a very considerable thickness; the sinking was, however, greater than the amount of deposit could compensate, and the rocks have therefore the appearance of a tolerably deep sea formation at the top of the Nummulitic series. Then again, we have a long and steady rising of the land, and in consequence a great denudation going on, a denudation which has caused the removal of a great deal of the Nummulitic formation, in localities where sea-currents, high-tides and other unfavourable circumstances assisted in the work of destruction. It is curious to notice on the top of the Nummulitic limestone, how the surface of the rock has been broken by the waves; how the fragments have been rolled and rubbed and then glued together again. This appearance is always seen as a bed of transition between the Nummulitic and the Miocene. A considerable time must have elapsed between the end of the deposition of the bed and the breaking up of it, as we must allow time for its solidification. But at any rate, here, at the beginning of the Miocene epoch, we had the Nummulitic limestone forming a nearly horizontal and far-reaching sea-coast, covered with a very thin sheet of water, rolling and polishing pebbles. But this conglomeratic layer is thin, and we very soon see a large quantity of mud and sand, and pebbles of far distant rocks, brought down to the sea.

96. Let us consider the kind of map we have at the beginning of the Miocene epoch, and we will have no difficulty in understanding the formation of the Miocene sandstone and conglomerates of the SubHimalayan and Sub-Afghan chains. We have an immense expanse of sea, north of the tropic of Capricorn, between the latitudes 90° W. and 90° E., for, in these days, the Andes had not yet surged up and most of South America was under water, as well as nearly the whole of Africa, Arabia, Persia and India. There were probably groups of islands where these continents now stand, but the immense, dry, thirsty plains and plateaux of these countries were then under the sea. There was therefore no impediment to the regular play of the Trade Winds, no monsoons or winds deviated by the rarifying power of arid deserts, but especially no chains of mountains to dry the S. E. trade-winds before their arrival at the equator, and their ascending to become upper currents with a direction to the N. E. At the tropic of Cancer, these winds, still charged with the whole of the

humidity they had sucked from the sea in the Southern Hemisphere, descend again and become under or lower currents, keeping their N. E. direction. Before proceeding far, these winds meet a couple of ranges of mountains forming a great everted V, opening to the south, and on these ranges they poured such a quantity of rain. that a denudation began to take place to an amount nowhere else exemplified. The only approach to this rain-fall is that now observed in Patagonia, a high country which happens to be situated in the Southern Himisphere, somewhat in a position analogous to that of the Himalaya in the Northern Hemisphere during the Miocene epoch. In Patagonia "Captain King found the astonishing rain-fall of "nearly thirteen feet (151 inches) in forty-one days; and Mr. "Darwin reports, that the surface water of the sea, along this part of "the South American coast, is sometimes quite fresh, from the vast "quantity of rain that falls."+

We are now therefore prepared to anticipate a formation composed of coarse debris of the older mountains, washed down by violent torrents; we understand how it is that the waters of the sea lost their saltness, and that marine shells deserted these regions, and are therefore not to be found as fossils, or are at any rate excessively rare. The continual and violent rushing of streams, charged with mud and boulders, did not allow of the development of fluviatile animals; and thus we find the lower Miocene a mass of clay, sand and large boulders, in beds considerably false-bedded and totally free of fossils, with the exception, in a few protected localities, of some bulrushes imbedded in salt. These torrents occasionally tore up forests from the mountain sides in their headlong course, and thus it is that we find here and there small niduses of semi-carbonized wood, interred in the sandstone. The masses of conglomerate, accumulated in certain places, are of tremendous size, and probably mark the exit from the hills of the principal torrents of the Miocene Himalaya. The deposit of this coarse debris of the old volcanic chain and of the several deposits which had become gradually accumulated round it, attains a thickness of no less than 5,000 feet, and probably in some places much more. This mass of

See for a general explanation of the routes of the winds and the causes which alter these routes, the work of Captain Maury, L. L. D., U. S. N. entitled, "The Physical Geography of the Sea and its Meteorology."

Maury's Physical Geography of the Sea and its Meteorolgy. Page 129

clay, sand and boulders could not fail to convert the sea, which we have seen was shallow, into dry land, and thus we have this overlapping of the Upper Micoene on the edge of the Lower which is represented at para. 11. The Lower Miocene was itself exposed to the denudating influence of the rain, and boulders of Lower Miocene sandstone are common in the Upper Miocene.

The Upper Miocene appears to have been altogether a fresh-water formation; I mean, an accumulation of materials brought down by rivers of large size, which, in all probability, wandered through the flat plains of the lower Miocene, and extended in deltas and marshes and creeks, just as the Ganges and the Indus are observed to do now-adays. We may fairly imagine these Miocene tracts to have resembled closely a modern Indian plain traversed by large inundating riversa thick jungle of high grass and small trees for the elephant, the mastodon, the monkey, and a host of other animals to dwell in, and on the sides of the large meandering rivers, wastes of sand and clay, shallow pools and quicksands for the delight of the crocodiles, the tortoises and the hippopotamus. On sands left dry by changes in the course of the rivers, or piled up in undulating hillocks by the winds, grew thinly planted trees, such as we now see in the sandy tracts of Scinde, to feed and shelter the camel, the giraffe, and innumerable deer of various species; and on intermediate lands, good pasture supported the horse, the ox and sivatherium.

In the districts of Rawul Pindee, of Jheelum, of Bunnoo, of Kohat, the Upper Miocene has a thickness of more than 2,000 feet; but in the Rajaori and Poonch provinces of the Maharajah of Jummoo's kingdom, the bed attains a much greater thickness.

Any one who travels through the plains of the Punjab will notice the great quantity of cows, of oxen and horses seen loose on the sand near every village, and will remark at the same time, that when a stream has cut through the sand and thus exposed a section, not a bone is seen buried under the surface. If, however, he comes to a marsh, such as the one near Guriwall, in Bunnoo, he will observe that the bones will remain perfectly preserved in the thick mud, saturated with kullur,* which forms the bottom of the

* Impure Sulphate of soda, with a little carbonate of soda and chloride of sodium, which impregnates, more or less, nearly the whole of the soil of the Pun jab, and effloresces on the surface after rain or irrigation.

marsh. Now this kullur appears to have existed in the soil of the Upper Miocene, as the sandstones of that age are often covered with an efflorescence of that salt; and, indeed, that now seen in the alluvium is derived from the disintegrating, decaying and washing away of the Miocene beds. The fossil bones are always found either in a dark clay-stone, which has a bitter taste when applied to the tongue, or in a light-coloured sandy claystone. It is therefore highly probable that the existence of a marsh or swamp is necessary to the preservation of bones and their fossilification. This accounts for the bones being found in beds of limited extent, whilst for many miles not one is to be discovered; but it also brings additional evidence that the Upper Miocene was deposited as a growing delta, similar to the Sunderbunds of the Ganges and to the creeks of the mouths of the Indus.

What a singular landscape this belt of land must have presented! If we remember that at least seven different species of elephants roamed in these jungles, some much larger than the living one, and with tusks nine feet and a half long; that the dinotherium had a skull three feet and nine inches in length; that the mastodon was 17 feet long from the tail to the end of the tusks; that the sivatherium was a gigantic four-horned antelope-like animal; that the crocodiles were much larger than they are at present, and that the tortoises had a shell measuring 20 feet across; we may wonder indeed at the strange appearance which the jungles must have presented!!

I have called this fossiliferous formation Upper Miocene. In placing it in the Miocene, I have adopted the general opinion of geologists, but it may be Pliocenic and not Miocenic. I have not succeeded yet in discovering shells in these beds, and without shells it is impossible to fix with certainty the age of the formation.

I have forgotten to notice, that during the whole of the Miocene epoch there was a slow and steady sinking of the land. This sinking allowed of the accumulation of materials to the great thickness I have indicated, but unlike that which took place during the Eocene period, it was not sufficient to keep the country under the sea, the quantity of sand and clay and boulders, deposited by the rivers, being more than adequate to compensate for the sinking. The country, however, by the sinking was kept to a very little height above the sea level, and

the inundations of great rivers added continually to the thickness of the deposit.

97. There is no evidence of any violent action having taken place during the Eocene and Miocene epochs. There had been risings and sinkings of the whole country, but these were imperceptible to the senses, and were probably not more active than the same phenomena which now occur in many parts of the world, unknown to the inhabitants. The belt of flat land had increased to a good breadth, and the coast had become sufficiently distant from the mountains to enable the animals to live in peace and plenty, away from the storms and torrents of the hills, when the whole of the portion of the earth we have been considering was raised into an immense vault, by the forcing up of granite assisted by gases. When the gases condensed or escaped, the arch settled down by fracturing its sides, and these faulted sides of the arch are now, what we call the Himalayas and the Afghan chains of mountains.

When the settle-down began to take place, and the sides of the arch or vault were being broken, the direction of the linear volcanoes of the Silurian epoch compelled the new fractures to conform to it. On the eastern slope of the vault, the fractures ran from N. W. to S. E., on the western slope from N. E. to S. W. As is generally the case in an anticlinal, the highest portion of the vault settled down again to a level much lower than the sides, and we have therefore, in the northern Punjab, low hills, whilst on each side we have mountains towering to the sky.

It is not necessary to enter here into all the details of the complications which the masses of porphyry, trachyte, granite and other rocks, which had been cooling ever since the middle of the Paleozoic epoch, caused in the upheaval of the Afghan-Himalayan vault and in its settle-down. These details have already been sufficiently indicated in paras. 81 to 87. But I will insist on the effect of these masses being forced up like wedges through the rocks which covered them, and by their filling up a great deal of the space once occupied by these covering rocks, they compelled these last to be either folded or broken into pieces and packed edgeways.

It is not necessary to imagine that the top of the vault was raised to the same height as we now see the great peaks of the Himalayas