^1/?' a)id Hydrogen hi/ lieating a Platitmm Wire, 313 



find out how the temperature to which the wire must be 

 raised in order that condensation may connnence did depend 

 iijion the expansion. The curves given represent the main 

 features of the resuUs obtained. The results ])lotted on 

 diaoram (1 were obtained in a diti'erent w\ay from the others. 

 Each point here represents the minimum expansion which 

 gives condensation when the wire is at the corresponding 

 temperature. 



The condensation-nuclei produced in hydrogen persist for 

 a much shorter time than they do in air. In hydrogen the 

 nuclei (which would have given a dense shower had the 

 expansion been made immediately after the heating-current 

 was cut off) will have all disappeared in five minutes after 

 the heating. It was often noticed though, that if an ex- 

 pansion be made immediately after the heating, those nuclei 

 not carried down by the first shower will persist for a much 

 longer time than would have been the case had the expansion 

 not been made. 



The following effect in hydrogen may also be mentioned. 

 If a large number of nuclei ha^e been produced by heating 

 the wire to a bright yellow heat, then raising the wire to a 

 dull red heat for a few seconds causes the nuclei to disappear 

 entirely. This effect is possibly due to the property which a 

 hot body has of removing the dust-particles from a vessel, 

 which has been explained by Lodge * as due to the bombard- 

 ment of the dust out of the air upon the cool wall-surfaces. 



New Form of Cloud- Chamber. 



It has been already mentioned that one of the experimental 

 difficulties was to measure the temperature of the wire before 

 it had attained a steady state, this being necessary to prevent 

 the heating of the gas in the apparatus. 



To be able to pass the current through the wire until it 

 attained a steady state would be advantageous in two ways. 

 In the first place the temperature could be more accurately 

 determined, and secondly, the time during which the current 

 was passing through the wire would be longer. To meet 

 these ends the apparatus shown in fig. 3 was constructed. 



The wire was fixed inside a separate glass balb B, con- 

 nected with the cloud-chamber A by means o£ a hollow 

 ground-glass tap, the internal diameter of which at the 

 narrow^er end was one centimetre. 



The diameters of the bulbs B and A were respectively 

 5 cms. and 4^ cms. The short tube connecting the larger 

 bulb and the tap was 3 cms. long and 1*2 cm. internal diameter. 

 * ' Nature,' vol. xxxi. p. 26o (1885). 



