Question of Valency in Gaseous Ionization. 769 



catches observed with this drop icas a single. When large 

 drops were used the chance that the same (3 particle would 

 ionize two adjacent molecules, and thus throw two separate 

 ions simultaneously upon the drop, became larger, and yet, 

 in general, we caught only singles even with drops which 

 were so large that a j3 particle in going the length of a 

 diameter of a drop would have to pass through at least 

 8 molecules. The drop shown in Table V. has a diameter 

 which is 20 times the mean free path of an air molecule, and 

 20 

 — =3*5 times the mean free path of an electron, if the 



latter be regarded as a point, yet out of 34 catches there is 

 but one which could possibly be a double. This means of 

 course that a /3 particle ionizes but a small fraction of the 

 molecules through which it actually passes — a conclusion 

 which can also be reached in other ways. With the use of 

 drops whose diameters were ten times the mean free path 

 of an electron we occasionally, as in our early experiments, 

 observed double changes when the field was on, but since 

 these were never observed with drops of one-tenth the size, 

 the conclusion is inevitable that these apparent doubly 

 charged ions were in fact two separate singly charged ions. 



The above observations represent only a small portion of 

 those which we have made within the past three months, but 

 they present the best evidence which we have been able to 

 find for the formation of doubly charged ions. We would, 

 however, point out this difference between our experiments 

 and those of Townsend and F ranch and Westphal. While 

 their measurements have to do with the charges carried by 

 ions at a considerable time after the formation of these ions 

 our experiments have to do solely with the measurement of the 

 charge which is freed from a neutral moleade in the act of 

 ionization. If two single positive charges attached themselves 

 after formation to a minute dust particle or other molecular 

 aggregate, and thus formed a doubly charged ion, our experi- 

 ments would not reveal the fact. Or, again, if X rays were 

 capable of detaching more readily an elementary charge from 

 an already singly charged ion than from a neutral molecule, 

 and thus forming double positives (a very improbable hypo- 

 thesis), our method would not be able to detect such doubles. 



We are extending investigations of the types herein 

 described to other gases and to vapours. 



§ 6. Conclusion. 



Our conclusion may be stated as follows : — Although we 

 entered upon this investigation with the expectation of proving 

 the existence of valency in gaseous ionization, we have instead 



