§8 



rnjir have risen in a slate of gas or vapour through the joints. But, without resorting to 

 Ihip liy|HJ[]ic*L* , Let uf guppose that, from tlio first , all the ingrcdicuti existed together 

 in Ihc naa«. The external portion of the cube clc. differ* more in mineralogical than in 

 tLi;jiiL..-.il din racier from the inlerfial. The tchurl nf fhe- former, compared wilh the fcltpir 

 of the latter, lm a great exec** of alumina ami u great dclicumoy qf silica, But this dif- 

 ference is compensated^ and Ike balance af ingredients restored, by the rjuarLx which nr> 

 eompantc* the whorl. IT we therefore reduce thu iulcrnal and nlcrnal portions lo their 

 constilucnta, we shall fii^-L ihat the essential difference is only about 1U per cent, and that 

 U consist* in Hie Inlter having about 9 pet cent vf oiidc of iron iuAicad of only 1 per 

 ceul, and in ha^anr; about 2 pi:r eiut of □ new ingredient, — boracic ticid. We cam hardly 

 err in attributing Che difference mainly to (he chemical aelion of the acid, which, whether 

 by iltjclf of ju combination wilh soda, is remarkable for it* fusibiLity , and its power of com- 

 municating thin property lo conipuundti. The tendency of certain ^uhituuecs in a £u$ed mas* 

 to relifc toward* the surface when other fluhs(ancc& arc there found for which I hey hare a 

 stronger nihility than tin: Dtln-r const i I ls>tH> of Ihc ma** i* well Lltuwn, But tlirrc is Ho 

 diihcully in conceiving lhal -ui li a I ranker may have been uininly mrchaliical. The boracjc acid 

 mny have retained a portion of the mailer with which it wa* En combination in a fluid 

 or viscid state, for some time subsequent lo Ihe cry^CLilliin I ion and partial solidification of 

 the frUpnr etc, and the internal pressure of the Rcmi -solidified ma*s alone may have 

 forced ibis Howards iho side* *ad cavued it to rj*c to the surface, ll is n .n-r Jai end thai 

 both feitpar nnd quart* remain in a viscid stale at tcmparainrc-s grcoliy infctluur to that 

 ai which they jirr fused, and hence there must hate been a decree el internal pressure Bub- 

 tequcnl lo crystallization^ The fugitive character of schorl is more than once poiijlrd out by Sir 

 IL P¥ ik Use in; in hi* He port. Amongst other instances of Alterations produced on sedimen- 

 tary rocki* by contact with granite be mem lions some slate* in which schurl has been Sntro- 

 dueeil between the laminae. A mnrt remarkable case occurs in a granite conMSlinrj of large 

 felspar cmlals in a base of schorl aod qumrU- At ionic plaeej fhc fukpar cry.ntali hafc 

 been decompnfied and replaced by crystals of schorlcro^infj each other in various direct ionj, 

 and the schorl in Ihc ^urrcunding bise ii evidently ddlcient. 



The aluJtidanec of quart t at the surface may he ■ I m- in roim? measure to the eireum stance 

 that ibe boracie acid , wheiher ei^ielled from tbc iDleriour on the cryiUallixaliou of the 

 fdpjwr and mica or derived from w j tljont through The joints , would, win' fever iu uljjmale 

 ]ocahCy was , be hogtile tn the Jo rtual ion of felspar and mica there. 



However wq may aeeouril for it , ihc fad of the tehorl laLmg; the place of fehpar and 

 mica in Ihc Arrenu/rm and Cormtatf fjranilc*^ appear i In- jui,ihii;..u:- to lljnt of horn- 

 blende rr.ptaeinre mica in ihr rjraniles and syeuilCH of Puio Ulin t and if we titeud th* 

 prcceedinjj ifKCulatiooj regarding the one transition lo Ihe olhcr, wo shall find ibem in har- 

 mony with the view* formerly expressed. We must in the first place eouflider ihe Island 

 ilwif as ihe mnrmit oF on* granilic bubble, of which much oT The cxlernal portions have 

 dittppcarcd A porlion of the bubble (iis mpcrjaecni rocksj whatftw ihcy were , havinrj 

 been *wt'pl away) \& now elevai^d above the sea r to as lo cipu^: a belt below high water 

 mark to lite action uf the wave* whieh arc working into the nucleus, and t in iheir pr** 



