578 
MESSES. GIBSON AND BARCLAY ON MEASUREMENTS OF 
meter used has been once ascertained, this mean may be obtained from a single reading 
without reversing the connexions. 
The following Table shows the values obtained for a and (3 in the measurement of 
the spherical condenser: — 
Date. 
a. 
1 3- 
Nov. 17, 1869 
211 
183 
197 
Nov. 17, 1869 
211 
186 
198-5 
Nov. 19, 1869 
211 
198 
Taking 198 as the mean reading and 1572 as the value of x, the value of the sliding 
condenser with its index at zero was 1374 scale-divisions. 
A condenser was now prepared having paraffine instead of air as the dielectric. This 
consisted of a flat circular brass box with a tube projecting from the centre of the lid, 
through which an electrode was carried from a brass disk imbedded in paraffine, with 
which the box and tube were filled, midway between the top and bottom of the box. 
The box was placed in water reaching nearly up to the top of the tube. For the value 
of this condenser at different temperatures the following results were obtained : — 
Date. 
Temp. 
a. 
P- 
2 * 
Value = -\-z. 
Nov. 17, 1869 
11-5 C. 
180 
168 
1542 
Nov. 18, 1869 
13-3 
215 
192 
203 
1577 
Nov. 18, 1869 
17 
212 
224 
1598 
Nov. 19, 1869 
12-5 
220 
232 
1606 
Nov. 19, 1869 
15-8 
228 
206 
217 
1591 
Nov. 22, 1869 
17-9 
235 
223 
1597 
Crevices which had formed in the paraffine when cooling, gradually admitted water 
so as ultimately to destroy the insulation of the disk. 
The increase of capacity thus caused is probably sufficient to account for the varia- 
tions shown in the Table, which, therefore, are not to be attributed to any alteration of 
inductive capacity due to change of temperature. Taking the first of the values, 1542 
scale-divisions, or 63-687 centimetres, as probably the most accurate, the following 
determination of specific inductive capacity of paraffine is obtained. The value of the box 
condenser with air instead of paraffine as the dielectric, obtained from measurement of 
it on the sliding condenser described in the Appendix, is 32-306 centimetres. These 
two values give F975 as the specific inductive capacity of paraffine when that of air is 
taken as unity, a correction having been made on account of three vulcanite pins used 
to support the insulated plate. 
The following form of condenser (see fig. 5) was next employed as being most likely 
to show effects of temperature undisturbed by other causes. Into a cylindrical brass 
vessel (a a) 15-5 centimetres in depth and 8-61 centimetres in internal diameter, melted 
