1904.] on Condensation Nuclei. 459 



enter through the filter the fog rapidly cleared, and only a fine rain 

 continued to be produced. While the apparatus was iu this condition 

 an X-ray tube was set iu action near the aluminium tube ; the rain 

 was succeeded by fog, which continued to pour out from the end of 

 the condenser so long as the X-rays were kept in action. Condensa- 

 tion nuclei are, as this experiment proves, produced in air exposed to 

 Eoutgen rays. Later experiments will, however, show that they 

 have entirely difi'erent properties from the ordinary dust nuclei. 



When air has been comj)letely freed from dust particles, so that a 

 slight expansion of the air (initially saturated with water vapour) 

 does not result in the formation of any drops, it is found that quite a 

 high degree of supersaturation may be brought about without the 

 appearance of a single drop. There is, however, a limit to the 

 supersaturation which can exist without condensation of the vapour 

 in drops resulting. To study this condensation in dust-free air, and 

 to measure the expansion required to produce the necessary degree 

 of supersaturation, a special form of expansion apparatus is required. 

 The lantern slide thrown on the screen shows the construction and 

 mode of working of the apparatus. The second slide is a photograph 

 of the machine in action, the exposure having been made immediately 

 after an expansion ; the cloud formed (in this case on nuclei pro- 

 duced by the action of radium) is plainly visible along the path of a 

 concentrated beam of light from a lantern. 



Let us now try an actual experiment with the expansion 

 apparatus. On making a slight expansion a cloud forms on the dust 

 particles which are present ; this slowly settles to the bottom of the 

 vessel. The air is allowed to contract to its original volume, and a 

 second expansion of the same amount made. The drops formed are 

 on this occasion comparatively few, and they fall rapidly ; the dust 

 particles have nearly all been carried down with the drops formed by 

 the previous expansion. The fewer the nuclei on which water 

 condenses the larger will be the share of water available for each 

 drop, and the more rapid will be the fall. The next expansion pro- 

 duces no drops. While the air is in the expanded condition, the 

 piston being at the bottom of the expansion cylinder, air is removed 

 from the cloud chamber by opening the connexion to the air-pump 

 until the pressure is about 13 or 14 cms. of mercury below that of 

 the atmosphere ; the piston is again allowed to rise by putting the 

 air space below it in communication with the atmosphere. The 

 next expansion is thus comparatively large, the pressure after the 

 expansion has taken place and the temperature has risen to its 

 original value being 13 cms. or more below the initial pressure. 

 Yet, in spite of the high degree of supersaturation reached, not a 

 drop of water is seen. Making the fall of pressure 16 cms., however, 

 we see on expansion a shower of drops. And although these drops 

 are few and large, falling therefore rapidly, yet, however often the 

 same expansion be repeated, the drops produced on expansion show no 

 diminution in number. Thus the nuclei removed with the drops are 



Vol. XVII. (No. 98.) 2 i 



