1 70 



HANDBOOK OF PHVSIOLCKJY 



NEUROPHYSIOLOGV 



I I I I I I I 



msec 



FIG. 19. Synaptic transfer in squid giant axons. The incom- 

 ing presynaptic spike elicits only a small membrane potential 

 change in the postsynaptic cell. The p.s.p. arises after a brief 

 latency and, if it attains the critical firing level, elicits a spike. 

 [From Bullock & Hagiwara (32).] 



need not be as large as is the excitatory one. It must 

 only be large enough to decrease the depolarizing 

 p.s.p. below the critical firing level for the spike, but 

 it can then produce dramatic effects since the absence 

 of conductile activity eliminates further transfer to 

 other cells and results in the disappearance of distant 

 actions within the organism. 



Synaptic Delay 



Synaptic latency, which was discussed above, in- 

 volves only the activity of the presynaptic terminals 

 and the response of electrically inexcitable synaptic 

 membrane. Synaptic delay includes not only the 

 latency but also the utilization time of electrical 

 excitability. This last involves the duration of the rise 

 of the depolarizing p.s.p. and of whatever further 

 depolarization this may develop in its excitatory ac- 

 tion on electrically excitable sites, and the consequent 

 time that is required for the p.s.p. (and the local 

 response) to reach the critical level for evoking a 

 spike. The rise time of the p.s.p. for this level may be 

 brief, about o. i to 0.3 msec. (figs. 6, 12), but can be 

 much longer (figs. 7, 9), particularly if the depolariz- 

 ing p.s.p. is liminal for discharge of the spike. Tem- 

 poral summation or facilitation, in which repetitively 

 evoked depolarization becomes larger, may then de- 

 crease the utilization time and thereby shorten the 

 synaptic delay (cf. 140). The shortening might also 

 occur because of decreased synaptic latency or 

 heightened synaptic excitability, effects which are 

 discussed in the next section of this chapter. 



The existence of synaptic delay has been a.scribed 

 chiefly to slowed conduction of the afferent impulse 

 in the fine terminals of the presynaptic fibers (cf. 



57, 140). That explanation is no longer tenable. 

 Strong electrical stimuli directly applied to the inner- 

 vated surface of the eel electroplaque, and therefore 

 to the nerve terminals, nevertheless cause a neurally 

 evoked response always after a considerable synaptic 

 latency (fig. 6). Further evidence may be derived 

 from figure ig and other data of similar nature which 

 show that the presynaptic spike arrives at the synap- 

 tic surface somewhat before the p.s.p. is elicited. Thus, 

 .synaptic latency and the utilization time involved 

 in the rise of the p.s.p. to the critical firing level are 

 probably the major factors in synaptic delay. 



Sujifrjuisition of Pustsyriafitic Potentials and Spikes 



The electrically inexcitable generators of p.s.p. 's act 

 independently of and in parallel with the electrically 

 excitable membrane that produces the spike (4, 

 48, 71). Thus, a p.s.p. can be evoked during the 

 spike, when a second response of the electrically 

 excitable membrane is impossible due to its absolute 

 refractoriness (figs. 6, 7). However, the combined 

 response depends upon the prevailing electrochem- 

 ical conditions of the cell. The p.s.p. may subtract 

 from as well as add to the spike, the former occiu'ring 

 when the spike itself carries the membrane potential 

 into the region at which the p.s.p. reverses as de- 

 scribed above (48, 136; cf 97). The conclusion that 

 the spike under certain conditions wipes out the 

 p.s.p. (cf 60, p. 30 ff) may therefore require revision. 

 A complicating factor that ma)' explain these find- 

 ings of Eccles and his colleagues is the distortion 

 produced in the spike when the latter is elicited in a 

 depolarized electrically excitable membrane (cf 

 95). An "undershoot' of apparently hyperpolarizing 

 phase then terminates the spike, even though it is 

 absent in the response evoked at the normal resting 

 potential of the membrane (fig. 10; cf 60, fig. 16). 

 The distortion is probably due (95) to excess of po- 

 tassium conductance over the sodium conductance 

 as in .squid giant axons (113). This excess would be 

 caused h\ increased sodium inactivation produced 

 by the depolarization. 



The foregoing remarks indicate that electrical and 

 physiological conditions of the soma membrane affect 

 the recording of celhdar potentials. The soma, how- 

 ever, is only one part of the cell, although it is the 

 one most easily accessible to microelectrodcs. Even 

 in neurons without dendrites, as is the case in tissue- 

 cultured dorsal root ganglion cells, the intracellularly 

 recorded response to stimuli may take on complex 

 forms (42). This indicates that activits in and the 



