,3S4^ ACTION OF HEAT MODIFIED 



fide of (he large tube, and efTeftually clofed its pores. In this 

 manner, many good experiments were made with barrels lying 

 horizontally in common muffles, (the arrangement juft de- 

 fcribed being reprefented in Fig. 13.) 

 Beft material for I was thus enabled to carry on experiments with this 

 ''^'^'" porcelain, to the utmoft that its ftrength would bear. But I 



was not fatisfied with the force fo exerted ; and hoping to 

 obtain lubes of a fuperior quality, I fpent much time in expe- 

 riments with various porcelain compofitions. In this, I fo 

 far fucceeded, as to produce tubes by which the carbonic 

 acid was in a great meafure retained without any internal 

 glaze. The beft material I found for this purpofe, was the 

 pure porcelain-clay of Cornwall, or a compofition in the 

 proportion of two of this clay to one of what the potters cali 

 Cornijh-ftone, which I believe to be a granite in a Rate of de- 

 compofition. Thefe tubes were {qvqw or eight inches long, 

 with a bore tapering from 1 inch to 0.6, Their thicknefs 

 was about O.'d at the breech, and tapered towards the muzzle 

 to the thinnefs of a wafer. 

 Improvement by • I now adopted a new mode of operation, placing the tube 

 ^^rticaV'^^ ^"''^ vertically, and not horizontally, as before. By obferving the 

 thin ftate of borax whilfl: in fulion, I was convinced, that it 

 ought to be treated as a complete liquid, which being fupport= 

 ed in the courfe of the experiment from below, would fecure 

 perfe£l tightnefs, and obviate the failure which often happen- 

 ed in the horizontal pofition^ from the falling of the borax to 

 the lower fide. 

 Particular de- In this view, (fig. 16.) I filled the breech in the manner 



defcribed above, and introduced into the muzzle forae borax 

 (C) fupported at the middle of the tube by a quantity of 

 filex mixed with the bottle glafs (B). I placed the tube, fo 

 prepared, with its breech plunged into a crucible filled with 

 land (E), and its muzzle pointing upwards. It was now my 

 objed to apply heat to the muzzle-half, whilft the other re- 

 mained cold. In that view, I conftrufled a furnace (figs. 14 

 and 15.) having a muffle placed vertically (c d,) furround- 

 ed" on all fides with fire (e e), and open both above (at c), 

 and below (at d). The crucible juft mentioned, with its 

 tube, being then placed on a fupport diredly below the ver- 

 tical muffle, (as reprefented in fig. J 4. al F) it was raifed, fo 

 ^lat the half of the tube ne:j^t the muzzle was introduced into 



the 



fcription of the 

 procei's, 



