﻿880 Dr. E. N. da C. Andrade on the Carriers of 



Lenard gives as the mobility calculated from this formula 



the value 2'3 — iu nitrogen at 2000° C. *, but the diameter 



sec. 



of the metal atom assumed by him is, according to work 

 which has since appeared, too large, and the value of e too 

 small. Taking for e the value 4*7 x 10 ~ 10 c.G.s. electrostatic 

 units (Planck), and calculating the radius of the strontium 

 atom from the value given by Becker f for the radius of the 

 sodium atom, we have for the mobility of a permanently 



charged metallic atom 17 — ^ / ■ (or for a molecule of 



, sec./ cm. v 



SrO 15 — - / '). Thus we must assume, as we have 



sec./ cm. J 



done above, th;it even the faster carriers cannot be always 

 positively charged, but alternate the charged state with a 

 neutral state caused by recombination with a free electron. ■ 

 Evidence has been given for the assumption that the 

 faster carriers come from a region near the edge of the 

 luminous streak. Now iu the streak itself we may assume 

 that the impact of the metallic atoms against one another is 

 the main cause of the liberation of an electron, and the con- 

 sequent rendering of an atom positive ; impact between a 

 metal atom and a molecule of the flame-gases can also be 

 sufficient to render an atom positive, but is a far less active 

 cause %. In consequence of the large number of free electrons 

 produced in the streak itself, the positive atom is quickly 

 neutralized again ; in fact from the velocity of the carriers 

 in the streak the atom can, on the average, be positive for 

 only about ^J^ of the time. Now a metallic atom w r hich 

 is near the edge of the streak may, while it is positively 

 charged, be driven out of or diffuse out of § the luminous 

 streak, where the metal concentration is high, into the free 

 flame remote from the positive electrode, where the number 

 of free electrons must be small. Hence the average positive 

 life of an atom which has arrived into the free flame near 

 the earthed electrode will be much longer than that of an 

 atom in the dense streak. From the calculated mobility of 

 a permanently charged metallic atom, a carrier of the faster 

 kind must be positive for about \ of the time. 



* P. Lenard, Ann. der Physik [4] xvii. p. 197 (1905). 



t A.. Becker, "Uber die Diffusion leuchtender Metalldampfe etc.," 

 Sitzungsberichte der Heidelberger Acad. 1911, p. 16. 



X E. N. da C. Andrade, " Some Experiments on the Electrical Be- 

 haviour of Metallic Vapours etc.," shortly to be published ; and P. Lenard, 

 Uber die Elektricitatsleitung und Lichtemission etc.," ' Sitzungsberic/de der 

 Heidelberger Acad. 1911. 



§ Cf. p. 869. 



