CHEMICAL EFFECTS OF PUDDLING. 131 



from the iron, which was much less friabK iron thus 



cleansed from its scoria gave us the following results : — 



PlTHt Anmlyd*. Swxmd Analytic Moan. 



Carbon 2421 2.444 



Silicium 188 0200 0.104 



Fourth Sample, taken at 1 20 p.m. 



As soon as the last sample had been taken out the damper 

 of the furnace was slightly raised, so as to admit a gentle 

 current of air, which did away with the smoke which had 

 been issuing from the puddler's door, and a clear and bright 

 flame was the result. This was done, no doubt, to facilitate 

 the oxidation of the carbon of the iron, and to increase this 

 action the puddler quickly agitated the mass. Under these 

 two actions the mass swelled up rapidly, and increased to at 

 or five times its original bulk, and at 1 20 p.m., the 

 mass being in full boil, this fourth sample was taken out. 

 Whilst cooling, it presented the interesting fact, that in 

 various parts of it small blue flames of oxide of carbon were 

 perceived, no doubt arising from the combustion of carbon 

 by the oxygen of the atmosphere. 



It is curious that this phenomenon was not observed in the 

 previous samples. It is due probably to the following causes: 

 Firstly, that the pig iron having been brought by the boil to a 

 state of minute division, offers a large surface to the action of 

 the oxygen of the air, and thus the combination of the oxygen 

 with the carbon of the iron is facilitated. And secondly, that 

 at this period the carbon seems to possess little or no affinity 

 for the iron, for one of us has often observed, that when pig 

 iron rich in graphite is puddled, the carbon is liberated from 

 the iron ; for if a cold iron rod is plunged into the mass of 

 melted iron in the puddling furnace, it is covered with iron 

 and abundant shining scales of graphite carbon. 



The appearance of this, No. 4 sample, was very interest- 

 ing; and the best idea that, we can give of it is, that 



