110 OVERLYING AND TRAP ROCKS. 



ralogist ought to be able to make the distinction, in 

 any rock. Like greenstone, by the predominance of 

 the felspars over the augit, it assumes the aspect of a 

 binary Syenite, to careless mineralogists ; while in mi- 

 nute intermixture, or else, as pure augit in minute 

 condensed crystals, it puts on the aspect of a basalt. 

 I have already said that many of the reputed basalts 

 are of this nature. Augit rock presents an interesting 

 analogy to the produce of volcanoes, in which this 

 mineral is far more often present than hornblende. 



I have equally been obliged to introduce the new 

 term HYPERSTHENE ROCK, for an important member 

 of this family, never before described, or even sus- 

 pected ; especially as possessing geological characters 

 very different from the other associated substances, 

 while consisting of mixtures of Hypersthene with 

 either kind of felspar. It is a rare rock, or else it has 

 ofted been overlooked, or confounded with ordinary 

 greenstones, as it had been in Scotland until the pub- 

 lication of The Western Islands. In Sky and in Aird- 

 namurchan, it forms extensive tracts, and probably 

 abounds in Greenland and Labrador. In general, it 

 presents the external large features of granite ; rising 

 into spiry summits, and exhibiting a durability very 

 rare among the members of this family. It is equally 

 disposed in huge convex beds, separated by actual 

 fissures, or by their indications ; while it also divides 

 into prismatic and cuboidal masses. It is also found 

 in lar^e spheroidal concretions, extending to fifty feet 

 and upwards in diameter ; but I have observed no 

 examples of the prismatic or columnar structure. It 

 occasionally exhibits the same tendency as granite, to 

 decompose in crusts ; and thus presents an occasional 

 schistose structure, which, as in that case, may be only 

 the result of weathering. 



