P lates G, 7.] Sir D avid B rew ster, in 1842, appears to have been the first to examine the spectra of salts by means of the oxygen and coal-gas flame, about 180 of which were deflagrated in a platinum spoon (Edinburgh ' Boy. Soc. Proc./ vol. 6, p. 145). Professor N orman L ockyer* has given us a map of metallic spectra at the tem perature of the oxygen and coal-gas blow-pipe (' Boy. Soc. Proc./ vol. 23, p. 120). The region observed in the case of twenty-two metals does not extend beyond wave

more »

... ength 4000 ; and, although we have both arc and spark spectra for the region up to wave-length 1800, we are still unacquainted with the spectra of elements and compounds obtained by means of flames at high temperatures in the ultra-violet region. ining Spectra with Flames at High Temperatures. In studying the spectra of flames there are many points worthy of consideration arising from the structure of the flame, the nature of the combustible, the heat evolved during combustion, and the temperature attained. The temperature of a candle-flame is high enough to give all the spectra capable of being produced by the oxy-hydrogen blow-pipe, for by such simple means we can melt W ollaston's platinum wires and produce the band spectrum of carbon. The reason for such a flame being practically useless for spectroscopic purposes does not arise from the temperature being too low, but from the area of maximum temperature being too small, so that the material to be tested and the support upon which it is held in the flame exercise * [The following quotation contains a passage which is perhaps the earliest reference to such sp e ctra :-" The pure earths, when violently heated, as has recently been practised by Lieutenant D rummond, by directing on small spheres of them the flames of several spirit lamps urged by oxygen gas, yield from th eir surfaces lights of extraordinary splendour, which, when examined by prism atic analysis, are found to possess the peculiar definite rays in excess which characterize the tints of flames coloured by th e m ; so th at there can be no doubt that these tints arise from the molecules of the colouring m atter reduced to vapour, and held in a state of violent ignition." * Light,' Sir.