FLAME SPECTRA AT HIGH TEMPERATURES. 
1(55 
second, and third series of lines measured, Liveinc4 and Dewar, always appear in 
these spectra; in some cases the fourth and fifth series are well rendered. 
Although the number of lines exhibited by some of the metals is large, yet the 
extent of spectrum is small compared with that yielded by condensed sparks. 
Typical band spectra are exhibited by sulphur, selenium, and tellurium. The first 
yields a continuous spectrum, in which a series of beautiful bands is seen, the second 
a series of fine bands occurring at closer intervals, the third is characterized by 
bands still closer together, and near the more refrangible termination of which four 
lines occurring in the spark spectrum of tellurium are visible. 
Thus we see that increase in atomic mass causes shorter periods of recurrence of 
bands, while we know that it causes greater periods in the recurrence of lines. 
It is also worthy of remark that the most volatile of these elements emits a 
continuous spectrum, with a band spectrum just emerging from it ; the second gives 
us a beautiful and purely a band spectrum, while the third least volatile and more 
metallic substance of largest atomic mass, and producing the densest vapour, yields a 
band spectrum, together with a line spectrum. Several metals, such as nickel, yield 
nothing but lines, others give us both lines and bands, as manganese and iron, while 
tin, lead, silver, and gold yield very beautiful band spectra. Metalloids and non- 
metallic elements are generally considered to be essentially different from metals, 
since they emit channelled or baiid spectra at one temperature and line spectra 
at another. It was, in fact, first laid down by Plucker and Hittorf that “ Thei'e 
is a certain number of ele)nentary substances, which, ivhen clijferently heated, furnish 
two kinds of spectra of quite a different character, not having any line or any band 
in common ” (‘ Phil. Trans.,’ vol. 155, p. 6). 
The discovery of this fact was of great importance, for it led to the conclusion that 
as one spectrum of an element is replaced by another and totally different spectrum 
of the same element, there must be an analogous change in the constitution of the 
ether, indicating a new arrangement of the gaseous molecules, and this implies either 
a chemical decomposition, or an allotropic condition of the vapour of the substance. 
Plucker and Hittorf concluded that the same matter, in two allotropic states, 
emitted different spectra, but the allotropy was dependent solely on temperature. 
Band spectra they designated spectra of the 1st Order, and Line spectra, spectra of 
the 2nd Order. The former have been fully recognized as the spectra of metal¬ 
loids, such as c.arbon, phosphorus, sulphur, selenium, and tellurium, but it seems to 
have been overlooked that Plucker and Hittorf observed spectra of the 1st Order- 
in the case of a few heavy metals, particularly lead and manganese. Metallic lead 
and its compounds were found to yield the same band spectrum in the oxy-hydrogen 
flame, and manganese exhibited a curious spectrum of the 1st Order, most similar 
to that of carbon, but with the lines composing the bands differently distributed. 
The well-known spark spectra of these elements are spectra of the 2nd Order. 
Lecoq de Boisbaudran has observed a beautiful spectrum of aluminium of the 
