596 Royal Irish Academy. 



the other. It appeared that but few sparks would pass from the 

 conductor to the first jar. If the last one was removed, and its chain 

 fastened to the next, the first jar would take a few more sparks, and 

 so on ; for it was found, that whenever the last jar in the series at any 

 time was removed, the same results followed ; and this was the case 

 when the last but one was removed, clearly proving that the capa- 

 city or aptitude of the first jar to take a charge was influenced and 

 diminished by the second, more so by the third, fourth, &c. Its apti- 

 tude was greatest when it was by itself, and not connected, as de- 

 scribed, with the others. 



" This result disappointed my expectations, so far as my intended 

 improvement on the electric battery was concerned ; and it also ap- 

 peared to point out the existence of a principle influencing the charge 

 of the electric jar, which was not recognized in the popular treatises 

 on electricity. I procured a number of glass plates with fixed and 

 moveable coatings. These plates were insulated and arranged with 

 and without coatings in every way that Brande's rule required, but 

 the general result was the same as that given above. 



" From numerous experiments made with these plates, I came to 

 the following general conclusions : — 



" 1 . That the actual quantity of the positive and negative electri- 

 cities which we can accumulate in the opposite surfaces of an electric 

 or non-conductor, as a plate of glass or dry ice, depends upon the 

 distance of these surfaces. 



" 2. Every case of charge of one jar or plate may be assimilated 

 to that of any number of jars or plates in a series such as Brande's, 

 by supposing the one jar or plate to be divided into the greater num- 

 ber, its thickness being the sum of the thicknesses of all the segments 

 or plates ; the inside of the first jar or surface of first plate in con- 

 tact with conductor, and outside of last jar or plate in contact with 

 the ground, being considered as the proper opposite surfaces of the 

 proper plate, and those on which the electricities evolved by the 

 friction of the cylinder and rubber of the electrifying machine are ac- 

 cumulated or heaped. 



" If we make a pile of the plates, coated or not, and charge the 

 outside surfaces by coating them, and connecting one with the cylin- 

 der and the other with the rubber of the machine, we find all the con- 

 ditions of the experiment complied with. There is no necessity for 

 any connexion with the ground, which in Brande's can act merely as 

 the conductor to convey the negative charge of the rubber to the ex- 

 treme surface. 



" Let us now unpack the pile, and we find that the charge of the 

 intermediate plates diminishes as we approximate towards the centre 

 of the pile, being greatest near the extremes. At equal distances the 

 charges are equal ; for the charges of the first plate but one, and the 

 last but one, will as perfectly neutralize each other as the charges of 

 the surfaces of the first and last. The same is found to be the case 

 with the surfaces of the third plates from each extreme, and so on of 

 the others ; but it is not the case with a second and a third, a first 

 and a fourth plate, and so on, no two uneouals as to place exactly 



