110 C. Bar us — Ionization of Water Nuclei, 



function of the potential gradient J£/(?\—9\); and so long as 

 the value of h has not been directly found, the assumed small 

 value of the exponential term is merely admitted for argument 

 but will be further interpreted in § 14. 

 The additional constants needed are 



(7=72-5/9 X 10 11 



A = -91, the voltage factor of the electrometer. 



The table then gives as the currents, 



dJEJ/dt — *120 volts /sec, for the positive charge, 

 = '059 " for the negative charge; 



so that the average current, '090 volt/sec, may be taken as 

 due to the ions, positive and negative, but neutral as a whole, 

 while the observed values of dE /dt are due to the excess of 

 negative ions. Hence 



(dE/dt) C 



16-7 e V 



9-7 X 10 s 



The number of ionized nuclei entering the condenser is thus 

 estimated at about a million per cubic centimeter, if each 

 nucleus carries one electron, and it will probably be indefi- 

 nitely larger, if the nuclei laden air-current V is swifter; for 

 the charge is rapidly lost in the conveyance tube e between the 

 receiver A and the condenser C, fig. 1. 



5. Comparison with Coronas. — The number of ions thus 

 computed is excessive, particularly in view of the number of 

 nuclei obtained from the order of the first corona. The 

 previous investigations show that this is number 9 in my series, 

 and corresponds to about 5,000 nuclei. 



Several inferences are thus suggested : either the charge of 

 each nucleus is about 200 times larger than the value assumed, 

 i. e., each nucleus on entering the condenser brings 200 elec- 

 trons ; or nuclei are lost at the outset at an enormously rapid 

 rate, so that only *005 of the original number remain when the 

 first condensation is made ; or finally, that each nucleus emits 

 200 electrons as a result of the peculiar radio-activity inferred 

 by J. J. Thomson (1. a). The second view may be disproved 

 by direct tests. My own preferences in this paper are for the 

 first, for the reasons to be given in the sequel. 



6. Potential of the Nucleus. — Assuming the nuclei to be 

 10" B C1 " in radius (they are probably smaller), their capacities 

 will be 10 _6 /9xlO n farads. Hence the potential of each is 



F=4-l volts. 



With the particles present even to the extent of a million 

 per cub. cm., this charge would be easily retained, since it 



