river. Further on they ascended the river Kobdo, which flows rapidly between wooded and hilly banks. Here, in contrast with the Tsagan-gol valley, the slopes of the hills are covered with trees, and there is plenty of rich meadows and pastures.

Continuing their journey up the Kobdo river, in a south-western direction, the party came to the spot where the river forms two small lakes, on the banks of which there are several Kirghiz villages. The lower Kobdo lake has an irregular elongated form, stretching 10 miles from east to west, with a width of about 5 miles. There are in it fifteen islands, of which some are from 2 to 2 miles long. Taking advantage of the boat which they brought with them, the explorers made a series of soundings of the lake, and visited the slands. The upper lake is connected with the lower by a rapid watercourse, 420 feet wide. The upper lake proved much larger than it was supposed to be; its length is about 16 miles, and its width 4 miles; its depth attains 18 fathoms. From these lakes the party travelled to Lake Dain-nor, and found on the way numbers of graves with ancient tombstones. The circumference of this last lake is 13 miles, and it is 4 miles wide. The pass Ak-korum is not very high, but difficult, on account of the great quantity of water which covers the stony ground and makes it slippery. Further on the party went along the valley of the river Saksai, which they found similar in nature to the Tsagan-gol; then they reached the Buyuntu, which flows first in a stony bed amidst thickets of bushes, and further on in a narrow and low gorge. Here and there small terraces, covered with trees, chiefly willows and ash, were seen along the banks. The population consists of Olot-Kalmuks. The gorge was followed for about 50 miles, when the valley suddenly widened, attaining a width of about 7 miles; the mountains on both its sides became very low, and a grand view opened upon a high snow-clad range, which runs parallel to the gorge to the right of it. A little further on this second party saw Kobdo, which they reached on August 28, having covered 378 miles in twenty days. At Kobdo the expedition stayed for nine days, buying provisions for its further six months' journey; and then, on September 6, they started towards the east-south-east, along the northern foot of the Gobi Altai.

THE MONTHLY RECORD.

EUROPE.

Oxford Degree for Physical Geography.-Some years ago the University of Oxford instituted the degree of Bachelor of Science to be awarded for original research. It has just been conferred on Mr. H. N. Dickson, of New College, University Lecturer on Physical Geography, whose research took the form of a comprehensive study of the temperature and salinity of the surface water of the North Atlantic, with special reference to seasonal variations and the influence exercised on the climate of Western Europe. The result is to throw much light on the changes in the circulation of the ocean according to the season, and on the interaction of sea-surface temperature and climate. The examiners were Sir John Murray and Prof. Odling.

The Birmingham Water-supply from Wales.-An interesting account of the Birmingham water-supply scheme, now in course of execution, appears in the fifteenth volume of the Proceedings of the Royal Institution of Great Britain,

being a paper read last year before the Institution by Mr. James Mansergh. The drainage area, over which control was obtained from Parliament, lies in the western corner of Radnorshire, with parts of neighbouring counties, including the basins of the Elan and Claerwen, western feeders of the upper Wye, the distance to Birmingham by the route chosen being 80 miles, as compared with 100 miles from Thirlmere to Manchester, and 66 from Vyrnwy to Liverpool. The projected Welsh scheme for London involves an aqueduct of 170 miles. The area of supply is 45,562 acres, while from the record of rainfall kept by the lord of the manor from 1871 onwards, the mean annual fall of a long series of years is taken as 68 inches, and that of three consecutive dry years as 55 inches. From this it is calculated that, allowing for evaporation and loss by overflow, a supply of seventyfive million gallons daily will be obtained, in addition to twenty-seven millions, the amount fixed as 66 compensation supply" for the river below the lowest dam. As the discharge at this spot has been estimated at only four and a half million gallons during very dry weather, the benefit to the river is evident, apart from the additional service done by obviating disastrous floods. The geological features of the area favour the operations for reservoir construction. At a point called Caban Coch, a little below the junction of the Elan and Claerwen, bands of hard grits and conglomerates cause a contraction of the valley, and point out the best spot for the construction of the lowest dam. This will be 122 feet high, and will impound the water for a distance of 4 miles up the Elan valley, and 2 miles up that of the Claerwen. Spots have been chosen for the erection of five more dams higher up the valleys, two in that of the Elan, and three in that of the Claerwen, the positions being determined by the fact that they afford the greatest impounding capacity with the least amount of structural work. An interesting and unusual feature is the provision of a "submerged " dam within the lowest reservoir, so as to provide the necessary head of water (770 feet above ordnance datum) for the Birmingham supply in times of drought, while leaving that contained in the rest of the reservoir, at a lower level, available for the "compensation " supply to the river. At such times the former supply will be derived from the upper reservoirs, a drought of 180 days being thus provided for. The drainage area above Caban Coch is twice that dealt with in the case of Vyrnwy, and four times that of Thirlmere. In time of high flood, the discharge may amount to 700,000 cubic feet a minute, and the surplus water will pour over the whole 600 feet of the dam, with a depth of about 3 feet at the crest. Almost the whole collecting area consists of mountain pasture or moorland; but in the lower parts of the valleys a few small farmsteads will be submerged, as well as the church and old manor-house of Nant-gwillt and Cwm Elan House, once the residence of Shelley.

Anthropogeography of Corsica.-An instructive study, by Prof. Ratzel, of the effect produced by the physical geography of Corsica on the course of its history, appears in the Annales de Géographie for July 15 last. The writer begins by examining the influence exercised by the position of the island with respect to neighbouring lands, and to the enclosed basin of the Tyrrhenian sea, the historical importance of which has always been greater than that of the more open sea to the west of Corsica. By its more northerly position, Corsica was further removed than Sardinia from the influence of pre-Roman civilization, and was long placed at a disadvantage from the prior attention directed to the latter. Naturally, the island is most nearly connected with the Tuscan coast of Italy, being further removed from France, and still further from Spain. Its relations, however, with the nearest part of the Italian coast (Pisa) lasted for a comparatively short period (1098-1348), being followed by Genoese domination, while the rivalry between Genoa and

France finally brought about its connection with the latter. The shifting of political ascendency thus followed a regular course from E.S.E. (Rome) to N.N.W. (Marseilles and France). Although an island, Corsica was too small and too near other land-areas to maintain an independent existence, but its insularity has been sufficient to impress a decided character of their own on the life and thought of the inhabitants. As regards the physical structure of the island, the dominating fact is its division, broadly speaking, between the granitic area of the west and south and the schistose area of the north-east. The line of division is marked by an important depression, parallel with the principal lines of elevation, which was covered by the sea during the Tertiary epoch. The main water-parting of the island, nearer the west coast in the north, and the east in the south, plays a very different rôle in the granitic and schistose areas, presenting greater facilities for communication across it in the latter. The character of the surface in the two regions is also very different, the north-eastern section being composed of undulating heights running generally north and south, while the granitic area is a land of mountains rising abruptly from the sea in parallel ridges, forming long and narrow valleys, and gulfs cutting deeply into the coast-line. Socially it is the land of shepherds and "signori," whilst historically it lies off the track of great events. In the extreme north and south are minor areas, distinct in many ways from the rest of the island. The region of the north-east, though in area only a fourth part of the island, forms in history practically the whole of Corsica. In it one of the most important positions is occupied by Corte, the radiating centre of the principal lines of traffic, and the point of contact between the mountains and the coastlands; and it contains as well a whole series of sites of historical importance. This region has a striking counterpart in the south-west of Sardinia, in which the co-ordination of geographical features, and the resulting historical phenomena, are strictly analogous. The concluding section of Prof. Ratzel's paper deals with the contrasts offered by the east and west coasts, and the reasons which make the latter, with its many indentations, an exception to the rule that a broken coastline favours the development of a country.

ASIA.

New Surveyor-General of India. The Indian Survey Notes for September, 1899, contain the announcement of the retirement from the post of SurveyorGeneral of Major-General C. Strahan, R.E., and of the appointment in his place of Lieut.-Colonel St. George C. Gore, previously Superintendent of Trigonometrical Surveys. General Strahan was the last officer in the active service of the Indian Government of the corps of engineers raised under the East India Company, the distinguished record of which is thus brought to a close by his retirement. He had served in the Survey Department for thirty-six years, during the last four and a half of which he had been at its head. An important part of his work had been connected with the topographical surveys of the Native States of Bundelkhand, Central India, etc., where the present system of plane-table surveying may be said to have been elaborated and brought to perfection. To this work he had devoted himself for no less than twenty years. The place of Colonel Gore as Superintendent of Trigonometrical Surveys is taken by Major S. G. Burrard, R.E., one of the officers entrusted a few years ago with the re-determination of the longitude of Madras. Another announcement is that of the death of Hira Singh, a native surveyor of the Department, who did much good work on and beyond the north-west frontier while attached to various military and other expeditions, including that of the Afghan Boundary Commission.

Rhinoceros Remains from Eastern Mongolia.-We have received from M. Obrucheff, the well-known Russian traveller, the copy of a pamphlet on the discovery of the remains of rhinoceros in Eastern Mongolia during his journey of 1892-94. The identity of the remains, which consist principally of a lower jaw with fragments of teeth, has only recently been established by Prof. Suess, whose report forms the first part of the brochure. They are of interest as establishing the fact that the sedimentary formation in which they occurred is a fresh-water deposit of at most Middle Tertiary age, and thus shedding valuable light on the geological history of Eastern Central Asia. M. Obrucheff gives a description of the geological features of the neighbourhood of the Khuldyin Gobi plateau, on the edge of which he discovered the remains. The plateau lies near the salt lake Iren-dabassun-nor, on the caravan route from Urga to Kalgan. Its border is much broken into isolated fragments, where the geological formation is easily determined. The upper layer is a whitish fine-grained conglomerate, beneath which is a whitish marl with intermixture of clay, etc., while the lowest layer is a brownish-red marl. The rhinoceros remains were found in the middle layer. From the general lie of the country, M. Obrucheff concludes that the strata of the Khuldyin Gobi, which occupy the lowest part of the whole region, form the oldest member of the extensive Gobi deposits of Eastern Mongolia. These being now proved, as above stated, to be of fresh-water origin, it is found that the ancient Inner Asiatic sea or series of lakes consisted from the beginning of fresh water, and existed during the latter half of the Tertiary period. These conclusions agree with those of Prof. Lóczy, the geologist of the Széchenyi expedition, though M. Obrucheff had before doubted the fresh-water character of the ancient sea, having taken the remains now described for those of a large fish.

The Geological Structure of the Malay Archipelago.-A concise summary of our present knowledge of the geology of the Malay archipelago, with especial reference to the main lines of tectonic structure, is given in the Journal of the Tokyo College of Science (vol. xi. part 2), by Prof. B. Koto. The basis of the description is Prof. Suess' chapter on the archipelago in 'Das Antlitz der Erde,' but Prof. Koto has extended and modified his account from the most recent researches. Beginning with the great Malayan arc, which extends from Burma to the vicinity of Timor and forms the outer portion of the archipelago, the writer traces in turn its various component parts, especially the great volcanic belt running through Sumatra, Java, etc. The outermost member he considers to run from Cheduba off the Arakan coast, through the Andamans and Nicobars to the line of islands which fringe the south-west coast of Sumatra, the volcanic belt being represented in the north by Narcondam and Barren island. In the east this seems to terminate in the island of Pantar, north-west of Timor, being separated by the latter and other islands near it (all remarkably free from volcanoes) from the volcanic arc of Banda, which may thus be regarded as independent of the great Malayan arc. Prof. Koto is inclined to cut short the Banda arc, which begins with Roma in the south-west, at the Banda group, believing that Wawani on Amboina cannot be established as an active volcano. On the final decision of this question it depends whether or not a connection exists between the volcanic chains of Banda and the Moluccas. Outside the Banda arc runs a parallel chain of islands from Kisser northwards-the shattered relic of an old mountain range, while a third external zone has also been indicated. This threefold arc, of which the inner zone is of young volcanic origin, has a striking counterpart in the lesser Antilles, and still more in the Riu-kiu islands. Within the great Malayan curve lie the three K-shaped islands of Borneo, Celebes, and Gilolo, the external resemblance of which is, however, only superficial. To the south their relation with the outer curve is

not clear, but their geologic lines converge unmistakably in the Philippine group, presenting a striking instance of "virgation." In the south-west Borneo possibly falls on the tectonic line which runs from the Malay peninsula through Banka and Billiton, while in its centre tectonic lines running from east to west have lately been discovered, and neo-volcanic rocks have unexpectedly been found to exist. The north-eastern peninsula of Celebes lies on the volcanic belt which, beginning in the Gulf of Tomini, runs north through Sangi, and in Mindanao joins the chain of the Moluccas to form the great volcanic belt of the Philippines. This is not considered by Prof. Koto to run the whole length of Luzon, but, after traversing its south-east section, to cross the sea to the extreme north-east corner. The other tectonic lines of the group are somewhat complicated. The map accompanying the paper is an excellent specimen of Japanese lithography.

AFRICA.

The Egyptian Problem.*-Although Mr. Silva White's recent work is written with avowed political motives, its aim being the advocacy of a British protectorate over Egypt, the political questions involved are so far bound up with geographical considerations that the book contains much matter of interest to geographers, as well as to the wider circles in which a subject so nearly affecting the future of the British Empire cannot fail to arouse attention. From first to last the author follows a continuous line of argument, in which he attempts to make each step the logical sequence from what has preceded it, the starting-point of the whole being the geographical description of the Nile valley. Mr. White has a striking and original way of presenting his facts, and his volume is decidedly suggestive, while, apart from its more definite object, it is of value as presenting a clear view of recent Egyptian history, its administrative system, and the progress it has made under British control. In the discussion of the physical factors, Mr. White's central idea is the organic unity of the Nile valley and the consequent need that it should be under one political control. He is, perhaps, inclined to lay too much stress on the isolation of the Nile valley from the rest of the continent, and to overlook the differences which exist between its component parts, which, but for the connection supplied by the river itself, could hardly have been brought into close relation on purely natural grounds. After dealing with Egypt proper, the author gives an interesting chapter on the Sudan, the section dealing with the "title-deeds of the Nile valley" being useful as bringing together the various international agreements on which the Anglo-Egyptian rights depend. Mr. White urges the importance of fixing a definitive frontier with Abyssinia, the absence of which is the one weak point in our settlement of pending questions. He regards the Lybian desert as placed within the Egyptian sphere by the recent agreement with France, but it is to be observed that Article III. of the Declaration merely conceded the French claims up to a certain line in this direction, without saying anything of the status of the country beyond it, which most naturally forms the Hinterland of Tripoli. Again, the dissatisfaction felt in Italy was not so much due, as Mr. White supposes, to the idea that Tripoli became a French sphere, as

*The Expansion of Egypt under Anglo-Egyptian Condominium.' By Arthur Silva White. London: Methuen. 1899.

+ Mr. White makes a curious slip in stating that the Nile basin would contain the area of a hundred British isles. For a hundred, ten should of course be read.

The trade route from Bengazi to Wadai renders this otherwise valueless tract

of some importance to the possessor of Tripoli.