Lecture-experiments in Electricity. 233 



the two sides of zero was 



252 divisions ; 



with two Grove's cells, the sum of the readings was 



507 divisions. 



Divisions, 

 With one DanielFs cell, the sum of the readings was 152 

 With another Daniell's cell, it was 155 



Sum .... 307 



With the two Daniell's cells connected in series, the sum of the 

 opposite deflections was 307 divisions. 



These numbers give, as the mean ratio of the electromotive 

 force of one Grove's cell to that of one Daniell's cell, 



507:307 = 1-65 : 1. 



According to Poggendorff, the ratio, as determined by his me- 

 thod of compensation, is 1*68 : 1. 



The mode of comparison by means of the galvanometer and 

 condenser may be rendered more accurate by increasing the ca- 

 pacity of the latter, so as to get larger readings and so diminish 

 the relative importance of the errors of observation. The above 

 numbers, however, which are of course given merely for the sake 

 of illustration, do not represent the limit of accuracy attainable 

 with the apparatus I employed : by simply altering the position 

 of the adjusting magnet of the galvanometer, so as to render the 

 suspended magnets more perfectly astatic, a deflection of 355 

 was obtained instead of 307. For proving to a class the way in 

 which the electromotive force of a galvanic battery depends upon 

 the mode in which the cells composing it are connected together, 

 and other fundamental facts of a like nature, this method can 

 easily be made abundantly accurate, and is probably as conve- 

 nient and rapid as any of the methods in common use. 



4. Method of demonstrating the existence of the Inverse and 

 Direct Extra-currents. — The only method of rendering distinctly 

 evident the retardation in the establishment of electric currents 

 in coiled conductors, or Faraday's extra-current on making bat- 

 tery-contact, which I have found described in any of the ordinary 

 text-books of physics, is one due to Edlund, and requires the use 

 of a differential galvanometer. By an arrangement of apparatus, 

 which may be regarded as a modification of that employed by 

 Edlund, it is easy to show the extra-current both on making 

 and breaking the circuit upon an ordinary galvanometer. This 

 arrangement will be understood by reference to the figure, where 



Phil, Mag. S. 4 Vol. 38. No. 254. Sept. 1869. R 



