[mclennan-petrie] spectra OF HELIUM, ETC. 19 



the shutter was opened with the electromagnet and the exposure then 

 made. This was generally of from nine to ten hours' duration. Some 

 photographs were also taken of the vacuum carbon arc, the spectro- 

 graph and the lamp, a drawing of which is shown in Fig. 3, being as 

 highly exhausted as possible. A current of 50 amperes was used, the 

 supply being the 110 volt D.C. main. In addition to the above some 

 photographs were taken of the lead spark in helium at atmospheric 

 pressure. The sparking chamber used in this case was made of glass 

 and is shown in Fig. 4. Previous to the taking of the photographs of 

 the spectra of the lead spark the helium was purified by being made 

 to circulate through the charcoal tubes Q and R. While this was 

 being done the spark discharge was made to pass between the elec- 

 trodes in order to drive out any gaseous impurity which might have 

 been occluded in them. During this preliminary sparking the shutter 

 was, of course, kept in the closed position. 



III. Results 



The plates were all carefully measured up and the mean wave- 

 lengths obtained from them, together with their relative intensities, 

 are given in Table I. The values given by Lyman^ for the wave- 

 lengths obtained by him with the spark spectrum of helium, as well 

 as those obtained by Millikan,- and by Millikan, Bowen and Sawyer^ 

 with the spark spectrum of carbon, are also included in the table. 



IV. Discussion of Results 



A point of interest which arises in considering the results given 

 in the table attaches to the wave-lengths X = 1931 A.U., X = 1657 A.U. 

 and X = 1561 A.U. These wave-lengths all came out strongly in the 

 spectra of helium, carbon and lead. The first and last were absent 

 from the spectrum of hydrogen, but X = 1657 A.U., though of weak 

 intensity, was clearly marked in the spectrum of this gas. These 

 three wave-lengths were strongly recorded in the spark spectrum of 

 carbon obtained by Millikan, and they were also obtained as wave- 

 lengths of strong intensity in the spectra of carbon obtained by 

 McLennan and Lang* and by McLennan, Ainslie and Fuller.^ While 

 working with fluorite or vacuum grating spectrographs several 



1 Lyman, Ast. Phys. Jl., Vol. XLIII, p. 89, 1916. 



2 Millikan, Ast. Phys. Jl., Vol. LH, No. 1, p. 47, 1920. 



3 Millikan, Bowen and Saw>'er, Ast. Phys. Jl., Vol. LIII, No. 2, p. 150, 1921. 

 * McLennan and Lang, Proc. Roy. Soc. 95, p. 272, 1919. 



6 McLennan, Ainslie and Fuller, Proc. Roy. Soc. 95, p. 327, 1919. 

 —24 



