SEQUENCE OF EVENTS 227 



active — COOH groups bury themselves in the water phase of the axis 

 cylinder. If an increase in hydrogen-ion concentration of the water 

 phase should now take place, it would diminish the attraction of the 

 — COOH group for the water, and thus cause a change in the surface 

 energy of the monomolecular array and consequently a change in 

 potential. 



When the nerve fiber becomes locally active a rearrangement in the 

 monomolecular array may take place. The normal difference of poten- 

 tial suffers a transient decrease followed by a slower restoration so that a 

 sensitive voltage-measuring device in contact with the point of excitation 

 on the external surface of the nerve will show a pulse-like variation of 

 potential with time. This transient lasts about 0.004 second. About 

 one tenth of this time is used for the potential to rise to its maximum 

 value. 



The change in the state of the fiber that permits the development of 

 this transient is not known. Any theory that may be proposed must be 

 based on the following well-established facts. The duration of the spike- 

 like pulse is constant for a restricted range of speeds of propagation and 

 fiber diameters (Table VI-1). If the spike appears at all, it appears 

 with the maximum size which the intrinsic properties of the nerve permit. 

 The nerve pulse is independent of the nature of the stimulus which starts 

 it. Its speed of propagation increases with the diameter of the fiber and 

 increases with a rise in temperature. If its amplitude is locally 

 depressed, it attains its full size when it emerges into a subsequent nor- 

 mal nerve section. The energy with which the action is maintained is 

 supplied locally. 



At various times, as the experimental data have accumulated, models 

 of nerve excitation have been developed to coordinate the existing data. 

 One of the early models designed to illustrate pulse propagation consisted 

 of a wire-core axis cylinder while the interstitial fluid was replaced by 

 an envelope of electrolytic solution. This was the so-called core con- 

 ductor model which, however, lacks the equivalence of a bioelectrical 

 membrane. Hermann used it to show that it duplicated the " passive " 

 electrical properties of the nerve. 



In spite of the perfected physical analogies attained by such models, 

 one can duplicate biophysical conditions only very approximately, by 

 substituting a polarizable organic interfacial membrane for the oxide film 

 and a fluid conductor of physiological composition for the metallic core. 

 Labes and Zain's [1927] model approached these conditions by using 

 collodion sacs filled with a neutral solution of potassium phosphate for 

 the cores, which were surrounded with sodium chloride solution isotonic 

 with the phosphate mixture. The inner and outer solutions of a series 

 of these collodion cells were placed in communication with each other 



