228 THE BIOPHYSICAL PROBLEM OF NERVE CONDUCTION 



through and with the external fluid by a series of fine glass capillaries 

 with which the semi-permeable nature of the membrane was simulated. 

 The iron- wire model developed by Lillie [1923] carries the analogy into 

 the dynamic stage, and it is generally accepted as the most complete 

 model of the propagated nerve pulse. 



Lillie's Iron- Wire Model of Nerve Propagation 



The model developed by Lillie to illustrate impulse transmission in a 

 nerve fiber uses a pure iron wire that had been " conditioned " in strong 

 nitric acid, which coated it with a layer of iron oxide. If such a wire is 

 then placed in dilute nitric acid, no further chemical activity will take 

 place; it is now said to be in its " passive " state. If the passive wire, in 

 its dilute nitric acid bath, is scratched so that the oxide coating is pene- 

 trated, the dilute acid will attack the exposed underlying clean iron 

 surface at this point. It is found that the exposed iron becomes electri- 

 cally positive with respect to the adjacent edge of the oxide-coated sur- 

 face. The electric eddy currents flowing across the edge of this surface 

 fracture reduce the oxide coating to iron and oxidize the adjacent free 

 iron surface, thus causing the exposed iron to become coated with an 

 inactive surface. 



This reduction-oxidation process travels along the wire, in both direc- 

 tions from the scratch, in the form of a local pulse-like electrical disturb- 

 ance with the speed of about 15 cm/sec. This speed may be decreased 

 by increasing the concentration of the acid. 



After the initial formation of the passive surface on the iron wire a pro- 

 longed phase follows, during which a reexcitation of the wire by scratch- 

 ing produces only imperfect oxidation-reduction pulses. By degrees, 

 however, the oxide-coated wire regains its original property of a 

 recovered passive wire. 



Similar properties are, as we have seen, exhibited by nerve fibers dur- 

 ing their refractory period and again in the relative refractory period in 

 which the speed of propagation gradually regains its normal value. 



If the wire is enclosed in a tube filled with nitric acid, the velocity of 

 transmission of the oxidation-reduction pulse is found to vary with the 

 diameter of the tube. In a narrow tube the resistance through the elec- 

 trolyte of neighboring points, between which the eddy currents flow, is 

 increased. Hence, the currents are smaller and the time required to 

 reduce the film is increased. This decrease in velocity with decrease in 

 diameter is also one of the properties of nerve fibers. 



The model is an excellent one provided that its limitations are realized. 

 It is not, however, a theory of nerve activity, and it does not suggest one. 



