus indicating that diffusion processes along the extracellular spaces were very slow.
5. The use of a saline in which cation concentrations have been elevated to the extracellular level resulted in normal resting potentials (64·6±3·3 mV.) and action potentials (90·9±7·2 mV.) in desheathed cords, despite the relatively high potassium concentration (17·1 mM./l.).
6. Recordings of the maximum rates of rise and rates of fall showed that there was no significant modification in the shape of the action potential in these different experimental conditions.
7. The values of the ‘sheath potential’ were very variable from one impalement to another and it is suggested that this potential might be related to variations of the microelectrode tip potential bathed in different ionic solutions.
8. The low resting potentials and high action potentials of giant axons in intact nerve cords may result from an excess of inorganic cations in the extracellular fluid.