CHAPTER VI. 



TIIK DirKUSlON OK NUCLKI OK LIKE OKIGIN I\ lUKKEltKNT MKDIA. 



1. lypevi of diffusion, phenomena. — In a surve}- ' of the diffusion rates of 

 nuclei derived from the same source and under otherwise like conditions, but sus- 

 pended in different media (/. e., in air saturated with different vapors), the occur- 

 i-euces may be classified with i-eference to two extreme types : the first includes 

 vapors derived from such strongly ionizing liquids as have been examined, like 

 water, ethyl alcohol, methyl alcohol, etc. In these cases the coronas obtained from 

 the same initial nucleation, on successive condensations by the exhaustion method, 

 ai-e always regularly anuulai-, or, at most, distorted in color only. With the alco- 

 hols the coronas may sl)o\v the colors of two successive coronas (in the sense de- 

 fined in Chapter II), in the upi)er and the lower halves of the same rings, but 

 there is no distortion of form appreciable. With watc vapoi- the coronas do not 

 even show color distortion. Examples of this kind arc given in Chapter IV, Jj^j 

 27, 28, and need not here be further treated. 



The second type of phenomena are observed with vapors dei'ived chiefly from 

 the non-ionizing liquids, like the hydrocarbons, gasolene, benzine, etc., carbon 

 disulphide, benzol, toluol, etc., in which the successive coronas derived fiom the 

 same original nucleation show gradually increasing distortion of form. If the 

 initial corona is annulai- and the distribution of nuclei therefore uniform, the fol- 

 lowing coi-onas pass through campanulate distortion," and finally become mere color 

 strata. Examples of this kind are also given in abundance in Chapter IV, § 10-26. 



The distinction between ionizing and non-ionizing solvents which is here 

 implied, is not, however, a perfectly I'igid rule. In treating such a liquid as acetone 

 (Chapter IV, § 29), one is already put at a loss in the inannei- above specified, and 

 the presence of ionizing power is to be regarded as incidental. 



2. Coiulitions of hoDiogeneous distribution. — The effect of precipitation is a 

 disturbance of the original homogeneous nucleation. The nuclei, as a whole, are 

 dropped with the fog and their distribution changed. Tiie degree to which the 

 uniform distribution will be kept up depends, therefore, on the rate of diffusion of 

 the nucleus. It follows that the nuclei of the first group must diffuse very rapidly, 

 so that all parts of the receiver continually contain the same number per cub. cm. 

 The effect of removal of nuclei by loading is thus quickly wiped out. On the 

 othei' hand, the diffusion of nuclei of the second group (non-electr(dyzing solvents) 

 is apparently slow by comparison. It is quite possible to obsei've the air in the 



' Am. ypiirn. (5), XIII, ]i|i. 400-402, 1902. ■' Chapter IV, §§ 7, 8, 9. 



13S 



