154 



TUE STUUOTUUE OK TIIE NUCLEUS. 



TMil.i: 11.— DIFFUSION. \V.\TER. EXHAUSTION, 76-6; cm. FOG NOT COMPACT, 

 STRINGS SEPARATED BY CLEAR AIR. 



\cm.Trks. 



(^ Nuclei put in below for 5 *'^- nt (S/ = 4"'- exhaustion after 

 j wailing 10-15 "^ Ascent of fog strings a.\ial. 



Nuclei put in above, for 5 "^ at fJ / = 4 ■"'" 



Nuclei added above for 5 '^"^- at rf / = 4 '''"■ Descent of foi: 



strings on outside of column. 

 Nuclei added below for 10 ^- at 4 "'■ exhaustion 5 «*^ later. 



Nuclei added below for 10 ^•^'^- at 4"'-, 5 ^'^^ waiting. Ascent 



axial. 

 Nuclei added above. 



At'tei- lliis a variety of expei'iments were luaile in wliicli the nuclei were allowed 

 to ditt'ii.se tluougli tubes and partitions. Thus sulphur nuclei added below a paste- 

 hoard partition, showed on exhaustion a dense fog below, l)ut oidy filiinents above, 

 which on .second exhaustion changed to a faint fog above, remaining dense below, 

 etc. Phosphorus nuclei behaved similarly. On 'successive exhaustion, the fog 

 above gradually grew denser, and eventually coronas of different orders appeared 

 in the two sections. Although filtered air was added above, some nuclei are lifted 

 by the exhaustion which c<)mi)licates the experiments. Hence the following tests 

 were varied ; the diffusion ves.sel was exhausted at the ujiper compartment and the 

 nuclei added there, while filtered air was introduced below. In these cases a thin 

 sheet of cotton .5 cm. thick and not even opaipie nearly stopped the diffusion of 

 nuclei. Three hours after the nucleation in the npper compartment, the corona 

 seen there was still of the centrally green order, whereas oidy diminutive normal 

 coronas were in evidence below the partition. 



A tulnilated partition, figure 3 above, was now installed, consisting of 10 



vertical tubes, «, a, in parallel, each 10 cm. long and .0 cm. in diameter, all 



other communication between the compartments being sealed off with cotton. It 

 was ascertained that the diffusion of atpieous nuclei could be estimated in this way 

 in terms of the sizes of the coronas above and below the partition AB, after the 

 lapse of a definite time, dated from the introduction of nuclei above the partition. 

 In measurements made 2, 5, Kt, and 60 minutes after nucleation, the coronas still 

 showed colored centers .above but were normal and of gradually increasing size 

 lu-low. The results i>rove the feasibility of experiments of this kind for water 

 but tliey were temi)orarily discontinued because of the difficulty of keeping 



vatini 



the walls of the diff'usion tower clear of precipitated moisture. 



