134: Prof. W. B. Ayrton on the Electrical 



slowly. The capacity was then measured (several observations 

 being made al each temperature) by charging the condenser 

 with 75 Darnell's cells joined in series, and discharging it 

 through an exceedingly delicate Thomson's reflecting galva- 

 nometer. The curves A B C D, E F G, H I K (fig. 2) repre- 

 sent the results obtained on three different days, distances 

 measured parallel to X representing temperature, the points 

 and X corresponding respectively with 0° C. and 100° C, 

 and distances measured parallel to Y representing capacity, 

 the zero-line for capacity for the curves A B C D, E F G, 

 HIK being below X by a distance equal to f- of Y. 

 For the curve G H I the zero-line is X. It will be seen at 

 once that these curves, obtained on different days, do not give 

 the same capacity for the same temperature (the numbers, 

 therefore, that have been calculated for the specific inductive 

 capacity are not given) ; but considering, first, the very small 

 capacities that had to be measured, and, secondly, the difficulty 

 of accurately determining the temperature of a non-heat- 

 conductor like wax, even when enclosed in the oil-bath, the 

 discrepancies in the curves are not to be wondered at. One 

 fact, however, is very striking in all the three curves ; and that 

 is the rise in capacity as the temperature very slowly falls 

 from about 80° to 60° C, and the subsequent diminution in 

 the capacity on a still further diminution of the temperature. 

 Probably, had experiments on capacity been made when cooling 

 from a much higher temperature, there would have been 

 observed, first, a gradual diminution in capacity due to cooling 

 down to about 80° C. (traces of this first diminution are seen 

 in the portion A B of the curve ABCD); then we have the 

 rise of capacity as the wax solidifies at about 60° C; and, 

 lastly, we see the subsequent rapid decrease on further cooling. 

 Now this is precisely in agreement with the changes known 

 to occur in the index of refraction for light ; and hence the 

 interest of these experiments. 



As there was always a small electromotive force in the wax 

 condenser, and as the vibrations of the galvanometer-needle 

 were, as usual, damped by the air-vane, I used for calculating 

 the capacity the formula developed by Prof. Perry and 

 myself for employment in such cases*, which is 



* * A Test for determining the Position of a partial Discontinuity, 

 without Earth-fault,' by Professors W. E. Ayrto n and John Perry, 

 p. 12. 



