DETERMINATION OF DIFFERENTIAL EXCITABILITY 279 



of nerve, I have obtained no resultant deflection whatever, 

 showing that such shocks induce exactly equal excitations 

 in an isotropic tissue. But if the excitability of one of the 

 two points be first abolished by killing, then a definite 

 resultant responsive current is obtained, from the excitable 

 living to the inexcitable dead. So perfect were in fact the 

 results secured by means of these equi-alternating electric 

 shocks, that I was desirous not only to detect, but also to 

 record photographically, the responses thus obtained. In 

 this a certain difficulty is experienced, inasmuch as the 

 alternating shocks are apt to render the recording spot of 

 light tremulous, and thus to spoil the photographic im- 

 pression. This may, however, be overcome by making the 

 alternation frequency so high, in reference to the period of 

 the needle or suspended coil of the galvanometer, that the 

 unsteadiness of the deflection ceases. 



I shall now describe the practical means employed to 

 obtain equi-alternating shocks of any frequency that may be 

 desired. This I have been able to do in several ways, and^ 

 among others, by using a Rotating Reverser. This consists 

 of an ebonite disc, on the periphery of which there are strips 

 of metal of equal breadth, and separated from each other by 

 equal distances. The odd strips (i, 3, 5, and so on) are 

 connected together and led to a metallic ring on the left of 

 the disc. The same is done with the even strips, which are 

 led to the right. The two electrodes of a battery are led 

 through a key, K, to these two metallic rings and are con- 

 nected with them by means of brushes. Thus one ring, with 

 all the odd strips, is connected, say, with the positive, and the 

 other, with all the even strips, with the negative pole of the 

 battery. The current is led ofT by a second pair of brushes, 

 placed diametrically opposite to each other on the disc, in 

 the primary circuit, P, of an induction coil (fig. 171). Let us 

 suppose the upper brush to be connected with an odd strip, 

 the lower will then be connected with an even. The current 

 in the primary coil now flows in one direction. When the 

 disc is rotated, so as to bring the next pair of strips in 



