.36 



HANDBOOK OF PHYSIOL(K;V 



NEUROPHYSIOLOGY I 



have been shown to be specialized (85), and it must 

 be supposed that it is in this region that the receptor 

 potential is generated. By recording across a barrier 

 surrounding the internode between the second and 

 third nodes of Ranvier, it was found possible to record 

 distinct phases of activity due to each of the first two 

 nodes if the thresholds of these nodes were raised by 

 anodal polarization. No phase of impulse activity 

 could be found attributalile to the nonmyelinated 

 terminal even though thresholds were raised by an 

 amount that, on theoretical grounds, should have 

 been quite adequate to reveal such impulse activity 

 if it existed. It therefore appears that after a mechani- 

 cal stimulus the impulse is set up at the first node of 

 Ranvier. 



Indirect evidence that impulses are not initiated in 

 the terminations of the afferent nerve fibers of certain 

 other preparations has already been considered in the 

 section on the effects of applied currents. 



Not only is there evidence that impulses are, in 

 some receptors at least, set up away from the termi- 

 nals in which the receptor potentials are generated, 

 but there is also evidence that such terminals are not 

 invaded by antidromically conducted impulses. In 

 the crayfish stretch receptor the receptor potential 

 is not abolished by an antidromic action potential; if 

 the impulse invaded the membrane that is involved 

 in the production of the receptor potential one would 

 expect a complete short circuiting of this membrane 

 and the temporary abolition of the receptor potential 

 (28). A similar observation has been made with the 

 olfactory mucous membrane of the frog (78); stimula- 

 tion of the olfactory nerve at different strengths and 

 frequencies had no effect on the response of the olfac- 

 torv membrane to an exciting .substance. As has 

 already been stated in the last section, an antidromic 

 impulse causes slightly less depression of the receptor 

 potential in the Pacinian corpuscle than does an 

 impulse set up by a mechanical pulse. It has already 

 been argued that an impulse initiated in this receptor 

 by a mechanical stimulus starts at the first node of 

 Ranvier; if an antidromically conducted impulse in- 

 vaded the nonmyelinated terminal then it would be 

 expected to produce a greater depression of the 

 receptor potential. This is not the case and it seems, 

 therefore, that antidromic impulses do not invade the 

 nonmyelinated terminal (18). 



Evidence that impulses set up by physiological 

 stimuli to receptors do not start in the receptor region 

 might simply mean that all-or-nothing impulses can- 

 not occur there during receptor activity. That anti- 

 dromic impulses do not invade the terminals might be 



a result of Ijlock at regions of low safety factor, though 

 from parallel situations elsewhere this does not seem 

 ver) likely. The most probable explanation of all 

 these results is that those regions of membrane that 

 are not in\aded are different from the rest of the 

 neuron surface and are not capable of producing a 

 regenerative all-or-nothing response. 



When a frog's muscle spindle is discharging at low 

 frequency small all-or-nothing potentials can be seen. 

 These are much smaller than the propagated impulse 

 and may occur in a number of discrete sizes (57). 

 They disappear if the frequency of discharge of full- 

 size impulses is increased and also if the receptor is 

 bombarded antidromically. After a full-size impulse 

 there is always a delay before the next all-or-nothing 

 event, whether full-size or small, but after one of the 

 small all-or-nothing potentials the interval may be 

 quite short. An explanation of these events may be 

 that impulses are set up in the terminal branches of 

 this type of receptor, but an impulse in a single branch 

 is unable to pass the regions of low safety factor that 

 occur where the branches join (57). A full-size im- 

 pulse would then only be set up if there were sufficient 

 synchrony in the activity of the terminal branches. 

 On the same argument all-or-nothing activity in a 

 single branch would fail to invade other branches and 

 therefore would not depress their activity, while a 

 full-size antidromic impulse would invade them all. 



EFFECT OF PROC.MNE .J^ND SODIUM LACK ON 

 RECEPTOR POTENTIALS 



In the frog's muscle spindle concentrations of pro- 

 caine from 0.1 to 0.3 per cent abolish impulse acti\ity 

 but leave the receptor potential apparently unaffected. 

 Higher concentrations of procaine reduce the ampli- 

 tude of the receptor potential, affecting the static 

 phase more than the dynamic (58). Similar results 

 can be obtained with the Pacinian corpuscle of the 

 cat. The impulse is abolished by concentrations of 

 o.i to 0.5 per cent procaine in the bathing fluid, but 

 if the procaine is washed out after aijout 10 min. 

 there is no reduction in the amplitude of the receptor 

 potential. Prolonged soaking in these concentrations 

 causes a reduction of the receptor potential amplitude 



(37)- 



Similar eff"ects can be obtained in both these prep- 

 arations if they are soaked in sodium-free solutions. 

 Ten minutes .soaking in such a solution abolishes 

 repetitive firing from the muscle spindle while thirty 

 minutes is enough to abolish the initial spike (58). 



