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



The first assumption is almost certainly justified, although 

 no direct measurements of velocity lost by collision with 

 metal molecules have been made. The second assumption 

 seems to agree with all experimental results, but the evidence 

 available does not exclude other hypotheses. Thus Akesson * 

 has suggested that electrons with velocity several times the 

 resonance potential may lose several " quanta " of energy at 

 one collision. His conclusions are based on experiments 

 with non-metallic gases, and the authors doubt whether he 

 has conclusively proved this important point. 



It has been found that the resonance and ionization 

 potentials of metals are related to frequencies in their line 

 spectra by the quantum equation Ye = hv, whtre V is the 

 potential, e the electron charge, h Planck's constant of action, 

 and v the frequency. 



The resonance potential is determined by the frequency of 

 a prominent line in the spectrum, the first line in a principal 

 or combination series in all cases so far observed. The 

 ionization potential is determined by the limiting frequency 

 o£ this series in all the metals studied except thallium. 

 These spectral relations are similar among metals of the 

 same group of the periodic table. 



The metals were vaporized in porcelain tubes 3x35 cm., 

 electrically heated. 



A cylinder of iron gauze served as a grid, and with lead 

 an outer cylinder of iron was used for the plate. With 

 calcium the plate consisted of an iron tube closed at the 

 bottom, thus protecting the porcelain from the hot metal. 

 A temperature of 1050° C. was required for a sufficiently 

 dense vapour of lead and 950° G. for calcium. At these 

 high temperatures all parts of the apparatus emit electrons, 

 and it is necessary to employ a cathode from which the 

 thermionic emission is relatively high. A hot molybdenum 

 wire coated with calcium oxide was satisfactory for calcium, 

 but it was found necessary to use an equipotential source for 

 lead. This consisted of a small steel tube coated with lime 

 and heated by an inner coil of molybdenum wire. A low 

 gas-pressure was maintained by means of mercury vapour 

 pumps of the type designed by one of the writers -j-. 



Discussion of He suits with Lead, 



Fig. 1 gives some of the curves of partial and total current 

 versus accelerating potential in lead vapour. Curves 4, 14 



* Lunds Universitets Arsskrift, NF. Avd. 2, Bd. xii. Nr. 11. 

 t Wash. Acad. Sci. vii. p. 477 (1917). 



