508 



Mr. W. H. Preece. On the Conversion of [Mar. 10, 



disturbances produced by some thermic action rather than by any 

 luminous effect. Several other experiments made by them confirm 

 this conclusion, notably those made upon crystals of sulphate of copper, 

 a substance which Mr. Crookes has shown to be highly opaque to rays 

 of low refrangibility.* 



Now, the questions arise, is this thermic action expansion and con- 

 traction of the mass due to its absorption of heat ? Or is it a disturb- 

 ance due to molecular pressure similar to that which produces the 

 rotation of the radiometer ? Or is it due to some other cause ? 



The argument against the first assumption when applied to hard 

 disks, is that time is a material element in such actions, and that the 

 rate of cooling of warmed diaphragms is too slow to admit of such 

 effects. Lord Rayleighj has questioned the validity of this argument, 

 and has shown that if the radiating power of the body experimented 

 on were sufficiently high a slow rate of cooling would be favourable to 

 rapid fluctuations of temperature. It became desirable to test this 

 point experimentally. The following apparatus was constructed for 

 the purpose. 



AB is a thin strip or wire 6 centims. in length of the substance to 

 be examined, fixed to a platinum " make and break " M, and adjusted 

 to a lever S, round whose axis is fastened a silk thread, the end of 

 which is attached to the strip or wire at A, and whose position could 



be adjusted by a screw 0. Any variation of tension due to expansion 

 or contraction of the wire would produce intermission in the electric 

 currents passing through the telephone T, which if periodically pro- 

 duced would result in sonorous vibrations in the telephone. Heat 

 from various sources and at various distances from bright lime-light to 

 dull heat from hot metallic surfaces, was allowed to fall through 

 rotating vanes intermittently on AB ; but notwithstanding every pre- 

 caution, and the many materials used, not more than six interruptions 

 per second could be produced, although the system was beautifully 

 sensitive to the smallest changes of temperature. 



* " Phil. Trans.," vol. 169, § 268. 

 f " Nature," January 20, 1881. 



