THE ELECTRICAL RESPONSE. 



419 



capillary tube. This photograph of what may be called a "normal" 

 or "comparison curve" having been taken, the reverser of the com- 

 pensator is turned over, and a second photograph of the same kind 

 is obtained with the same electromotive force acting in the opposite 

 direction. The first of these curves is reproduced in Fig. 231. Its 

 character is such that, at the moment corresponding to any point what- 

 ever in its course, the difference of potential acting on the electrometer 

 is to the total such difference (in this case O'Ol volt) as the distance 

 from that point to the asymptote a is to the distance of this line from the 

 starting point of the curve. From the latter distance, the true value 

 in volts of the dis- 

 tance (1) referred 

 to above, can be 

 readily calculated. 

 If, for example, the 

 difference in ques- 

 tion be 10 mm., and 

 the corresponding 

 difference of poten- 

 tial O'Ol volt, the 

 value of each milli- 

 metre is: then known 

 to be one millivolt. 

 To find (2), i.e. the 

 value of the angle 

 of inclination at 



FIG. 231. Comparison curve. It is of such a character that any 

 two ordinates at an angular distance of one degree from 

 each other are, as measured from the asymptote a, in the 

 proportion of 1 : 1 '3. The straight line touching the curve 

 is drawn as nearly as possible parallel to the steepest part 

 of the ascent of the diphasic curve, and is, where it touches 

 the comparison curve, 23 mm. from the asymptote. If it be 

 assumed that O'Ol volt corresponds to 10 mm., the greatest 

 acting difference of potential during the first phase is 23 

 millivolts. Adding to this estimate the difference due to 

 the previous rise of the mercury up to the point of contact, 



any point in the 

 variation curve, the 

 point in the com- 

 parison curve at 

 which the inclina- 

 tion is identical 

 must be sought. 

 The distance in 

 mm. of this point 

 in the comparison 

 curve from the 

 asymptote then 

 gives the corre- 

 sponding difference 

 of potential 



we have 25 millivolts, the height above the base line being 

 about 2 mm. The fact that the curve is inscribed as if 

 on the surface of a disc, must be noted. To obtain strictly 

 correct results, the corresponding points in the two curves 

 should be at the same distance from the centre of the disc on 

 which they are inscribed. If they are not so, a correction is 

 necessary. Thus, in the case figured, the inclination of the 

 experimental curve must be increased about 1 per cent. 



in 



millivolts. For reasons which need not be here entered upon, the com- 

 parison curves must be taken under the same conditions as the variation 

 curve to be referred to them, and the points compared in the two curves 

 must be at the same distance from the centre of rotation of the disc. 



The photographic record of a diphasic response is shown in Fig. 232 ; 

 the time of excitation, i.e. the moment at which the primary circuit is 

 broken, being indicated by the radial line. The ascending part of the 

 curve corresponds in the main to the first phase the descending to the 

 second. 1 The essential points to be determined by the investigation of 



1 Elaborate analyses of several curves kindly made for me by Mr. Burch by his 

 method, have been published elsewhere (see Journ. PhysioL, Cambridge and London, 

 1895, vol. xviii. p. 130). 



