192 MR. A. HARKER ON THE OVERTHRTTST [May I903, 



belt, to be described below ; but local brecciation is not an uncom- 

 mon incident, and it usually shows a certain orderly disposition. 

 In addition to gently inclined crush-bands, which are comparable 

 with surfaces of overth rusting, there are others with vertical posi- 

 tion, which are rather the analogues of normal faults. These may 

 be connected with the great Palaeozoic crust-movements of the 

 region, or they may be of Tertiary date : in the absence of all 

 formations between Torridoniau and Tertiary, the question cannot 

 be brought to a decisive test. These vertical crush-bands are not 

 widely distributed, and are best studied in the Kilmory district. A 

 good example crosses the river about a mile above the deserted 

 hamlet : this is about 15 feet wide. Others on the neighbouring 

 hills attain locally a greater breadth, but they usually die out in a 

 short distance when followed along their length. They are con- 

 spicuous owing to the bleaching of the red sandstone, an effect 

 which has been already remarked as a constant incident of thermal 

 metamorphism, though it is not strictly confined to those circum- 

 stances. Such a brecciated and bleached band usually shows, at 

 least in some part of its length or its width, an injection with 

 igneous material. The invading magma has probably been an 

 ordinary basalt, but it has been considerably modified by absorbing 

 silica, with a certain amount of alkalies, etc., from the sandstone. 

 Thin slices [10486, 10504 1 ] show a very intimate admixture of the 

 two rocks, abundant sand-grains in a partly-corroded state being 

 embedded in a matrix of igneous origin. The bulk of the latter 

 consists of slender felspar-crystals with a strong tendency to radiate 

 arrangement, as in many of the so-called ' variolites.' The extinction- 

 angles are quite low, indicating somewhat acid varieties, and it 

 is probable that alkali-felspars are present. The ferromagnesian 

 element is represented by numerous little pale-green or yellow ser- 

 pentinous pseudomorphs, apparently replacing a rhombic pyroxene. 

 Granules of epidote are also found ; but it is not clear whether 

 these are of motamorphic origin or due to subsequent alterations. 



The amount of igneous material in these injected crush-bands 

 varies greatly, affording every gradation from a basalt-dyke crowded 

 with fragments of sandstone to a brecciated grey sandstone free 

 from basalt ; and such variation may be observed within a short 

 distance along the length of a given band, or even across its breadth. 

 The sandstone, when not impregnated in the manner described, is 

 notably metamorphosed, often showing the quasi-spherulitic structure 

 already noticed in another connection (p. 191). The metamorphism 

 generally seems excessive, in comparison with the amount of igneous 

 material, and there is frequently very marked metamorphism in 

 places where no basaltic or other intrusion is to be detected. Thus, 

 on the slope west and south-west of Loch an Tairbh, about 1| miles 

 north-east of Kilmory, a pale band is traceable through the red 

 pebbly sandstones for nearly 500 yards in a south -south-westerly 



1 These numerals in brackets are the index-numbers of the rock-slices in the 

 Geological Survey Collection. 



