Measurement of Electromagnetic Radiation. 61 



Laboratory. The length of the oscillator was the same as th.it 

 of either of the wires in the tube, and it was placed so as to be in 

 the plane of the partition of the tuba, at a distance of about 

 30 centimetres (fig. 11). The best length of spark was about 

 half a centimetre. 



At first, on separating the knobs of the oscillators, so that 

 no spark passed, the instrument showed a large deflexion in 

 the direction of cold. This was due to electrostatic attraction 

 between the oscillator and the mirror. To prevent this action 

 a piece of tin-foil (E F G H, fig. 8) was pasted round the 

 tube at the level of the mirror, a hole being cut in it the 

 size of the lens, which effectually put a stop to this action. 



The instrument now indicated a considerable amount of 

 heat developed in the wire whenever the coil was worked. 



The effect of altering the capacity of the oscillator by adding 

 pieces of tin-foil to the ends was tried, but with no beneficial 

 result. 



A wet duster was hung between the oscillator and the in- 

 strument for the pnrpose of seeing whether the radiation would 

 pass through. The deflexion produced was only slightly less 

 than when no duster was interposed. 



In order to measure the heat developed by the induced 

 currents in the wire, the deflexion produced when the coil 

 was worked was noted ; then a steady current was sent 

 through the wire, and the resistance in the circuit altered till 

 an equal deflexion was obtained. The readings obtained with 

 the oscillator were somewhat varied, owing to small changes 

 in the character of the sparks. Another difficulty was caused 

 by the coil often refusing to work continuously for the time 

 necessary to get a reading (about 1\ minute), as the contact- 

 breaker became burnt by the sparks. 



The following are the deflexions obtained in centimetres :— 



45 ; 44 ; 37 ; 36 ; 35*6 ; 36*9 ; 44*8 ; 

 45-5 ; 40-2 ; 40'2 ; 39. 



Mean 40*4 centim. 



One Daniell cell working through a total external resist- 

 ance of 1 15 ohms gave the following deflexions : — 



41-0; 40-0; 39'3 ; 40-0; 41-0. 



Mean 40*3 centim. 



We may therefore say that the heating-effect due to the 

 oscillations (oscillator at 30 centim.) is equal to that pro- 



