September 19, 1901J 



NA TURE 



511 



of foliation both in the gneiss and dykes are more or less parallel 

 with the planes of movement or disruption. If the latter be 

 vertical or nearly horizontal the inclination of the foliation planes 

 is found to vary accordingly. 



Close to the well-defined disruption-planes, like those between 

 Scourie and Kylesku, the gneiss loses its low angle, is thrown 

 into sharp folds, the axes of which are parallel with the planes 

 of movement. The folia are attenuated, there is a molecular re- 

 arrangement of the minerals, and the resultant rock is a granu- 

 litic gneiss. Indeed, the evidence in the field, which has been 

 confirmed by the microscopic examination of the rocks by Mr. 

 Teall, seems to show that granulitic biotite and hornblende 

 gneisses are characteristic of the zones of secondary shear. A 

 further result of these earth-stresses is the plication of the 

 original gneisses in sharp folds, trending N. W. and S.E. and E. 

 and W. ; and the partial or complete recrystallisation of the 

 rocks along the old planes of mineral banding. 



In like manner, when the basic dykes are obliquely traversed 

 by lines of disruption, they are deflected, attenuated, and within 

 the shear zones appear frequently as phacoidal masses amid the 

 reconstructed gneiss. These phenomena are accompanied by 

 the recrystallisation of the rock and its metamorphosis into horn- 

 blende schist. .Similar results are observable when the lines of 

 movement are parallel with the course of the dykes. All the 

 stages of change from the massive to the schistose rock can be 

 traced — the replacement of pyroxene by hornblende, the con- 

 version of the felspar and the development of granulitic structure 

 with foliation. Here we have an example of the phenomena 

 developed on a larger scale by the post-Cambrian movements, 

 viz., the production of common planes of schistosity in rocks 

 separated by a vast interval of time, quite irrespective of their 

 original relations. For both gneiss and dykes have common 

 planes of foliation, resulting from earth-stresses in pre-Torri- 

 donian time. 



It is important to note also that linear foliation is developed 

 in the basic dykes where there has been differential movement 

 of the constituents in folded areas. In the case of the anticline 

 mapped by Mr. Clough, near Poolewe in Ross-shire, he has 

 shown that the linear foliation is parallel with the pitch of the 

 folds. All these phenomena tend to confirm the conclusions 

 arrived at by Mr. Teall, and published in his well-known paper, 

 " On the Metamorphosis of Dolerite into Hornblende Schist" 

 (Quart, /ourn. Geol. Soi., vol. xli. p. 133). 



The ultrabasic and acid rocks likewise occur in the schistose 

 form, for the peridotites pass into talcose schists and the granite 

 becomes gneissose. 



In connection with the development of schistosity in these 

 later intrusive rocks it is interesting to observe that where the 

 basic dykes merge completely into hornblende schist, and seem 

 to become an integral part of the fundamental complex, biotite 

 gneisses and granular hornblende gneisses prevail. Whatever 

 be the explanation, the relationship is suggestive. 



The unconformability between the Lewisian gneiss and the 

 overlying Torridon Sandstone, which was noted by Macculloch 

 and confirmed by later observers, must represent a vast lapse of 

 time. When tracing this base-line southwards through the 

 counties of Sutherland and Ross, striking evidence was obtained 

 by the Geological Survey of the denudation of that old land 

 surface. In the mountainous region between Loch Maree and 

 Loch Broom it has been carved into a series of deep narrow 

 valleys with mountains rising to a height of 2000 feet. In that 

 region it is possible to trace the orientation of that buried moun- 

 tain chain and the direction of some of the old river courses. 

 This remnant of Archtean topography must be regarded as one 

 of the remarkable features of that interesting region. 



In 1893 the various divisions of the Torridon Sandstone, as 

 developed between Cape Wrath and Skye, were tabulated by 

 the Geological Survey, which may here be briefly summarised. 

 They form three groups : a lower, composed of epidotic grits 

 and conglomerates, dark and grey shales with calcareous bands, 

 red sandstones, and grits ; a middle, consisting of a great suc- 

 cession of false-bedded grits and sandstones ; an upper, com- 

 prising chocolate-coloured sandstones, micaceous flags with dark 

 shales and calcareous bands. The total thickness of this great 

 pile of sedimentary deposits must be upwards of 10,000 feet, 

 and if Mr. Clough's estimate of the development of the lower 

 group in Skye be correct, this amount must be considerably in- 

 creased. Of special interest is the evidence bearing on the 

 stratigraphical variation of the Torridon Sandstone when traced 

 southwards across the counties of Sutherland and Ross. The 



NO. 1664, VOL. 64] 



lower group is not represented in the northern area, but south- 

 wards, in Ross-shire, it appears, and between Loch Maree and 

 .Sleat varies from 500 to several thousand feet in thickness. 

 These divisions of the Torridon Sandstone are of importance in 

 view of the correlation of certain sediments in Islay with the 

 middle and lower Torridonian groups which there rest uncon- 

 formably on a platform of Lewisian gneiss. 



In continuation of the researches of Dr. Hicks, published in 

 his paper "On pre-Cambrian Rocks occurring as Fragments in 

 the Cambrian Conglomerates in Britain" (Geo/. Ma:;., 1S90, 

 p. 516), Mr. Teall has specially investigated the pebbles found 

 in the Torridon Sandstone. The local basement breccias of that 

 formation have doubtless been derived from the platform of 

 Lewisian gneiss on which they rest, but the pebbles found in the 

 coarse arkose tell a different story ("Annual Report of the 

 Geological Survey for 1895," p. 20). He has found that they 

 comprise quartzites showing contact alteration, black and 

 yellow cherts, jaspers with spherulitic structures which indicate 

 that they have been formed by the silification of liparites of the 

 " Lea-rock " type and spherulitic felsites that bear a striking 

 resemblance to those of Uriconian age in Shrop.shire. The.se 

 interesting relics have been derived from formations which do 

 not now occur anywhere in the western part of the counties of 

 Sutherland and Ro.ss, and they furnish impressive testimony of 

 the denudation of the Archaean plateau in pre-Torridonian time. 



The.se Torridonian sediments, like the sandstones of younger 

 date, contain lines of heavy minerals, such as magnetite, ilmen- 

 ite, zircon, and rutile ("Annual Report of the Geological 

 Survey for 1S93," p. 263). The dominant felspar of the arkose 

 group is microcline, that of the basal group oligoclase. In the 

 calcareous sediments of the upper and lower groups fo.ssils 

 might naturally be expected, but the search so far has not been 

 very successful. Certain phosphatic nodules have been found 

 in dark micaceous shales of the upper group which have been 

 examined by Mr. Teall. From their chemical composition these 

 nodules might be regarded as of organic origin ; but he has 

 found that they contain spherical cells with brown-coloured 

 fibres, which appear to be debris of organisms (ibid., 1899, 

 p. 185). 



Early in last century the Torridonian deposits were referred 

 by Macculloch (Trans. Geol. Soc, ser. i, vol. ii. p. 450; 

 " The Western Isles of Scotland," vol. ii. p. 89) and Hay 

 Cunningham ( Transactions of the Highland and Agricultural 

 Society of Scotland, vol. xiii. 1839) to the " Primary Red 

 Sandstone," and by Murchison (Trans. Geol. Soc, ser. 2, 

 vol. iii. p. 155), Sedgwick, and Hugh Miller to the Old Red 

 Sandstone. The structural relations of the Torridon Sandstone 

 to the overlying series of quartzites and limestones were first 

 clearly shown by Prof. Nicol (Quart. Journ. Geol. Soc, vol. 

 xiii. p. 17), who traced the unconformability that separates them 

 for 100 miles across the counties of Sutherland and Ross. 

 When Salter pointed out the Silurian facies of the fossils found 

 in the Durness limestone by Mr. Charles Peach, the Torri- 

 donian formation was correlated with the Cambrian rocks of 

 Wales by Murchison {ibid., vol. xv. p. 353). The discovery of 

 the Olenellus fauna, indicating the lowest division of the 

 Cambrian system, in the quartzite-limestone series by the 

 Geological Survey in 1S91 (ibiJ., vol. xlviii. p. 227) demon- 

 strated the pre-Cambrian age of the Torridon Sandstone. 

 In view of that discovery, which proves the great antiquity o. 

 the Torridonian sediments, it is impossible to climb those 

 picturesque mountains in Assynt or Applecross without being 

 impressed with the unaltered character of these deposits. Yet 

 it can be shown that under the influence of post-Cambrian 

 movements they approach the type of crystalline schists. 



Before proceeding to the consideration of the Durness series 

 of quartzites and limestones and their relations to the Eastern 

 Schists, brief reference must be made to the controversy 

 between Murchison and Nicol regarding the sequence of the 

 strata. 



The detailed mapping of the belt between EriboU and Skye 

 by the Geological Survey has completely confirmed Nicol's con- 

 clusions (i) that the limestone is the highest member of the 

 Durness series; (2) that the so-called " Upper Huartzite " and 

 "Upper Limestone" of Murchison's sections are merely the 

 repetition of the lower quartzite and limestone due to faults or 

 folds ; (3) that there is no conformable sequence from the 

 quartzites and limestones into the overlying schists and gneiss ; 

 (4) that the line of junction is a line of fault indicated by 

 proofs of fracture and contortion of the strata. It is true that 



