which now runs along the Bandin ridge. This tongue of gneissose granite has thus almost completed an ellipse of outcrop, and from the quaquaversal dip can only indicate a continuous sheet beneath the schistose strata of Naori which are the highest beds visible of the series. Apparently below this continuous bed there are several more lenticular masses of the gneissose granite inserted between the schistose strata, but only outcropping on the south side of the synclinal. They rapidly thin out towards the north-west beyond Dujukatoli and gradually coalesce for the most part towards the east-south-east and join the south extension of the Dudatoli massif. South of Kainur there are one or two very thin beds which run for a short distance and thin out both ways. If they have any connection with the main mass of the rock it must be a subterranean one.

Schistose series.

I have already called attention to the like disposition of the similar rocks at Kalogarhi Mountain and at the Rama Serai,1 and I here lay especial stress upon the fact, because in a line of section from the Sub-Himalayan boundary to the snows running through Kalogarhi, Dudatoli, and Kedarnath, the only regular apparent synclinals of any importance are connected with the gneissose and schistose series. Nowhere except on the north side of a mass of gneissose granite is there ever a prevailing south dip of the strata. This may be either a coincidence or a necessary relation. I leave the question at present to be afterwards returned to. I now come to some of the most marked features in the schistose series, which forms as it were a groundwork for the gneissose granite. Over a large area these rocks consist of the more arenaceous, or quartz-schist type with thinner dividing beds of more argillaceous schists; the schistosity being not very marked. Foliation takes place in the majority of cases along the original bedding, as is well demonstrated, where the arenaceous and slaty types interbed. In some few cases however I found it crossing the bedding along incipient cleavage planes, especially in some quartz-schists; whilst films of mica were often visible along joint planes. In addition, wherever the rocks had been much crushed and cut by nearly parallel divisional planes and by slickensides their surfaces were also coated with a thin micaceous glaze. In some cases among the quartz-schists laminæ more or less micaceous had been cleaved across, and the mica plates re-arranged perpendicular to the lamina where they occur. To whatever cause this regional metamorphism be due, it is certain that it begins imperceptibly and continues with a minor degree of intensity over a large tract. The section from Shánkar on the Ramganga due north to near Hansuri on the east Nyar plainly demonstrates this. About a mile from any outcrop of gneissose granite as we approach the Dudatoli massif, in no matter what direction, there is a rapid, but gradual change sets in in the metamorphism of the schistose beds. The faint films of micaceous material assume by degrees the aspect of distinct layers of mica plates of considerable thickness. Vein quartz appears ramifying along the foliation planes. Garnets gradually assemble in the schist; first showing as minute pin-heads under a coating of what one may call mica-leaf, and gradually increasing in size and definiteness concomitantly with the mica until they reach an average size of peas, and rarely as large as filberts. On every side of the Dudatoli area can these most

'Records, XX., pt. 1, p. 30.

striking changes be observed. In the present paper I am only offering evidence collected in the field, and consequently confine my observations to what is visible under a pocket lens or to the eye.

Varieties of the schists.

The more intensely crystalline schists may, as a rule, be divided into two kinds from a structural point of view. The more common type is one in which the planes of foliation have a wavy or blistered appearance, and are completely covered with mica of a pale silvery hue. The leaves of mica are, as it were, welded together along the wavy irregularities of the foliation planes, forming an uninterrupted layer of the shining mineral. Under the lens the garnets are of a dull claret colour, and when not mere grains are of a more or less distinctly crystalline form. The break of continuity along the folia, caused by a garnet, is always overcome by the mica plates waving round it on both sides. Nevertheless, I also found other plates which ran undisturbed towards the garnet, and then stopped abruptly (see fig. 1). The quartz, which in this type is very subordinate, is of necessity present in lenticular layers between the waving sheets of mica. Secondary quartz, in large lenticular or nodular strings, enwrapped in a thin tissue of mica, is frequently met with. The other type of schist is a much more quartzose rock, and is regularly bedded with the former. The mica in this case does not cover in, in a complete manner, the foliation planes. Instead, it is disseminated in small flakes, and is of silvery hue, and also brown. At the same time, their arrangement is always parallel with the bedding, causing a genuine foliation. Garnets are scarcer than in the former rock.

The first of these two types is only rarely puckered or corrugated. Near Mafkori pass south-east of Dudatoli there is a variety in which the mica plates, with a leaden sheen, have the appearance of shrivelled tea-leaves. A rather singular structure arrested my attention in the schistose beds as they are advancing from the widely spread slightly schistose kind to the more intensely metamorphic border round the gneissose granite. The foliation surfaces, which were very regular, straight planes, shewed a set of parallel ridges and furrows running in the direction of the dip. They were very minute, and a good idea of the effect to the eye may be obtained from the aspect of a corrugated iron roof.

I may here emphasize two points-first, the schist found near the gneissose granite is entirely a thorough crystalline schist, a fact needing no microscope to demonstrate; and secondly, along a line of country, where rock is exposed at every step, it is seen that this culminating intense form graduates into a wide-spread less intense form, and that in turn graduates into ordinary slates and quartzites.

From the universality of the changes as the gneissose granite is approached, it is only reasonable to conclude either that the extra schistosity and the development of garnets were brought about by the introduction of the gneissose granite (in whatever way the latter was produced), or that they and the gneissose granite were the joint result of some more remote and subtle cause. In any case, the two rock slates are ← so inseparable that they may be classed as contemporary. It follows that whatever can be proved concerning the geological age of the one may be taken as evidence for the age of the other. The importance of this will be noticed in a paper to follow.

I was unable to gather any clue from the manner of occurrence of the garnets.

C

The bending round of the mica plates does not satisfy me as an argument for the priority of crystallisation of the garnets. I may, however, say that the latter in all their stages, from the minutest speck to the fully crystallised form, were always intact: there was no sign of crushing or drawing out of them with the foliation. Perhaps a microscopic examination may yield more decisive results. Turning now to the gneissose granite of the Dudatoli neighbourhood, it is necessary to say something of its mineralogical composition. In

Gneissose granite. this I shall be brief, partly because the rock seems to answer perfectly to much of the allied rock already so luminously described by Colonel McMahon, and partly because the object of this paper is rather to draw attention to the larger aspect of the rock in the field than to its microscopical character. At the same time, I may point out that its coarsely crystalline condition makes the examination under the lens not so hazardous and superficial a matter as it might otherwise appear. It must be borne in mind that I am speaking at present of the rock in this locality of Dudatoli only.

The rock as a whole may be said to be eminently felspathic, and distinguished from an ordinary granite or gneiss by the presence of schorl crystals. These latter occur in all manner of conditions, from large to small, from perfect prisms to others which are manifestly severed portions of a single prism, and held together like a string of beads. The schorl is the most completely crystallised of the minerals; the micas, black and white, come next, often constituting the only perfectly crystallised minerals in the [rock. Orthoclase, in rectangular prisms sometimes twinned, is not typical in the rock, but usually appears as in an "augen" or kernel gneiss. The quartz as in all granites is always without crystalline form and fills the interstices between the other minerals. In many cases minute granules of quartz and small portions of schorl are contained in the larger porphyritic orthoclase crystals and "eyes." Garnets occur very sparingly, sometimes singly and sometimes in nests. Kyanite, Beryl, and other accessory minerals I am confident are not present in this locality. This outline is sufficient to shew the general mineralogical resemblance between this rock and others described by numerous writers on other parts of the Himalaya. In themselves the minerals are not so important as in their mode of arrangement which varies so conspicuously in different parts of the same mass of gneissose granite that it requires very special mention. In describing these I may add that they are paralleled nearly always in the Chor and Kalogarhi Mountains; but at Kedarnath we come upon rocks of an altogether different structure, and none of the remarks to follow will consequently apply to the gneissose rock of the snowy range where I have at present seen it. The latter better agrees with Stoliczka's "Central gneiss" or what Fouqué and Lévy call "gneiss granulitique." I hope to describe it later.

The Dudatoli gneissose granite for the purpose of classification easily divides up Varieties of the gneisinto three types—(A) FOLIATED, (B) SEMI-FOLIATED, (C) NON-FOLIATED. These three all graduate one into the other. (A) may be sub-divided into (1) Tabular foliated, (2) Lenticular-Tabular foliated.

sose granite.

1 See Minéralogie Micrographique, Fouqué and Lévy, Manche III., fig. 2.

2 See Mem. G. S. I., V., pt. I, p. 12.

3 Minéralogie Micrographique, p. 175.

(A) (1) is a variety not often seen in this part, and then only near the junction with the schists. It is the most decidedly fissile of any of the varieties, It is built up of continuous straight foliæ of felspar and quartz, with intermediate foliæ of mica in continuous films. These each run their own course at least for a great distance without coinciding, the mica being very generally muscovite only. To the weather and to the hammer this rock behaves more like a schist than a granite (see fig. 2). (A) (2) is much more common, especially in the thinner bands of the rock, which occur near the outskirts of the gneissose granite area. In it the felspar is still undifferentiated into eyes or crystals; but the foliæ of felspar and quartz swell out and thin again (see fig. 3), foreshadowing the perfect eyes of the prevailing forms of the gneissose granite (see B). In using the words "undifferentiated" and "foreshadowing," I by no means imply that the lenticular-tabular is necessarily an embryonic condition of the augen. It seems as likely that the reverse is the case, for in some quartzites of coarse grain, between Rudarprayag and Agastmundi, I have seen the grains, originally rounded, absorbing smaller portions of quartz at each end and so taking on a lenticular appearance, which with the mica films developed coincidently gave the appearance of a lenticular-tabular quartz schist (see fig. 5).

(B) may be sub-divided into (1) Augen, (2) Porphyritic-augen.

In (B) (1) the different mineral layers are no longer distinct. On the contrary, the mica plates of two layers unite with each other on each side of an "eye," thus cutting up the felspathic layer into a number of isolated eyes (see fig. 4). The long axes of the eyes being parallel give the semi-foliated character to the rock. Connected with this sub-division is a rock, which on the foliation planes shews eyes of felspar blotched and drawn out in the direction of dip. It and the lenticular-tabular quartz schist mentioned above seem to indicate differential movements of the particles of the rock, and are paralleled by structures in other basic rocks, which I hope to describe in another number.

(B) (2) only differs from (B) (1), by containing, in addition, larger, more blunted eyes at intervals, very often turned in directions other than parallel with the foliation and sometimes approximating to a rectangular outline. This sub-division is the most developed of any at Dudatoti, the Chor and Kalogarhi.

(C) in all respects resembles a normal granite, and is usually porphyritic, the porphyritic crystals being always sharply rectangular and often twinned. They are oriented in all directions.

General facts of distribution.

The result of a careful examination of the Dudatoli area is to shew that the subdivision (A) (2) is nearly always found in the thinner bands of the rock, and in the thicker ones near the junction with the schists. The division B is found impartially in all but the thinnest bands. The division C is only found in those parts where the bands of gneissose granite have united together to form a wide continuous mass, and then only in small quantities compared with the division B. These results may be shortly stated by saying that the more the rock loses its bedded appearance the more it approaches a massive and perfectly granitic form; whilst wherever it alternates rapidly by interbedding with the schists, the foliated and semi-foliated types predominate.

I must now advert to a consideration of the inter-relations between the schists Inter-relations beand the gneissose granite. These are of such great imtween the schists and portance in understanding by how much the latter has acted the gneissose granite. functionally as an intrusive rock, are so amply demonstrated by close work with the hammer, and may hence be said to be the fieldgeologist's speciality, that I make no apology for treating them fully in this place. The gneissose granite of Dudatoli, in no case that I know of, breaks through the schists, disturbing them: no violent contortions, no puckerings of the foliation planes ever take place as distinct results of the intrusion of the gneissose granite. The difference between this rock and a genuine intrusive granite with large masses of muscovite, which very rarely occurs in the Dudatoli area, and which by the fact that it has no passage forms towards the gneissose granite may be considered to be entirely distinct and doubtless of another geological age, is most marked. A block of mica-schist containing the latter is penetrated irregularly by an amoebiform mass of it which at every thrown-out process or vein has crumpled and tortured the schist into utter compliance with its own irregularities of shape. Compare this with the thin band of gneissose granite near Hansuri, thoroughly exposed in the river section. Its upper and lower surfaces are perfect planes fitting in with similar planes in the mica schist. If the former rock had been sawn in a mason's yard and fitted in with artistic precision, it could not present a more composed aspect, both in its own regular structural planes and in those of the mica schist among which it lies. And yet the two rocks are so unlike that a pencil point may be placed precisely on the junction line. Or consider again what looks on the map a contradiction to the statement of its non-eruptive character, viz., the blunted process near Gulek. At Burari the dip of the foliation of the gneissose granite is S.-W. 30°; and that of the schists in contact is perfectly parallel. Following the junction boundary north-west, there is, near Risti, a more westerly dip in the gneissose granite; and this is also conformed to by the schistose beds. East of Gulek the dips of both have worked round towards the north-west, still retaining their concordancy. Similarly, if the boundary line be further examined, turning towards the north and north-west below Dobri trigonometrical station, it will equally be seen that there is no sudden intrusion, no erupting of the one rock among the other, but on the other hand that each has its foliation planes perfectly parallel with the other, and there is no contortion or puckering of the mica-schist whatever. Without going into detail over the whole map, I may summarily state that everywhere, except where a fault is manifest, the lie of the one rock coincides completely with the lie of the other: such an event as the foliation lines of one crossing at right angles the foliation of the other is unknown.

That the foliation represents true bedding, is placed beyond a doubt by the numerous river sections where from a great height the interbedding of the two rocks can be traced conforming to the trend of the foliation planes. Thus the gneissose granite is insinuated among the schists; and if it is intrusive from a foreign source, and not inborn, it must have acted on the principle of the wedge and parted the schists with wonderful precision along very great distances. At Kalogarhi the same remarks apply.