88 



HANDBOOK OF PH%SIOLOGY 



NEUROPHYSIOLOGY 



each of the pools, a nonpolarizable electrode is im- 

 mersed. The electrodes in the lateral pools are directly 

 grounded and the one in the middle pool is grounded 

 through a resistor of o. i to 0.3 megohms. The cur- 

 rents produced by the fiber in response to an electric 

 shock applied to the fiber near its cut end are recorded 

 by amplifying the IR drop across the resistor. 



If the myelin sheath were a perfect insulator of 

 electricity, no flow ol current should be recorded with 

 this arrangement. Actually, a relatively strong flow of 

 current is observed through the myelin sheath. As can 

 be seen in the records of figure 11.^, the membrane 

 current led through the myelin sheath has clear 

 double peaks of an outward flow, followed by a long 

 phase of a weak inward current. 



When a node of Ranvier is introduced into the 

 middle pool (fig. iiE), an entirely different result is 

 obtained. The flow of current through the membrane 

 of the fiber in the middle pools is triphasic, first out- 

 ward, then inward and finally outward (weak). 

 Comparing the two records in figure 11, it is found 

 that a strong flow of inward current takes place only 

 at the nodes of Ranvier. Since the total amount of 

 current leaving a fiber at any moment has to be equal 

 to the sum of the current entering the fiber at the 

 same moment, the peaks of the outward current 

 through the myelin sheath (record A) should corre- 

 spond roughly to the peaks of inward current at the 

 neighboring nodes (Ni and N2). The effects of more 

 distant nodes are naturally far smaller than those of 

 the neighboring nodes. 



That the first peak in record .-1 of figure 1 1 is caused 

 by the response at node Ni and the second peak by 

 the response at N2 has been shown in the following 

 manner. When a few drops of cocaine-Ringer's solu- 

 tion are introduced in the lateral pool in which N2 is 

 immersed, the height of the second peak is immedi- 

 ately reduced. When the same cocaine-Ringer's solu- 

 tion is applied to the portion of the nerve fiber in the 

 middle pool, no change in the current is observed. 

 Finally, v\hen the narcotizing solution is introduced 

 gradually into the pool of Ni, the height of the first 

 peak is gradually reduced, while the second peak re- 

 mains unchanged until it disappears suddenly at the 

 moment when the propagation of the impulse is 

 blocked. 



Further evidence indicating that electric responses of 

 a myelinated nerve fiber are evocable only at the 

 nodes of Ranvier has been obtained by narcotizing 

 the portions of the fiber located in the lateral pools 

 and stimulating the fiber through two of the elec- 

 trodes (124, 132). When there is one node in the 



middle pool (as in the diagram of fig. 11 B), a full- 

 sized action current can be recorded from a short (i 

 mm) nonnarcotized portion of the nerve fiber. But, 

 when no node is left in the normal Ringer's solution 

 in the middle pool (as in fig. i lA), no action current 

 can be elicited from the fiber. 



The size 01 the membrane action potential at the 

 node was estimated by Tasaki & Takcuchi (135) by 

 measuring the action current and the resistance of 

 the single fiber preparation. Huxley & Stampfli (67) 

 estimated it by compensating the action current with 

 an external voltage source (assuming that the myelin 

 sheath is a perfect insulator). Later, a direct method 

 of recording the action potential of the nodal mem- 

 brane was developed (128). All the.se indirect and 

 direct methods give a figure between 95 and 115 mv 

 at the peak of activity. Later, we shall discuss the 

 difTerence between the shape of the nodal action 

 potential and that of the squid action potential. 



If one assumes that the rnyelin sheath behaves like 

 a condenser with a parallel resistance as shown by 

 the diagram of figure g.^, the flow of current through 

 the myelin sheath should be described by equation 

 (4-1) in the preceding section. The voltage I' in the 

 equation can be either an applied voltage or an 

 action potential developed at the nodes. The two 

 peaks in the current flowing through the myelin 

 sheath (fig. 11. -1^, therefore, are indicative of the 

 situation in which the voltage inside the myelin 

 sheath rises in two steps, one step at the beginning of 

 the action potential at Ni and the other step when 

 X> is also activated. Actually, the time interval be- 

 tween the two peaks is close to the internodal con- 

 duction time discussed previously on p. 79. 



It requires a slight mathematical treatment of the 

 data to separate the current led through the myelin 

 sheath into its capacitative and ohmic components 

 and to determine the absolute values for the capacity, 

 (■„,, and the resistance ;„,, of the myelin sheath (125). 

 Although this method of measuring the membrane 

 capacity and the resistance is not as direct as that 

 for the squid axon, the accuracy of the measurement 

 is fairly high (the probable error being about 10 per 

 cent). The results of recent measurements of these 

 membrane constants are listed in the uppermost 

 column of table i. The observed values of f,„ and r,,. 

 were converted into the values for myelin sheath of a 

 unit area (represented by capitalized figures) by using 

 equations (4-4) and (4-5) in the preceding section. 



The capacity and the resistance of the nodal mem- 

 brane given in table i were determined by measuring 

 the current through node (Ni) in the middle pool of 



