OF AN ELECTRIC DISCHARGE. 



29 



11. ^2 variable, Wj constant. 



It will be seen that in the first table a satisfactory agreement 

 exists between the numbers of the third and fifth columns. In 

 the second table the differences are certainly more considerable ; 

 when, however, we reflect how difficult it must be practically to 

 fulfil the theoretic conditions required by the formula, particu- 

 larly that of complete insulation, and that even in this case the 

 formula is to be regarded as only approximately true, the dif- 

 ferences are not calculated to create a doubt as regards the 

 theory ; and it must be also remarked here that all the numbers 

 of the fifth column are greater than what my formula gives 

 them, while to approximate to that of Dove they must be 

 smaller. 



Still less suited to a comparison with the theory is the second 

 of the above-mentioned series of experiments. In these, three 

 or four batteries, or single jars, were used as elements, which 

 while charging were united to form a cascade battery. The 

 discharge was then effected with the first alone, or with the first 

 and second, and so forth ; the amount of heat generated in the 

 connecting wire being each time observed. 



Certain difficulties present themselves here, from which it 

 must be remarked the other experiments were not totally free, 

 but could not here exert the same amount of influence as in the 

 cases now under consideration. Among these the fact deserves 

 particular attention, that every new battery added necessitated 

 an increase in the length of the connecting arc. In the series 

 of experiments just considered, when the number of jars of the 

 battery was increased, the new jars were set close beside the old 



