The Panjab, North-West Frontier Province, and Kashmir

River Valleys.—The cold weather traveller who is carried from Delhi to Ráwalpindí over the great railway bridges at points chosen because there the waters of the rivers are confined by nature, or can be confined by art, within moderate limits, has little idea of what one of these rivers is like in flood time. He sees that, even at such favoured spots, between the low banks there is a stretch of sand far exceeding in width the main channel, where a considerable volume of water is running, and the minor depressions, in which a sluggish and shallow flow may still be found. If, leaving the railway, he crosses a river by some bridge of boats or local ferry, he will find still wider expanses of sand sometimes bare and dry and white, at others moist and dark and covered with dwarf tamarisk. He may notice that, before he reaches the sand and the tamarisk scrub, he leaves by a gentle or abrupt descent the dry uplands, and passes into a lower, greener, and perhaps to his inexperienced eye more fertile seeming tract. This is the valley, often miles broad, through which the stream has moved in ever-shifting channels in the course of centuries. He finds it hard to realize that, when the summer heats melt the Himalayan snows, and the monsoon currents, striking against the northern mountain walls, are precipitated in torrents of rain, the rush of water to the plains swells the river 20, 30, 40, or even 50 fold. The sandy bed then becomes full from bank to bank, and the silt laden waters spill over into the cultivated lowlands beyond. Accustomed to the stable streams of his own land, he cannot conceive the risks the riverside farmer in the Panjáb runs of having fruitful fields smothered in a night with barren sand, or lands and well and house sucked into the river-bed. So great and sudden are the changes, bad and good, wrought by river action that the loss and gain have to be measured up year by year for revenue purposes. Nor is the visitor likely to imagine that the main channel may in a few seasons become a quite subsidiary or wholly deserted bed. Like all streams, e.g. the Po, which flow from the mountains into a flat terrain, the Panjáb rivers are perpetually silting up their beds, and thus, by their own action, becoming diverted into new channels or into existing minor ones, which are scoured out afresh. If our traveller, leaving the railway at Ráwalpindi, proceeds by tonga to the capital of Kashmír, he will find between Kohála and Báramúla another surprise awaiting him. The noble but sluggish river of the lowlands, which he crossed at the town of Jhelam, is here a swift and deep torrent, flowing over a boulder bed, and swirling round waterworn rocks in a gorge hemmed in by mountains. That is the typical state of the Himalayan rivers, though the same Jhelam above Báramúla is an exception, flowing there sluggishly through a very flat valley into a shallow lake.

The Indus Basin.—The river Sindh (Sanskrit, Sindhu), more familiar to us under its classical name of the Indus, must have filled with astonishment every invader from the west, and it is not wonderful that they called after it the country that lay beyond. Its basin covers an area of 373,000 square miles. Confining attention to Asia these figures, large though they seem, are far exceeded by those of the Yangtsze-Kiang. The area of which a description is attempted in this book is, with the exception of a strip along the Jamna and the part of Kashmír lying beyond the Muztagh-Karakoram range, all included in the Indus basin. But it does not embrace the whole of it. Part is in Tibet, part in Afghánistán and Biluchistán, and part in Sindh, through which province the Indus flows for 450 miles, or one-quarter of its whole course of 1800 miles. It seems likely that the Jamna valley was not always an exception, or at least that that river once flowed westwards through Rájputána to the Indian ocean. The five great rivers of the Panjáb all drain into the Indus, and the Ghagar with its tributary, the Sarustí, which now, even when in flood, loses itself in the sands of Bikaner, probably once flowed down the old Hakra bed in Baháwalpur either into the Indus or by an independent bed now represented by an old flood channel of the Indus in Sindh, the Hakro or Nara, which passes through the Rann of Kachh.

The Indus outside British India.—To the north of the Manasarowar lake in Tibet is Kailás, the Hindu Olympus. On the side of this mountain the Indus is said to rise at a height of 17,000 feet. After a course of 200 miles or more it crosses the south-east boundary of the Kashmír State at an elevation of 13,800 feet. From the Kashmír frontier to Mt Haramosh west of Gilgit it flows steadily to the north-west for 350 miles. After 125 miles Leh, the capital of Ladákh, is reached at a height of 10,500 feet, and here the river is crossed by the trade route to Yarkand. A little below Leh the Indus receives the Zánskar, which drains the south-east of Kashmír. After another 150 miles it flows through the basin, in which Skardo, the principal town in Baltistán, is situated. Above Skardo a large tributary, the Shyok, flows in from the east at an elevation of 8000 feet. The Shyok and its affluent, the Nubra, rise in the giant glaciers to the south-west of the Karakoram pass. After the Skardo basin is left behind the descent is rapid. The river rushes down a tremendous gorge, where it appears to break through the western Himálaya, skirts Haramosh, and at a point twenty-five miles east of Gilgit bends abruptly to the south. Shortly after it is joined from the west by the Gilgit river, and here the bed is about 4000 feet above sea level. Continuing to flow south for another twenty miles it resumes its westernly course to the north of Nanga Parvat and persists in it for 100 miles. Our political post of Chilás lies in this section on the south bank. Fifty or sixty miles west of Chilás the Indus turns finally to the south. From Jálkot, where the Kashmír frontier is left, to Palosí below the Mahaban mountain it flows for a hundred miles through territory over which we only exercise political control. Near Palosí, 812 miles from the source, the river enters British India. In Kashmír the Indus and the Shyok in some places flow placidly over alluvial flats, and at others with a rapid and broken current through narrow gorges. At Skardo their united stream is said, even in winter, to be 500 feet wide and nine or ten feet deep. If one of the deep gorges, as sometimes happens, is choked by a landslip, the flood that follows when the barrier finally bursts may spread devastation hundreds of miles away. To the north of the fertile Chach plain in Attock there is a wide stretch of land along the Indus, which still shows in its stony impoverished soil the effects of the great flood of 1841.

The Indus in British India.—After reaching British India the Indus soon becomes the boundary dividing Hazára and Pesháwar, two districts of the North West Frontier Province. Lower down it parts Pesháwar from the Panjáb district of Attock. In this section after a time the hills recede on both sides, and the stream is wide and so shallow that it is fordable in places in the cold weather. There are islands, ferry boats and rafts can ply, and the only danger is from sudden freshets. Ohind, where Alexander crossed, is in this section. A more famous passage is at Attock just below the junction of the Kábul river. Here the heights again approach the Indus on either bank. The volume of water is vastly increased by the union of the Kábul river, which brings down the whole drainage of the southern face of the Hindu Kush. From the north it receives near Jalálábád the Kunar river, and near Charsadda in Pesháwar the Swát, which with its affluent the Panjkora drains Dír, Bajaur, and Swát. In the cold weather looking northwards from the Attock fort one sees the Kábul or Landai as a blue river quietly mingling with the Indus, and in the angle between them a stretch of white sand. But during floods the junction is the scene of a wild turmoil of waters. At Attock there are a railway bridge, a bridge of boats, and a ferry. The bed of the stream is 2000 feet over sea level. For ninety miles below Attock the river is confined between bare and broken hills, till it finally emerges into the plains from the gorge above Kálabágh, where the Salt Range impinges on the left bank. Between Attock and Kálabágh the right bank is occupied by Pesháwar and Kohát and the left by Attock and Mianwálí. In this section the Indus is joined by the Haro and Soán torrents, and spanned at Khushálgarh by a railway bridge. This is the only other masonry bridge crossing it in the Panjáb. Elsewhere the passage has to be made by ferry boats or by boat bridges, which are taken down in the rainy season. At Kálabágh the height above sea level is less than 1000 feet. When it passes the western extremity of the Salt Range the river spreads out into a wide lake-like expanse of waters. It has now performed quite half of its long journey. Henceforth it receives no addition from the east till the Panjnad in the south-west corner of the Muzaffargarh district brings to it the whole tribute of the five rivers of the Panjáb. Here, though the Indian ocean is still 500 miles distant, the channel is less than 300 feet above the sea. From the west it receives an important tributary in the Kurram, which, with its affluent the Tochí, rises in Afghánistán. The torrents from the Sulimán Range are mostly used up for irrigation before they reach the Indus, but some of them mingle their waters with it in high floods. Below Kálabágh the Indus is a typical lowland river of great size, with many sandy islands in the bed and a wide valley subject to its inundations. Opposite Dera Ismail Khán the valley is seventeen miles across. As a plains river the Indus runs at first through the Mianwálí district of the Panjáb, then divides Mianwálí from Dera Ismail Khán, and lastly parts Muzaffargarh and the Baháwalpur State from the Panjáb frontier district of Dera Ghází Khán.

The Jhelam.—The Jhelam, the most westernly of the five rivers of the Panjáb, is called the Veth in Kashmir and locally in the Panjáb plains the Vehat. These names correspond to the Bihat of the Muhammadan historians and the Hydaspes of the Greeks, and all go back to the Sanskrit Vitasta. Issuing from a deep pool at Vernág to the east of Islámábád in Kashmír it becomes navigable just below that town, and flows north-west in a lazy stream for 102 miles through Srínagar, the summer capital, into the Wular lake, and beyond it to Báramúla. The banks are quite low and often cultivated to the river's edge. But across the flat valley there is on either side a splendid panorama of mountains. From Báramúla the character of the Jhelam suddenly changes, and for the next 70 miles to Kohála, where the traveller crosses by a fine bridge into the Panjáb, it rushes down a deep gorge, whose sides are formed by the Kajnág mountains on the right, and the Pír Panjál on the left, bank. Between Báramúla and Kohála there is a drop from 5000 to 2000 feet. At Domel, the stage before Kohála the Jhelam receives from the north the waters of the Kishnganga, and lower down it is joined by the Kunhár, which drains the Kágan glen in Hazára. A little above Kohála it turns sharply to the south, continuing its character as a mountain stream hemmed in by the hills of Ráwalpindí on the right bank and of the Púnch State on the left. The hills gradually sink lower and lower, but on the left side only disappear a little above the cantonment of Jhelam, where there is a noble railway bridge. From Jhelam onwards the river is of the usual plains' type. After dividing the districts of Jhelam (right bank) and Gujrát (left), it flows through the Sháhpur and Jhang districts, falling finally into the Chenáb at Trimmu, 450 miles from its source. There is a second railway bridge at Haranpur on the Sind Ságar line, and a bridge of boats at Khusháb, in the Sháhpur district. The noblest and most-varied scenery in the north-west Himalaya is in the catchment area of the Jhelam. The Kashmír valley and the valleys which drain into the Jhelam from the north, the Liddar, the Loláb, the Sind, and the Kágan glen, display a wealth of beauty unequalled elsewhere. Nor does this river wholly lose its association with beauty in the plains. Its very rich silt gives the lands on its banks the green charm of rich crops and pleasant trees.

The Chenáb.—The Chenáb (more properly Chínáb or river of China) is the Asikní of the Vedas and the Akesines of the Greek historians. It is formed by the union of the Chandra and Bhága, both of which rise in Lahul near the Báralácha pass. Having become the Chandrabhága the river flows through Pángí in Chamba and the south-east of Kashmír. Near Kishtwár it breaks through the Pír Panjál range, and thenceforwards receives the drainage of its southern slopes. At Akhnúr it becomes navigable and soon after it enters the Panjáb district of Siálkot. A little later it is joined from the west by the Tawí, the stream above which stands Jammu, the winter capital of Kashmír. The Chenáb parts Siálkot and Gujránwála on the left bank from Gujrát and Sháhpur on the right. At Wazírábád, near the point where Siálkot, Gujrát, and Gujránwála meet, it is crossed by the Alexandra railway bridge. Leaving Sháhpur and Gujránwála behind, the Chenáb flows through Jhang to its junction with the Jhelam at Trimmu. In this section there is a second railway bridge at Chund Bharwána. The united stream runs on under the name of Chenáb to be joined on the north border of the Multán district by the Ráví and on its southern border by the Sutlej. Below its junction with the latter the stream is known as the Panjnad. In the plains the Chenáb cannot be called an attractive river, and its silt is far inferior to that of the Jhelam.

The Ráví.—The Ráví was known to the writers of the Vedic hymns as the Parushní, but is called in classical Sanskrit Irávatí, whence the Hydraotes of the Greek historians. It rises near the Rotang pass in Kángra, and flows north-west through the southern part of Chamba. Below the town of Chamba, it runs as a swift slaty-blue mountain stream, and here it is spanned by a fine bridge. Passing on to the north of the hill station of Dalhousie it reaches the Kashmir border, and turning to the south-west flows along it to Basolí where Kashmír, Chamba, and the British district of Gurdáspur meet. At this point it is 2000 feet above the sea level. It now forms the boundary of Kashmír and Gurdáspur, and finally near Madhopur, where the head-works of the Bárí Doáb canal are situated, it passes into the Gurdáspur district. Shortly after it is joined from the north by a large torrent called the Ujh, which rises in the Jammu hills. After reaching the Siálkot border the Ráví parts that district first from Gurdáspur and then from Amritsar, and, passing through the west of Lahore, divides Montgomery and Lyallpur, and flowing through the north of Multán joins the Chenáb near the Jhang border. In Multán there is a remarkable straight reach in the channel known as the Sídhnai, which has been utilized for the site of the head-works of a small canal. The Degh, a torrent which rises in the Jammu hills and has a long course through the Siálkot and Gujránwála districts, joins the Ráví when in flood in the north of the Lyallpur district. But its waters will now be diverted into the river higher up in order to safeguard the Upper Chenáb canal. Lahore is on the left bank of the Ráví. It is a mile from the cold weather channel, but in high floods the waters have often come almost up to the Fort. At Lahore the North Western Railway and the Grand Trunk Road are carried over the Ráví by masonry bridges. There is a second railway bridge over the Sídhnai reach in Multán. Though the Ráví, like the Jhelam, has a course of 450 miles, it has a far smaller catchment area, and is really a somewhat insignificant stream. In the cold weather, the canal takes such a heavy toll from it that below Mádhopur the supply of water is mainly drawn from the Ujh, and in Montgomery one may cross the bed dryshod for months together. The valley of the Ráví is far narrower than those of the rivers described in the preceding paragraphs, and the floods are most uncertain, but when they occur are of very great value.

The Biás.—The Biás (Sanskrit, Vipasa; Greek, Hyphasis) rises near the Rotang pass at a height of about 13,000 feet. Its head-waters are divided from those of the Ráví by the Bara Bangáhal range. It flows for about sixty miles through the beautiful Kulu valley to Lárjí (3000 feet). It has at first a rapid course, but before it reaches Sultánpur (4000 feet), the chief village in Kulu, some thirty miles from the source, it has become, at least in the cold weather, a comparatively peaceful stream fringed with alder thickets. Heavy floods, however, sometimes cover fields and orchards with sand and boulders. There is a bridge at Manálí (6100 feet), a very lovely spot, another below Nagar, and a third at Lárjí. Near Lárjí the river turns to the west down a bold ravine and becomes for a time the boundary between Kulu and the Mandí State. Near the town of Mandí, where it is bridged, it bends again, and winds in a north-west and westerly direction through low hills in the south of Kángra till it meets the Siwáliks on the Hoshyárpur border. In this reach there is a bridge of boats at Dera Gopípur on the main road from Jalandhar and Hoshyárpur to Dharmsála. Elsewhere in the south of Kángra the traveller can cross without difficulty on a small bed supported on inflated skins. Sweeping round the northern end of the Siwáliks the Biás, having after long parting again approached within about fifteen miles of the Ráví, turns definitely to the south, forming henceforth the dividing line between Hoshyárpur and Kapúrthala (left bank) and Gurdáspur and Amritsar (right). Finally above the Harike ferry at a point where Lahore, Amritsar, Ferozepur, and Kapúrthala nearly meet, it falls into the Sutlej. The North Western Railway crosses it by a bridge near the Biás station and at the same place there is a bridge of boats for the traffic on the Grand Trunk Road. The chief affluents are the Chakkí, the torrent which travellers to Dharmsála cross by a fine bridge twelve miles from the railhead at Pathánkot, and the Black Bein in Hoshyárpur and Kapúrthala. The latter is a winding drainage channel, which starts in a swamp in the north of the Hoshyárpur district. The Biás has a total course of 390 miles. Only for about eighty miles or so is it a true river of the plains, and its floods do not spread far.

The Sutlej.—The Sutlej is the Shatadru of Vedic hymns and the Zaradros of Greek writers. The peasant of the Panjáb plains knows it as the Nílí or Ghara. After the Indus it is the greatest of Panjáb rivers, and for its source we have to go back to the Manasarowar lakes in Tibet. From thence it flows for 200 miles in a north-westerly direction to the British frontier near Shipkí. A little beyond the Spití river brings it the drainage of the large tract of that name in Kángra and of part of Western Tibet. From Shipkí it runs for forty miles in deep gorges through Kunáwar in the Bashahr State to Chíní, a beautiful spot near the Wangtu bridge, where the Hindustan-Tibet road crosses to the left bank. A little below Chíní the Báspa flows in from the southeast. The fall between the source and Chíní is from 15,000 to 7500 feet. There is magnificent cliff scenery at Rogí in this reach. Forty miles below Chíní the capital of Bashahr, Rámpur, on the south bank, is only 3300 feet above sea level. There is a second bridge at Rámpur, and from about this point the river becomes the boundary of Bashahr and Kulu, the route to which from Simla passes over the Lurí bridge (2650 feet) below Nárkanda. Beyond Lurí the Sutlej runs among low hills through several of the Simla Hill States. It pierces the Siwáliks at the Hoshyárpur border and then turns to the south, maintaining that trend till Rúpar and the head-works of the Sirhind canal are reached. For the next hundred miles to the Biás junction the general direction is west. Above the Harike ferry the Sutlej again turns, and flows steadily, though with many windings, to the south-west till it joins the Chenáb at the south corner of the Multán district. There are railway bridges at Phillaur, Ferozepur, and Adamwáhan. In the plains the Sutlej districts are—on the right bank Hoshyárpur, Jalandhar, Lahore, and Montgomery, and on the left Ambála, Ludhiána and Ferozepur. Below Ferozepur the river divides Montgomery and Multán from Baháwalpur (left bank). The Sutle; has a course of 900 miles, and a large catchment area in the hills. Notwithstanding the heavy toll taken by the Sirhind canal, its floods spread pretty far in Jalandhar and Ludhiána and below the Biás junction many monsoon canals have been dug which inundate a large area in the lowlands of the districts on either bank and of Baháwalpur. The dry bed of the Hakra, which can be traced through Baháwalpur, Bikaner, and Sindh, formerly carried the waters of the Sutlej to the sea.

The Ghagar and the Sarusti.—The Ghagar, once a tributary of the Hakra, rises within the Sirmúr State in the hills to the east of Kálka. A few miles south of Kálka it crosses a narrow neck of the Ambála district, and the bridge on the Ambála-Kalka railway is in this section. The rest of its course, till it loses itself in the sands of Bikaner, is chiefly in Patiála and the Karnál and Hissár districts. It is joined by the Umla torrent in Karnál and lower down the Sarustí unites with it in Patiála just beyond the Karnál border. It is hard to believe that the Sarustí of to-day is the famous Sarasvatí of the Vedas, though the little ditch-like channel that bears the name certainly passes beside the sacred sites of Thanesar and Pehowa. A small sandy torrent bearing the same name rises in the low hills in the north-east of the Ambála district, but it is doubtful if its waters, which finally disappear into the ground, ever reach the Thanesar channel. That seems rather to originate in the overflow of a rice swamp in the plains, and in the cold weather the bed is usually dry. In fact, till the Sarustí receives above Pehowa the floods of the Márkanda torrent, it is a most insignificant stream. The Márkanda, when in flood, carries a large volume of water, and below the junction the small channel of the Sarustí cannot carry the tribute received, which spreads out into a shallow lake called the Sainsa jhíl. This has been utilized for the supply of the little Sarustí canal, which is intended to do the work formerly effected in a rude way by throwing bands or embankments across the bed of the stream, and forcing the water over the surrounding lands. The same wasteful form of irrigation was used on a large scale on the Ghagar and is still practised on its upper reaches. Lower down earthen bands have been superceded by a masonry weir at Otu in the Hissár district. The northern and southern Ghagar canals, which irrigate lands in Hissár and Bikaner, take off from this weir.

Action of Torrents.—The Ghagar is large enough to exhibit all the three stages which a cho or torrent of intermittent flow passes through. Such a stream begins in the hills with a well-defined boulder-strewn bed, which is never dry. Reaching the plains the bed of a cho becomes a wide expanse of white sand, hardly below the level of the adjoining country, with a thread of water passing down it in the cold weather. But from time to time in the rainy season the channel is full from bank to bank and the waters spill far and wide over the fields. Sudden spates sometimes sweep away men and cattle before they can get across. If, as in Hoshyárpur, the chos flow into a rich plain from hills composed of friable sandstone and largely denuded of tree-growth, they are in their second stage most destructive. After long delay an Act was passed in 1900, which gives the government large powers for the protection of trees in the Siwáliks and the reclamation of torrent beds in the plains. The process of recovery cannot be rapid, but a measure of success has already been attained. It must not be supposed that the action of chos in this second stage is uniformly bad. Some carry silt as well as sand, and the very light loam which the great Márkanda cho has spread over the country on its banks is worth much more to the farmer than the stiff clay it has overlaid. Many chos do not pass into the third stage, when all the sand has been dropped, and the bed shrinks into a narrow ditch-like channel with steep clay banks. The inundations of torrents like the Degh and the Ghagar after this stage is reached convert the soil into a stiff impervious clay, where flood-water will lie for weeks without being absorbed into the soil. In Karnál the wretched and fever-stricken tract between the Ghagar and the Sarustí known as the Nailí is of this character.

The Jamna.—The Jamna is the Yamuna of Sanskrit writers. Ptolemy's and Pliny's versions, Diamouna and Jomanes, do not deviate much from the original. It rises in the Kumáon Himálaya, and, where it first meets the frontier of the Simla Hill States, receives from the north a large tributary called the Tons. Henceforth, speaking broadly, the Jamna is the boundary of the Panjáb and the United Provinces. On the Panjáb bank are from north to south the Sirmúr State, Ambála, Karnál, Rohtak, Delhi, and Gurgáon. The river leaves the Panjáb where Gurgáon and the district of Mathra, which belongs to the United Provinces, meet, and finally falls into the Ganges at Allahábád. North of Mathra Delhi is the only important town on its banks. The Jamna is crossed by railway bridges between Delhi and Meerut and between Ambála and Saháranpur.

Changes in Rivers.—Allusion has already been made to the changes which the courses of Panjáb rivers are subject to in the plains. The Indus below Kálabágh once ran through the heart of what is now the Thal desert. We know that in 1245 A.D. Multán was in the Sind Ságar Doáb between the Indus and the united streams of the Jhelam, Chenáb, and Ráví. The Biás had then no connection with the Sutlej, but ran in a bed of its own easily to be traced to-day in the Montgomery and Multán districts, and joined the Indus between Multán and Uch. The Sutlej was still flowing in the Hakra bed. Indeed its junction with the Biás near Harike, which probably led to a complete change in the course of the Biás, seems only to have taken place within the last 150 years[2].

CHAPTER IV

Aravallí System.—In the southern and south-eastern districts of the Panjáb there are exposures of highly folded and metamorphosed rocks which belong to the most ancient formations in India. These occupy the northern end of the Aravallí hills, which form but a relic of what must have been at one time a great mountain range, stretching roughly south-south-west through Rájputána into the Bombay Presidency. The northern ribs of the Aravallí series disappear beneath alluvial cover in the Delhi district, but the rocks still underlie the plains to the west and north-west, their presence being revealed by the small promontories that peep through the alluvium near the Chenáb river, standing up as small hills near Chiniot in the Sháhpur, Jhang, and Lyallpur districts.

The Salt Range in the Jhelam and Sháhpur districts, with a western continuation in the Mianwálí district to and beyond the Indus, is the most interesting part of the Panjáb to the geologist. It contains notable records of three distinct eras in geological history. In association with the well-known beds of rock-salt, which are being extensively mined at Kheora, occur the most ancient fossiliferous formations known in India, corresponding in age with the middle and lower part of the Cambrian system of Europe. These very ancient strata immediately overlie the red marls and associated rock-salt beds, and it is possible that they have been thrust over bodily to occupy this position, as we have no parallel elsewhere for the occurrence of great masses of salt in formation older than the Cambrian.

The second fragment of geological history preserved in the Salt Range is very much younger, beginning with rocks which were formed in the later part of the Carboniferous period. The most remarkable feature in this fragment is a boulder-bed, resting unconformably on the Cambrian strata and including boulders of various shapes and sizes, which are often faceted and striated in a way indicative of glacial action. Several of the boulders belong to rocks of a peculiar and unmistakable character, such as are found in situ on the western flanks of the Aravallí Range, some 750 miles to the south. The glacial conditions which gave rise to these boulder-beds were presumably contemporaneous with those that produced the somewhat similar formation lying at the base of the great coal-bearing system in the Indian peninsula. The glacial boulder-bed thus offers indirect evidence as to the age of the Indian coal-measures, for immediately above this bed in the Salt Range there occur sandstones containing fossils which have affinities with the Upper Carboniferous formations of Australia, and on these sandstones again there lie alternations of shales and limestones containing an abundance of fossils that are characteristic of the Permo-Carboniferous rocks of Russia. These are succeeded by an apparently conformable succession of beds of still younger age, culminating in a series of shales, sandstones, and limestones of unmistakably Triassic age.

There is then an interruption in the record, and the next younger series preserved occurs in the western part of the Salt Range as well as in the hills beyond the Indus. This formation is of Upper Jurassic age, corresponding to the well-known beds of marine origin preserved in Cutch. Then follows again a gap in the record, and the next most interesting series of formations found in the Salt Range become of great importance from the economic as well as from the purely scientific point of view; these are the formations of Tertiary age.

The oldest of the Tertiary strata include a prominent limestone containing Nummulitic fossils, which are characteristic of these Lower Tertiary beds throughout the world. Here, as in many parts of North-Western India, the Nummulitic limestones are associated with coal which has been largely worked. The country between the Salt Range plateau and the hilly region away to the north is covered by a great stretch of comparatively young Tertiary formations, which were laid down in fresh water after the sea had been driven back finally from this region. The incoming of fresh-water conditions was inaugurated by the formation of beds which are regarded as equivalent in age to those known as the Upper Nari in Sind and Eastern Baluchistán, but the still later deposits, belonging to the well-known Siwálik series, are famous on account of the great variety and large size of many of the vertebrate fossil remains which they have yielded. In these beds to the north of the Salt Range there have been found remains of Dinotherium, forms related to the ancestors of the giraffe and various other mammals, some of them, like the Sivatherium, Mastodon, and Stegodon, being animals of great size. On the northern side of the Salt Range three fairly well-defined divisions of the Siwálik series have been recognised, each being conspicuously fossiliferous—a feature that is comparatively rare in the Siwálik hills further to the south-east, where these rocks were first studied. The Siwálik series of the Salt Range are thus so well developed that this area might be conveniently regarded as the type succession for the purpose of correlating isolated fragmentary occurrences of the same general series in northern and western India. To give an idea as to the age of these rocks, it will be sufficient to mention that the middle division of the series corresponds roughly to the well-known deposits of Pikermi and Samos.

Kashmir deserves special mention, as it is a veritable paradise for the geologist. Of the variety of problems that it presents one might mention the petrological questions connected with the intrusion of the great masses of granite, and their relation to the slates and associated metamorphic rocks. Of fossiliferous systems there is a fine display of material ranging in age from Silurian to Upper Trias, and additional interest is added by the long-continued volcanic eruptions of the "Panjál trap." Students of recent phenomena have at their disposal interesting problems in physiography, including a grand display of glaciers, and the extensive deposits of so-called karewas, which appear to have been formed in drowned valleys, where the normal fluviatile conditions are modified by those characteristic of lakes. The occurrence of sapphires in Zánskar gives the State also an interest to the mineralogist and connoisseur of gem-stones.

Of this kaleidoscopic assemblage of questions the ones of most immediate interest are connected with the Silurian-Trias succession in the Kashmír valley, for here we have a connecting-link between the marine formations of the Salt Range area and those which are preserved in greater perfection in Spití and other parts of the Tibetan highlands, stretching away to the south-east at the back of the great range of crystalline snow-covered peaks.

In this interesting part of Kashmír the most important feature to Indian geologists is the occurrence of plant remains belonging to genera identical with those that occur in the lower part of the great coal-bearing formation of Peninsular India, known as the Gondwána system. Until these discoveries were made in Kashmír about ten years ago the age of the base of the Gondwánas was estimated only on indirect evidence, partly due to the assumption that glacial conditions in the Salt Range and those at the base of the Gondwánas were contemporaneous, and partly due to analogy with the coal measures of Australia and South Africa. In Kashmír the characteristic plant remains of the Lower Gondwánas are found associated with marine fossils in great abundance, and these permit of a correlation of the strata with the upper part of the Carboniferous system of the European standard stratigraphical scale.

Kashmír seems to have been near the estuary of one of the great rivers that formerly flowed over the ancient continent of Gondwánaland (when India and South Africa formed parts of one continental mass) into the great Eurasian Ocean known as Tethys. As the deposits formed in this great ocean give us the principal part of our data for forming a standard stratigraphical scale, the plants which were carried out to sea become witnesses of the kind of flora that flourished during the main Indian coal period; they thus enable us with great precision to fix the position of the fresh-water Gondwánas in comparison with the marine succession.

Spití.—With á brief reference to one more interesting patch among the geological records of this remarkable region, space will force us to pass on to consideration of minerals of economic value. The line of snow-covered peaks, composed mainly of crystalline rocks and forming a core to the Himálaya in a way analogous to the granitic core of the Alps, occupies what was once apparently the northern shore of Gondwánaland, and to the north of it there stretched the great ocean of Tethys, covering the central parts of Asia and Europe, one of its shrunken relics being the present Mediterranean Sea. The bed of this ocean throughout many geological ages underwent gradual depression and received the sediments brought down by the rivers from the continent which stretched away to the south. The sedimentary deposits thus formed near the shore-line or further out in deep water attained a thickness of well over 20,000 feet, and have been studied in the tahsíl of Spití, on the northern border of Kumáon, and again on the eastern Tibetan plateau to the north of Darjeeling. A reference to the formations preserved in Spití may be regarded as typical of the geological history and the conditions under which these formations were produced.

Succession of Fossiliferous Beds.—In age the fossiliferous beds range from Cambrian right through to the Tertiary epoch; between these extremes no single period was passed without leaving its records in some part of the great east-to-west Tibetan basin. At the base of the whole succession there lies a series of schists which have been largely metamorphosed, and on these rest the oldest of the fossiliferous series, which, on account of their occurring in the region of snow, has been named the Haimanta system. The upper part of the Haimanta system has been found to contain the characteristic trilobites of the Cambrian period of Europe. Over this system lie beds which have yielded in succession Ordovician and Silurian fossils, forming altogether a compact division which has been distinguished locally as the Muth system. Then follows the so-called Kanáwar system, which introduces Devonian conditions, followed by fossils characteristic of the well-known mountain limestone of Europe.

Then occurs a break in the succession which varies in magnitude in different localities, but appears to correspond to great changes in the physical geography which widely affect the Indian region. This break corresponds roughly to the upper part of the Carboniferous system of Europe, and has been suggested as a datum line for distinguishing in India an older group of fossiliferous systems below (formed in an area that has been distinguished by the name Dravidian), from the younger group above, which has been distinguished by the name Aryan.

During the periods that followed this interruption the bed of the great Eurasian Ocean seems to have subsided persistently though intermittently. As the various sediments accumulated the exact position of the shore-line must have changed to some extent to give rise to the conditions favourable for the formation at one time of limestone, at another of shale and at other times of sandy deposits. The whole column of beds, however, seems to have gone on accumulating without any folding movements, and they are consequently now found lying apparently in perfect conformity stage upon stage, from those that are Permian in age at the base, right through the Mesozoic group, till the time when Tertiary conditions were inaugurated and the earth movements began which ultimately drove back the ocean and raised the bed, with its accumulated load of sediments, into the great folds that now form the Himálayan Range. This great mass of Aryan strata includes an enormous number of fossil remains, giving probably a more complete record of the gradual changes that came over the marine fauna of Tethys than any other area of the kind known. One must pass over the great number of interesting features still left unmentioned, including the grand architecture of the Sub-Himálaya and the diversity of formations in different parts of the Frontier Province; for the rest of the available space must be devoted to a brief reference to the minerals of value.

Rock-salt, which occurs in abundance, is possibly the most important mineral in this area. The deposits most largely worked are those which occur in the well-known Salt Range, covering parts of the districts of Jhelam, Sháhpur, and Mianwálí. Near the village of Kheora the main seam, which is being worked in the Mayo mines, has an aggregate thickness of 550 feet, of which five seams, with a total thickness of 275 feet, consist of salt pure enough to be placed on the table with no more preparation than mere pulverising. The associated beds are impregnated with earth, and in places there occur thin layers of potash and magnesian salts. In this area salt quarrying was practised for an unknown period before the time of Akbar, and was continued in a primitive fashion until it came under the control of the British Government with the occupation of the Panjáb in 1849. In 1872 systematic mining operations were planned, and the general line of work has been continued ever since, with an annual output of roughly 100,000 tons.

Open quarries for salt are developed a short distance to the east-north-east of Kálabágh on the Indus, and similar open work is practised near Kohát in the North West Frontier Province, where the quantity of salt may be regarded as practically inexhaustible. At Bahádur Khel the salt lies at the base of the Tertiary series, and can be traced for a distance of about eight miles with an exposed thickness of over 1000 feet, sometimes standing up as hills of solid salt above the general level of the plains. In this area the production is naturally limited by want of transport and the small local demand, the total output from the quarries being about 16,000 tons per annum. A small quantity of salt (generally about 4000 tons a year), is raised also from open quarries in the Mandí State, where the rock-salt beds, distinctly impure and earthy, lie near the junction between Tertiary formations and the older unfossiliferous groups.

Coal occurs at numerous places in association with the Nummulitic limestones of Lower Tertiary age, in the Panjáb, in the North West Frontier Province, and in the Jammu division of Kashmír. The largest output has been obtained from the Salt Range, where mines have been opened up on behalf of the North Western Railway. The mines at Dandot in the Jhelam district have considerable fluctuations in output, which, however, for many years ranged near 50,000 tons. These mines, having been worked at a financial loss, were finally abandoned by the Railway Company in 1911, but a certain amount of work is still being continued by local contractors. At Bháganwála, 19 miles further east, in the adjoining district of Sháhpur, coal was also worked for many years for the North Western State Railway, but the maximum output in any one year never exceeded 14,000 tons, and in 1900, owing to the poor quality of material obtained, the collieries were closed down. Recently, small outcrop workings have been developed in the same formation further west on the southern scarp of the Salt Range at Tejuwála in the Sháhpur district.

Gold to a small amount is washed from the gravel of the Indus and some other rivers by native workers, and large concessions have been granted for systematic dredging, but these enterprises have not yet reached the commercially paying stage.

Other Metals.—Prospecting has been carried on at irregular intervals in Kulu and along the corresponding belt of schistose rocks further west in Kashmír and Chitrál. The copper ores occur as sulphides along certain bands in the chloritic and micaceous schists, similar in composition and probably in age to those worked further east in Kumáon, in Nipál, and in Sikkim. In Lahul near the Shigrí glacier there is a lode containing antimony sulphide with ores of zinc and lead, which would almost certainly be opened up and developed but for the difficulty of access and cost of transport to the only valuable markets.

Petroleum springs occur among the Tertiary formations of the Panjáb and Biluchistán, and a few thousand gallons of oil are raised annually. Prospecting operations have been carried on vigorously during the past two or three years, but no large supplies have so far been proved. The principal oil-supplies of Burma and Assam have been obtained from rocks of Miocene age, like those of Persia and the Caspian region, but the most promising "shows" in North West India have been in the older Nummulitic formations, and the oil is thus regarded by some experts as the residue of the material which has migrated from the Miocene beds that probably at one time covered the Nummulitic formations, but have since been removed by the erosive action of the atmosphere.

Alum is manufactured from the pyritous shales of the Mianwálí district, the annual output being generally about 200 to 300 tons. Similar shales containing pyrites are known to occur in other parts of this area, and possibly the industry might be considerably extended, as the annual requirements of India, judged by the import returns, exceed ten times the native production of alum.

Borax is produced in Ladákh and larger quantities are imported across the frontier from Tibet. In the early summer one frequently meets herds of sheep being driven southwards across the Himalayan passes, each sheep carrying a couple of small saddle-bags laden with borax or salt, which is bartered in the Panjáb bazars for Indian and foreign stores for the winter requirements of the snow-blocked valleys beyond the frontier.

Sapphires.—The sapphires of Zánskar have been worked at intervals since the discovery of the deposit in 1881, and some of the finest stones in the gem market have been obtained from this locality, where work is, however, difficult on account of the great altitude and the difficulty of access from the plains.

Limestone.—Large deposits of Nummulitic limestone are found in the older Tertiary formations of North-West India. It yields a pure lime and is used in large quantities for building purposes. The constant association of these limestones with shale beds, and their frequent association with coal, naturally suggest their employment for the manufacture of cement; and special concessions have recently been given by the Panjáb Government with a view of encouraging the development of the industry. The nodular impure limestone, known generally by the name of kankar, contains sufficient clay to give it hydraulic characters when burnt, and much cement is thus manufactured. The varying composition of kankar naturally results in a product of irregular character, and consequently cement so made can replace Portland cement only for certain purposes.

Slate is quarried in various places for purely local use. In the Kángra valley material of very high quality is obtained and consequently secures a wide distribution, limited, however, by competition with cheaply made tiles.

Gypsum occurs in large quantities in association with the rock-salt of the Salt Range, but the local demand is small. There are also beds of potash and magnesian salts in the same area, but their value and quantity have not been thoroughly proved.