Experiments ref peeling Heat, 23 



nearly three quarters of an inch in length. This experiment 

 has fince been frequently repeated, and always fucceeds ill 

 nearly the fame manner. 



This ftone remains ftill in the ftate in which the experi- 

 ment left it. Its figure and polifli are quite gone, and it is 

 ftudded all over with carbonaceous matter of a fomewhat 

 metallic brilliancy, which, on examination with a magni- 

 fier, appears to be plumbago, in union with the furface of 

 the diamond. 



Ametlyjisi 



III. An amethyft, weighing i~ carat, in 40" broke into 

 feveral pieces. At the end of i' 55" the fragments were 

 withdrawn and again weighed, as there was reafon to fupect 

 that fome of them had flown off. Some of them, weighing 

 ~|ths of a carat, were then taken and expofed to the heat for 

 1' 15", when they became fufed into one mafs, which, on 

 cooling, again broke into feveral pieces. 



The colour of the ftone is gone; the fragments are femi- 

 tranfparent, and very much refemble fragments of cryftals of 

 a pure white fait; fuch, for inftance, as purified nitre. The 

 fufed mafs might perhaps have been preferved entire, if the 

 precaution of cooling it flowly, by placing it under a muffle, 

 or in a crucible, in a furnace, and fuffering the fire to die 

 away gradually, had been taken ; but this was not attempted. 



IV. Another amethyft, weighing half a carat, expofed to 

 the heat for 5', was fufed into a globule which is nearly 

 fpheroidal and pretty fmooth, but full of fmali bubbles and 

 cracks, fimilar to thofe in glafs which is thrown, when 

 heated, into water. The whole is colourlefs and tranfparent 

 like flint glafs, except a portion on one fide, which, having 

 probably been leaft heated, looks like white enamel. The 

 ftone loft T ' r part of its weight. 



Sapphires. 



V. An oriental fapphire, weighing 1-5^ carat, though the 

 heat was applied with caution, by expofing it firft to the 

 action of a lamp urged by a common blow-pipe, broke into 

 three pieces, but fecmed to have undergone no other change 



Vol. VIII. E at 



