Positive Ionization from Hot Salts. 673 



stage by Mr. C. J. Davisson*, who has made several im- 

 portant improvements in the apparatus which I had used, 

 and has made a very thorough examination of the emission 

 from the salts of the alkaline earth metals. He finds that 

 all the ions emitted by these salts fall into two groups : 



(1) those which have values of ejm corresponding to an atom 

 of the earth metal which has lost a single electron, and 



(2) those which have values in the neighbourhood of those 

 corresponding to K + and Na + , or intermediate between them. 

 In point of fact nearly all the ions of the second group were 

 quite close to the value corresponding to K + . The experi- 

 ments with barium salts were particularly instructive. With 

 a single exception all the salts tried furnished only ions of 

 the type Ba + . No evidence could be obtained for the 

 emission of ions of the type Ba ++ , Sr ++ , Ca ++ , and Mg + + 

 from any of the salts, although they were looked for most 

 carefully by Mr. Davisson. 



The discovery by Garrett f that the pressure of the sur- 

 rounding gas had an important effect on the emission of 

 positive ions from hot aluminium phosphate seemed to render 

 it probable that at pressures high enough to cause a con- 

 siderable increase in the thermionic current, part at any rate 

 of the positive ions might be expected to be charged atoms 

 or molecules of the surrounding gas rather than atoms of the 

 basic metal. It seemed that this question could be definitely 

 settled by making measurements of e/m for the ions from 

 such salts in gases at low pressures. According to the curves 

 given by Garrett the ionization from aluminium phosphate 

 is very much greater at a pressure of one-twentieth of a 

 millimetre of air or carbon dioxide than it is in a good 

 vacuum. Strictly speaking the method of measuring ejm 

 assumes that the number of ions which undergo collisions 

 between the electrodes should be negligible, but it seemed 

 likely that the method would give fairly satisfactory results 

 up to pressures of this order. I therefore suggested to Mr. 

 Davisson that he should make measurements of the value of 

 ejm for the positive ions emitted by aluminium phosphate in 

 different gases at different pressures %. He found that in 

 air and hydrogen the value of ejm was practically the same 

 at all pressures below about one-tenth of a millimetre, and 

 was identical with the value in carbon dioxide at pressures 

 less than about one-thirtieth of a millimetre. The value of 

 ejm which he found corresponded very closely with that for 



* Phvs. Rev. vol. xxxii. p. 620 (1011). 

 t Phil, Mag. [6] vol. xx. p. 573 (1910;. 

 X C. J. Davisson, Phvs. Rev. vol. xxxii. p. 620 fl911). 



