Frederiofc Guthrie on Eutcafa* 17!) 



in the proportion found In § 220i namely 82*9 of nitrate of 

 .soiliiini and 67*] of nitrate or potassium, the specific gravity 

 of the alloy would be 2*149, The density Pound shows that 

 there basbeenan expansion <>n the two associating themselves 

 together to form the alloy. A.ndthis expansion is presumably 

 in near connexion with the lowering of the melting-point, 



§ ^!>. The generalization of Berthollet Is only one, aspect 

 of the question of distribution. When a compound l>o<ly ato* 

 and another compound body by being brought together in 

 watery solution give rise to an insoluble body ay and a 

 soluble one ba, we have no more right <<> say that the change 

 is brought about in virtue of the insolubility of ay than that it 

 is due to the solubility ofba, Strictly speaking, it is due i<> 

 the difference in solubility <>f the two. For the solubility <>r 

 Hir one is .is much concornod with the separation of the two 

 now compounds as in the insolubility of the other, [t is on 

 this aocount that eutexia, without water, and as it exists in the 

 case of two fused metals <>«■ anhydrous salts, may and must 

 have •■! determining influence <>«> the arrangement in ratio or 

 kind of the constituents, 



§ -I'M). Geological and Mineralogical Significance ofEutetcia* 



\ low i lie temperature of liquefaction <>r rocks may !"■ lowered 

 by the introduction of even small quantities <>i" water will be 

 discussed in my next memoir, Lt has been trery diligently 

 considered l>y Daubree, Let us briefly examine the question 

 of the interaction of rocks and their constituents, without, <>r 

 oecei sity, the intervene i<>n of water. 



In reviewing the analyses of ETelspatbio minerals, most of 

 which, unless weathered after solidification, appear to be 

 ei entially anhydrous, we find them to be silicates or potassium 

 mid aluminum, in which, f<> use the common expression, the 

 potassium is partly replaced by sodium, calcium, magnesium, 

 and manganese, the aluminum by iron. That water is not 

 essential to the formation of this kind of mineral, is proved 

 bj the occurrence of it in copper and iron furnaces, Mica, on 

 the other band, appears always '<> contain water, sometimes 

 at much as three or four per cent#,with s lessor percentage of 

 silica than felspar; it contains generally a far larger pro- 

 portion of aluminum and of iron ; the abundance of the latter 

 element m.-iy indeed be considered, together with its defi- 

 oienoy in silica, as its distinguishing characteristic, chemically 



Speukin^, from orUioola.sn. 



Bilioa ili'- great earth^cid, «■« title which i< shares with 

 earbonic :i<i<l, whoso oilier realm is the air, and i<> some 

 ( .ten! will) bydrochleric acid, whose proper region is thi 

 - appears both in the anhydrous form - * 1 * * I wiih as mUofa as 



" a K 2 



