128 Mr. C. T. B. Wilson. Condensation Nuclei 



(Expansion experiments probably furnish one of the most delicate 

 methods of detecting these rays.) 



3. Ultra-violet light acting on moist air or oxygen produces nuclei 

 which, when the radiation is weak, require quite as great a degree of 

 supersaturation to cause water to condense on them as those produced 

 by X-rays. With stronger radiation, however, the nuclei appear to 

 grow, the expansion required to produce a cloud now depending on the 

 strength of the radiation and on the time for which the gas has been 

 exposed to the rays before expansion. With .very strong ultra-violet 

 light the growth of the nuclei, even in unsaturated air, continues till 

 they become visible (as stated in a preliminary note).* The pheno- 

 mena are then like those observed by Tyndall with certain vapours 

 exposed to ordinary light. That nuclei are produced when ultra- 

 violet light enters an expansion apparatus through a quartz window 

 was discovered by Lenard and Wolff, but they believed them to 

 .arise from disintegration of the quartz. That these nuclei arise not 

 .at the quartz, but throughout the volume of the air exposed to the 

 rays, is capable of experimental proof in a variety of ways. In hydro- 

 gen even strong ultra-violet light produces comparatively few nuclei, 

 these requiring also as great a degree of supersaturation as the nuclei 

 produced by X-rays in order that water may condense on them. 



4. Sunlight produces in air nuclei which require large expansions 

 {#2/1^ about 1'25) in order that water may condense on them. 



5. Certain metals in moist air produce nuclei, always requiring great 

 supersaturation in order that condensation may take place on them. 

 The supersaturation required is generally as great or greater than 

 that required in the case of X-ray nuclei. In the presence of amalgam- 

 ated zinc dense fogs are obtained with expansions, which in the absence 

 of the metal only result in the formation of a very few drops. Clean 

 surfaces of zinc or lead have a similar but much slighter effect ; with 

 copper or tin it is inappreciable. These phenomena are obviously 

 closely connected with the effects which these metals exert on a photo- 

 graphic plate, studied by Russell and others. 



6. Ultra-violet light acting on a negatively electrified zinc plate pro- 

 duces condensation nuclei, as was proved by the steam jet experiments 

 of Lenard and Wolff. The nuclei, however, are not, as these obser- 

 vers supposed, produced by disintegration of the metal, for expansion 

 experiments show that they are identical with the nuclei produced by 

 X-rays with respect to the degree of supersaturation required to cause 

 condensation to take place on them, and therefore entirely unlike dust 

 particles. In hydrogen the maximum number of drops in the fogs 

 which result on expansion is obtained with comparatively weak fields ; 

 no nuclei are produced when the zinc is positively electrified. 



7. The discharge from a pointed platinum wire in moist air or hydro- 



* ' Camb. Phil. Soc. Proc.,' vol. 9, p. 392, 1898. 



