92 Bancs — Geometric Sequences of the Coronas' of Cloudy 



From equations (1) and (2), j3 = ZAk(\oge)/B; and if k = 

 18 cm./min., i?=15 cm., then ^L = '001. Thus the absorption 

 velocity found from these experiments is but 1/1000 of the 

 value found when the saturated emanation is forcibly passed 

 through fine bore tubes less than '5 cm, in diameter. A cor- 

 relative result was instanced in my earlier experiments showing 

 the relatively small value of h found in passing the emanation 

 through a wide tube, 5 cm', in diameter, though I was not at 

 the time quite clear as to the reason. With the necessarily 

 small mean free path of the nucleus, which being large in com- 

 parison with molecules owes its slow motion chiefly to the 

 unfavorable bombardment of many molecules, the result stated 

 is precisely what is to be expected. It is in harmony with the 

 preservative effects of the dilution of the saturated emanation. 



In general, therefore, the absorption velocity of the nucleus 

 depends on the violence with which the contact of the nucleus 

 and the solid boundary is promoted. In a bundle of fine tubes 

 of a given area the absorption is enormous, caet. par., as com- 

 pared with the single tube of the same area. Nevertheless the 

 value of Jc seems to reach a superior limit in case of extreme 

 agitation. This maximum of k is the value which I interpreted 

 as being the true nuclear velocity and used with this meaning 

 (after allowing for the motion in all directions) in the inter- 

 pretation of my electrical experiments. 



10. Estimated size of water particles. — The rate of subsi- 

 dence of the fog is not a good criterion* of diameter, because 

 this datum is complicated by the evaporation of water particles 

 (apparent subsidence at the top), and by their inevitable growth, 

 remembering that the coronas are all fleeting phenomena. Some 

 notion of their size, and this an upper limit, may be obtained. 

 If the fog subsides at the rate of l cm /sec, the radius of the 

 particle will be r— -0009 cm . For any other velocity expressed 

 in terms of this normal rate, r=9xl0~ 4 X \/v. Now the rates 

 are never a small fraction of cm./sec, so that the radii are not 

 liable to be much below say 10~ 4Cm , a datum which at first 

 sight is surprisingly large but is corroborated by the following 

 independent estimates. 



It has been shown above that in the case of spheres the 

 moisture precipitated per cub. cm. of air partially exhausted 

 as stated, is 79X10" 8 grams, and that with 5X10 4 nuclei per 

 cub. cm. in the saturated emanation, this is equivalent to an 

 initial diameter of the water particles of about 2'5/10 4cm . This 

 datum is an order of values like the preceding, whereby two 

 results are to this degree confirmed, viz : the order of size of 

 particles producing axial color and the number of particles 

 estimated for the saturated emanation. 



* Air cleared of fog by the warmer walls of the receiver after exhaustion, also 

 rises to the top. 



