1. A technique is described for recording the bioelectric activity of the squid giant axon during and following alteration of the internal axonal composition with respect to ions or other substances. 2. Experimental evidence indicates that the technique as described is capable of measuring changes in local bioelectric activity with an accuracy of 10 to 15 per cent or higher. 3. Alterations of the internal K+ or Cl- concentrations do not cause the change in resting potential expected on the basis of a Donnan mechanism. 4. The general effect of microinjection of K+ Rb+, Na+, Li+, Ba++, Ca++, Mg++, or Sr++ is to cause decrease in spike amplitude, followed by propagation block. 5. The resting potential decreases when the amplitude of the spike becomes low and block is incipient. 6. The decrease in resting potential and spike amplitude may be confined to the immediate vicinity of the injection. 7. At block, the resting potential decreases up to 50 per cent, but injection of small quantities of divalent cations may cause much larger localized depolarization. 8. The blocking effectiveness of K+, Na+, and Ca++ expressed as reciprocals of the relative amounts needed to cause block is approximately 1:5:100. Rb+ has the same low effectiveness as does K+. Li+ resembles Na+. Ba++ and Mg++ are approximately as effective as Ca++. 9. Microinjection of Na+ may cause marked prolongation of the spike at the injection site as well as decrease in its amplitude. 10. The anions used (Cl-, HCO3-, NO3-, SO4-, aspartate, and glutamate) do not seem to exert specific effects. 11. A tentative explanation is offered for the insensitivity of the resting potential to changes in the axonal ionic composition. 12. New data are presented on the range of variation, in a large sample, of the magnitude of the resting potential and spike amplitude.

1. Current flow outward through the caudal, reactive membrane of the cell causes direct stimulation of the electroplaque. The electrical response in denervated as well as in normal preparations recorded with internal microelectrodes is first local and graded with the intensity of the stimulus. When membrane depolarization reaches about 40 mv. a propagated, all-or-nothing spike develops. 2. Measured with internal microelectrodes the resting potential is 73 mv. and the spike 126 mv. The latter lasts about 2 msec. and is propagated at approximately 1 M.P.S. 3. The latency of the response decreases nearly to zero with strong direct stimulation and the entire cell may be activated nearly synchronously. 4. Current flow inward through the caudal membrane of the cell does not excite the latter directly, but activation of the innervated cell takes place through stimulation of the nerve terminals. This causes a response which has a latency of not less than 1.0 msec. and up to 2.4 msec. 5. The activity evoked by indirect stimulation or by a neural volley includes a prefatory potential which has properties different from the local response. This is a postsynaptic potential since it also develops in the excitable membrane which produces the local response and spike. 6. On stimulation of a nerve trunk the postsynaptic potential is produced everywhere in the caudal membrane, but is largest at the outer (skin) end of the cell. The spike is initiated in this region and is propagated at a slightly higher rate than is the directly elicited response. Strong neural stimulation can excite the entire cell to simultaneous discharge. 7. The postsynaptic potential caused by neural or indirect stimulation may be elicited while the cell is absolutely refractory to direct excitation. 8. The postsynaptic potential is not depressed by anodal, or enhanced by cathodal polarization. 9. It is therefore concluded that the postsynaptic potential represents a membrane response which is not electrically excitable. Neural activation of this therefore probably involves a chemical transmitter. 10. The nature of the transmitter is discussed and it is concluded that this is not closely related to acetylcholine. 11. Paired homosynaptic excitation discloses facilitation which is not present when the conditioning stimulus is direct or through a different nerve trunk. These results may be interpreted in the light of the existence of a neurally caused chemical transmitter or alternatively as due to presynaptic potentiation. 12. The electrically excitable system of the electroplaque has two components. In the normal cell a graded reaction of the membrane develops with increasing strength of stimulation until a critical level of depolarization, which is about 40 mv. 13. At this stage a regenerative explosive reaction of the membrane takes place which produces the all-or-nothing spike and propagation. 14. During early relative refractoriness or after poisoning with some drugs (eserine, etc.) the regenerative process is lost. The membrane response then may continue as a graded process, increasing proportionally to the stimulus strength. Although this pathway is capable of producing the full membrane potential the response is not propagated. 15. Propagation returns when the cell recovers its regenerative reaction and the all-or-nothing response is elicited. 16. Excitable tissues may be classified into three categories. The axon is everywhere electrically excitable. The skeletal muscle fiber is electrically excitable everywhere except at a restricted region (the end plate) which is only neurally or chemically excitable. The electroplaque of the eel, and probably also cells of the nervous system have neurally and electrically excitable membrane components intermingled. The electroplaques of Raia and probably also of Torpedo as well as frog muscle fibers of the "slow" system have membranes which are primarily neurally and chemically excitable. Existence of a category of invertebrate muscle fibers with graded electrical excitability is also considered. 17. In the eel electroplaque and also probably in the cells of neurons, tests of the mode of neural activation carried out by direct or antidromic stimulation cannot reveal the neurally and chemically activated component. The data of such tests though they appear to prove electrical transmission are therefore inadequate for the detection and study of the chemically initiated process.

No abstract