[ 6r ] 



He mufl; know that fuperfaturation, a flight contamination 

 with certain heterogeneous fubftances, a variation in the 

 temperature, a variable degree of agitation, a difference in the 

 quantity of the menflruum, or in the time of its difparitlon, 

 befides many other unknown circumflances, daily produce 

 different efFecfts in the cryl^allization of falts, the bodies in 

 which this operation has been moft; attended to. 



" But water (continues he) being the general medium in which 

 " bodies colledted at the bottom of the fea are always con- 

 " tained, if thofe mafTes of colledled matter are to be confolidated 

 " by folution, it mufl be by the difTolution of thofe bodies 

 " in that water as a menflruum, and by the concretion or 

 " cryflallization of that difTolved matter, that the fpaces firft 

 " occupied by water in thofe maffes are afterwards to be filled 

 " with a hard and folid fubflance ; but without fome other power 

 " by which the water contained in thofe cavities and endlefs 

 " labyrinths of the flrata fhould be feparated in proportion 

 " as it has performed its tafk, it is inconceivable how thofe 

 " mafTes, however changed from the flate of their firfl fub- 

 " fidence, fhould be abfolutely confolidated, without a particle 

 '■ of fluid water in their compofition." Abflradling from his 

 own gratuitous hypothefis, it is very eafy to fatisfy our author 

 on this head ; the concreting and confolidating power in mofl 

 cafes arifes from the mutual attradlion of the component particles 

 of flones to each other ; if thefe particles leave any interflices, 

 thcfe are filled with water which no way obf.rucls their folidity 

 v>'hcn the points of contadl arc numerous ; hence the decrepitation 



of 



