1082 
PROFESSOR W. N. HARTLEY ON 
kinds of metal, but in no case could it exceed the temperature of the “ boil,” or that 
moment when the spectrum is most intense. 
It must be remembered that the composition of the pig iron used, and the more or 
less rapid rate at which it is blown, undoubtedly influence the temperature. The 
greater the mass of material operated upon, the greater will be the rise in tempera¬ 
ture. For instance, a greater heat evolution and a higher temperature would result 
from the combustion of 10 cwt. of silicon in 12 tons of pig iron during a blow of 
12 minutes, than from 1 cwt. of silicon in 5 tons of pig iron blown for 20 minutes. 
There is one significant fact to be observed in the spectra of the flame photographed 
during the “ boil” and the ‘‘finishing stage” which bears upon the temperature of the 
metal. When the oxyhydrogen-flame spectra of manganese, manganic oxide, iron, 
and ferric oxide are photographed, the number of the lines and bands in the spectra 
are not more numerous than with a Bessemer flame spectrum of only one half minute’s 
exposure, though the above spectra may have received any exposure from 30 to 80 
minutes. 
Marshall Watts observed (‘Phil. Mag.,’ 1870) that the sodium lines 5681 and 
5687 may be employed as an index of temperature, since they are present in the 
spectrum of any flame containing sodium the temperature of which is hot enough to 
melt platinum, but do not appear at lower temperatures. The Bessemer flame does 
not show this double line, but only the I) lines; neither does it show the lithium 
orange line, which appears at a somewhat lower temperature. 
We cannot conclude from this that the flame is not hot enough to produce these 
lines, because in such a case we have to deal, not only with the temperature, but the 
quantity of material present, and the relative brilliancy and consequent visibility of 
the two pairs of sodium lines. 
The proportion of sodium in the Bessemer flame is evidently very small, from the 
narrowness and want of intensity of the D lines, and the fact that they were not seen 
reversed in any spectrum. Hence, though the temperature may be high enough, the 
quantity of material present is not sufficiently large to yield the lines 5681 and 5687. 
The quantitative relations of the different lines have really not been investigated in 
flame-spectra, except and alone so far as they apply to total extinction of all lines, 
which in the case of sodium refers to the D lines only, and this in flames no hotter 
than that of a Bunsen burner. If we apply the same line of reasoning to the appear¬ 
ance of the reversed hydrogen lines in the red and the blue, it may be stated that the 
line in the red, corresponding to solar line C, never appears in any hydrogen or hydro¬ 
carbon flame burnt v/ith air or with oxygen. It invariably appears in sparks passed 
through steam, and it also comes out as a reversed line under suitable conditions in 
the Bessemer spectrum. 
The conditions of its appearance do not depend upon an alteration in temperature, 
but on the presence of a sufficient amount of water-vapour in the blast. 
This certainly seems to point to a higher temperature than that accorded to the 
