C. Bancs — Possibility of a Colloidal State of Gases. 401 



nuclei into pure air is a branched design, like the roots of a 

 tree. Finally, unless the vapor tension of the liquid is too 

 small (as in petroleum), only normal coronas'* need be expected 

 to occur in the cases of the vapors of non-electrolytes. 



3. It follows from all this that the nuclei derived from the 

 same source and under otherwise like conditions must be rela- 

 tively very small in case of the vapor of an ionizing liquid, 

 and relatively very large in case of the vapor of a non-ioniz- 

 ing liquid. So far as its apparent origin is concerned, the same 

 nucleus differs enormously in size with the medium in which 

 it is suspended. It must consist, moreover, of clusters of 

 many molecules, the clusters having a given average size for a 

 given vapor. 



4. The next inference to be drawn is some notion of the 

 chemical composition of these nuclei. For convenience merely, 

 they are usually supplied by introducing into the receiver air 

 which has passed over phosphorus, or burning sulphur, of a 

 flame, or glowing charcoal, etc. They may be obtained, how- 

 ever, without putting anything material into the receiver, by 

 passing the X-rays through it. So far as the behavior of the 

 resulting nucleus here in question is concerned, there is no 

 qualitative difference. Hence one must look upon the gaseous 

 contents of the receiver as containing the stuff out of which 

 the nuclei are made. The inquiry is thus narrowed down to 

 this: which of the gases involved (air and vapor) is made col- 

 loidal in the manner specified. If the air be selected, then 

 there is no immediate reason for the enormous difference of 

 size of nuclei of the same origin in electrolytic and non-elec- 

 trolytic solvents. If the vapor be selected, then the difference 

 of size corresponds to the electro-chemical differences of the 

 liquids. But it is quite premature to attempt further decision 

 than to state that clusters of gaseous molecules of specific aver- 

 age size are aggregated in each medium. 



I have already called attentionf to the similarity of this 

 aggregating behavior of the nucleus to that of particles of 

 clay suspended in the corresponding liquids, or to many col- 

 loids. It does not, therefore, seem unreasonable to look upon 

 the nucleated gas in the receiver as possessing properties which 

 may be summarily referred to as a colloidal state of the gas in 

 question. 



5. The agency which holds the nucleus or molecular cluster 

 together is presumably the electron. One should, therefore, 

 anticipate greater conductivity in case of the much more 

 mobile nuclei of the first type (vapor of ionizing liquids) than 

 in the sluggishly moving nuclei of the second type (non-ioniz- 

 ing liquids), though it is not certain that the electron resides 



*This Journal, xiii, pp. 81-94, 1902. f Science, xv, p. 517, 1902. 



