1904.] on Condensation Nuclei. 461 



bination and diffusion will continue, and the number of ions in the 

 vessel will diminish rapidly. 



Experiment shows that, while in the absence of an electric field, 

 quite a considerable fog is formed when an expansion, slightly 

 exceeding 1 • 25, is effected 10 seconds after the rays have been cut 

 off, with 200 volts between the upper and lower plates the same 

 expansion, allowed to take place 3 or 4 seconds after the stopping 

 of the rays, produces only a very slight shower. Or, again, if the 

 rays be kept on all the time the resulting fog is very much less 

 dense with the electric field acting than without it. These results 

 are easily explained if we assume that the condensation nuclei are 

 the ions, and apply the result obtained by purely electrical methods, 

 that the ions travel about 1 • 6 cm. per second in a field of 1 volt 

 per cm. The nuclei causing the rain-like condensation without 

 exposure to Rontgen or radium rays are also removed by the action 

 of an electric field ; we have thus the direct proof that they also are 

 ions. Recent experiments have proved that a charged conductor 

 suspended within a closed space loses its charge by leakage through 

 the air, and that the conduction shows all the peculiarities of that met 

 with in an ionised gas. And, indeed, it appears that this ionisation 

 is due to the action of radiation of the radium type from the walls of 

 the vessel and from outside the vessel. The condensation method of 

 detecting ions is, it may be pointed out, a very delicate one ; a single 

 ion if present in the vessel will be detected. 



The positive and negative ions are not alike in their power of acting 

 as condensation nuclei. In most of the experiments shown to-night 

 the negative ions alone have in fact come into action. The positive 

 require a considerably greater expansion in order that water may 

 condense upon them. The final volume must for the positive ions be 

 about 1-31 times the initial instead of only 1-25, corresponding to a 

 six-fold instead of a four-fold supersaturation. 



To demonstrate the difference between the positive and negative 

 ions the same form of apparatus is used as in the previous experi- 

 ment. Instead, however, of a difference of potential of 200 volts, 

 only two or three volts are applied between the plates. And in this 

 experiment only a thin layer close to the lower plate is exposed to 

 the action of the rays. Under these conditions, if the upper plate is 

 the positive one, the negative ions will be attracted upwards out of 

 the ionised layer, and will occupy the greater part of the volume of 

 the vessel, while the positive ones will have only a short distance to 

 travel before reaching the lower plate. If the rays be cut off before 

 the expansion is made it is easy to arrange the interval to be of such 

 a duration that all the positive ions have been removed, while only a 

 small fraction of the negative ions have reached the upper plate 

 before the expansion takes place. Thus we can try the effect of 

 expansion when the vessel is charged with practically negative ions 

 only. By reversing the electrical field the action of positive ions, 

 almost free from negative ions, can be studied. When the expansion 



2 I 2 



