force was collected. Jingis Khan, who had, after the capture of Samarkand, encamped his army in the fertile country between this town and Naksheb, now detached a division under two of his sons against Khowaresm, while he ordered the forces which had traversed Khorassan to form a girdle round the southern edge of the desert. When the sons of Muhammad discovered that they were being hemmed in, they tried to escape, Jellal-ud-din made a brilliant charge and managed to get away, but the other two sons were captured and beheaded and their heads were shewn about on spears. We are told that the peasants of the canton of Vesht were generally enriched by the number of precious stones captured from the Khorazinans, which the Mongols, who did not know their value, sold them at an absurdly small price.

Meanwhile the Mongol army marched upon Khowaresm, the Khogend of other writers, the modern Khiva, the capital of that rich cluster of cities that then bordered the Oxus, a river very like the Nile in forming a strip of green across two sandy deserts which bound it on either hand. The Kankalis were then its garrison. Jutchi, the eldest son of Jingis, commanded the Mongols, he summoned the inhabitants to surrender, offering them easy terms. His father, he told them, had made him a present of their country, and he wished the city to preserve its beauty and prosperity. The summons was without avail, and the siege proceeded. For lack of stones the Mongol catapults were served with balls made out of the neighbouring mulberry trees, hardened by being soaked in water. The quarrels of Jallchi and his brother Zagatai interfered with the progress of the siege, discipline was loosened, and the Mongols after six months labour had lost a great number of men. Jingis, when he heard of the quarrelling, appointed a younger son, Ogotai, to superintend the work. It was now pushed on with vigour; the Mongols at length assaulted the town, fired its buildings with naptha, and after seven days of desperate street-fighting captured it, they sent the artizans and skilled workmen into Tartary, set aside the young women and children as slaves, and then made a general massacre of the rest of the inhabitants. They destroyed the city, and then submerged it by opening the dykes of the Oxus. (Eber el Ethir, see D'Ohsson i. 270). The ruins are probably those now known as Old Khogend.

Having summered his horses in the beautiful meadows of Nakhsheb, Jingis advanced upon Termed, which was taken by assault and its inhabitants slain. An incident of the capture is worth repeating, an old woman on the point of being killed, said she had a magnificent pearl which she would give them if they spared her, when they demanded it she told them she had swallowed it, upon which she was disembowelled. Jingis ordered the other corpses to be dealt with the same way and searched for similar treasure. He then wintered at Séman, whence he conquered Badakshan, and sent an army to ravage Khorassan. Jelal-ud-din had taken refuge at Ghizin; Jingis advanced in pursuit. The great city of Balkh, the cradle of the earliest tra

ditions of the Aryan races, submitted to him, but he was afraid to leave it behind him. On pretence of numbering its inhabitants he enticed them out of the city and then slaughtered them, the city itself was reduced to ashes.

To Tuli, a son of Jingis, had been assigned the conquest of Khorassan, whose cities had been submissive enough to the Mongols when in pursuit of the sultan Muhammad, as we have already related. Khorassan was then one of the richest and most prosperous regions on the earth's surface; its towns were very thickly inhabited, and it was the first and most powerful province of Persia. The Mongol invasion altered all this, and the fearful ravage and destruction then committed is almost incredible, the following brief facts will give a little idea only. Tuli was preceded by a corps of 10,000 men under the command of Togachar. These advanced upon Nessa, the siege of which has been told by one of its contemporary chieftains, Muhammad of Nessa, after fifteen days pounding from twenty catapults, which were served by prisoners, a breach was made, the walls were stormed, the inhabitants ordered to evacuate the city, they were then told to lie down side by side, and were tied together with cords, then the Mongols destroyed the whole, men, women and children, with showers of arrows. This horrible hecatomb destroyed 70,000 people. The historian Muhammad, with many fugitives, had taken refuge in an inpregnable fort called Kharendu. When the Mongols saw they could not take it they consented to retire on the payment of 10,000 cotton garments. According to their custom, they massacred the two old men who had volunteered on the dangerous errand of carrying this booty to their camp. The Mongols then scattered about Khorassan, everywhere making the peasantry serve them as sappers, etc., in their attacks on the towns, of these Sebzevar was taken by assault and 70,000 people killed, the same fate overtook the inhabitants of Car and Nocam.

(To be continued.)

-0

ON THE PHYSICAL GEOGRAPHY OF THE

HIMALAYA.

Continued from page 165.

Now, that the Himalaya really was, at one time, in great part submerged; that the vast mass of detritus from the Himalaya at present spread over the plains in its vicinity was so spread by the ocean when the founts of the deep were broken up; that this huge bed of detritus, every where forthcoming, is now found in unequal proportion and distribution and state of comminution; as for example, deeper piled within than without the Sandstone range and the embaying spurs, and also, more gravelly and abundant to the north-west, more sandy and scant to the south-east; and, lastly, that the

*Captain Herbert has given statements of its depth to the westward, where there is a Sandstone range. To the eastward, where there is none, I found it on the right bank of the Tishta, under the mountains, 120 feet; at fifteen miles lower down, 60 to 70 feet; at fifteen miles still further off the mountains, 40 to 50 feet. There was here no interruption to the free spread

Gangetic plain really now has a great oblique dipt from the Sutlej at Ruper to the Brahmaputra at Gwálpárá, whereby all the Himalayan feeders of the Ganges are in the plains so much bent over to the eastward-these are presumptions relative to the past, as legitimate as the extant facts suggesting them are incontrovertible; and we have but to observe how, at the grand epoch adverted to, the action of general causes was necessarily modified by the peculiar features of the scene, as above indicated, in order to come at a just conception of the aspect and character of the lower Himalayan region, all along the line of the mountains. Thus the longitudinal trough parallel to the mountains, and exclusively denominated the Tarai by Captain Herbert, may to the north-west have been caused by the set of the subsiding oceanic current from north-west to south-east; but however caused, it exists as a palpable definite creature, only beneath the Thakorain and Kúmáun, is faintly traceable beneath Népál, and is wholly lost beneath Sikim and Bhútán. But the great bed of débris is everywhere present, and with no other distinctions than those pointed out, whether it be divided into Bháver and Dhún, by the Sandstone range, as is usually the case west of the Méchi, or be not so divided owing to the absence of that range, as is always the fact east of the Méchi. Again, every where there is, at that point where this vast bed of gravel and sand thins out, a constantly moist tract, caused by the percolation of hill waters through the said bed, and their issue beyond it; and that constantly moist tract is the Tarai, whether it runs regularly parallel to the line of mountains and be distinctly troughed, as to the Westward is the case, or whether there be no such regularity, parallelism, or of troughing, as to the Eastward is the case.

Why that vast mass of porous débris, which every where constitutes the appropriated domain of the Saul forest, and that imporous trough outside of it, which every where constitutes its drain, should as far Eastward as the Méchi, be both of them developed parallelly to each other and to the

of the detritus, and I followed one continuous slope and level-the main high one. The country exhibited, near the rivers especially, two or three other and subordinate levels or terraces, some marking the effect at unusual floods of extant fluviatile action, but others unmistakeably that of pristine and oceanic forces. I measured heights from the river. I could not test the sub-surface depth of the bed. There was everywhere much more sand than gravel, and boulders were

rare.

+ Saharunpúr is 1,000 feet above the sea; Múradábád 600; Gorakpur 400; Dumdanga 312; Rangpur 200; Gwalpárá 112. My authorities are As. Res. vol. xii., J.A.S. B. No. 126, Royle's Him. Bot., Griffith's Journals, and J. Prinsep in epist. The oblique dip to the plains towards the east seems to be increasing, for all the Himalayan rivers descending into the plains, as they quit their old channels, do so towards the east only. I would propose, as an interesting subject of research, the formal investigation of this fact, grounding on Rennell's maps and noting the deviations which have occurred since he wrote. The Tishta which fell into the Ganges now falls into the Brahmapútra.

line of the mountains, whilst beyond the Méchi Eastward to Assam (exclusive) they should exhibit little or no such parallellism, but should rather show themselves plainwards, like an irregular series of high salient and low resalient angles resting on the mountains, or like small insulated plateau, or high undulated plains,† surrounded in both the latter cases by low swampy land analogous to the Tarai, it would require a volume to illustrate in detail. I have given a few conspicuous instances in the foot-notes. For the rest, it must suffice to observe that such are the general appearances of the Bháver and Tarai to the Westward and to the Eastward; and that the general causes of the differences have been pretty plainly indicated above where the necessary effects of the sandstone range, of the mountain spurs, and of the Eastern dip of the plains upon those oceanic forces, to which all phænomena of the region owe their origin, have been suggested.

Throughout Assam, from Gwálpárá to Saddia, Major Jenkins assures me there is neither Bháver nor Tarai; and if we look to the narrowness of that valley between the Himálaya and the mighty and impetuous Brahmaputra, and consider moreover the turmoil and violence of the oceanic current from the N.W., when its progress was staid by the locked-up valley of Assam, we shall be at no loss to conceive how all distinctive marks of Bháver and Tarai should here cease to be traceable.‡

It will be observed that, in the foregone descriptions of our Himalayan rivers, I have not adverted (save casually in one instance, in order to correct an error as to the true name of the Káli) to their partial Trans-Himálayan sources. And I confess it seems to me, that perspicuity is by no means served by undue insistency on that feature of our rivers. Captain Herbert was thus led to travel beyond his proper limits with a result by no means favourable; for, it appears to me, that he has confounded rather than cleared our conceptions of Central Asia as the

Parbat Jowár, on the confines of Assam and Rangpúr, is one of the most remarkable of these small plateau. It is considerably elevated, quite insulated, remote from the mountains, and covered with saul, which the low level around exhibits no trace of. Parbat Jowar is a fragmentary relic of the high level, or Bháver, to which the saul tree adheres with undeviating uniformity.

+ Conspicuous instances occur round Dinajpúr and north-west and north-east of Siligori in Rangpur, where are found highly undulated downs, here and there varied by flat-topped detached hillocks, keeping the level of the loftiest part of the undulated surface. Looking into the clear bed of the Tishta, it struck Dr. Hooker and myself at the same moment, how perfectly the bed of the river represented in miniature the conformation of these tracts, demonstrating to the eye their mode of origination under the sea.

The climate of that portion of the Eastern Himálaya, which is screened from the south-west monsoon by the mountains Sonth of Assam, is less humid than the rest, precisely as are the inner than the outer parts of the whole chain. The fact, that much less snow falls at equal heights in the humid Eastern than in the dry Western Himalaya, depends on other causes Darjeeling has uot half as much snow as Simla.

Bám-i-dúnya (dome of the world) by attempting to detach therefrom that most characteristic part of it, the plateau of Tibet, because certain Indian rivers have (in part) Tibetan sources! My theory of water-sheds does not incline me to venture so far into regions too little known, to allow of the satisfactory settlement of the question, and the less so, inasmuch as the rivers I have to speak of would not afford so plausible an excuse for so doing, as if I had to treat of the Indus, Sutlej,* and Bráhmapútra alias Sánpú.† The Arun and the Karnáli, though they draw much water from Tibet, draw far more from the "pente meridionale" of the Himálaya, or the ghát line and all South of it; and this is yet more true of the Ganges, the Mónas and Tishta, though they also have partial Trans-Himálavan sources. To those sources of the several Himalayan (so I must call them) rivers above treated of, I will now summarily advert:

The Monas.-It is by much the largest river of Bhútán, which State is almost wholly drained by it. It has (it is said) two Tibetan sources, one from Lake Yámdotsó vel Palté vel Yarbroyum, which is a real lake, and not an island surrounded by a ring of water as commonly alleged-the other, from considerably to the West of l'alté. These feeders I take to be identical with Klaproth's Mon-tchú and Nai-tchú vel Lábnak-tchú, strangely though he has dislocated them.

The Tishta is also a fine river, draining the whole of Sikim, save the tracts verging on the plains. The Tishta has one Tibetan source, also, from a lake, viz., that of Chólamú. To speak more precisely, there are several lakelets so named, and they lie close under the N. W. shoulder of Pow

Recte Salaj vel Satrudra.

+ Dr. Gutzlaff, once read a paper before the Geographical Society of London, and reverted to Klaproth's notion, that the Sánpu is not the Brahmaputra. But Mr. Gutzlaff overlooked J. Prinsep's important, and I think decisive argument on the other side, viz., that the Brahmapûtra discharges three times more water than the Ganges, which it could not do if it arose on the north-east confines of Assam, notwithstanding the large quantity of water contributed by the Mónas. Yárû or Yerû (Erû) is the proper name of the river we call Sánpf, which latter appellation is a corruption of the word Tsangpo, referring either to the principal province (Tsang) watered by the Yáru, or to the junction therewith, at Digarchi, of another river called the Tsang, which flows iuto the Yáru from the Nyenchhen chain or Northern boundary of Southern Tibet. vel Arú is the proper spelling. But words beginning with the vowels á and é, take initial y in speech. take this occasion to observe, in refereuce to the Yamdo lake above mentioned, that it is not, as commonly described and delineated in our maps, of a round shape, but greatly elongated and very narrow. It is stated to me on good authority to be eighteen days' journey long (say 180 miles), and so narrow in parts as to be bridged. It is deeply frozen in winter, so as to be safely crossed on the ice, whereas the Erú river is not so, owing to the great force of its current a circumstance proving the rapid declivity of the country watered by this great river.

Erû

Tsang po means simply 'river,' and should not be called Sanpo but Tsang po.-J.S.]

The Arún is the largest of all the Himalayan rivers, with abundant Cis-Himàlayan and three Trans-Himalayan feeders. One, the Western, rises from the "pente septentrionale" of the Himalaya, in the district of Tingri or Pékhu; another, the Northern, from a place called Dúrré; and a third, the Eastern, from the undulated terraced and broken tract lying N. and a little W. of Chólamú and S. of Kambala, or the great range which bounds the valley of the Yárú on the S. from W. of Digarchi to E. of Lhása.

The Karnali is much larger than the Alpine Ganges, and nearly equal to the Arún, perhaps quite so. It drains by its feeders the whole Himalaya between the Nanda-dévi and Dhoula-giri peaks, and has itself one considerable Tibetan source deduced either from the north face of Himachal near Momonangli, or from the east face of that crescented sweep, whereby Gángrí nears Himachal, and whence the Karnáli flows eastward to the Taklakhár pass.

The Ganges also has of late been discovered to have one Tibetan feeder, viz., the Jáhnavi, which after traversing a good deal of broken country in Gnári, between the Sutlej and the Himálaya, passes that chain at the Nilang Ghát to join the Bhágarathi.‡

I will conclude this paper with the following amended comparative table of Andean and Himálayan peaks, Baron Humboldt having apprised me that Pentland's measurements, as formerly given by me, have been proved to be quite erroneous, and Colonel Waugh having recently fixed Kangchan and Chumalhári with unrivalled precision and accuracy:

Chief peaks of Andes, Feet. Chief peaks of Himalaya. Feet.

N.B.-Dévadhúnga vel Bhaíravthán vel Nyanam, half way between Gosain-thán and Kangchan, is 29,002, determined in 1856. Kang-chan, abounding in snow.' Chúmálhári, 'holy mountain of Chúmá. These are Tibetan words; the other names are Sanskritic, but set down in the Prakritic mode, e. g., Jamnavatari equal to Jamnoutri, etc.

*The valley of the Yárú is about sixty linear miles from the Sikim Himálaya (Láchén and Donkia passes); but the intermediate country, called Damsen, is so rugged, that it is ten stages for loaded yáks from the one terminus to the other. Damsen is stated to be one of the most rugged and barren tracts in the whole of Utsang or Central Tibet, a howling wilderness. —Hooker.

Moorcroft's Travels, J. A. S. B. No. 126, and I.J.S. Nos. 17-18.

§ Humboldt, in his Aspects of Nature, has given some further corrections of those heights. There are three peaks superior to Chimbarazo, but inferior to Aconcagua.

Tsandagotta or Chandragupta (Sandracotta) who is said to have died in the year B.C. 357, or 186 years after the death of Gautama. Chandragupta came from the line of Maureya, a place not named in the list, and was placed on the throne of Palibothra through the instrumentality of "Janneka;" a Brahmin from Taxila or Tekkasila. This brings me to another question which requires discussion, viz., who were the "Yavana or Yona"? The name in full is similar to "Yavan" of the Jewish books, and, when contracted, to Iw of the Greeks, but these people appear in India before the time of Alexander. Were these people Ionians? and had they anything to do with the rise and spread of Buddhism? One of the missionaries sent by Asoka to Aparanta (Burma?) was called Yonaka Dharma Rakshita.

R. F. ST. ANDREW ST. JOHN.

Review.

Peking, Jeddo, and San Francisco. The Conclusion of a Voyage round the World. By the Marquis de Beauoir. Translated from the French by Agnes and Helen Stephenson. With fifteen engravings from photographs. London: John Murray. 1872.

This is a spirited little volume of jottings by the way over a wide field of travel. The author has all the vivacity of the Frenchman, and the practical observation of the English traveller. But that which enhances the value of the book in our eyes is the pretty engravings which bring the scenes before us with perfect truthfulness. There is little need for the author to prefix the appropriate motto: "J'étais là; telle chose m'advint," for there is evidence of this without stint in the book itself. What could be more exact than the pictures of the kango and its fair occupant, or the mandarin's "cabriolet," or the Japanese colonel in all his glory of regimentals? The author leads us then to contemplate the vast Wellingtonias of California, and the great wooden bridge of the Pacific Railway.

Amid so much that is true and interesting we cannot be surprised to find occasional slips. Where information is picked up from stray acquaintances in foreign climes we sometimes find that a trifle of exaggeration tinges the communications, and the unwary traveller having little time to sift matters, commits himself to sundry errors. As an instance of this he tells us that the Tokaido is a "long road which traverses the whole of Nippon, from the

south-west point of Nangasaki to the north-east extremity at Hâcodâde." This is incorrect, for the Tokaido extends merely from Yedo to Miako (the capital), and again Nangasaki is in the island of Kiusiu, and not in Nippon, the large island of Japan. But so much exactness is perhaps not needful in a work of this nature. Most of the descriptions are well done, and are pleasant reading. Our traveller was badly treated on an excursion to Nang-kao near Peking; a smart piece of writing details a scene in this adventure. The author says:

"For one short moment I no longer know what is happening, except that a great Mongol pins me by the collar of my coat while three others take my necktie, money, the chain of my watch, and I know not what besides (luckily I had had time to break the ring of my watch and slip it into my boot). The chaplain's gold creates an enormous excitement; a few dollars found on the prince collects fifteen howling scoundrels in a bunch hanging from his pockets. Louis, who carries the Chinese bank-notes, look like the distributor of advertisements on our boulevards."

We notice several inaccuracies of spelling in the Chinese and Japanese words, but these will not much interfere with the intelligibility of the story. The Marquis de Beauvoir's experience of Chinese carts, and his visit to the Japanese arsenal are shewn in the following extracts:—

"They are curious machines, these Chinese carts; a kind of litter of blue cotton is attached in very shaky fashion to an axletree about three feet long, and to two heavy wheels; you cannot lie down because it is too short, nor can you put in a seat to sit upon for it is too low. On the other hand, it is an extremely light carriage and can go anywhere. I ensconce myself as best I can in it, with the help of a sack of bran to act the part of springs; my muleteer seats himself on the near shaft, jumping to the ground each moment to excite his beasts noisily and even cruelly; the shaft mule will only pay attention to his voice, and our safety depends on its humour; its harness consists only of two very long traces, fastened together to the axletree near the left wheel; it can only pull, therefore, on one side, and always trots sideways.

For the first hour we are perfectly bewildered. The high-road-if that name may be given to such a remnant of the dark ages as this track-varies from six feet wide, in narrow places, to a hundred and fifty or two hundred feet, in the open country, and near the villages this sea of dust is strewn with thousands of corners of paving-stones, or old brickbats, which send one up into the air like a ball off a racket. It is in these places that the obstinate muleteers take a particular pleasure in urging their animals to full speed, and you may suppose what clouds of dust our caravan then raises! We are almost choked, and, when I venture to open my eyes, I can see neither the cart in front of me nor even my own shaft-mule nor the sun, which is merely a dull red speck in this strange mist. Any one who has not experienced similar joltings and such innumerable bruises, can have no conception