OF FLAMES EESULTIXG FROM BESSEMER PROCESSES. 
503 
vanadium is present both in the Cleveland ore, in the iron smelted from it, and in the 
basic slag. The basic Bessemer steel appears to be free from vanadium. This 
element gives no spectrum in the oxyhydrogen flame by which its presence may 
be recognised, but we have detected it by chemical reactions in solutions of the ii’on 
and of the basic slag. 
The oxy hydrogen flame spectra of all the elements named above have been 
studied in the course of this investigation, as well as those of nearly all of the other 
elements. 
The basic steel contains the following elements : Iron, gallium, manganese, lead, 
copper, nickel, cobalt, silver, chromium, calcium, sodium, and potassium. The 
chromium, cobalt, and silver are present in minute traces. 
The Bessemer flame spectrum does not contain any lines of strontium, chromium, 
nickel, cobalt, or lead, elements present in the raw inaterials which are easily 
recognised by lines in the oxyhydrogen flame spectra. These are doubtless present 
in too small quantities to give lines strong enough to appear on the strong continuous 
spectrum ; some of the weaker iron lines are also lost in the continuous spectrum in 
many of the Bessemer flame spectra photographed. 
Some attempts were made to reproduce the more complex part of the spectrum of 
the Bessemer flame in the laboratory. Bolls of filter joaper containing a powdered 
mixture of compounds of iron, manganese, })otassium, sodium, and calcium were 
burnt in tlie oxyhydrogen flame, and the spectrum photographed in the same instru¬ 
ment as was used at Middlesbrough. After a few experiments, in which different 
proportions of tlie constituents were used, a mixture was arrived at which gave a 
very fair copy of the Bessemer flame spectrum. 
Conclusions. 
(1.) Line si^ectra are not observed in the open-hearth furnace. 
The open-hearth furnace, though at a temperature as high as that of the Bessemer 
converter, does not show in a })hotograph the spectra of any of the metals known to 
be present, but a band of continuous rays extending from the red into the ultra-violet. 
The reason is that the atmosphere of the furnace is an oxidising one, and, moreover, 
the oxidation becomes greater when the furnace doors are opened. This was shown 
by observations on a cnpola furnace ; the metallic lines were visible only when the 
Iflast was turned off, that is to say, when the atmosphere had a reducing action. 
(2.) The phenomena of the “ basic” Bessemer blotv differ considerabhj from those of 
the “ acid ” process. 
First, from the very commencement of blowing a luminous flame appears. It is 
visible as soon as the lime dust has been blown out of the vessel. We have proved 
that it might proceed from possibly three sources : ignited lime dust, ignited alkali 
metals, which give a continuous spectrum, and hydrocarbons from the tar mixed with 
