208 



HANDBOOK OF PHYSIOLOGY 



NEUROPHYSIOLOGY I 



discharge is left unchanged; the same effect is pro- 

 duced by the addition of magnesium ions (4, 19, 

 20, 21, 41, 63). Calcium appears to exert a specific 

 facilitatory action on the release of acetylcholine by a 

 nerv'e impulse, and the action of magnesium may 

 then be accounted for by a competition with calcium 

 for the reactive site. This antagonistic relation be- 

 tween calcium and magnesium at the terminals is in 

 contrast to their common action in raising the 

 threshold depolarization for the initiation of an action 

 potential in a nerve or muscle fiber. By the with- 

 drawal of calcium or addition of magnesium or both, 

 the endplate potential can be reduced to a small 

 fraction of its normal size and can be made to ap- 

 proach in amplitude the spontaneous miniature po- 

 tential. When this is done, the amplitude of the 

 response to successive nerve impulses is seen to fluc- 

 tuate widely, in contrast to its constancy under 

 normal conditions or when the response is reduced to 

 any degree by treatment with curare. With the 

 junction sufficiently deprived of calciimi, the response 

 occurs intermittently. When the proportion of fail- 

 ures is large, the responses to a series of ner\e im- 

 pulses have a distribution of amplitudes similar to 

 that of the spontaneous discharge. At a somewhat 

 lower level of depression, the distribution shows 

 several peaks, corresponding to small integral multi- 

 ples of the mean of the spontaneous potentials. It is 

 evident that the endplate potential under these condi- 

 tions is composed of a variable whole number of 

 miniature endplate potentials, the fluctuation being 

 due to variation in number and in size of units. The 

 result of an analysis of this fluctuation for the proba- 

 bility of occurrence of different numbers of units can 

 be accurately fitted by a Poisson distribution. This 

 implies that there is no interaction between units, 

 the probability of occurrence of each being unaffected 

 by the number of units composing the response. 



With the distribution in this form, relatively large 

 fluctuations in response would occur only when the 

 number of contributing units is small. As the number 

 increases, the amplitude of fluctuation relative to the 

 mean amplitude of response will vary in inverse 

 proportion to the square root of the number of units, 

 while the additional dispersion due to variation in 

 amplitude of individual units will become pro- 

 gressively less significant. Fluctuations occur in the 

 curarized endplate potential, evoked under condi- 

 tions in which the release of acetylcholine from the 

 nerve terminals is normal, and these can be attributed 

 to a variation in the number of units around such a 

 magnitude as would be predicted roughly from the 



size of the normal response (68). The probability 

 that the normal endplate potential is composed of 

 these units is greatly strengthened by the observa- 

 tion that an increase in the calcium ion concentra- 

 tion beyond that normally present in the bathing 

 fluid produces a further increase in the size of the 

 response, which is entirely attributable to an increase 

 in the release of acetylcholine from the nerve ter- 

 minals, and which is presumably due to an increase 

 in the probability of release of individual units of 

 acetylcholine (13, 29, 43, 52). The curarized end- 

 plate potential can in this way be increased two or 

 three times in size. 



In contrast with the effect on the response to a 

 nerve impulse, changes in the calcium concentration 

 in cither direction from normal are usually found 

 to have no effect on the frequency of spontaneous 

 miniature potentials. Calcium withdrawal (or mag- 

 nesium addition) does, however, reduce the fre- 

 quency when this has first been raised by the presence 

 of a high concentration of potassium ions or by a 

 current applied to depolarize the nerve terminals. It 

 thus appears that the depletion of calcium ions has a 

 similar action in presenting an increase in the proba- 

 bility of a unit of acetylcholine being released during 

 a given time interval by a maintained depolariza- 

 tion, as in reducing the probability of its release by a 

 nerve impulse. 



Another procedure which modifies the number of 

 units responding to a nerve impulse is the previous 

 activation of the nerve. In the curarized amphibian 

 muscle, the second of two closely spaced nerve im- 

 pulses elicits a larger endplate potential than does the 

 first (30, 43, 72). With continued repetitive stimula- 

 tion of the nerve, the individual responses increase 

 progressively until a steady condition is attained. By 

 this means the size of the response may be increased 

 to two or three times that elicited by an isolated im- 

 pulse. (This increase is in addition to the summation of 

 electrical changes in the postjunctional structure, the 

 later responses adding to the potential change remain- 

 ing from previous responses.) In the case where two im- 

 pulses are set up in the nerve the potentiation of the 

 response to the second is found to be greatest at the 

 shortest interval of time at which the ners-e will con- 

 duct. The effect falls gradually as the interval be- 

 tween the nerve impulses is increased, the response 

 having returned to its unpotentiated size at an inter- 

 val of about 100 msec. That this potentiation is a 

 prejunctional phenomenon and moreover that it in- 

 volves a change in the number of units of acetylcho- 

 line released is revealed by studying the effect under 



