100 
MESSES. W. DE LA EUE AND H. W. MtfLLER ON THE 
each phase by means of an electrometer ; from these observations, data may be derived 
with respect to the current and the apparent resistance of the tube at each epoch. 
In using a condenser of this capacity certain precautions are necessary, for it several 
times has happened that there is no indication of a current until a sudden jump 
takes place which destroys the vacuum tube ; this was remedied by placing a safety 
wire of platinum, O' 002 inch in diameter and 10 inches long, between the tube and one 
terminal of the condenser. Such a wire was found to permit of only 4 per cent, of the 
charge passing, and though frequently deflagrated by this amount of charge during 
the course of the experiments, it has effectually protected the tube by thus breaking 
off the connexion. W e are indebted to our friend Professor Stokes for the suggestion 
of the introduction of resistances during the charging of the condenser and also for 
that of the safety wire. 
Fig. 26 is a diagram of the arrangement. C, the condenser, one side of which is in 
Fig. 26. 
connection permanently with the battery at A, the other side is in connection with 
the battery at Z only when the key K (corresponding to K K', fig. 9, page 63) is pressed 
down ; this connection is either through the adjustable fluid resistance Fit" or direct, 
without intervening resistance, when the key K' (corresponding to S S', fig. 9) is pressed 
down ; T T' is the tube to be experimented with ; the current from the condenser has 
to pass through the safety wire, the tube, and the adjustable coil resistance in the 
megohm, or the adjustable fluid resistances Fit, FP 7/ [see also fig. 10, page 64], which 
may be plugged out by the plugs P and P' ; or through both coil resistances and fluid 
resistance if so required. When K is raised, the charged condenser supplies the current. 
We have already stated* that a charge of 47 '5 microfarads, with .1080 cells, 
deflagrates 10’5 inches (2 6‘67 centims.) of platinum wire, 0'005 inch (0‘127 millim.) in 
diameter ; when the condenser of 42 *8 microfarads is charged up, with 3240 cells, the 
charge deflagrates the same length of wire, but 0‘0125 inch (0‘317 millim.) in diameter. 
* Proc. Roy. Soc., vol. xxiii. p. 366. 
