122 EFFECTS of HEAT 



fire was allowed to burn out till the barrel was black. The me- 

 tal was difcharged irregularly. Towards the end, the inflam- 

 mable air produced, burnt at the muzzle, with a lambent flame, 

 during fome time, arifing doubtlefs from hydrogen gas, more 

 or lefs pure, produced by the decompolition of the water. The 

 upper pyrometer indicated 36 , and the lower one 19 . The 

 chalk which lay in the outer part of the large tube was in a 

 Hate of marble. The inner tube was united to the outer 

 one, by a liar of fufed matter, black at the edges, and fpread- 

 ing all round, furrounding one of the fragments of porcelain 

 which had fallen by accident in between the tubes. The in- 

 ner tube, with the Harry matter adhering to it, but without 

 the coated fragment, feemed to have fuftained a lofs of 1 2 per 

 cent., on' the original carbonate introduced. But, the fub- 

 flance furrounding the fragment being inappreciable, it was 

 impoflible to learn what lofs had been really fuftained. Exa- 

 mining the little tube, I found its edges clean, no boiling over 

 having taken place. The top of the fmall lump of chalk had 

 funk much. When the little tube was broken, its contents gave 

 proof of fuiion in fome parts, and in others, of the nearefl ap- 

 proach to it. A ftrong action of ebullition had taken place all 

 round, at the contact of the tube with the carbonate : in the 

 heart, the fub fiance had a tranfparent granular texture, with little 

 or no cryflallization. The fmall piece of lump-chalk was united 

 and blended with the rammed powder, fo that they could fcarce- 

 ly be diftinguifhed. In the lower part of the carbonate, where 

 the heat mull have been weaker, the rod had acted more feebly 

 on the tube, and was detached from it : here the fubltance was 

 firm, and was highly marked in the fracture with cryftalline 

 facettes. Wherever the carbonate touched the tube, the two 

 fubllances exhibited, in their mixture, much greater proofs of 

 fuiion than could be found in the pure carbonate. At one 

 place, a ftream of this compound had penetrated a rent in 



the 



