ClIArTKU IV. 



(iKXKKAL SI'ltVEV OK (UNDKNSATIO.V I.\ THE VAI'OIts (IK lONI/.ING A.NJ) NUN-IO>IZING 



The subjects to he treated pioiuisciiuiisly in tlie present cha[)ter, are the be- 

 liavioi' of uuclei in vapors of iioiiioiiiziiig liquids, the form of coronas observed, the 

 intensity of axial color effects, the production of nuclei spontaneously, by shaking 

 li(|uids, etc., and the distribution of the initial nucleation as evidenced on exhaustion 

 by the initial fogs. The niethods used, though efficient for exploration, are not 

 adai)ted to secure results of pi-ecisiou. The next chapter will show that the merest 

 traces of solute will mateiially modify the behavior of a nominally pure solvent; 

 and hence the liquids of the following paragraphs should in critical instances be re- 

 ijaided as dilute solutions. 



GRADED CONDENSATION. 



1. It would ho difficult to read the admirable work on the relation of rain 

 and atmospheric electricity which has issued from the Cavendish Laboratory, with- 

 (jut being convinced of the strength of the arguments put forth. That in a repe- 

 tition of these researches, in particular of the experiments of C. T. R. Wilson ' on 

 the comparative efficiency as condensation luiclei of positively and negatively 

 charged ions, one would Init reproduce his i-esults, admits of no doubt. 



In so im[)ortant a question, however, it is none tlie less desirable to reach 

 identical conclusions from entirely different methods of approach. It has been 

 part of my purpose to be driven to like inferences; in other words to I'each a point 

 in my \v(irk where I should have to abandon the nucleus as an agency which for 

 purely mechanical or thermodynamic i-easons facilitates condensation, and be com- 

 pelled to recognize the special activity due to its chai'ge. 



I had hoped to accomplish this in the following experiments'' with benzine 

 when contrasted with the corresponding behavior of water; but the results, con- 

 trary to my expectation, are so curious and pronounced an accentuation of the 

 nuclear theory that it seems worth while to specially describe them. 



2. The work originated in the following point of view: If the action pro- 

 moting con<l('nsation is in any <lcgi'ee of a chemical nature (such suppositions have 

 Ijeeii nuide ; the production of hydrogen su[ieroxide, for instance, has been sug- 

 gested), then there should l>e a marked ililfereiice in the efficacy of the same nucleus 



' C 'P. R. Wilson, /'//i7. Tram., London, vol. cxciii., jip. 289-308, 1899. 

 ' Cf. Science, xv., |i|). 175-178, 1902. 



70 



