ON THE INDUCTIVE CAPACITY OF A GOOD SPRENGEL VACUUM, ETC. 199 



rod and the otlier to tlie inside of the g]ass tube. To give, in a small 

 space, length to these glass rods, for obtaining surface insulation, they 

 are made zigzag, and in the form of a flat spiral. To the thin platinum 

 rods are attached two fine platinum wires, which form the two electrodes 

 of the condenser, and where these pass out through the glass tube, glass 

 is fused on to the wire both inside and outside, as in the figure, to increase 

 the surface insulation. 



The area that one set of aluminium cylinders exposes 

 to the other is about 1800 square centimetres, so that the 

 electrostatic capacity is about 450 centimetres in absolute 

 units, or ^o^ of a microfarad. 



A small spiral glass tube connects the condenser with 

 a three-fall-tube Sprengel pump, and as the internal ca- 

 pacity of the condenser is large, it was thought desirable End of Condenser, 

 to attach to the pump an Alvergniat or Geissler arrange- 

 ment to enable the pressure to be rapidly reduced to about 10 cen- 

 timetres of mercury. A barometer gauge and a McLeod gauge are 

 attached to the pump. The entire glasswork in the apparatus was made 

 by Mr. Gimingham, and the Committee desire to express their thanks 

 for the assistance he has so kindly given. 



Method of Experimenting. — In the accompanying figure, A is the alu- 

 minium condenser just described, the interior not being shown in the 

 figure ; B is Sir William Thomson's ' sliding condenser,' kindly lent by 

 him to the Committee. This, as is well known, consists of a brass tube, 

 I, about 38 centimetres long and 5'08 centimetres in external diameter, 

 attached at one end to an ebonite collar, d d, by 

 which it is supported and insulated. Outside and 

 inside this brass tube, but without touching it, 

 slide two other tubes, T, and T^, electrically con- 

 nected with the outer tube T-^ and with the earth. 

 The motion of the tube Tj forms a coarse adjust- 

 ment, and that of the tube T2 a fine adjustment of 

 the capacity of the condenser. On account of the 

 action of the edges it would be somewhat difficult 

 to calculate the whole capacity of this condenser 

 for any given position of the tubes, but it is com- 

 paratively easy to calculate the change of capacity 

 produced by moving either tube a known distance 

 measured on the fine linear scale engraved on each 

 of the sliding tubes. 



As the capacity of this condenser, when the 

 tube r, is in its mean position, is considerably less 

 than that of the aluminium condenser, another air- 

 condenser, C, having a fixed capacity about equal to the difference, was 

 constructed. B, then, could be adjusted so that its capacity, together with 

 that of C, was equal to that of A, when A contained air at ordinary pi'es-. 

 sure. Then any change in the capacity of A, produced by exhausting the 

 air, could be measured by finding the new adjustment of B necessaiy ta 

 produce balance. 



The mode of testing the equality of capacities was suggested by one 

 of the Committee, Dr. Lodge, and consisted of a modification of Prof. 

 Hughes' Induction Balance. Z is a coil of wire of about 3 ohms' resist- 

 «,u<5e, in which the current from two or three Grove's cells, P, flows in- 



'fTOf"^' 



