376 Dr. J. Burdon-Sanderson and F. J. M. Page. [May 13, 



metric circuit is shown in the woodcut. When the keys K, K 3 , 

 K 3 , and K 5 are open, and the switch w is thrown over to the left, as 

 represented, and the rheotome closed, any current coming from mto/ 

 passes through the compensator (C), the rheotome (R), and the 

 galvanometer (G). 



Before making each observation, K is closed, and the slider of the 

 compensator moved until the needle of the galvanometer is at zero : 

 this having been done, K is opened and the rheotome set in motion, 

 when it is found that its revolutions are without effect on the galva- 

 nometer. 



The binding screws connected with the outer pools of the rheotome 

 are connected with the key h, from one of the blocks of which a wire 

 passes through the primary coil of the induction apparatus, and then 

 through a second apparatus for the purpose of short-circuiting the 

 closing induction currents, to one pole of a Grove's battery of two 

 cells, the other wire leading to the opposite pole. The terminals of 

 the secondary coil of the induction apparatus are connected with the 

 exciting electrodes in such a way that the rotations of the rheotome, 

 eacli of which makes and breaks the. primary current of the induction 

 apparatus, are without effect on the preparation, excepting when the 

 key h' is opened for the purpose of making an observation. 



The rate of revolution of the rheotome which is best adapted for the 

 purpose is J revolution per second. The period to be investigated 

 usually lasts two seconds after each excitation. This period is conve- 

 niently divided into 20 periods of 0"'l each. The " closing time " 

 of the rheotome, i.e., the time during which it closes the galvano- 

 metric circuit each revolution, is 0"'l. 



As the purpose of the observation is to investigate the condition of 

 the led-off surfaces during each successive period, twenty observations 

 must be made, for each of which the rheotome must be set, so that its 

 closing time coincides with one of these 0"'l periods. This is effected 

 with the aid of a table, one column of which shows the periods, the 

 other the reading of the rheotome scale which corresponds to it. As 

 each observation requires three-quarters of a minute to make and 

 record it, and must be repeated to ensure accuracy, the investigation 

 of a variation cannot be completed in less than a quarter of an hour. 

 The following example will serve to illustrate the form which the 

 results assume : — 



Table I. 



Time after excita- 











































tion, at which galva- 











































nometric circuit was 











































opened, the pei iod of 





























// 





// 













0-1 0-2 



0-3 



0-4 



0-5 



0-6 



0*7 



0-8 



0*9 



1-0 



1-1 



1-2 



i'-3 



1-4 



1-5 



1-6 



1-7 



1-8 



\-9 



2-0 





-42J-3 























































+ 3 



+ 30 



+ 12 



+ 1 







