Dr. W. M. Watts on the Spectra of Carbon. 261 



litres. litres. 



3*78 in 3-78 carbonic anhydride. 

 8-10 in 16-20 carbonic oxide. 

 0-57 



12-45 



But 7944 litres of nitrogen are mixed in air with 21 litres of 

 oxygen. Hence 21 — 12*45 = 8-55 litres of oxygen have com- 

 bined with iron. 



The heat produced by the combustion is as follows : — 



grs. grs. Thermal units. 



2-03 carbon burning to CO 2 evolve 203 x 8080= 16402 

 8-69 „ CO „ 8-69x2474= 21499 



32-08 iron „ Fe 3 O 4 „ 32-08 x 1582= 50778 



88679 



The products of combustion and their specific heats are as 

 follows : — 



grs. 



7*43 CO 2 x 0216 = 1-60 



20-27 CO x 0-248 = 5-03 



44-26 Fe 3 4 x 0-152* = 6-73 



99-92 N x 0-244 =24-38 



0-82 x 0-218 = 0-18 



37-92 

 and the temperature of the flame is therefore 

 88679 



The result of this calculation is, of course, open to the same 

 objection as all calculated flame-temperatures, that no allowance 

 can be made for dissociation. It is too high also for another 

 reason — that a very considerable part of the heat produced is 

 expended in heating up the molten metal itself, which is im- 

 mensely hotter at the end of the blow than it is at the beginning. 

 If we assume that, together with the quantities given above, we 

 have 300 grs. iron heated up from 1000° C. to the temperature 



» 3X6-4+4X4 =ntl 



at. wt. Fe 3 O l 

 t This calculation represents 10 grs. carbon burnt for 32 grs. iron. As- 

 suming the pig-iron to contain 3 per cent, carbon, this would give a loss of 



32 iron for — — X 10=333 pig iron, or about 1.0 per cent. The average 

 o 



loss from all causes is reckoned, I believe, at about 15 per cent. 



