NEURON PHYSIOLOGY INTRODUCTION 



67 



the EPSP is not suddenly switched off after the summit 

 of the EPSP, but that, as shown in the analysis of 

 figure 5/I (broken line), a small residual current 

 continues to flow and thus delays the repolarization 

 during the decline of the EPSP (continuous line). It 

 will be appreciated that the EPSP's of figures 4 and 

 ^A are produced by the operation on the neuron of 

 the postsynaptic currents generated by many synaptic 

 knobs that have been activated simultaneously by 

 the afferent volley. 



By passing an extrinsic current across the neuronal 

 membrane it has been possible even to reverse the 

 potential across it, its interior then being po.sitive to 

 the exterior. When this occurs, the EP.SP is also 

 reversed in sign (cf. Grundfest, Chapter V, fig. 35), 

 which indicates a reversal of the postsynaptic currents 

 shown in figure 5^ and of the ionic flux across the 

 subsynaptic membrane (17). The effects on the EPSP 

 of diminution and reversal of the membrane potential 

 and of changes in the ionic composition of the neuron 

 are explicable by the postulate that the activated sub- 

 synaptic membrane becomes permeable to all small 

 ions, such as Na"*", K"*" and Cl~. The time course of 

 this permeability change is given by the broken line 

 of figure ^A, and its effect on the membrane potential 

 can be derived from the electrical diagrain of figure 

 5C. A similar investigation on the endplate potential 

 of the neuromuscular junction (24, 26; Fatt, Chapter 

 VI) has shown that reversal occurs at a membrane 

 potential of about — 1 5 mv, which would be close to 

 the liquid-junction potential between the muscle fiber 

 and its environment. More accurate investigations on 

 the EPSP may likewise reveal that a battery of about 

 — 1 5 mv should be inserted in the synaptic component 

 of the diagram in figure 5C. 



It can now be taken as established that transmission 

 across synapses occurs not by the spread of electrical 

 currents, but by the specific chemical substances 

 which impulses cause to be liberated from the pre- 

 synaptic membranes (29, 38, 43). These substances 

 alter the ionic permeability of the subsynaptic mem- 

 brane and consequently initiate specific ionic fluxes 

 across this membrane. These fluxes in turn are re- 

 sponsible for the postsynaptic currents that cause the 

 transient depolarizations or hyperpolarizations of the 

 postsynaptic membrane which are produced respec- 

 tively by excitatory or inhibitory action (16, 1 7). Since 

 it gives the time course of the ionic permeability 

 change, the broken line of figure 5.-I may be taken to 

 give the time cour.se of action on the subsynaptic 

 membrane of the brief jet of excitatory transmitter 

 substance that a presynaptic impulse causes to be 



emitted from the presynaptic knob. .Acetylcholine is 

 the transmitter substance at a few types of central 

 synapse, but the excitatory transmitter has not yet 

 been identified for the great majority. 



Impulses can also be generated in a nerve cell by 

 another method that is of particular value in relation 

 to the problem of locating the site at which impulses 

 arise in nerve cells. When the a.xon of a nerve cell is 

 stimulated, an impulse travels antidromically up to the 

 nerve cell and usually invades it, generating an anti- 

 dromic spike potential as in figure 6A. When thus 

 recorded by a microelectrode in the soma, the anti- 

 dromic spike potential has two main components, as 

 shown by the step on the rising phase which is greatly 

 accentuated in the electrically differentiated record 

 lying immediately below the potential record in 

 figure 6.4. Evidence from recent intensive investiga- 

 tions (i, 7, 13, 39, 40, 46) can all be satisfactorily 

 explained by the postulate that the initial small spike 

 is generated by the impulse in the initial segment of 

 the neuron (axon hillock plus nonmeduUated axon), 

 while the later large spike is produced when the 

 impulse invades the soma-dendritic membrane (13, 

 46). The two spikes may therefore be called the IS 

 and SD spikes. 



When the neuronal spike potentials generated by 

 synaptic or direct stimulation are recorded at suffi- 

 cient speed, they are likewise seen to be compounded 

 of IS and SD spikes, particularly in the differentiated 

 records (fig. 6B), though the separation is always less 

 evident than with the corresponding antidromic 

 spike potential. It must therefore be postulated that 

 the EPSP produced by the activation of synapses 

 covering the soma and dendrites is effective not by 

 generating an impulse in these regions, but by the 

 electrotonic spread of the depolarization to the initial 

 segment, as is illustrated by the lines of current flow 

 in figure 6C. By recording the impulse discharged 

 along the motor nerve fiber in the ventral root it is 

 found that usually this impulse started to propagate 

 down the meduUated axon about 0.05 msec, after the 

 initiation of the IS spike, i.e. the meduUated axon is 

 usually excited secondarily to the initial segment (14). 

 The critical level of depolarization for generating an 

 impulse thus gives the threshold for the IS mem- 

 brane, as marked by the horizontal arrow labelled 

 IS in figure 65, and not of the SD membrane. An 

 approximate measure of the threshold for the SD 

 membrane is given by the membrane potential ob- 

 tained at the first sign of inflection produced by the 

 incipient SD spike, as is indicated by the differentiated 

 records in figure 6.4 and B. This potential is measured 



