istán, it appears that the north side of the Sikaram in the Saféd Koh range is composed of transition rocks, amongst which magnesian and calcareous beds predominate.

From the western flank of the same hill and from an elevation of about 10,000 feet, Dr. Aitchison brought specimens of unaltered shales with fucoid markings. A similar rock is in situ on the south side of the peak. From the Shalinar stream, east side of the Paiwar Kotal, a pebble of Lithodendron limestone was brought. The latter is possibly of carboniferous age. It would appear then that beds of paleozoic age accompanied possibly by older mesozoic strata exist in the Saféd Koh range and form the highest part of it. Pebbles of paleozoic rocks, probably of carboniferous age, have been found by Major N. Vicary 1 near the Khaibar mouth near Jamrúd and both the locality last named and the Sikaram beds seem to belong to one belt of palæozoic strata, of which the Attock section is the eastern continuation.

The group of altered rocks between Jagdallak, Gandamak, and Ali Boghán may be outliers of the same formation.

Glacial. Recent formations.-In my former paper, I had occasion to mention some of the deposits and traces of former glaciers in Afghán-Túrkistán. Since then I have seen by far the most perfect instances of recent glacial action, when crossing the Hindu Kúsh by the Chahárdar pass in October 1886. The road which leads from Chápdarra camping ground on the north side of the Hindu Kush to the top of the pass ascends a narrow straight valley, bounded on each side by steep cliffs, some of them crowned with perpetual snow. The bottom of the valley itself is greatly choked and partially filled with debris, which might be simply the detritus from the hillsides. Large cones and fans of fragmentary material descend from each small ravine on both sides. So far only the configuration of the valley, its nearly straight course and absence of larger side streams, would suggest the former presence of glaciers. But on reaching an elevation of 12,000 feet, one suddenly comes to a huge mass of debris, which closely resembles the recent accumulations near the lower end of a glacier. Large blocks, some of them of immense dimensions, are loosely mingled with angular fragments of every size and the whole is arranged like a dam across the valley. The hillsides (gneiss) are polished and grooved and the blackened surfaces glisten and shine in the distance like metal. All the larger blocks show extensive grooving and deep ice-scratches on their polished sides. This mass of debris lies at the base of a terrace filling the valley. The former glacier, of which this is the end moraine, was on the upper and raised portion of the valley. The latter bears the remarkable appearance of an ice-worn trough; it is wider than the valley below, and its base is now partially filled by finer debris, through which a small stream winds its way amidst a series of swampy pools. It is within the area of perpetual snow and the latter with frozen patches of ice lies on the hillsides and in sheltered depressions.

The valley looks as if the glacier had only quite recently left it. Moraines and glacial silt still lie as they were deposited. The head and catchment area of the valley close to the top of the pass (14,100 feet) is still rather thickly covered with frozen snow.

1 On the Geology of the Upper Punjab and Peshawar, Quart. Journ. Geol. Soc., VII, p. 38. Rec., Vol. XIX, p. 263.

Glacial traces on the south slope of the Hindu Kush-The descent from the Chahárdar pass to the Deh-i-Tang lies down a narrow valley of much the same character as the one just described. But the most interesting feature in connection with it is, that in this valley there are some small glaciers still remaining. Near the head of the valley, just south of the Chahárdar pass, at an elevation of 12,050 feet above sea-level, several small side ravines join; I noticed three of them were still filled with glaciers, and though they were very small, the moraine accumulations near their lower ends were enormous. Especially the one from the right side shoots off an enormous cone of large fragments, amongst which there are some very good examples of ice-scratched blocks.

Recent conglomerates.-Both in Túrkistán and the neighbouring South-western Badakhshan deposits of recent and sub-recent conglomerates, sands and clays are largely developed. The hills which skirt the cretaceous anticlinals between Haibak and Dahána Ghori are formed by these deposits which attain there a great thickness. Similarly the valley of the Súrkh-áb is partially filled by them.

The valleys belonging to the Kabul river drainage south of the Hindu Kush are to a large extent lined with terraces of conglomerates, as, for instance, the wide terraces of Siáh-Gird, Chahárdeh, etc.

1

These conglomerate terraces form quite a feature in the landscapes of the road east of Kabul, amongst which I may mention the terraces of Gandamak and Nimlah Bagh.

I believe these accumulations belong to the same age as the Indus gravel beds, which are seen to skirt the hills the whole way from Peshawar to Sind.

In the next number of the "Records " I intend giving a geological map of Afghánistán and part of Persia with a summary of the geological structure and mineral resources of Afghánistán.

CALCUTTA, 23rd December 1886.

Physical Geology of West British Garhwal; with Notes on a Route Traverse through Jaunsar Bawar and Tiri-Garhwal, by C. S. MIDDLEMISS, B.A., Geological Survey of India.

Introduction..

PART I.

In Part 2 of the Records for 1885 I described a fossiliferous zone of pretertiary age amongst the old mountain-building rocks which form part of the Lower Himalaya of British Garhwal. When that preliminary notice was published I had only been working for a short time in the district, and consequently the area treated of was confined, and no generalizations could be made. Since then, having spent another field season there, I am

able to make some additions to our knowledge of the stratigraphy of those parts. But, inasmuch as this part of the Himalaya is divided from ground which has been already geologically surveyed in Jaunsar Bawar by a broad strip of almost completely unknown country, viz., part of Tiri-Garhwal, I am still compelled to put aside for the present all definite.correlation between the rock systems displayed in Jaunsar and those upon which I have now been engaged. In consequence of this I shall still adhere to the method observed in my preceding paper of naming the series, when fossils are absent, after their prevailing lithological character; trusting to time, and a wider experience, to eventually make them one with the old established Himalayan formations.

British Garhwal has been topographically surveyed on the one-inch scale, and the principal object of my first season's work there was to settle down on some part of it where the strata showed signs of falling into a natural order, and then work from that point as closely as the advantages of a large-scale map would admit.

But, before doing so, I went through Jaunsar Bawar (the next British possession west-north-west of British Garhwal) for the purpose of making myself as thoroughly acquainted as possible with that already mapped district. From its no great distance from Garhwal it was thought that it might have many points of similarity, and so give one new to Himalayan work a useful basis to go upon.

From the northern extremity of Jaunsar Bawar I rapidly crossed by Tiri through native territory until I struck my own working district at Srinagar in British Garhwal. Before coming to the main object of this paper I shall therefore briefly set down a few somewhat disjointed notes referring to my route traverse: not because they have any intrinsic value, but because of their possible bearings on past or future work.1 The southern parts of Jaunsar Bawar were examined by me in some detail with especial regard to the position of the Mandhali series in its Jaunsar Bawar. numerous: unconformable appearances near Chakrata. It is not my province to describe these rocks, with the questions arising from which Mr. Oldham is now engaged, but the numerous examples about which I have notes, from their containing clear blebs of quartz, fragments of felsites and a great deal of felspar and felspathic material scattered about in the matrix, gave me the idea of a rock produced by the degradation of felsites or granitic rocks and produced perhaps during a time of intermittent volcanic activity. North of Chakrata, the high Deoban ridge up to Mandhali, and some way beyond, was wrapped in deep snow, the product of a late storm, at the time of my visit; and work was therefore curtailed in a great measure. What I saw of the Mandhalis there left no doubt in my mind as to their identity with the rocks classed as the same a little south of Chakrata. At Tiutar (near the Tonse) I made a careful examination of the Chakrata series to the east of the Konain-Mudhaul fault, in order to solve, if possible, the question of inversion, and whether the igneous rock was intrusive or inter-bedded. From

To make these route observations more intelligible it may be noted that Mr. Oldham's Bawars are taken by him to be about on the horizon of the base of the infra-Krol group of the Simla section, the Mandhalis being older still, but newer than the Deoban, limestone (see Vol. XVIII, p. 4):

* See R. D. Oldham's "Note on the Geology of Jaunsar and the Lower Himalaya." Rec. Geol. Sury. India, Vol. XVI, 1883, p. 193.

the presence of what appeared to be thin ash beds, associated with the igneous rock, I at first felt sure both must be inter-bedded, but until microscopic sections are cut there is a possibility that they may turn out to be merely crushed and pressurefoliated diorites. With regard to inversion, the lie of the beds as contrasted in deep gorges, and on the neighbouring mountain spurs south-west of Tiutar, inclined me in favour of the supposition of inversion; for to the eye there seems to be a steepening of the beds in towards the mountain; but, on the other hand, the limestone which lies apparently immediately below the igneous rock is a very whiteygrey, compact, and marble-like rock, such as could have been produced by the contact metamorphism of the igneous rock. In addition, south-west of Tesar Khera, where the limestone contains some magnetic iron in small crystals, the contact of the igneous rock has in places altered the latter, so that it now lies as amorphous lumps filling lacunæ among the joints of the limestone. From these. facts the limestone must be older than the igneous rock, and, if the latter is interbedded, their present position must be the original one, and the inversion theory cannot be maintained.

Tiri-Garhwal.

Bawars.

From Tiutar I rounded the north end of Jaunsar Bawar and left British territory at the head of the Khunigadh river which divides Jaunsar from Tiri-Garhwal. From the Khunigadh pass I descended, in a south-east direction, to Porohla. Except for a mile or so from the pass, where there were some diorites and ash (?) beds, there was nothing met with but Bawar quartzites during the whole descent. They lie dipping slightly towards the north-east and never exceeding an angle of 20°. Their colour is white, and, when seen through a lens, they appear to be made up of little angular fragments of clear quartz and apparently nothing else. They form the exceedingly precipitous ridge west of Gundalho, lying between the two branches of the Kamalada river, which meet at Porohla. The steep bare walls into which they weather utterly barred my progress at a point on the map above the of Gundalho; but, with their low dip towards the north-east, they appeared to continue much higher up towards the main range running parallel to the Tonse. I have no doubt that these are identical with the Bawar quartzites of Mr. Oldham. Porohla lies in a widened and flattened valley, given over to cultivation, and called the Rama Serai, lying to the north-north-east from the village along the present line of the river.

Gneissose rock.

Along its bed there must be an unseen junction between the Bawar quartzites just mentioned, and another rock which is petrologically a gneiss. Unfortunately both it and the Bawars themselves, in this region, have so weathered at the surface that in the few exposures that I could examine in a hasty march I found no reliable junction that gave me a clue as to the nature of the boundary. This excessive weathering into a fine gravel of the separated crystals of the gneiss and the disintegrated grains of the quartzite, is probably the cause of the flat cultivated stretch of the Rama Serai, the valley of which has become in this locality choked with the products of disintegration. I had the gneiss with me, gradually emerging from its weathered covering, up to, and beyond Kumalo; but, when the position indicated on the map by the in Kumalo was reached, it came to an end, and Bawar quartzite and some schists

continued up to the ridge, trending north from Saulda peak. The eastern boundary of this gneissose rock runs a little east of north to beyond Dokri, from the point just mentioned. This boundary is as difficult to unravel as the western boundary, for the uniform Bawars show but little dip, except where they rise in escarpments, or when schists come in among them, and even then the results are discordant.

With regard to the gneiss itself, its foliation planes, when visible, are roughly horizontal just as the Bawars are. It will thus be seen that stratigraphically no position can be assigned to it from the evidence before us, for it is not known whether the boundaries are natural or faulted; and, from the nearly horizontal dip found in every exposure, an average actual dip cannot be deduced. Then again, I have no evidence to bring forward as to whether the rock, provisionally named a gneiss, is by origin a gneiss or a granite. Petrologically it is identical with the rock of the Chor mountain, which has received much attention from Indian geologists. That rock, by Colonel McMahon's microscopical evidence, and by the still more convincing evidence of large and numerous included fragments of schist and quartzite found by Mr. Oldham and myself during the season's work of 1883-84, is by origin a granite; but whether the same can be said of this rock depends on how far an exact likeness between two rocks of this kind is to be deemed conclusive of their identity. At the same time, though willing myself to withhold judgment until more extensive mapping has been done, there is no doubt that the two rocks in so far as their mineral characters go are the same. There are the same quartz and felspar, with sometimes a predominance of the former; there are pale and black micas; and there are schorl crystals, developed here and there, and occurring abundantly in cracks and veins, just as they did in the gneissose granite of the Chor. In some places I found a decomposed greenstone, probably a dyke of diorite, in the gneissose rock.

From the east edge of the gneiss I crossed the ridge north of Saulda peak and followed a tributary of the Banale river, west of Kanal. The whole of the west side of this ridge was chiefly composed of the Bawar quartzites, showing no evidence of dip, except a general horizontality. At the summit of the pass, and on the east side, schistose beds came on, and continued until some inter-bedded diorites and dioritic ashes appeared below them. Then came more schists, and finally I struck in the stream bed a massive limestone of the Deoban type.

It will thus be seen that, on the east side of the gneiss, we have a set of outcrops inversely arranged to those of the west side. The following will illustrate this :Bawar Schists and GNEISS. quartzites. igneous products.

WEST. Deoban Limestone.

Schists and Bawar

igneous products.

quartzites.

EAST.

Deoban

Limestone.

Now, the steep descent from the ridge into the tributary of the Banale, leaves the impression that the approximately horizontal beds to the east of the gneiss rank in order of superposition from the limestone below to the Bawar quartzite above; that is, we may assume an ascending series on that side of the gneiss from the limestone upwards. The west side of the gneiss is not so clear, but, from the undoubted identity of the quartzite with the Bawars there is the exact appearance there also of an ascending series towards the gneiss.

A seemingly absurd corollary follows from this: We appear bound to believe