i. THE ACID BOSSES OF SKYE

It is in the island of Skye that the granophyre and granite bosses attain their largest dimensions and afford, on the whole, the most complete evidence of their structures and their relations to the other parts of the volcanic series (Map VI.). They cover there a total area of about 25 square miles, and form characteristic groups of hills from 2000 to 2500 feet in height. On the south-east side, three conspicuous cones (the Red Hills) rise from the valley of Strath (Beinn Dearg Mhor, Beinn Dearg Bheag and Beinn na Caillich). A solitary graceful pointed cone (Beinn na Cro) stands between Strathmore and Strathbeg, while to the north-west a continuous chain of connected cones runs from Loch Sligachan up into the heart of the Cuillin Hills. Their conical outlines, their smooth declivities, marked with long diverging lines of screes, and their pale reddish or reddish-yellow hue, that deepens after a shower into glowing orange, mark off these hills from all the surrounding eminences, and form in especial a singular contrast to the black, spiry, and rugged contours of the gabbro heights to the west of them.

Besides this large continuous mass, a number of minor bosses are scattered over the district. Of these the largest forms the ridge of Beinn an Dubhaich, south of Loch Kilchrist. Several minor protrusions lie between that ridge and the flank of Beinn Dearg. Others protrude through the moory ground above Corry; several occur on the side of the Sound of Scalpa, about Strollamus; and one, already referred to, lies at the eastern base of Blath Bheinn. In the neighbouring island of Raasay, a large area of granophyre likewise occurs, which will be described with the Sills in later pages.

In so extensive a district there is room for considerable diversity of composition and texture among the rocks. As already stated, in some places, more particularly in the central parts of the hills, the acid material assumes the character of a granite, being made up of a holocrystalline aggregate of quartz, orthoclase, plagioclase, hornblende and biotite, without granophyric structure, and thus becomes a hornblende-biotite-granite (quartz-syenite, granite-syenite of Zirkel, or amphibole-granitite of Rosenbusch). By the development of the micropegmatitic structure and radiated spherical concretions, it passes into granophyre. By the appearance of a felsitic groundmass, it shades off into different varieties of quartz-porphyry or rhyolite, sometimes with distinct bi-pyramidal crystals of quartz.[387] This change, which here and there is observable along the edge of a boss, is sometimes accompanied with an ample development of spherulitic and flow-structures. As it is convenient to adopt some general term to express the whole series of varieties, I have used the word granophyre for this purpose.

[Footnote 387: The best account yet published of these varieties in Skye is that by Prof. Zirkel, _Zeitsch. Deutsch. Geol. Gesellsch._ xxiii. (1871) p. 88.]

That the large area of these rocks in Skye was the result of many separate protrusions from distinct centres of emission may be inferred, I think, not only from the varieties of petrographical character in the material, but also from the peculiar topography of the ground, and perhaps from the curious relation which seems, in some instances at least, to be traceable between the external features and apparent internal structure of the hills. It will be seen from the Map (No. VI.) that in the area lying to the east of Strath More the granophyre is broken up into nearly detached portions by intervening patches of older rocks. There can be little doubt that the mass of Beinn na Caillich and the two Beinn Deargs is the product of a distinct orifice, if not of more than one. Beinn na Cro, lying between its two deep bounding glens, is another protrusion. The western cones stand so closely together that their screes meet at the bottoms of the intervening valleys. Yet each group is not improbably the result of emission from an independent funnel, like the separate domite puys of Auvergne.

But, though I believe this large area of granitoid rock to have proceeded not from one but from many orifices, I have only here and there obtained, from the individual hills themselves, indications of an internal structure suggestive of distinct and successive protrusions of material from the same vent of discharge. On the outer declivities of some of the cones we may detect a rudely bedded structure, which will be subsequently referred to as well displayed in Rum (p. 403). This structure is specially observable along the east side of Glen Sligachan. Down the northern slopes of Marsco the granophyre (here in part a hornblende-biotite-granite) is disposed in massive sheets or beds that plunge outwards from the centre of the hill at angles of 30° to 40°. On the southern front of the same graceful cone, as well as on the flanks of its neighbour, Ruadh Stac, still plainer indications of a definite arrangement of the mass of the rock in irregular lenticular beds may be noticed. These beds, folding over the axis of the hill, dip steeply down as concentric coats of rock. The external resemblance of the red conical mountains of Skye to the trachyte puys of Auvergne was long ago remarked by J. D. Forbes,[388] and in this internal arrangement of their materials, indefinite though it may be, there is a further resemblance to the onion-like coatings which Von Buch and Scrope remarked in the structure of the interior of the Grand Sarcoui.[389]

[Footnote 388: _Edin. New Phil. Jour._ xl. p. 78.]

[Footnote 389: Von Buch, _Geognostische Beobachtungen auf Reisen durch Deutschland und Italien_, vol. ii. (1809) p. 245; Scrope, _Geology and Extinct Volcanoes of Central France_, 2nd edit. p. 68. Von Buch regarded the external form of this Puy as having been determined by its internal structure.]

Where the contour of the cones is regular, and the declivities are not marked by prominent scars and ribs of rock, this monotony of feature betokens a corresponding uniformity of petrographical character. But where, on the other hand, the slopes are diversified by projecting crags and other varieties of outline, a greater range of texture and composition in the material of the hills is indicated. This relation is well brought out on the western front of Marsco, where numerous alternations of granitoid and felsitic textures occur. On many declivities also, which at a distance look quite smooth, but which are really rough with angular blocks detached from the parent mass underneath, an occasional basalt-dyke will be observed to rise as a prominent dark rib. A good example of this structure is to be seen on the south front of Beinn na Caillich. Where a group of dark parallel dykes runs along the sides of one of these pale cones, it sometimes produces a curiously deceptive appearance of bedding. A conspicuous illustration may be noticed on the southern front of Beinn Dearg Meadhonach, north from Marsco. When I first saw that hillside I could not realize that the parallel bars were actually dykes until I had crossed the valley and climbed the slopes of the hill.[390]

[Footnote 390: The difference of contour and colour between the ordinary reddish smooth-sloped "syenite" and the black craggy "hypersthene rock" and "greenstone" in the Glamaig group of hills caught the eyes of Von Oeynhausen and Von Dechen (Karsten's _Archiv_, i. p. 83).]

Good evidence of successive protrusions of the acid rock within the great area of the Red Hills may be found on the south side of Meall Dearg at the head of Glen Sligachan, where the granophyre is traversed by a younger band or dyke of fine-grained spherulitic material about ten feet broad. The rock exhibits there the same beautiful flow-structure with rows of spherulites as is to be seen along the contact of the main granophyre mass with the gabbro on the same hill, which will be afterwards described. This dyke, vein or band, though possibly belonging to the same epoch of protrusion as the surrounding granophyre, must obviously be later than the consolidation of the rock which it traverses.

Occasionally round the margin of the granophyre a singular brecciated structure is to be seen. I have found it well marked on weathered faces, along the flanks of Glamaig and of Marsco, and Mr. Harker has observed many examples of it on the north side of the granophyre mass of the Red Hills. When the rock is broken open, it is less easy to detect the angular and subangular fragments from the surrounding matrix, which is finely crystalline or felsitic.

The actual junction of the eruptive mass with the surrounding rocks through which it has ascended is generally a nearly vertical boundary, but the granophyre sometimes plunges at a greater or less angle under the rocks that lie against or upon it. On the north side of Glamaig, for instance, the prophyritic and felsitic margin of the great body of eruptive rock descends as a steeply inclined wall, against which the red sandstones and marls at the base of the Secondary formations are sharply tilted. On the south side of the area a similar steep face of fine-grained rock forms the edge of the granophyre of the great southern cones, and plunges down behind Lias limestone and shale, Cambrian limestone and quartzite, or portions of the Tertiary volcanic series. Where the granophyre cuts vertically through the gabbro, the latter rock being more durable is apt to rise above the more decomposable granophyre as a crag or wall, and thus the deceptive appearance arises of the basic overlying the acid rock. As above mentioned, there seems every reason to believe that this peculiarity of weathering has given rise to or confirmed the mistaken impression that the granophyre is older than the gabbro.

There can be no doubt, however, that along many parts of the boundary-line the acid eruptive mass extends underneath the surface far beyond the actual base of the cones, for projecting knobs as well as veins and dykes of it rise up among the surrounding rocks. This is well seen along the northern foot of Beinn na Caillich. But of all the Skye bosses none exhibits its line of junction with the surrounding rocks so well and continuously as Beinn an Dubhaich. This isolated tract of eruptive material lies entirely within the area of the Cambrian limestone, and its actual contact with that rock, and with the basalt-dykes that traverse it, can be examined almost everywhere. The junction is usually vertical or nearly so, sometimes inclining outwards, sometimes inwards. It is notched and wavy, the granite sending out projecting spurs or veins, and retiring into little bays, which are occupied by the limestone. The subdivisions of the latter rock have recently been traced by Mr. Harker up to one side of the granite and recognized again on the other side, with no apparent displacement, as if so much limestone had been punched out to make way for the uprise of the acid boss. The older dykes, too, are continuous on either side of the ridge. The granite is massive and jointed, splitting up into great quadrangular blocks like an ancient granite, and weathering into rounded boulders. Its granitic composition and texture are best seen where the mass is broadest, south of Kilbride. Towards its margin, on the shore of Camas Malag, the granophyric structure appears, especially in narrow ribbons or veins that run through the more granitic parts of the rock. These may be compared with the much larger dyke of spherulitic rock above noticed as traversing the granophyre of Meall Dearg.

Immediately to the south of Camas Malag the junction with the limestone is well displayed, and the eruptive rock, which is there granitic in character, sends out into the limestone a vein or dyke about two feet broad, of closer grain than the main body of the boss, but still distinctly granitic in structure. The junction on the north side is equally well seen below the crofts of Torran. Here the rock of the boss, for a few yards from its margin, assumes a fine-grained felsitic aspect, and under the microscope presents a curious brecciated appearance, suggestive of its having broken up at the margin before final consolidation. Portions of the already crystallized granite seem to be involved in a microgranitic base. The rock has here truncated a number of basalt-dykes which intersect the Cambrian limestone. To one of these further reference will be made in the sequel.

On the surface of the mass of Beinn an Dubhaich, a few little patches of limestone occur to the south of Kilchrist Loch. Considering the nearly vertical wall which the granophyre presents to the adjacent rock all round its margin, we may perhaps reasonably infer that these outliers of limestone are remnants of a once continuous limestone sheet that overlay the eruptive rock, and hence that, with due allowance for considerable denudation, the present surface of the boss represents approximately the upper limit to which the granophyre ascended through the limestone. The actual facts are shown in Fig. 347.

All round the margin of this boss, the limestone has been converted for a variable distance of a few feet or many yards into a granular crystalline marble. The lighter portions of the limestone have become snowy white; but some of the darker carbonaceous beds retain their dark tint. The nodules of chert, abundant in many of the limestones, project from the weathered faces of the marble. The dolomitic portions of the series have likewise undergone alteration into a thoroughly crystalline-granular or saccharoid rock. The most thorough metamorphism is exhibited by portions of the limestone which are completely surrounded by and rest upon the granite. The largest of these overlying patches was many years ago quarried for white marble above the old Manse of Kilchrist. I have shown by lithological, stratigraphical and palæontological evidence that this limestone, instead of belonging to the Lias, as was formerly believed, forms a part of the Cambrian or possibly the very lowest Silurian series, being a continuation of the fossiliferous limestone of western Sutherland and Ross-shire.[391] Mr. Clough and Mr. Harker, in the progress of the Geological Survey in Skye, have ascertained that the distinctive characters of the three groups of strata into which the limestone can be divided may be recognized even through the midst of the metamorphism.[392]

[Footnote 391: _Quart. Journ. Geol. Soc._ vol. xliv. (1888) p. 62.]

[Footnote 392: _Annual Report of Director-General of the Geological Survey for 1895._]

The generally vertical line of separation between the rock of Beinn an Dubhaich and the contiguous limestone has been taken advantage of for the segregation of mineral veins. On the southern boundary at Camas Malag, a greenish flinty layer, from less than an inch to two or three inches in width, consisting of a finely-granular aggregate of some nearly colourless mineral, which polarizes brilliantly, coats the wall of the granophyre, and also both sides of the vein which proceeds from that rock into the limestone. But the most abundant and interesting deposits are metalliferous. Fragments of a kind of "gossan" may be noticed all along the boundary-line of the boss, and among these are pieces of magnetic iron-ore and sulphides of iron and copper. The magnetite may be seen in place immediately to the south of Kilbride. A mass of this ore several feet in diameter sends strings and disseminated particles through the surrounding granophyre, and is partially coated along its joints with green carbonate of copper.

From the Skye area important evidence is obtainable in regard to the relation of the acid eruptions to (1) earlier eruptive vents filled with agglomerate; (2) the bedded basalts of the plateaux; (3) the bosses, sills and dykes of gabbro and dolerite; and (4) the great system of basic dykes.

(1) _Relation of the Granophyre to older Eruptive Vents._--The granophyre of Beinn na Caillich and the two Beinn Deargs has invaded on its north-eastern side the Cambrian limestone and quartzite, and has truncated the sheets of intrusive dolerite and gabbro that have there been injected into them. But to the south-west it rises through the great Strath agglomerate already described, and continues in that rock round to the entrance into Strath Beg. The eruptive mass is in great part surrounded with a ring of agglomerate, as if it had risen up a huge volcanic chimney and solidified there, though probably there were more than one vent in this agglomerate area. Again the thick mass of agglomerate north of Belig is interposed between the bedded lavas and the great granophyre mass which extends northwards to Loch Sligachan. On the west side of the Blaven ridge, a number of masses of agglomerate are found on both sides of Glen Sligachan, along the border of the same great tract of acid rock.

With regard to the relation of the granophyre of the Red Hills to the great agglomerate of Strath, we may infer that the granophyre has not risen exactly in the centre of the old funnel, but rather to the north of it, unless we suppose, as already suggested, that some of the agglomerate belongs to the cone that gathered round the eruptive orifice. It is interesting to observe, however, that granophyre, from the same or from another centre of protrusion, has likewise risen along the outer or southern margin of the agglomerate, generally between that rock and the limestone, but sometimes entirely within the agglomerate. The distance between the nearest part of this ring of eruptive rock and the edge of the boss of Beinn an Dubhaich is under 400 yards, the intervening space being occupied by limestone (or marble), much traversed by north-west basalt-dykes. Most of these dykes do not enter the rocks of the vent, and are abruptly truncated by the mass of Beinn an Dubhaich. The probable structure of this locality is shown in Fig 348.

The masses of agglomerate which further westward so curiously follow the margin of the great granophyre bosses, and those which are entangled in that rock and in the gabbro, probably indicate, as already suggested, the position of a group of older volcanic funnels which provided facilities for the uprise of the basic and acid magmas. The group of vents which, as we have seen, probably rose out of the plateau-basalts, and first served for the rise of the masses of gabbro, has by the subsequent protrusion of the granophyres been still further destroyed and concealed.

The granophyre intrusions in the great Strath agglomerate have lately been mapped and described by Mr. Harker. As regards their internal structure and composition, this observer remarks that compared with the normal granophyres of the Red Hills and other bosses of the district, these smaller intrusive masses are darker and manifestly richer in the iron-bearing minerals, and have a slightly higher specific gravity. But in their general characters they agree with the other granophyres. The most interesting feature in them is the evidence they afford that they have enclosed and partially dissolved fragments of basic rocks. To this evidence further reference will be made on a later page (see p. 392).

(2) _Relation of the Granophyre to the Bedded Basalts of the Plateaux. Metamorphism of the Basalts._--On the north-west side, the granophyre of Glamaig and Glen Sligachan mounts directly out of the bedded basalts. These latter rocks, which rise into characteristic terraced slopes on the north side of Loch Sligachan, appear on the south side immediately to the west of Sconser, and stretch westwards round the roots of Glamaig into the Coire na Sgairde. As they approach that hill they assume the usual dull, indurated, splintery, veined character of their contact metamorphism, and weather with a pale crust. Some of them are highly amygdaloidal, and between their successive beds thin bands of basalt-breccia, also much hardened, occasionally appear. Veins of granophyre become more numerous nearer the main mass of that rock. The actual line of junction runs into the Coire na Sgairde and slants up the Druim na Ruaige, ascending to within a few feet of the top of that ridge. A dark basic rock lies on the granophyre, the latter being here finer grained and greenish in colour, and projecting up into the former.[393] There is so much detritus along the sides and floor of Glen Sligachan that the relations of the two groups of rock cannot be well examined there. But the basalts, which present their ordinary characters to the north of the Inn, are observed to become more and more indurated, close-grained, dull and splintery, as they draw nearer to the granophyre of Marsco. This part of the district furnishes the clearest evidence of the posteriority of the great cones of Glamaig and its neighbours to the plateau-basalts which come up to the very base of these hills.[394]

[Footnote 393: I think it probable that some of the greenish portions of the granophyre along this part of the junction-line will be found to have had their structure and composition altered by having incorporated into their substance a proportion of the bedded basalts through which they have been disrupted.]

[Footnote 394: The dykes of granophyre in these basalts are referred to at p. 444.]

Round the eastern group of cones some interesting fragments of the once continuous sheet of plateau-basalts remain, and show the same relation of the acid protrusions on that side. One of these lies on the granophyre of the flanks of Beinn na Caillich, a little to the west of the loch at the northern base of that hill. Another of larger size forms a prominent knob about three-quarters of a mile further west, and is prolonged into the huge dark excrescence of Creagan Dubha, which rises in such striking contrast to the smooth red declivities of the granophyre cones around it. This prominence at its eastern and northern parts consists of highly indurated splintery basalt in distinct beds, some of which are strongly amygdaloidal. The bedding is nearly vertical, but with an inclination inwards to the hill. Towards the south-west end a thin band of basalt-breccia makes its appearance between two beds of basalt. Its thickness rapidly increases southward until it is the only rock adhering to the granophyre. Beyond the foot of the hill, limestone and quartzite occupy for some distance the bottom of Strath Beg, much invaded by masses of quartz-porphyry. At the summit of Creagan Dubha abundant veins run into the basic rocks from the granophyre, which is here finer grained towards the margin; and there are likewise veins of quartz-porphyry which, though their actual connection with the main mass of granophyre cannot be seen, are no doubt apophyses from it.

This outlier of altered basalt and breccia appears to me to be a fragment of the plateau-basalts which once overlay the Cambrian and Jurassic rocks of Strath Beg, and were disrupted by the uprise of the granophyre. It continues to adhere to the wall of the eruptive mass that broke up and baked its rocks. Its breccia, passing southward into a coarse agglomerate, may be a product of the same vent or group of vents that discharged the great agglomerate mass above Kilbride and Kilchrist. I have already (p. 282) referred to what appears to be another outlier of the basalts on the south side of Beinn Dearg.

On the northern and southern flanks of Beinn na Cro, similar evidence may be observed of the posteriority of the granophyre to the basic rocks. Round the northern base of the hill a continuous tract of plateau-basalts, dolerites and gabbros forms the ridge between Strathmore and Strathbeg. There is an admirable section of the relation of the two groups of rock on the eastern side of the western glen. Along the lower part of the declivity, coarsely-crystalline gabbros, like some of those in the Cuillin Hills, are succeeded by sheets of dolerite and basalt, the whole forming an ascending succession of beds to the summit of the ridge. The edges of these beds are obliquely truncated by the body of granophyre, which slants up the hill across them and sends veins into them. They are further traversed by basalt dykes, which here, as almost everywhere, abound (Fig. 349). On the south side of Beinn na Cro, highly indurated black and grey Lias shales and sandstones have been tilted up steeply and indurated by the eruptive rock of the hill; and at one place some 800 feet above the sea, a little patch of altered basalt, lying on the shale, but close up against the steep declivity of granophyre, forms a conspicuous prominence on the otherwise featureless slope.

Reference has already been made to the mass of fine-grained hornblende-granite which runs for several miles at the base of the volcanic series on the eastern side of the Blaven group of hills. Mr. Harker has traced a great development of granophyre on the west side of these hills, where the acid rock sends apophyses both into the bedded basalts and into the gabbros.

Combining the results of observations made not only in Skye but in Mull, Rum and Ardnamurchan, I shall here give a fuller account of the metamorphism of the basalts, to which frequent allusion has been made as one of the evidences of the posteriority of the eruptive bosses of rock round which it occurs.[395] The field-geologist observes that the basalts, as they are traced towards these bosses, lose their usual external characters. They no longer weather into spheroidal blocks with a rich brown loam, but project in much jointed crags, and their hard rugged surface shows when broken a thin white crust, beneath which the rock appears black or dark bluish-grey, dull and splintery. They are generally veined with minute threads or strings of calcite, epidote and quartz, which form a yellowish-brown network that projects above the rest of the weathered surface. Where they are amygdaloidal, the kernels no longer decay away or drop out, leaving the empty smooth-surfaced cells, but remain as if they graduated into the surrounding rock by an interlacing of their crystalline constituents. They then look at a distance more like spots of decoloration, and even when seen close at hand would hardly at first betray their real nature.

[Footnote 395: Many years ago I was much struck with the evidence of alteration in the igneous rocks of Mull, and referred to it in several papers, _Proc. Roy. Soc. Edin._ (1866-67) vol. vi. p. 73; _Quart. Journ. Geol. Soc._ xxvii. (1871) p. 282, note. The subject was more fully discussed in my memoir in the _Trans. Roy. Soc. Edin._ vol. xxxv. (1888) p. 167, from which the account in the text is taken. Prof. Judd has more recently referred the alteration to solfataric action (_Quart. Journ. Geol. Soc._ xlvi. 1890, p. 341). As already mentioned, I have been unable to detect evidence of such action. The alteration is always intimately connected with the presence of intrusive masses, and it affects indifferently any part of the basalt-plateaux which may chance to lie next to these masses. The bedded lavas can be traced step by step from their usual unaltered condition in the plateaux to their metamorphosed state next to the eruptive rocks. The nature or degree of the metamorphism has doubtless somewhat varied with the composition and structure of the rocks affected, and with the character and mass of the eruptive material; but it is certainly not confined to the older parts of the plateaux, nor to any supposed pre-basaltic group of andesites. I have found no evidence that such a group anywhere preceded the plateau-basalts. The andesites, so far at least as my observations go, were erupted at intervals during the plateau period, and alternate with the true basalts. The greatest accumulation of them lies not below but above the general body of the basalts, in the "pale group" of Mull. Nor even if the term "propylite" be adopted for these altered rocks, can it be applied to any special horizon in the volcanic series. The alteration of the basic rocks by the granophyre of St. Kilda will be described in the account of that island in Chapter xlvii.]

From the specimens collected by me among the Inner Hebrides up to the year 1888, I selected two dozen which seemed to be fairly typical of these altered rocks, and placed thin slices of them for microscopic examination in Dr. Hatch's hands. His notes may be condensed into the following summary. One of the most frequent features in the slides is the tendency in the component minerals to assume granular forms. In one specimen from Loch Spelve, Mull, the rock, probably originally a dolerite, shows only a few isolated recognizable crystals of plagioclase and augite, the whole of the rest of the rock consisting of roundish granules embedded in a felspathic matrix. The felspar crystals are sometimes broken up into a mosaic, though retaining their external contours. Besides the granules, which are no doubt augite, a few grains of magnetite are scattered through the rock, aggregated here and there into little groups. In another specimen, taken from the junction with the granophyre in Glenmore in the same island, parts of the augite crystals are converted into granular aggregates associated with large grains and patches of magnetite. The latter mineral also assumes in some of the rocks granular and even globular shapes suggestive of fusion.

The felspars, which in most of the basic rocks are usually remarkably clear and fresh, show marked kaolinization in some of these altered masses. Minute dusky scales of kaolin are developed, sometimes also with the separation of minute grains of quartz. The augite shows frequent alteration to hornblende, proceeding as usual from the exterior inward. In some cases only an envelope of uralite appears round the augite, while in others only a kernel of the original mineral is left, or the whole crystal has been changed. In many cases the altered substance appears as minute needles, blades and fibres of actinolite. Occasionally, besides the green hornblende, shred-like pieces of a strongly pleochroic brown hornblende make their appearance. Serpentinous and chloritic substances are not infrequent. Epidote is sometimes abundant. The titaniferous iron has commonly passed more or less completely into leucoxene. Here and there a dark mica may be detected.

Since the year 1888 I have continued the investigation of this subject, and have especially studied the metamorphism of the bedded basalts on the western shores of Loch Scavaig, where, as already described, they are truncated by vertical beds of gabbro, and are traversed by basalt-dykes and by abundant veins of fine-grained granophyre. The alteration here effected affords excellent materials for study, as the very same sheets of basalt can be followed from the normal conditions outside to the altered state within the influence of the metamorphic agent. The alternations of amygdaloidal and more compact sheets can still be recognized, although their enclosed amygdales have in places been almost effaced. They show the dull, indurated, splintery character, with the white weathered crust, so distinctive of this type of contact-metamorphism. They are traversed by numerous sills and veins of gabbro. As has been already suggested, although no large mass of granophyre appears here at the surface, the alteration of the basalts is probably to be attributed not so much to the influence of the gabbro, as to the abundant acid sills, dykes and veins, for there may be a considerable body of granophyre underneath the locality, the dykes and veins being indications of its vicinity.

In the summer of 1895 I examined the locality with much care, and collected some typical specimens illustrative of the conditions of metamorphism presented by different varieties of the bedded basalts. Thin slices cut from these specimens were placed in Mr. Harker's hands for microscopical examination, and he furnished me with the following notes regarding them.

"In hand-specimens the bedded basalts from the neighbourhood of the gabbro of Loch Scavaig [6613-6618] do not appear very different from the normal basalts of this region. The most conspicuous secondary mineral is yellowish-green epidote in patches, and especially in the amygdales.

"The texture of the rocks varies, and the slices show that the micro-structure also varies, the augite occurring sometimes in small ophitic plates, sometimes in small rounded granules. The chief secondary change in the body of the rock is shown by the augite, which is seen in various stages of conversion to greenish fibrous hornblende. Some round patches seem also to consist mainly of the latter mineral, and are probably pseudomorphs after olivine. Here the little fibres are confusedly matted together, without the parallelism proper to uralite derived from augite. No fresh olivine has been observed. The felspar and magnetite of the basalts show little or no sign of metamorphic processes, unless a rather unusual degree of clearness in the felspar crystals is to be regarded in that light.

"The contents of the metamorphosed amygdales are not always the same. Epidote is usually present in some abundance, and in well-shaped crystals. It has a pale citron tint in the slices, with marked pleochroism; but a given crystal is not always uniform in its optical characters. Frequently the interior is pale, and has a quite low birefringence. This is probably to be regarded as an intergrowth of zoisite in the epidote, and there are a few distinct crystals of zoisite seen in some places.

"In the slide which best exhibits these features [6613] the crystals of epidote are in part enwrapped and enclosed by what are doubtless zeolitic minerals. At least two of these are to be distinguished. One, very nearly isotropic, and with a pale-brownish tint, is probably analcime. Associated with this is a colourless mineral with partial radiate arrangement and with twin lamellation; the birefringence is somewhat higher than that of quartz, and the γ-axis of optic elasticity makes a small angle with the twin-line. These characters agree with those of epistilbite. In other parts of the same large amygdale, the epidote crystals are embedded in what seems to be a felspar. This latter mineral is rather obscure, and twin-lamellation is rarely to be detected; but it seems highly probable that felspar has here been developed by metamorphic agency at the expense of zeolites which once occupied the amygdale. I have observed undoubted examples of this in metamorphosed basalts from other parts of Skye, _e.g._ from Creagan Dubha, near the granophyre mass of Beinn Dearg.[396] The felspar occurs there in the same fashion, and in the same relation to epidote [2700, 2701]. In the specimens now described the chief minerals in the metamorphosed amygdales are those already named: others occur more sparingly, associated with them. In some cases there is a grass-green, strongly pleochroic, actinolitic hornblende, accompanied by a little iron pyrites [6615].

[Footnote 396: Compare _Trans. Roy. Soc. Edin._ vol. xxxv. p. 166.]

"Epidote and various hornblendic and augitic minerals are characteristic products in the metamorphism of amygdaloidal basalts in other regions: felspar with this mode of occurrence I have not seen except in Skye, where it seems to connect itself naturally with the abundance of zeolites in the amygdales of the non-metamorphosed lavas. It is to be observed that in these basalts from Loch Scavaig the alteration is shown especially in the amygdales, the body of the rock not being greatly affected: this indicates a not very advanced stage of metamorphism. The production of uralitic hornblende, rather than brown mica, from the augite and its decomposition-products, seems to be characteristic of the metamorphism of basaltic as distinguished from andesitic rocks, and is well illustrated by a comparison of the two sets of lavas near the Shap granite."[397]

[Footnote 397: _Quart. Journ. Geol. Soc._ vol. xlix. (1893) p. 361.]

Mr. Harker, who is at present engaged in mapping the central region of Skye, has had occasion to go over a number of the localities (Creagan Dubha, etc.) originally cited by me, and, while corroborating my general conclusions regarding them, has been able to obtain much fresh evidence regarding the nature and extent of the metamorphism which the bedded basalts have undergone. The results of his investigations will be published when the Geological Survey of Skye is further advanced.

(3) _Relation of the Granophyre to the Gabbros._--That the granophyres invade the gabbros has been incidentally illustrated in the foregoing pages. But as the mutual relations of the two rocks in the island of Skye have been the subject of frequent reference in previous writings of geologists, it is desirable to adduce some detailed evidence from a region which has been regarded as the typical one for this feature in the geological structure of the Inner Hebrides. No geological boundary is more easily traced than that between the pale reddish granophyre and the dark gabbro. It can be followed with the eye up a whole mountain side, and can be examined so closely that again and again the observer can walk or climb for some distance with one foot on each rock. That there should ever have been any doubt about the relations of the two eruptive masses is possibly explicable by the very facility with which their junction can be observed. Their contrasts of form and colour make their boundary over crag and ridge so clear that geologists do not seem to have taken the trouble to follow it out in detail. And as the pale rock undoubtedly often underlies the dark, they have assumed this infraposition to mark its earlier appearance.

I will only cite one part of the junction line, which is easily accessible, for it lies in Glen Sligachan immediately to the south of the mouth of Harta Corry. The rounded eminence of Meall Dearg, which rises to the south of the two Black Lochs, belongs to the granophyre, while the rugged ground to the west of it lies in the gabbro. The actual contact between the two rocks can be followed from the side of Harta Corry over the ridge and down into Strath na Creitheach, whence it sweeps northward between the red cone of Ruadh Stac and the black rugged declivities of Garbh Beinn. There is no more singular scene in Skye than the lonely tract on the south side of Meall Dearg. The ground for some way is nearly level, and strewn with red shingle from the decomposing granophyre underneath. It reminds one of some parts of the desert "Bad lands" of Western America. Grim dark crags of gabbro, with veins from the granophyre, rise along its western border, beyond which tower the black precipices of the Cuillins, while the flaming reddish-yellow cones of Glen Sligachan stand out against the northern sky.

Having recently described in some detail the relations of the boss of granophyre at this interesting locality, I will only here offer a brief summary of the chief features.[398] The granophyre of Meall Dearg forms a marginal portion of the great mass of the Red Hills. It has broken across the banded gabbros, and also cuts an isolated boss of agglomerate in the ridge of Druim an Eidhne. Its line of junction is nearly vertical, but along part of its course the wall of gabbro rises higher than that of the more decomposable granophyre. Hence the origin of the black crags that crown the red slopes of granophyre debris. Seen from a distance the basic rock seems to rest as a great bed upon the acid mass.

[Footnote 398: See _Quart. Journ. Geol. Soc._ vol. 1. (1894) p. 212.]

The younger date and intrusive nature of the granophyre are well shown by the change in the texture of the mass as it approaches the rocks against which it has cooled. The ordinary granophyric characters rapidly pass into a fine-grained felsitic texture, and this change is accompanied with the development of a remarkably well-defined flow-structure and of rows of spherulites which run parallel to the boundary wall. In a ravine on the west side of Meall Dearg, the lines of flow-structure and rows of large spherulites are seen to be arranged vertically against the face of gabbro.

Further proof of the later date of the protrusion of the granophyre is supplied by abundant felsitic dykes and veins which traverse the gabbro, and some of which can be seen to proceed from the main body of granophyre. These intrusions will be described in the next chapter, in connection with the dykes and veins of the acid rocks.

Additional evidence as to the posteriority of the granophyre to the gabbro has recently been obtained by Mr. Harker from a study of the internal structure and composition of the masses of these rocks which have been intruded into the agglomerate above Loch Kilchrist in Strath. He has found that the granophyre has there caught up from some subterranean depth portions of gabbro, and has partially dissolved them, thereby undergoing a modification of its own composition. "The gabbro-debris," he remarks, "has been for the most part completely disintegrated by the caustic or solvent action of the acid magma on some of its minerals. Those constituents which resisted such action have been set free and now figure as xenocrysts [foreign crystals], either intact or more or less perfectly transformed into other substances. At the same time the material absorbed has modified the composition of the magma, in the general sense of rendering it less acid." Mr. Harker has traced the fate of each of the minerals of the gabbro in the process of solution and isolation in the acid magma, which, where this process has been most developed, is believed by him to have taken up foreign material amounting to fully one-fourth of its own bulk, derived not from the rocks immediately around, but from a gabbro probably at a considerable depth beneath.[399]

[Footnote 399: _Quart. Journ. Geol. Soc._ vol. lii. (1896) p. 320.]

(4) _Relation of the Granophyre to the Basic Dykes and Veins._--Reference has already been made to the fact that the "syenite" bosses of Skye cut off most of the basalt-dykes, but are themselves traversed by a few others.[400] The locality that furnished me with the evidence on which this statement was originally made nearly forty years ago affords in small compass a clearer presentation of the facts than I have elsewhere met with. The sections described by me are visible at the eastern end of the boss of Beinn an Dubhaich, Strath; but similar and even better examples may be cited from the whole northern and southern margins of that eruptive mass. On the north side an extraordinary number of dykes may be traced in the Cambrian limestone from the shores of Loch Slapin eastwards. They have a general north-westerly trend, but one after another, as I have already remarked, is abruptly cut off by the granophyre. As an example of the way in which this truncation takes place, I may site a single illustration from the northern margin of the eruptive mass, near Torrin. It might perhaps be contended that the numerous dykes which traverse the limestone and stop short at the edge of the acid rock, are not necessarily older than the granophyre, but may actually be younger, their sudden termination at the edge of the acid boss being due to their inability to traverse that rock. That this explanation is untenable is readily proved by such sections as that given in Fig. 350, where a basic dyke (_b_) 9 or 10 feet broad running through the Cambrian Limestone (_a_ _a_) is abruptly cut off by the edge of the great granophyre boss. Not only is the dyke sharply truncated, but numerous pieces of it, from an inch to more than a foot in length, are enclosed in the granophyre. The latter is well exposed along the shore of Loch Slapin in an almost continuous section of nearly a mile in length. The contrast therefore between the development of dykes within and beyond its area cannot but arrest the attention of the observer. Though I was on the outlook for dykes in the granophyre, I found only one. Yet immediately beyond the eruptive boss they at once appear on either side up to its very edge, where they suddenly cease. The conclusion cannot be resisted that the protrusion of the acid rock took place after most of the dykes of the district had been formed, but before the emission of the very latest dykes, which pursue a north-west course across the boss (Fig. 348).

[Footnote 400: _Ante_, p. 173, and _Quart. Journ. Geol. Soc._ vol. xiv. (1857) p. 16.]

Some sections on the southern margin of Beinn an Dubhaich complete the demonstration that such has been the order of appearance of the rocks. Near the head of the Allt Lèth Slighe (or Half-way Burn), where the granite has pushed a long tongue into the limestone, a north-west basalt-dyke is abruptly cut off by the main body of the boss and by the protruded vein (Fig. 351). Besides this truncation, the acid rock sends out strings and threads of its own substance into and across the dyke, these injected portions being as usual of an exceedingly fine felsitic texture.

Similar evidence may be gathered from the area of the great granophyre cones further north. The profusion of basalt-dykes in the surrounding rocks stops short at the margin of that area. The comparatively few dykes which cross the boundary pursue a general north-west course through the granophyre, and, as already remarked, from their dark colour, greater durability and straightness of direction, stand out as prominent ribs on the flanks of the pale cones which they traverse.