76 Drs. Mohler, Foote, and Stimson on Ionization and 



The data lead to a mean value for the resonance potential 

 of 1*26 volts; for the ionization potential 7*93 volts. 



No line series have been found in the lead spectrum, nor 

 is there anything known of the lines appearing at low 

 voltage. Applying the quantum relation Ve=hv ; or 



X= — ^— where V represents volts and X Angstrom units ; 



the resonance potential gives X = 9800 A. with a possible 

 error of 800 A. The single line spectrum of lead, if such 

 exists, should be in this region. Thermopile measurements 

 of the lead spectrum by Randall * show an isolated group of 

 strong lines near this point, and the shortest wave-length 

 line of the group \ = 10291 falls well within the limits of 

 our prediction. 



The ionization potential corresponds to X—1550. This 

 may be the limit of a series of which X= 10291 is the first 

 line, but it is noticeable that the frequency ratio between 

 the first line and limit of such a series is much greater than 

 in the usual type. In nearly all known series the ratio 

 of frequencies is between two and three, while this is nearly 

 seven. In the case of thallium alone we found that ioniza- 

 tion was not determined by the limit of a principal series, 

 but our results have shown that there is little basis for 

 reasoning by analogy when we are dealing with metals in 

 different columns of the periodic table. If X= 10291 is the 

 single line spectrum we are able to compute an accurate 

 value for the resonance potential. The above data thus 

 give V= 1-198 volts. 



Discussion of He suits with Calcium. 



Fig. 2 illustrates typical curves of total and partial currents 

 in calcium vapour. Curves 8, 18, and 22 are total currents, 

 and curves 1, 3, 9, 14, and 15 partial current. In all 

 11 total current curves and 11 partial current curves were 

 obtained. The observations were made at temperatures 

 between 800° C. and 900° C. The shape of the curves with 

 steps of about 2 volts and intermediate inflexions which are 

 'ess prominent can be explained on the assumption that there 

 are two types of inelastic collision without ionization at 

 approximately 1*9 volts and 2*9 volts, of which the first 

 is the more probable. Thus in curve 14 the inflexions at a, 

 c, and e are due to electrons undergoing one, two, and three 



* Astrophys. Journ. xxxiv. p. 1 (1911). 



