Properties of Bees' -ivax and Lead Chloride. 135 



where Q is the quantity of electricity discharged through the 

 galvanometer, H the horizontal intensity of the galvanometer- 

 field, I the half length of the needle, G the galvanometer- 

 constant;, P the periodic time of vibration, L the logarithmic 

 decrement, ,i\ the first sudden swing on discharge, x r the 

 deflection that would be produced by the small constant 

 current, and which is equal to 



#! + Dx 2 

 1 + D ' 



where # 2 is the second swing and D the decrement. 



It might, of course, be at once objected that the rise in 

 capacity as the wax cools from about 80° to 60° C. is perhaps 

 not due to any change in the specific inductive capacity, but 

 merely indicates that the distance between the copper plate of 

 the condenser was slightly diminished by the wax shrinking 

 on solidifying. This solution, however, is improbable, since, 

 although a sudden expansion of the wax on solidifying (if 

 such an expansion existed) might have separated the plates, 

 it is unlikely that the contraction which really occurred 

 could have brought them nearer together than the thickness 

 of the paper by which they were separated when the wax 

 was liquid. Nevertheless, partly to obtain additional evi- 

 dence on this point, and partly to measure the specific 

 resistance (or resistance per cubic centimetre) of bees'-wax 

 at different temperatures, I made ten distinct sets of expe- 

 riments, occupying many days, on the conductivity of wax. 

 In these experiments the wax condenser was heated up to 

 about 130° C, and a current sent through it with the 75 

 Daniell's cells in series. The temperature was kept constant 

 at about 130° C. until the galvanometer-deflection had reached 

 its maximum, when it was considered that the wax had ac- 

 quired the temperature indicated by the thermometer. The 

 temperature was then allowed to fall very slowly, and fre- 

 quent readings of the galvanometer and thermometer were 

 taken. The results obtained are shown on fig. 3, temperature 

 being measured parallel to X, the points and X corre- 

 sponding with 30° and 130° C; distances measured parallel 

 to O Y represent conductivity on such a scale that for the 

 curve L M, representing the conductivity between 115° and 

 65° C, O S corresponds with a resistance of 67,735 megohms 

 per cubic centimetre ; and for the curves X P, Q R (which 

 are drawn on a larger scale) the distance O T represents a 

 resistance of 186,000 megohms, — the zero-line for conductivity 

 for all three curves being O X. The point P, corresponding 

 to a resistance of 37,000,000 megohms per cubic centimetre, 



