260 Dr. W. M: Watts on the Spectra of Carbon. 



nary pressure, but that when the pressure of the gas is increased 

 the temperature of the spark rises. When, then, we find that 

 on gradually diminishing the pressure the same spectrum is 

 given until the pressure falls to about 100 millims., and then 

 suddenly changes to the third form, we can hardly resist the 

 conclusion that this third form of carbon-spectrum is due to 

 carbon rendered luminous below 1500° C. The result that 

 the temperature of the discharge in a vacuum-tube may be 

 below 1500° C. is certainly unexpected, but it can hardly be 

 rejected* unless we give up the attempt to account for the dif- 

 ferences in the spectra of the same element by differences in the 

 temperature of ignition. We may, of course, suppose the exist- 

 ence of allotropic modifications of carbon-vapour, but we have no 

 proof of the existence of such. 



The explanation of the Bessemer- flame is extremely difficult. 

 I have endeavoured to obtain some approximation to the tempe- 

 rature of the flame both by calculation and by experiment. The 

 calculation is based upon the composition of the gas issuing from 

 the convertor. A sample of the gas collected from the converter 

 at the Steel- works at Crewe was analyzed by Mr. C. R. A. Wright, 

 B.Sc, and gave the following result : — 



Carbonic anhydride . . . 3*78 



Carbonic oxide 16*20 



Oxygen 0*57 



Nitrogen 79*44 



99-99 



The temperature is calculated on the assumption that the 

 oxygen of the air is used up in burning the carbon of the cast 

 iron to carbonic oxide and carbonic anhydride, and in burning 

 the iron to ferroso-ferric oxide. 



litres. grs. grs. 



3*78 carbonic anhydride weigh 7*43 and contain 2*03 carbon. 

 16-20 carbonic oxide „ 20-27 „ 869 „ 



0-57 oxygen „ 082 



79*44 nitrogen „ 99*92 



The total volume of oxygen contained in the gaseous products 

 of combustion is 



* Wullner (Pogg. Ann. Dec. 1868) regards the temperature in a hydro- 

 gen vacuum-tube as at a maximum when the tension is about 30 millims., 

 being lessened either by increase or diminution of the pressure. 



