its place in the Classification of Elementary Substances. 5 



In none of these observations of the characteristic 

 yellow line of terrestrial and solar helium, would any 

 account appear to have been taken of the influence of 

 pressure and diffusion with other gases in varying the 

 width of spectral lines, especially on the more refrangible 

 side of the spectrum. I have already shown in my paper 

 on the spectrum of thallium (Proc. Roy. Soc, 1893) that 

 the expansiveness and brightness of the C line of hydrogen, 

 at atmospheric pressure, masked completely one of the two 

 principal lines in the visible spectrum of thallium for more 

 than thirty years, so that the sharp red line in the arc- 

 and spark-spectrum of this element is not mapped in the 

 recent tables of Thalen, and Kayser and Runge.* 



I have recently repeated some of the experiments of 

 Ramsay and Lockyer on helium obtained by the distilla- 

 tion method from Norwegian cleveite, pitchblende, and 

 other minerals containing uranium. The result of these 

 experiments confirms the conclusion that the differences 

 in the determination of the wave-lengths . of the com- 

 ponents of the characteristic yellow line are due to the 

 same cause which masked the red line of thallium. 



The apparatus with which the experiments were made 

 is shown in Plate I., one-fourth the actual size. It 

 consists of a small steel cylinder heated from below by a 

 Bunsen burner, or the oxyhydrogen-fiame. A bent iron 

 tube, of small bore, connects the cylinder with an air- 

 pump and a glass sparking-receiver in which the spectra of 

 the gases are produced. The mouth of the receiver is 

 plugged with a stopper of caoutchouc, through which are 

 thrust a pair of iron wires terminating with platinum 

 electrodes. Vacuum-gauges are mounted on the pump for 

 measuring the amount of rarefaction in the cylinder and 

 sparking-receiver to a small fraction of an inch of mercury. 



* Uebey die Spectren der Elemente, Pt. VI., Berlin, 1892. 

 Brit. Assoc. Report, 1893, p. 403. 



