344 SCIENCE PROGRESS 



of junction of these types of rock run parallel with the general 

 " flow-structure " of the gneiss, and the contrast between the 

 exceptionally micaceous and exceptionally quartzose portions 

 seems to imply some local differentiation in the mass. 



Here, again, elongated blocks of hornblendic rock lie in the 

 gneiss, sometimes like pebbles, sometimes like very flat lenses ; 

 their direction of elongation is that of the prevalent structure 

 in the gneiss. At their margins they fade off into the gneiss 

 by the formation of layers of dark mica, whereby the more 

 felspathic rock around them becomes darker where they are 

 abundant. Epidote, sphene, and garnet are then found in it 

 as minerals of the junction-zones. 



Cutting across the gnarled structure of the gneiss runs a 

 series of dykes and veins of pure granite, consisting mainly of 

 quartz and felspar. These clearly represent cracks that have 

 arisen subsequently to the crush or flow that produced the 

 gneiss. Molten material, capable of crystallising as granite, 

 oozed up into them from below, and the quartz and felspar 

 often separated out so slowly that coarse " graphic granite " 

 has been formed. At other times crystallisation is on a delicate 

 scale, and the material of the veins looks almost like sandstone. 

 The fact that this later granite contains red felspars has been 

 observed also in far-distant localities, but this as yet has no 

 known significance. 



More important, perhaps, than what we have hitherto noticed 

 is the occurrence of a mass of limestone, quite large enough 

 to quarry into, enveloped as it were in the gneiss. The more 

 quartzose and felspathic bands of the gneiss appear as sheets 

 and veins cutting the limestone, and crystalline calcium-silicates 

 have freely arisen in the latter, as they would at the contact 

 with a once molten or " igneous " rock. Lime-garnets, brown 

 and nodular, with attempts at regular crystal-faces, are often con- 

 spicuous on the surfaces of junction. They may, indeed, be as 

 much as five centimetres across, and their size seems a measure 

 of the intensity of interaction, combined with the maintenance of 

 similar conditions throughout a long period of time. 



These features may lead us to regard the gneiss itself as 

 a once molten mass that has become streaky during flow ; 1 



1 Cf. the memorable views of G. P. Scrope, as far back as 1825 (Considerations 

 on Volcanos, p. 226); and of Charles Darwin, Geological Observations on South 

 America (1846), Minerva edition, p. 440. 



