MODIFIED by COMPRESSION. 143 



tube was perfectly clean. Its weight with its contents, feem- 

 ed to have fuffered no change from what it had been when 

 firfl introduced. Attending, however, to the balance with 

 fcrupulous nicety, a fmall preponderance did appear on the 

 fide of the weight. This was done away by the addition- 

 of the hundredth of a grain to the fcale in which the carbo- 

 nate lay, and an addition of another hundredth produced in it 

 a decided preponderance. Perhaps, had the tube, before its in- 

 troduction, been examined with the fame care, as great a diffe- 

 rence might have been detected ; and it feems as if there had 

 been no lofs, at leaft not more than one hundredth of a grain, 

 which on 10.95 grains, amounts to 0.0912, fay 0.1 per cent. 

 The carbonate was loofe in the little tube, and fell out by ma- 

 king. It had a yellow colour, and compact appearance, with 

 a ftony hardnefs under the knife, and a ftony fracture ; but with 

 very flight facettes, and little or no tranfparency. In fome 

 parts of the fpecimen, a whitifh colour feemed to indicate partial 

 calcination. On examining the fracture, I perceived, with the 

 magnifier, a fmall globule of metal, not vifible to the naked 

 eye, quite infulated and fingle. Poilibly the fubftance may have 

 contained others of the fame fort, which may have compenfated 

 for a fmall lofs, but there could not be many fuch, from the 

 general clean appearance of the whole. In the fra&ure, I faw 

 here and there fmall round holes, feeming, though imperfect- 

 ly, to indicate a beginning of ebullition. 



I made a number of experiments in the fame manner, 

 that is to fay, with the muzzle of the barrel upwards, in fome 

 of which I obtained very fatisfactory remits; but it was 

 only by chance that the fubftance efcaped the contamina-- 

 tion of the funble metal; which induced me to think of ano- 

 ther mode of applying the comprefling weight with the 

 muzzle of the barrel downwards, by which I expected to re- 

 peat, with a determinate weight, all the experiments formerly 



made 



