204 INFLUENCE OF TEMPERATURE ON BIOLOGICAL SYSTEMS 



Ringer or cocaine-Ringer. The partitions between the compartments meas- 

 ured about 0.5 mm in thickness. After the single fiber preparation was 

 passed through small holes in the partitions, these holes were sealed with 

 vaseline. 



In all three of these methods of recording the action current, the input 

 resistance R (300 K in the figure) was situated in such a position that 

 the longitudinal current, I, fiowing through the internode between Ni and 

 No produced an IR drop across this resistance. The voltage at the input 

 of the preamplifier varied directly with the value of R in a wide range of 

 R, indicating that the action current I is independent of the input resistance 

 R. Since, however, the response time of the input stage of the preamplifier 

 increased with increasing R (due to the capacity of the recording elec- 

 trode and of the preamplifier input to ground), it was not possible, under 

 ordinary experimental conditions, to increase R above 1 megohm with- 

 out losing high-frequency components in the response. When the ex- 

 citability of the node No was eliminated by the use of cocaine, action cur- 

 rents that roughly parallel the membrane potential at Ni were obtained. 



Measurement of Action Potential. The isolated fiber in the dissected 

 region of the nerve was suspended across the gap between a glass plate 

 and a small electrode of the Ag-AgCl type (fig. 2A). The air gap meas- 

 ured 0.5-1 mm. The portion of nerve on the glass plate was immersed in 

 normal Ringer. The electrode that was immersed in the pool of Ringer 

 on the glass plate was grounded (through 500 ohms). 



The cathode-follower used for this purpose had an effective input ca- 

 pacity of about 1 fi/xf or less. The details of a high input-impedance pre- 

 amplifier especially designed for this purpose in this laboratory will be 

 presented elsewhere (16). The grid electrode was completely enclosed in 

 a metal shield driven by the output of the preamplifier which had a gain 

 of unity. After eliminating the excitability of the node (N2) on the grid 

 electrode wath an isotonic KCl solution, both the resting and the action 

 potential of the fiber could be measured directly without any appreciable 

 distortion or reduction. The resistance of the preparation (measured by 

 using a square voltage pulse across the 500-ohm resistor and the 22-meg- 

 ohm resistor at the input) was between 30 and 60 megohm. 



The action potential of the individual nerve fiber can also be recorded 

 by the use of a submicroscopic microelectrode. But, at present, the results 

 obtained with submicroscopic electrodes are less reliable than those ob- 

 tained by the method described above. 



Measurement of Membrane Resistance and Capacity. The principle 

 of the method used for measuring the resistance and capacity of the mye- 

 lin sheath and the nodal membrane is illustrated in figure 2B. The single 

 fiber preparation was mounted on a platform consisting of three pools of 



