Again, in the largest of the vents, that near Scarlet Point, still clearer proof of successive eruptions and dislocations within a volcanic chimney may be noticed. At one point the accompanying section (Fig. 188) has been laid bare by the waves. The oldest accumulation is a fine green granular tuff (a), rudely and faintly arranged in layers inclined at high angles, like the fine materials in many of the vents of the basin of the Firth of Forth. This peculiar stratification, due not to the assortment of materials in water, but to the deposition of coarser and finer detritus by successive explosions, and to subsequent slipping or tilting, is a characteristic feature of the detritus which has filled up ancient volcanic funnels. A later explosion from some adjacent part of the same vent has given rise to the discharge of a coarse agglomerate (b), which with blocks sometimes six feet long, overspreads the earlier material. A third detrital accumulation in the same vent, consisting of a firm brecciated tuff (c) with much calcite in its matrix, has been brought down by a slip (f) which cuts across both of the previous deposits. A broad dyke (d) of vesicular diabase (augite-porphyry) traverses the vent, and is probably later than any of the other rocks in the section.

I will conclude this account of the Manx Carboniferous volcanic rocks with a brief reference to the intrusive masses which form a prominent feature of the coast-line. From the picturesque headland of Scarlet Point the broad dyke which forms that promontory may be traced for some distance westwards. Several other parallel dykes run in the same direction which, it will be observed, is also that of the chain of vents. It might be said that the vents are, as it were, strung together by a line of dykes. These eruptive masses traverse both the agglomerates and the bedded tuffs. They probably belong, therefore, to a comparatively late part of the volcanic history. That they are truly intrusive and not lava-flows is, I think, clearly shown by their vertical walls which descend through the surrounding rocks, and by the greater closeness of their texture, as well as the diminution in the size of their vesicles along the contact surfaces. But it must be admitted that in their remarkably developed vesicular structure they look more like streams of lava than ordinary dykes.

It is this structure which gives to these dykes their peculiar interest. Bands of vesicles, from an inch or less to several inches in breadth, run along the dykes parallel to the outer walls. Unlike the familiar rows of little amygdaloidal cells in ordinary basalt dykes, such as those of the Tertiary series in Scotland, these vesicles, though small and pea-like in the narrower bands towards the margins of the dykes, became so large, numerous, and irregular in the broader and more central bands, that the rock passes there into a rough slag.

While the intrusive material has for the most part risen in the form of dykes, in one part of the coast-section, a little to the west of Scarlet Point, it has been injected as a sill among the bedded tuffs.[58] A section taken at this locality gives the structure represented in Fig. 189. On the north side of the great dyke, the strata of tuff which dip under it, roll over and support an outlying sheet of the same material. The slaggy structure of parts of this sill give it some resemblance to a true lava-flow. But it is the same structure which can be seen in the dykes, while the closer grain along the contact-surface further connects it with these intrusions.

There is, however, a peculiarity about the development of the vesicular structure in this sill which I have not observed anywhere else. If we examine the southern side of the crag near its eastern end we observe that the successive bands of vesicles are arranged in the same direction as the surface of contact with the underlying tuffs, precisely as they are ranged in dykes parallel to the bounding walls. So far the structure is quite normal. But, moving a few yards westwards, we find that the bands begin to curve, and, instead of following the contact surface, strike it first obliquely and then at right angles, until we have the structure shown in Fig. 191. The bands here vary from less than an inch to more than a foot in breadth, and where broadest assume a slaggy texture. I sought in vain for any evidence of subsequent disturbance such as might have truncated these parallel rows of vesicles and pushed the rock bodily over the tuffs. The perfect parallelism of the bands with the surface of the tuff at the east end, and the absence of all trace of a thrust-plane at the base of the sill, seem to show that, though the rows of vesicles were undoubtedly at first arranged parallel to the surfaces between which the intrusion took place, the mass, before completely consolidating and coming to rest, was ruptured, and a portion of it was driven onwards at right angles to its previous line of movement.

A consideration of the singularly slag-like structure of the injected masses in the tuffs and agglomerates leads to the conclusion that though what we now see of these rocks did not actually flow out at the sea-bottom in streams of lava, it was intruded so close to the surface that the imprisoned vapours had opportunity to expand, as in superficial outflows.[59] This inference is in accord with that derived from an examination of the necks, wherein we find evidence of the probable survival of parts of the actual craters and volcanic cones.

As the records of the earliest eruptions during the Carboniferous Limestone period in the district of the Isle of Man are concealed, so also those of the last of the series lie under the sea. Where the highest visible tuffs overlie the Poyll Vaaish limestones they show no change in the nature of the materials ejected, or in the energy of eruption. They lie so abruptly on the dark calcareous deposits as to show that a considerable pause in volcanic activity was followed by a violent explosion. The same abundant grey-green pumice, the same kind of loose crystals of felspar, the same type of lava-blocks and bombs as had characterized the foregoing eruptions remained as marked at the end. But the further volcanic records cannot be perused, and we are left to speculate whether the coast-sections reveal almost the whole chronicle, or if they merely lay before us the early chapters of a great volcanic history of which the main records lie buried under the waves of the Irish Sea.

4. EAST SOMERSET

Various limited outcrops of igneous rocks have long been known to occur in the eastern part of Somerset. The largest of these lies in the midst of the Old Red Sandstone, on the crest of the axis of the Mendip Hills, between Downhead and Beacon Hill. Smaller patches occur in the Carboniferous Limestone near Wrington Warren, on the north side of Middle Hope, on Worle Hill and at Uphill. These rocks have been mapped as intrusive, though some of them have been described as conglomeratic or as volcanic breccias. While some of the masses are probably intrusive, others appear to be truly contemporaneous with the deposition of the Carboniferous Limestone. The highly vesicular basalt of Middle Hope looks much more like a superficial lava than an intrusion. Mr. Aveline gave a section showing three alternations of limestone and "igneous rock" at Middle Hope. A recent examination of that coast-line by Mr. A. Strahan shows that there are undoubted tuffs interstratified with the calcareous strata. There is thus proof that one or more small volcanic vents were in eruption on the floor of the Carboniferous Limestone sea in the neighbourhood of Weston-super-Mare.[60]

5. DEVONSHIRE

The change from the typical Old Red Sandstone of South Wales to the Devonian system of Devonshire, to which I have already referred, is hardly more striking than the contrast between the Carboniferous formations of these two areas.[61] The well-marked threefold subdivisions of Carboniferous Limestone, Millstone Grit and Coal-measures, so persistent throughout Britain, and nowhere more typically developed than in South Wales, are replaced in a distance of less than forty miles by the peculiar "Culm-measures" of Devonshire—a series of black shales, grey sandstones and thin limestones and lenticular seams of impure coal (culm), which are not only singularly unlike in original characters to the ordinary Carboniferous formations, but have been made still more unlike by the extensive and severe cleavage to which the Palæozoic rocks of Devon and Cornwall have been subjected. That these Culm-measures are truly Carboniferous is made abundantly clear by their fossil contents, though it has not yet been possible to determine how far they include representatives of the great stratigraphical subdivisions in other parts of the country.

It is to De la Beche that geology owes the first intimation of the occurrence of interstratified igneous rocks in the Carboniferous series of Devonshire. As far back as the year 1834, in his singularly suggestive treatise, Researches in Theoretical Geology, this eminent geologist expressed his opinion that not only were the "trappean" bands regularly intercalated in the sedimentary series and continuously traceable with the general stratification, but that they occurred at various localities in such a manner as to raise the suspicion that these points may mark some of the centres of eruption. He particularly cited the example of Brent Tor as a remarkable volcanic-looking hill, composed in part of a conglomerate "having every appearance of volcanic cinders."[62]

In his subsequently published Report on the Geology of Cornwall, Devonshire and West Somerset, De la Beche dwelt in more detail on the results of his study of these rocks, which he had traced out on the ground and expressed upon the maps of the Ordnance Geological Survey.[63] Hardly any additions have since been made to our knowledge of the field-relations of the rocks. It is to the maps and Report of De la Beche that we must turn for nearly all the published information on the subject. I shall therefore give here a summary of what can be gathered from these publications.

In tracing the limits of the Culm-measures, De la Beche found that no well-defined line could be drawn between these strata and the "grauwacke" or Devonian formations underneath. The Carboniferous series lies in a great trough, of which the axis runs nearly east and west, so that the lowest members of the series rise along the northern and southern margins. But De la Beche was struck with one remarkable contrast between the two opposite sides of the trough—a contrast which marks the Devonian as well as the Carboniferous formations of this region. On the south side an abundant and persistent group of intercalated bands of igneous, or as he called them, "trappean," materials can be followed along the whole line of boundary, while no such group occurs on the north side. He found these bands to be lenticular, traceable sometimes for a number of miles, then dying out and reappearing on the same or other horizons. He mapped them the whole way from Boscastle on the west to near Exeter on the east, and found that though the individual sheets might be short, the trappean zone was continuous as far as the southern margin of the Carboniferous series could be seen, except where it had been broken through by the great granitic mass of Dartmoor. He ascertained that the intercalated trappean rocks are not confined to the Culm-measures, but occur also in the contiguous portions of the "grauwacke" or Devonian system.

But further, he clearly recognized that the bands of igneous material which he mapped included both "greenstones," together with other varieties of massive eruptive rocks, and also volcanic ash or tuff, though he did not attempt to separate these out upon the maps, but contented himself with representing them all under the same colour. He admitted that some doubt might be entertained as to the age of the greenstones, for some of them might be intrusive and therefore later than the sedimentary deposits between which they lie. But he contended that there could be no uncertainty with regard to the trappean ash or tuff, which being regularly interstratified in the Carboniferous series, must be contemporaneous with it. He pointed out that many of the greenstones, as well as fragments in the conglomerates or ashes, were highly vesicular and must originally have been in the condition of pumice.

As an illustration of the centres of eruption from which these materials were ejected, De la Beche drew special attention once more to the conspicuous eminence of Brent Tor and the rocks in its neighbourhood. His remarks on this subject are well worthy of being quoted—"The idea that in the vicinity of Brent Tor a volcano has been in action, producing effects similar to those produced by active volcanoes, forcibly presents itself. That this volcano projected ashes, which, falling into adjacent water, became interstratified with the mud, silt and sand there depositing, seems probable. That greenstones and other solid trappean rocks constituted the lavas of that period and locality, here and there intermingled with the ash, appears also a reasonable hypothesis. Upon the whole there seems as good evidence as could be expected that to the north and north-west of Tavistock, ash, cinders and liquid melted rocks were ejected and became intermingled with mud, silt and sand during this ancient geological epoch, corresponding with the phenomena exhibited in connection with volcanoes of the present day, more particularly when they adjoin or are situated in the sea, or other waters where ejected ashes, cinders and lava can be intermingled with ordinary mud, silt and sand."[64]

It remains for some future observer to fill up the outlines thus sketched by De la Beche, by tracing the respective areas of lavas and tuffs, distinguishing the various petrographical types, separating the intrusive from the interstratified sheets, identifying the necks and bosses that may mark centres of eruption, and expressing these various details upon maps on a sufficiently large scale.

A serious difficulty in this research arises from the effect of the profound alteration which has been produced on the igneous rocks by the cleavage of the region. Many of the "greenstones" have been so cleaved as to become slaty or almost schistose. De la Beche recognized this change and wrote of the "schistose trappean ash." A result of this metamorphism has been to impart to rocks originally massive the same fissile structure as the adjacent slates possess; and in this condition it is often hardly possible to distinguish between "greenstone" and fine-grained "ash." There can indeed be little doubt that among these Carboniferous volcanic rocks, as we have seen to be the case with those of the Devonian system in the same region, many lavas or sills have been mapped as tuffs.

The chief additions to our knowledge of the Carboniferous volcanic group of Devonshire since the time of De la Beche have been made by Mr. F. Rutley, Mr. W. A. Ussher and General M'Mahon. Mr. Rutley[65] has endeavoured to trace the respective areas occupied by the different varieties of volcanic rocks in the district around Brent Tor, near Tavistock, and to show the probable connection of the successive bands of lavas and tuffs with a central vent of discharge situated at that hill. He believes that these bands occur on four different horizons in the sedimentary series. He has studied the microscopic structure of the rocks, which in his view include "amphibolites, gabbros, basalts, pitchstones and schistose ashes, or clastic rocks of a doubtful nature."[66]

Mr. Ussher has re-mapped the tract of Culm-measures on the east side of the Dartmoor granite, besides visiting some of the other areas outside of the granite mass. While confirming the general accuracy of De la Beche's survey, he has been able to improve the mapping by inserting more detail, separating especially the tuffs from the "greenstones." The latter have been found by him to be mostly dolerites, some of which, from their parallelism the bands of tuff, may be in his opinion contemporaneous lavas, though the majority of them are evidently intrusive. The tuffs are regularly interstratified among the Culm-measures, their most important band in this district having an average breadth of about 100 yards, and being traceable for at least two miles, possibly considerably further.[67] In going over this tract with Mr. Ussher I was led to regard many of the sheets of diabase (dolerite) or gabbro as true sills and bosses. Most of them occur as short lenticular or oval patches tolerably numerous, but not traceable for more than a short distance, though a connection may often exist which cannot be detected by the scanty evidence on the surface. One sheet which has been followed by Mr. Ussher from Combe to beyond Ashton, a distance of nearly two miles, presents in the centre a somewhat coarsely crystalline texture which rapidly gives way to a much closer grain, and the rock then becomes highly vesicular. It is overlain with dark Culm-shales and bands of fine shaly tuff, passing upward into a granular tuff. Some layers of this tuff assume a finely foliated appearance by the development of pale leek-green folia, which show slickensided surfaces parallel with the bedding. The rock then presents one of the usual appearances of schalstein. This structure seems obviously due to mechanical movement along the planes of stratification.

Bands of black chert and cherty shale are interpolated among the tuffs, which also contain here and there nodular lumps of similar black impure earthy chert—an interesting association like that alluded to as occurring in the Carboniferous volcanic series of the Isle of Man, and like the occurrence of the radiolarian cherts with the Lower Silurian volcanic series already described.[68]

The volcanic belt in the valley of the Teign can be followed for about two miles. It is undoubtedly interstratified among the dark Culm-measures, which are distinctly seen dipping under and overlying it.

General M'Mahon has recently shown what may be done by careful and detailed examination of the ground broadly sketched in by De la Beche. He chose for study a strip of "greenstone" shown on the Geological Survey Map to extend for about three and a half miles along the north-west margin of the Dartmoor granite. He has found that what is represented under one wash of colour on that map includes both tuffs and lavas. The tuffs, in spite of the alteration which they appear to have undergone from the proximity of the great granite mass, are found by microscopic investigation to be made up of fine volcanic dust containing minute lapilli of various lavas. Sometimes as many as six or seven different kinds of lava may be represented in the same microscopic slide. These include felsitic or rhyolitic and trachytic rocks together with fragments of dark glassy lava full of magnetite dust. With the tuffs are intercalated sheets of felsite and trachyte. In the same district coarse volcanic agglomerate occur, made up of blocks of different lavas and pieces of different sedimentary rocks.[69]

These observations are of special interest, inasmuch as they point to the eruption of a much more acid series of volcanic lavas and tuffs than had previously been known to exist in the Culm-measures. Until the ground has been more accurately mapped, it is impossible to say whether these rocks are older or younger than those that lie around Brent Tor, a few miles to the south-west. General M'Mahon has noted the presence of more basic eruptive rocks in the same district. He specially cites the occurrence of mica-diorite, of basaltic lavas altered into a serpentinous mass, and of a dolerite which may possibly mark the actual vent of the old Brent Tor volcano. His observations on the influence of the Dartmoor granite in inducing new mineral rearrangements in the igneous rocks of the Culm-measure series are full of interest.