MODIFIED by COMPRESSION. 147 



placing a heavy mafs, (fig. 40.), fo as to ad diredly and 

 fimply upon the muzzle of the barrel ; this mafs being guided 

 and commanded by means of a powerful lever, (a b). For 

 this purpofe, I procured an iron roller, weighing 3 cwt. 7 lb,, 

 and fufpended it over the furnace, to the end of a beam of 

 wood, refting on a fupport near the furnace, with a long arm 

 guided by a rope (c c) and pulley (dj, by which the weight 

 could be raifed or let down at pleafure. 



With this apparatus I made fome tolerable experiments ', but 

 I found the weight too light to afford certain and fteady refults 

 of the beft quality. I therefore procured at the foundry a 

 large mafs of iron ff), intended, I believe, for driving piles, and 

 which, after allowing for the counterpoife of the beam, gave a 

 direcT: preffure of 8.1 cwt. ; and I could, at pleafure, diminifh 

 the compreffing force, by placing a bucket (e) at the extremity 

 of the lever, into which I introduced weights, whofe effect 

 on the ultimate great mafs, was known by trial. Many bar- 

 rels failed in thefe trials : at laft, I obtained one of fmall bore, 

 inch 0.54, which gave two good refults on the 2 2d of June 

 1804. 



No. 8. — Wishing to afcertain the leaft compreffing force by 

 which the carbonate could be effe&ually conftrained in melt- 

 ing heats, I firft obferved every thing flanding firm in a heat 

 of above 20 ; I then gradually threw weights into the bucket, 

 till the compreffing force was reduced to 2 cwt. Till then, 

 things continued fteady ; but, on the preffiire being flill further 

 diminifhed, metal began to ooze out at the muzzle, with in- 

 creafing rapidity. When the preffure was reduced to i~ cwt. 

 air rulhed out with a hilling noife. I then flopped the experi- 

 ment, by pouring water on the barrel. The piece of chalk 

 had loft 12 per cent. It was white and foft on the outfide, but 

 firm and good in the heart. 



T 2 No. 9. 



