June 11, 1909] 



SCIENCE 



925 



quantities being apparently a linear one 

 between the limits of the absolute humid- 

 ity represented by 0.5 and 19.0 (grms./m^), 

 corresponding to relative humidities of 4 

 and 100 per cent. The mean mobilities for 

 these values of the humidity, from results 

 so far obtained, are 1/1280 and 1/3370, 

 respectively. In other words the mobility 

 for an absolute humidity of 2.4 is twice as 

 great as that for a humidity of 15.4 

 (grms./m^). The observations are not 

 regular enough to show if there is any dif- 

 ference between the mobilities of the 

 positive and negative ions. Owing to 

 ionization being caused by phosphorus, it 

 is not advisable to use phosphorus pent- 

 oxide as the drying agent in such experi- 

 ments, and calcium chloride has been 

 employed in all cases. 



The intermediate ion has been under ob- 

 servation for only a comparatively short 

 time. The measures so far made, how- 

 over, show that the mobility is largely af- 

 fected by change of the humidity of the 

 air, the magnitude varying from one fif- 

 teenth to about one tenth of that number 

 as the absolute humidity alters from 0.5 to 

 15 (grms./m=') at a temperature of about 

 22° C. To this statement there is a limita- 

 tion, the extent of which I do not as yet 

 fully know — in air in its natural state with 

 the absolute humidity between 14 and 16 

 grms./m^, at 22° C. when the ionization 

 due to this class of ions is relatively weak, 

 the mobility, at least of the positive ions, 

 is of the order of 1/65, while with strong 

 ionization the value is only about half as 

 great. Unless the limitation just men- 

 tioned provides an exception, on further 

 investigation, no definite difference be- 

 tween the mobilities of the positive and 

 negative ions of this class can be deduced 

 from the observations. 



The facts just described prove that there 

 is a definite connection between the ions 

 and the water vapor of the air, and open 



up an interesting field for speculation as 

 to the development and structure of elec- 

 trified clusters, and as to the nature of the 

 resistance which they experience in drift- 

 ing through the crowd of molecules. The 

 basis of the structure is, of course, the 

 molecular ion, which, it is well known, 

 originates from effects associated with 

 radio-active transformations occurring in 

 the air, the ionization being primarily due 

 to the presence of radium and thorium in 

 the material of the earth's surface. The 

 growth to more complex structure appar- 

 ently occurs by the collection of water 

 molecules round the molecular ion, owing 

 to the influence of its charge. 



Seemingly from a consideration of the 

 experimental results, we must recognize at 

 least two forms of electrified molecular 

 aggregation in the air which are stable 

 under ordinary conditions. As the mobil- 

 ities depend on the humidity, it might not 

 unreasonably be supposed that the inter- 

 mediate and large ions represent stages in 

 the development of the small ions into 

 visible drops of water, which occurs if the 

 air becomes sufficiently supersaturated. 

 It seems, therefore, curious that the large 

 ions are not separately apparent as con- 

 densation nuclei in cloud experiments. 



Mr. C. T. R. Wilson" has shown that in 

 such experiments the presence of a mod- 

 erate electrical field prevents the forma- 

 tion of drops if the expansion ratio does 

 not exceed the value 1.27. This proves 

 that the nuclei for these small expansions 

 are ions which can be removed by the field 

 before the expansion takes place. I have 

 carefully repeated the observations, with 

 an apparatus similar to that described by 

 Mr. Wilson, in order to determine if the 

 effect of the electric field varies with the 

 time it is on before expansion, and find the 

 full effect whether the interval is one sec- 

 ond or twenty minutes. With the fields 



"Wilson, PUl. Mag., June, 1904. 



