386 



SCIENCE 



[N. S. Vol. LII. No. 1347 



in McGill University, Montreal, to date from 

 October 1. Dr. Macallum will continue his 

 work as chairman of the Research Council 

 until a successor has been named. 



Me. Florian A. Cajori, formerly captain in 

 the food section of the Sanitary Corps and on 

 duty in Jugo-iSlavia "with the American Relief 

 Administration, has completed his graduate 

 work at Yale University and accepted a posi- 

 tion as instructor in physiological chemistry 

 at Leland Stanford, Jr., University, in Cali- 

 fornia. 



. Professor J. T. Wilson, F.R.S., Challis pro- 

 fessor of anatomy in the University of Sydney, 

 has been appointed to the chair of anatomy at 

 the University of Cambridge. 



Dr. Wilhelm Weest, professor of physics at 

 Wiirzburg, has been appointed to succeed Pro- 

 fessor Wilhelm Ebntgen, who recently retired 

 from the chair of physics at the University of 

 Munich. 



DISCUSSION AND CORRESPONDENCE 



THE SPECTRUM OF MERCURY VAPOR 



To THE Editor of Science: In your issue 

 of September 10, Professor C. D. Child calls 

 attention to a greenish glow discharge through 

 mercury vapour whose spectrum shows a con- 

 tinuous band throughout the greater part of 

 the visible spectrum, with the ordinary lines 

 superimposed. This summer, in experiment- 

 ing on the electrodeless discharge of certain 

 vapors, the -wTiter observed a similar appear- 

 ance. The method used was one previously 

 employed by Kowalski.^ A small quantity of 

 mercury was introduced into a highly ex- 

 hausted Pyrex bulb some 12 cm in diameter. 

 The bulb, surrounded by the primary coil of 

 a Tesla high-frequency outfit, was placed in 

 an electric oven and the appearance of the 

 discharge (if any) observed as the tempera- 

 ture was gradually increased. 



In common with Kowalski the writer ob- 

 served two distinct types of discharge. The 

 first, a dazzling white ring discharge occurred 

 at temperatures several degrees above and 

 below 90° C, and showed the ordinary bright 



1 J. Kowalski, Physik. Zeit., 15, 225, 1914. 



line spectrum. The second, a diffuse dis- 

 tinctly greenish glow filling the whole bidb, 

 took place at higher temperatures and was 

 visible until a temperature in the neighbor- 

 hood of 200° C. was reached. The spectrima 

 of this latter type showed a continuous band 

 with superimposed lines, an appearance simi- 

 lar to that described by Professor Child, but 

 at the higher temperatures only the line 6Jf61 

 was visible. The writer's observations agree 

 with those of Kowalski, who compares the 

 appearance he observed with an exactly simi- 

 lar one recorded by A. Kalahne.^ 



Professor Child states that the "radiators" 

 giving rise to the continuous band are un- 

 charged, and suggests that the source of this 

 type of radiation has to do with the forma- 

 tion of clusters of two or three atoms which 

 may be fomaed when mercury vapor is con- 

 densing. Professor Kowlaski ascribes the two 

 appearances noted above to two ionization 

 stages ("lonisierungsstufe")- It would seem 

 that a possible explanation is the following. 

 At the lower temperatures, because of the 

 greater mean free path, even in the case of 

 an electric field of relatively small intensity, 

 sufficient energy is communicated to an atom 

 on collision to produce ionization. During 

 recombination the line spectrum is emitted. 

 At the higher temperatures, because of the 

 relatively small mean free path (the vapor 

 pressure of mercury at 160° is roughly 

 twenty-five times that at 90°), but little 

 energ-y is communicated on collision and but 

 little, if any, ionization occurs. The line 

 spectrum accordingly is feeble or absent. 

 Some electrons, however, are displaced from 

 their normal orbits, and in their return to 

 their normal positions, radiation is emitted. 

 Normally such a radiation would also give 

 rise to spectral lines, but we may assume that 

 in the case of the mercury atom with its 

 numerous electrons, the frequent atomic im- 

 pacts occurring at high temperatures alter 

 the natural periods to such an extent that the 

 emission is continuous over a vidde range. 



The writer has imder way an extended 

 study of the electrodeless discharge of certain 



2 A. Kalahne, Wied. Ann., 65, 815, 1898. 



