THE structdrp: of the nucleus. 161 



fogs subsided on the positive side many times as i-apidly as tliey did on the 

 negative side, or that tlie negative ions are in correspondingly greater numVjer. 

 The effect is increasingly marked for smaller supersaturations. 



On extending my work with shaken nuclei in the last cha[)ter to solutions of 

 non-conductors in non-conductois, such as naphthalene ;ind of parafiine in benzol, 

 etc., I obtained results leading to the interpretation already summarized for aqueous 

 saline solutions. The purport of the present chapter is the same. The nucleus 

 is to be regarded as an exceedingly small droplet of very concentrated solution, 

 which pei'sists inasmuch as the decreased vapoi' pressure due to solution, at a cer- 

 tain specific radius is exactly counterbalanced by the increased vapor pressure 

 due to convexity. 



If this is true, then it seems doubtful to my mind, whether the experiments 

 of C. T. R. Wilson on the specific condensation effect of ionization can further be 

 regarded as crucial. 



If one introduces nuclei or makes nuclei by aid of the X-rays, in what is 

 virtually the acid and the alkaline side of a battery, even if the ionized moist 

 air is the electrolyte, one is conveying nuclei into, or making nuclei out of, 

 different media. The stuff out of wliich solutes are to be fashioned may be avail- 

 able in different degrees on the two sides. Whatever chemical effect is produced 

 out of Og, Ng, and HgO on one side by the rays, need not at all be the same as on 

 the other side, any more than the effect of shaking a veiy dilute solution need be 

 the same as the effect of shaking a stronger solution, where the results have been 

 shown to be enormously different as to the number, the velocity, and persistence 

 of nuclei produced. Hence from the accumulating evidence which I have brought 

 forward, I am led to infer that the two species of nuclei in Wilson's experiment 

 are for mei'e chemical reasons liable to be of dift'erent degrees of permanence, sizes, 

 and numbers, quite apart from the electric circumstances involved. One can not 

 therefore affirm that thedift'erence (respectively positive and negative) of ionization 

 is the immediate and sole cause of the difference of precipitation rates specified, or 

 briefly that negative ions precipitate more effectively than do positive ions, because 

 both a dift'eren'ce of ionization and a chemical difference is involved. 



I refei-, of course, to inmiediate causes. Remotely aflinities and cohesions 

 have the well-known electrical relations ; but with remote causes I am not here 



concerned. , . . 



Finally, if a marked difference in efficiency as condensation nuclei ot positive 

 and negative ions is granted, then any ionized emanation neutral as a whole, like 

 that fix.m phosphorus, should produce two groups of nuclei. On condensation 

 there should be two groups of coronal particles, interpenetrating and subsiding 

 through each other in the way I have frequently instanced in other experiments 

 (Chapter V § 37). No such effect has been observed. Phosphorus nuclei are rather 

 remarkable'for their identity, and the regular coronas observed even after 25 or 50 



exhaustions. , , . . , 



While these conclusions as to the origin of the different nuclei involve a 

 theoretical difference from Wilson's deductions, they are not at variance with his 



