C. Bar us — Ionization of Water Nuclei. Ill 



requires about 1,500 volts to spark across the intervening space 

 of -02 cm . With only 5,000 or 6,000 particles present the con- 

 dition would be enhanced over five times. 



There seems then to be no reason for departing from the 

 point of view in this paper even if the number of nuclei com- 

 puted from coronas is somewhat low. As the nuclei neither 

 vanish coiccidently with the ions nor are present in similar 

 orders of numbers, a sharp distinction must be drawn between 

 nuclei and ions. 



7. Behavior of Phosphorus. — It will be interesting to insert, 

 by way of comparison, the corresponding treatment of phos- 

 phorus nuclei. The last equation for n cannot here be assumed, 

 for Ohm's law is roughly obeyed. The conductions obtained 

 above depend on the intensity of the ionizer, and on the air 

 current, P, and the conduction a=S, found elsewhere, may be 

 assumed for V=2'5 liters/min., as an average case. 



The current must here be written 



E 



di=27r(r l> -{-r^kne- / — — ^clL 



for the element, dl, while the number of nuclei, n, follows the 

 same equation as above, if the spontaneous decay of nuclei in 

 the lapse of time is ignored. The exponential term may again 

 be omitted. The full result* after integration is 



2-3 a=z —(dE/dt)/E= — (^/d r 0A=^ = ^(l-e-- 3 ^^ 2 +^/ F )(2) 



V 2 1/ ^ 



Hence, 



(r^ — rJC 2'3 a 



16-7 Ve 



The constants are 



Thus 



r 2 = 1-05, r l = -32, centim., 

 a = -3/60, (referred to seconds) 

 C=8/\0 11 farads, 

 e = 2-3/10 19 coulombs, 

 F=2-5 lit/min. 



n = -1 X 10\ 



In my former workf with a condenser of totally different 



* Experiments with ionized air, p. 67. 



f Experiments with ionized air, p. 69 ; cf. pp. 53, 91, 93. Owing to an 

 error in estimating the capacity of the electrometer, the results there given 

 are too low and have since been corrected as in the text. This makes the 

 phosphorus emanation richer in nuclei than air ionized by the X-rays, a con- 

 dition otherwise probable. 



