390 THE CANADIAN NATURALIST. Dec] 



granitoid rocks just noticed.^ In all attempts to define and 

 classify compound rocks, it should be borne in mind that they are 

 not definite lithological species, but admixtures of two or more 

 mineralogical species, and can only be arbitrarily defined and 

 limited. 



§ 3. Having thus defined the mineral composition of granitic 

 rocks, we proceed to notice their structure. Gneiss has the same 

 mineral elements as granite, but is distinguished by the more or 

 less stratified and parallel arrangement of its constituents, and 

 lithologists are aware that in certain varieties of gneiss, this 

 structure is scarcely evident, except on a large scale, so that the 

 distinction between gneiss and granite rests rather on geognosti- 

 cal than on lithological grounds. To the lithologist, in fact, the 

 granitoid gneisses are simply more or less stratiform granites, while 

 it belongs to the geologist to consider whether this structure has 

 resulted from a sedimentary deposition, or from the flowing of a 

 semi-fluid heterogeneous mass giving rise to a stratiform 

 arrangement. 



§ 4. The rocks having the mineralogical composition of 

 granites present a gradual passage from the coarse structure of 

 ordinary micaceous hornblendic and binary granites to finely 

 granular and even impalpable mixtures of the constituent minerals, 

 constituting the rocks known as felsite, eurite and petrosilex. 

 These rocks are often porphyritic from the presence of crystals of 

 orthoclase, and sometimes of crystals or grains of quartz imbedded 

 in the finely granular or impalpable paste. These felsites and 

 felsite-porphyries are, in very many cases at least, stratified or 

 indigenous rocks, and they are sometimes found associated with 

 o-ranular aa'^resates of diff'erent dciirees of coarseness, which show 

 a transition from true felsites into granitic gneisses. The resem- 

 blances in ultimate composition between felsites, granites and 

 granitic gneisses are so close that it cannot be doubted that their 

 difl'erences are only structural. 



§ 5. Felsites and felsite-porphyries are well known in eastern 

 Massachusetts, at Lynn, Saugus, Marblehead and Newburyport, 

 and may be traced from Machias and Eastport in Maine, along 

 the southern coast of New Brunswick to the head of the Bay of 

 Fuudy, with great uniformity of type, though in every place subject 



* Amer. Jom-nal of Science, II, sxxviii, 95, See also Zirkel^ Fetro- 

 grci])hie, ii, HO. 



