COMPARISON OF ERUPT! VP: WITH METAMORPHIC ROCKS. 117 



accepted theory. But the relations between the metamorphic and eiuptive 

 rocks constitute a pending question. 



It will be unnecessary here to enter very minutely into a discussion 

 of these relations, and, indeed, a full discussion would require a very long 

 and copious review of the existing state of lithological science. It will be 

 sufficient to state in a summary manner those points of comparison which 

 immediately concern the subject in hand. The conclusion to which this 

 comparison tends is thai; a large proportion of the igneous rocks have the 

 petrogi'aphic characters which we ought to expect would result from the 

 fusion of certain groups of metamorphic stratified rocks. There are three 

 points of view from which the comparison may be made ; these are with 

 reference, first, to chemical constitution ; second, to mineral components ; 

 third, to mechanical texture. 



1st. Metamorphic and igneous rocks compared with respect to chemical 

 constitution. — The eruptive rocks are highly complex compounds, and always 

 contain certain constituents which may be called essential constituents. 

 These are silica, alumina, lime, soda, potash, and magnesia — six in number. 

 Iron in the form of some oxide is almost always present, but since it is 

 occasionally absent, or found in exceedingly small quantity, it cannot be 

 regarded as a universal and essential constituent. Silica is always the 

 dominant ingredient, and though the quantity of it varies greatly, yet the 

 variation is within tolerably definite limits, almost never exceeding 80 per 

 cent., and almost never falling below 45 per cent. The remaining five 

 constituents likewise vary, but always within tolerably naiTOw limits. 

 Thus alumina rarely falls below 13 per cent, and rarely exceeds 26 per 

 cent. Lime rarely exceeds 14 per cent, magnesia 10 per cent, soda 9 per 

 cent., and potash 8 per cent. The variations in the relative proportions of 

 these constituents is sufficiently wide to give well-marked specific or even 

 generic diff'erences in the kinds of volcanic products ; but the variations 

 are so limited and the relative proportions subject to such moderate depart- 

 ures from normal ratios, that the whole category of eruptive rocks possess 

 at least ordinal if not family likenesses. Turning now to the metamoi-phics 

 we find a far wider range of chemical constitution. Thus we have quartzites 

 which are almost pure silica ; we have crystalline limestones and dolomites 



