1 94 Effects of Heat modified ly Compression. 



cold, and apply the requisite heat to the carbonate lodged \xt 

 the breech. 



I thus- expected to confine the carbonrc acid ; but the at- 

 tempt was attended with considerable difficulty, and has led 

 to the employment of various devices, which I shall now 

 shortlv enumerate, as thev occurred in the course of prac- 

 tice. The simple application of the principle was found in- 

 sufficient, from two causes: 1st, The carbonic acid being 

 driven from the breech of the lube towards the muzzle, 

 among the pores of the pounded silex, escaped from the 

 compressins; force, by lodging itself in cavities which were 

 comparatively cold : 2dly, The glass of borax, on cooling, 

 was always found to crack ver}7 much, so that its tightness 

 could not be depended on. 



To obviate both these inconveniences at once, it occurred 

 to me, in addition to the first arrangement, to place some 

 borax (fig. 10. C) so near the breech of the tube as to un- 

 dergo heat along with the carbonate (A) ; but interposing 

 between this borax and the carbonate a stratum of silex (B), 

 in order to prevent contamination. I trusted that the borax 

 in a liquid or viscid state, being thrust outwards by the ex- 

 pansion of the carbonic acid, would press against the silex 

 beyond it (D), and totally prevent the elastic substances from 

 escaping out of the tube, or even from wandering into its 

 cold pari?. 



In some respects this plan answered to expectation. The 

 glass of borax, which can never be obtained when cold, 

 without innumerable cracks, unites into one continued viscid 

 mass in the lowest red heat; and as the stress in these ex- 

 periments begins only with redness, the borax, being heated 

 at the same time with the carbonate, becomes united and 

 impervious as soon as its action is necessary. Many good 

 results were accordingly obtained in this way. But I found, 

 in practice, that as the heat rose, the borax began to enter 

 into too thin fusion, alid w as often lost among the pores 

 of the silex, the space in which it had lain being foui>3 

 empty on breaking the tube. It was tlicreforc found neces- 

 • , 8 sary 



