416 PKOF. T. G. BONNET ON A SEEIES OF EOCKS 



and some of the hornblende appears secondary. If the rock be meta- 

 morphic in the ordinary sense, it belongs to this group ; but I have 

 great doubts whether it is not a true igneous rock. 



77 (Valley south of Ben Arnaboll, p. 396), hornblendic gneiss with 

 some microcline ; 78 (ibidem, p. 396), rather crushed, rather rich in 

 microcline, containing little besides felspar and quartz. 



79 (West of Cnoc Chaorinie, p. 388) rather rich in white mica, no 

 hornblende or black mica. 



80 (Craig Dhu, pp. 374, 378). Composition resembling 78, much 

 crushed, with a little viridite infiltrated in the cracks ; greatly resem- 

 bles the so-called igneous rock near the fault in Glen Laggan described . 

 in my paper (Quart. Journ. Geol. Soc. vol. xxxvi. p. 93), and, like it, is 

 certainly one of the granitoid gneisses of the Hebridean series. 



81 (Burn of Calda, p. 378) contains, with some hornblende, a 

 mineral resembling that described in 62, but more characteristically 

 a diallage ; part of it has been converted into uralite. Even diallage, 

 so far as my experience goes, is rare in gneisses ; though, looking at 

 the mode of its occurrence in igneous rocks, one would not be sur- 

 prised to find it. Some grains of structure generally similar are 

 replaced by an earthy -looking rather serpentinous mineral. If this 



' rock be metamorphic it is certainly Hebridean ; but I feel doubts 

 whether it may not be an igneous rock which has been deprived 

 by local crushing of its most characteristic structures. 



Rocks probably crushed in situ. 



Several of the above rocks exhibit some indications of crushing 

 and recementation ; but in the group to which I now proceed this 

 is very conspicuous, and sometimes has so much obscured the ori- 

 ginal structure that it is difficult to pronounce upon the true cha- 

 racter of the rocks. In most cases, however, I feel tolerably certain 

 that they belong to the Hebridean series. The cementing agent is 

 chalcedonic quartz, but other minerals, as viridite, sericite (?), and 

 iron-oxides are associated with it. The felspar is frequently greatly 

 decomposed and replaced by various earthy-looking granular minerals, 

 some of which may possibly be related to staurolite or andalusite. 

 The percolation of water through the pulverized mass, while it 

 favoured the deposition of quartz &c, would be likely to remove the 

 alkaline constituents of the felspars. The deposition of micaceous 

 or chloritic minerals on the surfaces of minute fracture under 

 continuous pressure often gives the rock the macroscopic aspect of a 

 dull-coloured normal mica-schist; but when it is examined microscopi- 

 cally, we see how superficial is this resemblance, and the true alliance 

 of the rock with the ordinary gneisses is revealed. To this group 

 belong 63 (2 miles S.S.E. Ullapool, pp. 364, 378), 73 (Dhuloch Beg 

 Burn, 378, 381) ; 74 (2 miles S.S.E. Ullapool, pp. 364, 378) contains 

 some calcite and dolomite (?). A secondary mineral has crystallized 

 with it, which appears to be felspar, and the same borders some of 

 the broken and decomposed original grains of that mineral. 



83 (Balloch, Ben More), 84 (Burn of Calda), 85 (ibidem), 86 

 (ibidem) (p. 378). Though parts of the slides bear a most remarkable 



