though it is highly probable that some at least belong to the same. epoch. I think it would be a most interesting point to study, whether the Central Indian Mountains participated in the great upheaval of the Afghan-Himalayan system, and to what extent they did so. Such a subject is not, however, to be discussed here, en passant. We must know more of what is buried under the alluvial sands and clays of the Punjab and the desert of Ajmeer, before we can decide on the relations of the Himalaya and Central Indian Mountains. The study of the Miocene beds appears the most likely sort of research to lead to interesting results. Could we once show satisfactorily that the plains of Northern India have been one day, and that not long ago (geologically speaking) a rugged country covered with Miocene hillocks and ridges, we should soon get an insight into the participation of the Central Indian Mountains in the great Afghan-Himalayan upheaval, and also into the nature of the soils and sub-soils of Upper India.

90. Let us now endeavour to sketch a geognostic history of the Afghan-Himalayan system of mountains, in accordance with the observations and hypotheses recorded in this paper.*

In the days of the Silurian epoch, the centre of Asia may be assumed to have been a sea uniting the Arctic to the Indian Ocean. In the middle of this sea, an archipelago of volcanic islands and subaqueous volcanoes existed, displaying great activity and ejecting into the sea an immense quantity of matter.

The position of these volcanoes and subaqueous vents is now represented by the porphyritic masses of Kaj-Nag, of Kistwar and Badrawar, by the summits of the catenated chains of Kashmir, &c., &c. The volcanoes were linear in their arrangement; one line, that of Kaj-Nag, Badrawar and Kistwar being continued far towards the south-east; and it is probable that the peaks of Chor, of Dodatoli and others in the same districts, are volcanic peaks on the same fissure. Another line or rather series of lines is that of the catenated chains in Kashmir, with a probable S. E. extension in the range of mountains which separate Lahool from Chumba. Another line again is that of Drass and Karghyl, at the back of the Ser and Mer

* A few unavoidable repetitions which occur in this portion of the paper will, I hope, be excused,

chain, and which is continued far towards the S. E., forming numerous and considerable volcanic mountains which appear as islands and promontories above the flat plain of the great Thibet plateau, through which the Sutlej runs.

These lines or fissures had a direction N. W.-S. E. and were all parallel, but the activity of the volcanoes was not the same on all the lines or in different parts of each line. Thus, in the line of Kaj-Nag and Badrawar, Chor and Dodatoli, the north-western end of the line is eminently distinguished and marked by very numerous and very long volcanoes, whilst the eastern one only gave passage to a few vents separated from each other by considerable intervals. On the other hand, (on another line) in Ladak, the volcanoes appear to have been small and few, whilst the eastern ends of the fissures appear to have been marked with many volcanoes of great size and activity. No volcanoes appear to have existed in that portion of the Silurian sea, where we now have the high mountains of Kailas and Karokoram; but where the Kuen Luen chain was at a later age to appear, it seems, that one or two lines of linear volcanoes did exist at the beginning of the Paleozoic epoch.

How long, how many thousands of years these volcanoes kept at their work, it is impossible even to guess. Their activity was immense, and it appears that in the waters which bathed the shores of the volcanic archipelago, too many outlets kept continually pouring out hot ejecta and roxious vapours to have allowed life to be present. We have seen that there is considerable evidence of the sea-bottom having been frequently heated enough to become cellular and amygdaloidal, and a reference to the section of the Tukt-i Suliman in Kashmir will, I think, leave little doubt of the frequency, the violence and the abundance of the discharges of lava, of lapilli, of ashes, and of hot liquid mud, We therefore find no Silurian fossils in Kashmir, and the slates and sandstones which are interbedded with the volcanic ejecta are completely deprived of fossils. This want of organic life did not, however, affect those portions of the sea which were sufficiently distant from the subaqueous craters and volcanic islands to escape the destructive effects of ejected materials; and we find, therefore, in the Karokoram chain and also in the Himalaya, between the Sutlej and the Kali, large beds of Silurian rocks with the usual fossils. These rocks are, as we have

seen, slates and shales which have until now proved azoic, but covered in by limestone rich in forms of the older Palæozoic period.

I need hardly say that the azoic slates, shales and sandstones which are interbedded with the ashes and amygdaloids in Kashmir are of Silurian date; if we wish, therefore, to colour a map of Kashmir solely in regard to the age of the rocks, we should have to colour all the ashes, slates, &c. Silurian. As the volcanic ejecta much predominate in quantity over the azoic slates and sandstone, I have not coloured the mass solely by age, but rather in view of the nature of the rocks.

But the Himalayan lines of insular volcanoes were not the only ones in that portion of the Silurian sea which we are considering; other linear volcanoes were directed from the N. E. to the S. W. in the longitudes and latitudes where we now find the great Afghan mountains. We know very little of these mountains: we have seen, however, that volcanic rocks of a granitoid appearance form the ranges of hills between Yeusofzaie and Bonneyr, and that clinkstone, granular and porphyritic, is quarried at Jellalabad. Dr. Bellew also tells us that he noticed volcanic rocks amongst the southern spurs of the Sufed Koh.* He also mentions that sharp earthquakes are frequent in the valley of the Korum, and it is reported by the Povindas who trade through the Gulwaira Pass, that a city situated at the back of the Suliman chains has been destroyed by a terrific earthquake. I need not point out the usual relation of severe earthquakes with accumulations of volcanic porphyries, in countries where no active volcanoes have been known to exist for several geological ages past. Then we have seen that the summits of the main chain of mountains, in the Vuzeeri country, are mostly composed of volcanic rocks; but the greatest amount of evidence is

*After crossing the hill-pass of Hazrah-Shutur-Gardan, the road lies through a gorge, and a stream or rivulet flows to the westward; in the bed of this rivulet pebbles of porphyry, hornblende and syenite (?) were seen; the surface of the soil was also covered with similar pebbles.. Near the top of the Shinghai Kothul, the volcanic rocks were again seen: Dr. Bellew says: "The surface was strewed with great blocks and fragments of porphyry and syenite, the latter was of various shades, from yellowish-green to greenish-brown, and its fragments shone with a vitreous lustre and broke with a similar frac ture." Chapter II. Narrative of a Mission to Kandahar. The above descrip. tion of syenite does not look much like syenite, it is nearly certain that the rock observed was a hypersthene rock.

derived from the boulders brought down by torrents and from those formerly carried down and now imbedded in the Miocene conglomerates which fringe the base of the Afghan mountains. These boulders and pebbles are mostly greenstone, felstone, trachyte, and porphyry identical with the Himalayan hornblende rock; and that peculiar variety of amygdaloidal greenstone, pierced with gas-vents, which has been described at No. 4 of the section of the Tukt-i-Suliman in Kashmir, para. 18, occurs in great abundance. (See also Pl. x. figs. 1. 1a.)

There can be, therefore, no possible doubt that the Afghan mountains were at the Silurian epoch an archipelago of volcanic islands and subaqueous volcanoes; indeed, they were merely another group of the same great archipelago; but the fissures or lines on which the vents were situated had a direction N. E., S. W.

Towards the end of the older Palæozoic epoch, the volcanoes appear to have subsided in violence, and allowed the waters of the neighbouring sea to cool. They did not do so, however, until they had ejected so much lava, scoriæ, lapilli, ashes, and debris of the inside of the earth that a great bar, a bar going from the North-west to the South-east and studded with the island-cones of half extinguished volcanoes, had been formed across the sea. A similar bar was produced by the Afghan group of volcanoes, directed N. E., S. W. and the two bars formed a gigantic V, with the angle pointing to the north. On these bars the sea was shallow; neither was it likely to be very deep between the two branches of the V. The end of the great activity of the volcanoes appears to have been marked by the breaking out of a great number of fumaroles or hot springs, depositing an immense quantity of silica, and forming thick beds of quartzite, sometimes pure and clear as glass, sometimes white and opaque as porcelain. We must not forget also, that all analogy points to a general rising of the sea bottom at the north-east of the Himalayan volcanic bar, not as a break, but as a gradual and slow upheaval of the earth's crust under the pressure of viscid granite.

But even these last efforts of the great volcanoes, these bursts of vapours and hot waters, became rare and intermittent, and animals made their appearance in the creeks and bays of the sea between the islands. It was then the dawn of the Carboniferous epoch, and all

over the great bars of volcanic debris a calcareous mud was deposited, teeming with the remains of animals, with the glimmering shells of the Producti, with large flat Orthida, and innumerable Bryozoa and numerous Encrinites which grew luxuriantly on the half chalky, half clayey, fœtid bottom of well protected island seas, gulfs and channels. And so it went on for years and years, until the sea became too shallow for Producti and Orthida to live in, and too easily disturbed to its very bottom to suit the delicate Bryozoa. These animals retired to greater depths on either side of the great bar, and in their stead appeared small Cucullæ, globular Terebratulæ, with here and there, on sandy banks, colonies of large Cardinia or Anthracosia, gibbose and smooth Aviculo-pectens, or radiated ones of great size. In calm waters, flat and large species of Goniatites basked in the sun in company with small Orthoceratida and large species of Bellerophon. Earthquakes were, however, frequent and terrible, raising and depressing large tracts of sea-bottom, folding and undulating the newly formed beds of limestone, so that most of the shells are found broken, and many of them are deformed to a wonderful extent.

Many changes occurred in the sea; clay and sand had been brought down in large quantities from the volcanic islands, and many of the creeks and inland seas were turned into swamps. Long shelving coast-lines extended from island to island, and many groups of the great archipelago were probably united by a low land into larger insular countries. The genera Cucullæa, Cardinia and Aviculo-pecten, and small Brachiopoda disappeared; and in their stead myriads of Gasteropoda, especially the Pyramidellida, living with numerous corals, made their appearance. As the islands joined more and more into larger dry lands, and approached nearer to a long strip of land supporting numerous peaks of extinct volcanoes, the rain-fall increased more and more, sand, mud and gravel accumulated in thicker beds at the mouth of the mountain torrents which now became rivers, and on the swampy shores forests of calamites and other trees grew up, whilst, out at sea, the mollusks and other animals continued to thrive at various depths, according to their kind. What has now become of these forests of calamites? Have they been buried in sands by oscillations of the coast and converted into coal? If they have, has the coal been denuded at a subsequent period? or has some portion of it escaped removal and