FLAME SPECTRA AT HIGH TEMPERATURES. 
211 
nitrate. Strong lines are visible in the continuous spectrum or at its edge. Liveing 
and Dewar obtained their magnesia by burning the metal and holding the ash in the 
oxy-hydrogen flame. 
The line X 2852 is common to both spectra. 
3. As to any possible relation of emission to absorption spectra, it may be remarked 
that Roscoe and Scecuster found that there was apparently none in the case of sodium 
and potassium (‘Roy. Soc. Proc vol. 22, p. 362, 1874), though the spectra were care¬ 
fully measured. 
Lockyer and Chandler Roberts (‘Roy. Soc. Proc.,’ vol. 23, p. 344, 1875) observed 
several channelled absorption spectra of metals by volatilizing them in tubes filled 
with hydrogen. No measurements were made, probably on account of the difficulties 
involved, and consequently the absorption spectra cannot be compared with channelled 
emission spectra of the same elements. 
Channelled absorption spectra were observed in the vapours of silver, manganese, 
chromium, antimony, bismuth, and selenium. 
Continuous absorption was tioticed in copper, cadmium, iron, cobalt, nickel, tin, 
lead, gold, and palladium. 
4. The spectra of sulphur, selenium, and tellurium were carefully investigated by 
Salet (‘Ann. Chim. Phys.’ [4], v'ol. 28, p. 47, 1873; also ‘ Traite Elementalre de 
Spectroscopie,’ p. 221), but only so far as the visible region, chieffy the green and 
blue rays.- There are, however, many bands in the spectra of selenium and tellurium, 
which lie in the ultra-violet region, which appear on my photographs and have been 
measured. 
5. According to a recent photograph, aluminium foil, wdien burnt, yields a beautiful 
channelled spectrum. 
I have ro express my thanks to Mr. Hugh Ramage, F.I.C., Assistant Chemist, 
Royal College of Science, for the care with which he has photographed many of these 
spectra, and otherwise rendered me valuable assistance. 
W. N. H., Sept. 29, 1893.] 
2 P. 2 
