J2J. OBSERVATIONS ON BASALT, &C. 



merit of the molecules of the glafs-. It commences by the 

 formation of fibres perpendicular to the furface of the glafs, 

 and penetrating into it. At nearly the fame time, fmall ra- 

 diated globules are formed in the interior of the glafs, and 

 the union of thefe with the fibres, by their mutual increafe, 

 forms the whole into a new fubftance; and, if the requifite 

 temperature be longer maintained, the fibres difappear, and 

 the whole becomes fine-grained, and almoft compact. This 

 fubftance, from the improved ftate of its aggregation, is much 

 ftronger and more tenacious than before, and is not fufible at 

 a heat fufficient to fufe the glafs it was formed from ; but, if 

 that aggregation be once deftroyed, the glafs refulting from 

 its fufion is equally fufible with the original glafs; and a re- 

 petition of the procefs will again form Reaumur's porcelain, 

 which may be again fufed, and fo on repeatedly, for the 

 quantity of alkali evaporated during the operation is ex- 

 hardening; tremely fmall. The hardnefs and brittlenefs of metals rapidly 

 3n"" S ' an * cooled > conlrafied with the foftnefs and tenacity refulting 

 from their gradual refrigeration, are all analagous inftances ; 

 and all the procefles in which annealing is employed, and 

 more remarkably the tempering of fteel, the proofs of (he 

 internal motions and arrangements of the particles of matter, 

 at temperatures very much below the heat is requifite for their 

 fluidity. 

 Oyftallization Whatever doubts may arife refpecting the formation of the 

 of a mats of the ft a ] s lnere f ee ms no reafon to fuppofe that their gradual 



fimc nature J ' . . 



throughout j increafe would ceafe, till all the molecules belonging to that 



fpecies were exhaufied, if the temperature favourable to their 



generation was continued. If the mafs was entirely conipofed 



of one fpecies of molecules, it would be refolveu into an 



aggregation of cryftals of the fame fubftance ; and probably 



by a ftill farther continuation of the procefs of arrangement, 



into one cryftal, which, though it might not poffefs a regular 



external form, would be perfed in its interna! firucuire. 



of particles dif- But, if the mafs contains two diftinft fpecies of molecule 1 ;, 



fering from each different remits muft take place, which will be modified by the 



other. , / , . . , 



proportional quantities or the components. As it has been 



demonftrated by Berthollet, that the attraction of maffes of 



Generally, the matter are relatively as their quantities, it follows, that unlefs 



niofi: abundant a ver ,, potent counteracting caufe be exerted, the rnoft abuiv- 

 ingredient will J f ° 



crjftallizefirft. dant ingredient in the mixture will be the fir ft to cryfiailize. 

 2 But 



