304 Effects of Heat modified by Compression. 



the square barrel, the heat had been much higher than was 

 supposed at the time, frcm the indication of the pyrometer 

 placed on the breech of the barrel ; and that in some of 

 them, particularly in the last, it must have risen at least as 

 high as in the present experiment. 



On the Slsfuf April 1805, a similar experiment was 

 made with a nc^v barrel, bored in a square bar of old sable 

 of about two. inches and a half in diameter, having its angles 

 merely rounded, the inner tube being filled with chalk. The 

 heat was maintained during several hours, and the furnace 

 allowed to burn out during the night. The barrel had the 

 appearance of soundness, but the metals came off' quietlv, 

 and the carbonate was entirely calcined, the pyrometer in- 

 dicating 63°. On examination, and after beating off the 

 smooth and even scale of oxide peculiar to the old sable, 

 the barrel was found to have yielded in its peculiar manner; 

 that is, by the opening of the longitudinal fibres. This ex- 

 periment, notwithstanding the failure of the barrel, was one 

 of the most interesting I had made, since it afforded proof 

 of complete fusion. The carbonate had boiled over the lips 

 of the little tube, standing, as just described, with its mouth 

 upwards, and had run down to within half an inch of its 

 lower end : mor-t of the substance was in a frothy state, with 

 large round cavities, and a shining surface ; in other parts 

 it was interspersed with angular masses, which have evi- 

 dently been surrounded by a liquid in which they floated. 

 It was harder, I thought, than marble; giving no effer- 

 vescence, and not turning red like quicklime in nitric acid, 

 which seemed to have no effect upon it in the lump. It 

 was probably a compound of quicklime with the substance 

 ot the tube. 



With the same barrel repaired, and with others like it, 

 many similar experiments were made at this time with great 

 success ; but to mention them in detail, would amount 

 nearly to a repetition of what. has been said. I shall take 

 notice of only four of them, which, when compared toge- 

 ther, throw much light on the theory of these operations, 

 and hkewise seem to establish a very important principle in 



seolocv. 



