of compact sheets and rough amygdaloids, as in those wonderful sections of old lavaflows along the coasts of the Western Islands and the north of Ireland. In the railway cutting at the mouth of the Mossgiel Tunnel the melaphyres are found in their lower parts to be interstratified with and to rest upon beds of volcanic ash and brick-red sandstone. These strata are well bedded, and are made up of thin alternations of red gravelly traptuff, or peperino, fine ash, and, more or less, ashy sandstone-the whole pointing to a period of intermittent eruption when showers of dust and lapilli fell upon and became interstratified with ordinary sediment. In the river section at Ballochmyle, described by Mr. Binney and Professor Harkness, a similar red angular tuff (the "breccia of these geologists) forms a bed at least 30 feet thick, passing up into the red sandstones, as will be more fully pointed out in the sequel. It rests upon the melaphyres, which can be traced to within a few yards from it when the river is low, though the actual line of junction is not here seen.

Such are the characters of this group of volcanic rocks round the whole of the ring with which they girdle the Permian basin of the Ayr. From the bedded, slaggy, amygdaloidal character of the melaphyres, from their interstratification with red sandstone and ashy beds, from the tuff which is found both below them and above, it is evident that they are not intrusive masses thrust up among the paleozoic formations, carrying with them little or no clue to their date, but that they have been poured out at the surface, and must be of the same geological age as the strata with which they are associated. Outside the volcanic ring there occurs a number of small rounded hills or hillocks, consisting of a coarse red volcanic agglomerate (d in Fig. 1). This rock is unstratified and presents a very tumultuous appearance. is made up of fragments of various melaphyres of all sizes, up to masses a yard or more in length, angular, subangular, and rounded, imbedded in a gritty, ferruginous, felspathic paste, in which scattered crystals of augite, melanite, and black mica occur. As a rule,

It

FIG. 1.-SECTION THROUGH THE PERMIAN BASIN OF AYRSHIRE.

a Carboniferous sandstones and shales.

c Beds of felspathic amygdaloidal Trap (Melaphyre), with interbedded e Brick-red Permian sandstones.

these hillocks rise conspicuously above the neighbouring ground, and when the geologist has familiarized himself with them in one part of the district, his eye soon detects their form in other localities not before visited. They are not outliers of a deposit once covering here a greater space. Though surrounded by Carboniferous sandstones, and shales, they do not lie upon them; on the contrary, they descend vertically through the coal measures like so many huge pipes, sometimes standing on lines of fault, while the coal near them is altered. They are, in short, true volcanic "necks," each representing a former crater or focus of eruption. In one case, that of Helenton Hill, near Monkton, I found the sides of the neck still partially crusted with a mass of rough scoriaceous melaphyre-the remnant of the slaggy scoria which once coated the walls of the volcanic orifice. These necks vary in size from only a yard or two to 500 or 600 yards in diameter, and though their present conspicuous, blunted, conical outline is due, of course, to much subsequent denudation, one can readily enough imagine them to be cones of tuff, marking the position of volcanos that have only recently become extinct.

II. That these necks come indiscriminately through the faulted Upper Coal-measures, Carboniferous limestone and lower or calciferous sandstones, may be taken as good evidence that they are later than the Carboniferous period. A very brief examination of the ground is enough to raise a suspicion that they may have been connected with the trap-tuff already noticed. And the more they are studied, the more probable does this connexion become, until we are convinced that they can only be the vents through which the ash and trap were ejected. The material of which the necks consist is identical with that forming the gravelly tuff or peperino, save that it is much coarser and unstratified. But this is a distinction which might be expected to exist between the material which consolidated at the actual focus of eruption, and that which was thrown to some distance and was stratified under water along with ordinary sandy sediment. In each case the paste is dull dirty-red, felspathic, gritty or gravelly, derived from the trituration of the fragments imbedded in it. which are all pieces of different melaphyres, having the same characters as those which form the sheets in the ring.

As the bedded volcanic rocks of this part of Ayrshire lie upon Carboniferous strata, and are overlaid with Permian, they must be either of Carboniferous or of Permian age. The mode of occurence of the necks affords a presumption that the whole volcanic series must be later than Carboniferous times, and this inference is fully borne out by the relation of the rocks of the encircling ring to the Permian sandstones of the basin. No feature more speedily arrests the attention of a visitor to the valley of the Ayr than the great contrast between the general aspect of the Permian sandstones and those of the Carboniferous series below. The latter are of dull purplish red or grey colours, thin-bedded, interstratified with endless seams of red purple or grey shales and nodular marls. The former, on the other hand, are marked out at once by the strange brilliance

of their colour, ranging from a brick red to bright orange, by their abundant false-bedding, their thick beds, their entire freedom from shale of any kind, and their clear granular quartzose texture. Now, beds of sandstone having this character, and identical in every respect with the recognised Permian sandstones of the basin, lie among the fine ash and peperino below the sheets of trap, similar beds are intercalated between the beds of trap, they occur again in the ashy deposits overlying the trap, and there they can be seen to form part of the ordinary sandstones of the Permian basin. This latter junction is singularly instructive. It is well seen along the banks of the river Ayr at Ballochmyle where bed after bed can be studied from the massive brick-red sandstones down into the tuff that overlies the melaphyres. The subjoined woodcut (Fig. 2) illustrates this section. At the bottom lies the dull-red stratified tuff-a truly volcanic rock consisting of nothing but angular, subangular, and rounded fragments of different felspathic traps, imbedded in a triturated paste of the

Fig. 2.-SECTION OF PERMIAN SANDSTONE AND TUFF. RIVER AYR, BALLOCHMYLE.

same materials. Some of the stones are slaggy lumps of rock, like volcanic bombs, and in one case I found the air-cells pulled round the spheroidal surface of the mass- -the result perhaps of the whirling of the bit of melted lava through the air at the time of ejection. In the upper part of the tuff, thin lenticular seams of brick-red sandstone, sometimes mingled with ashy material, make their appearance, and these increase in number until they shade up into the main mass of the Permian sandstones. The gradual cessation of the tuff is full of interest, for we find that even after having passed over the highest of its beds we still meet with occasional nests of volcanic lapilli and single stones in the red sandstones. In this section we see how the igneous forces with intermittent and continually decreasing showers of ash and stones, finally died out.

It thus appears that the volcanic rocks of this part of Ayrshire lie upon, are interstratified with, and are covered by red sandstones, which have a distinctive character throughout as parts of one series marked off from the Carboniferous sandstones on which they rest. It is admitted that these red sandstones are Permian; it follows that

the volcanic rocks described in this paper and lying near the base of these sandstones afford now the proof of actual volcanic eruptions in Britain during the continuance of the Permian period.

.

I have spoken only of the Ayrshire basin, but evidence of the contemporaneous intercalation of volcanic rocks is not less clear among the Permian sandstones of Nithsdale. The felspathic trap laid open in the railway cutting at Drumlanrig is identical in its mineralogical characters and general aspect with the Ayrshire melaphyres. Its upper surface has the usual slaggy amygdaloidal character of the higher part of a lava-flow, and it is covered with gently inclined red sandstones containing a few beds of gravelly tuff or breccia like that of Ballochmyle. In the Carron Water there are some excellent sections of these rocks showing the red sandstones, sometimes sprinkled with volcanic bombs and often interlaced with bands of fine ash, angular tuff and even coarse volcanic breccia. Along with these proofs of igneous action occur beds of vesicular and amygdaloidal melaphyre, like those of the Ayr, and showing the same radiating veins of horizontally stratified red sandstone which fill up cracks in the original lava-form mass. The course of the Carron Water furnishes ample proof of the existence of contemporaneous volcanic rocks in the Permian series of the south of Scotland. The same evidence may be traceable southwards towards Dumfries, but I have not yet had an opportunity of examining that part of the district.

Throughout the Carboniferous tracts of the Lowland Valley of Scotland there are many igneous rocks which must be later than even the Upper Coal-measures, but of which the geological date cannot be approximately fixed. Some of these rocks are comparable with parts of the Ayrshire series, and it becomes an interesting question whether they may not belong to the same period. The determination of an actual date among the post Carboniferous igneous rocks is one which may be of great use in working out the geological history of the broad Lowland valley. And it is likewise not without its interest, as it enables us to connect the British type of the Permian system by another link with that of the centre of Europe.

It may be mentioned, in conclusion, that the Permian basins of the south-west of Scotland are traversed by a set of intrusive doleritic traps, sometimes as irregular bosses (fin Fig. 1), or as dykes and beds. These rocks have no relation to the Permian volcanic group, further than that they came through it as well as through the sandstones. The dykes belong to that remarkable N.W. or N.N.W. series which runs across Scotland and the north of England. I called attention to these dykes some years ago,' and suggested that they might be as late as the Middle Oolite. Since that time additional evidence has been accumulating, and I believe it will be possible to show good grounds for believing that they are not only as late as the Oolitic period, but may even be of Tertiary age.

1 [Trans. Roy. Soc., Edin., xx, p. 650.]