-182 
PEOFESSOIi W. N HAKTLEY AND ^lE. IL EAMAGE OX THE SPECTEA 
For 3 and 4 the spectrograph was pointed as in C. 
Plate 1 was very sensitive to the greenish-yellow and yellow rays, but it shows no 
ti'ace of the sodium lines. The spectra are very strong in the blue and violet, but 
fade gradually towards tlie extreme violet, and thence to the wave-lengths recorded 
above. 
It must 1)6 remembered that the dame playing across the hearth of the furnace, 
although at a very high temjoerature, is oxidising, and, as will be seen later (j). 498), it 
should, at the most, give only feeble spectra of such elements as lithium, jDotassium, 
calcium, and, we should infer, iron and probably manganese. The carbon monoxide 
as it escapes from the bath of metal and slag would, however, tend to intensify the 
spectra of these metals, and care was taken to turn the collimator in a direction 
parallel to the surface of the hath and as near to its surface as the opening for the 
door would allow. 
The temperature of the gas must be higher than the temperature of the walls of 
the furnace which are heated by it, hence the fact of the sodium line being bright is 
in accordance with Kirchhoff’s law. 
Ohservations on the Flames from a Cupola Furnace. 
While at Crewe Works, eye observations were made on the spectra of the gases 
passing the opening of a cupola througli which the charge was introduced. Pig-iron 
was being melted in the cuj^ola for the “ acid ” Bessemer jirocess, while the blast was 
turned on the flame was bluish, and lines of sodium, lithium, and potassium were 
detected; the latter were very weak. When the Boot’s blower stopped, the flame 
l)ecame smaller and whiter, and the lines of the above elements became stronger ; 
the ends of the two strongest bands of manganese were also seen. There was, 
doubtless, a large j)roportion of reducing gas in the flame in the latter case. 
Description of the “ Blow ” 
and “ Over-blow” in the 
Middleshroiajh. 
“ Basic ” Bessemer Process at 
The vessel is first charged with about 2 tons of lime in lumps, and then with 
12 tons of fluid “mixer metal.” The blast is turned on, and the vessel rotated into a 
nearly vertical position. 
The iron used for the Ijasic Bessemer process in the works is a mixture of fluid 
metals coming direct from the blast furnaces with molten pig-iron from a cupola, the 
two fluids being completely mixed and in such proportions that the composition of 
the mixture is practically constant; it is thus rendered suitable for uniform treat¬ 
ment. This, which is technically known as “ mixer metal,” is poured out of a 
Hoerder mixer into a ladle, which delivers it into the converter; this “mixer metal” 
usually contains— 
