river, but they are all placed in localities well protected from the tides; I do not think they are increasing in size on the whole, though no doubt small changes take place now and then. The higher ground at some parts of the coast mentioned above also tells against a delta formation, unless it can be shown that these hills were formerly separate islands, which in course of time have become one with the mainland. The really dry land is, as far as Rewa town on the three southern branches, not more than 2 or 3 feet above ordinary high-water mark, and even a very heavy flood does not cause an extra rise of one foot down here, although it may swell the river at Viria to 47 feet above the usual level, and some 18 or 20 feet at Nansori.

The Laucala branch of the river is the most important, owing to its being the shortest waterway between the capital, Suva, and the Rewa district, the most populous and best cultivated part of Fiji. The place where this branch breaks away from the main river is called Hellsgate; it is a difficult narrow to pass through without running aground, as the channel is continually changing from one side to the other of the little island called Kata-na-elewa. Nearly opposite Hellsgate lies Rewa, the largest native town in Fiji; it has a Roman Catholic chapel and also a Wesleyan church. The men are known as good canoe-builders, and the women make some of the best crockery ware in the whole group. Opposite the town the river is about 400 yards broad, but from here upwards it diminishes pretty regularly in width, the banks generally increasing in height. The land on both sides continues quite flat, and is cultivated by Europeans or natives, nearly all the way up to Nansori, where the first hills are met with. Mould resting on clay layers is the usual formation of the soil found on the flat land along the river banks. On the chart will be noticed a red dotted line drawn almost parallel to the river bed at Vuci-Maca; this signifies an older bank of higher elevation. Such old banks are found in several other places, and even now they often confine the river during very heavy floods. The hills are nearly all of a uniform height (from 80 to 100 feet), and are evidently formed by pluvial denudation; most of the fertile soil has been washed off them on to the plains or into the river, leaving exposed steatite (soap-stone) or some other heavy clay resting on a hard sedimentary base-rock. Trees grow well enough on them, but as yet they have not been much cultivated for useful purposes. Having become narrower and deeper just opposite the Nansori Sugar Mills, the river again widens out opposite Elliston, where a large sand-bank has been formed and is now half-covered with grass. Another old river bank, higher than the one at Vuci-Maca, commences near Elliston, and extends beyond the native village at Verata. A little further up, at Dreke-in-kelo, the river makes a very sharp turn to the westward, forming quite a right angle with its previous course. From the mouth to this point is called "Lower Rewa," from here to Bau Levu is "Middle Rewa," and the rest is "Upper Rewa.” During heavy freshets a perfect whirlpool is formed at Dreke-in-kelo, and the depth of water on this spot considerably increased. Boats have no chance whatever of getting out, if once drawn into it, and even steamers under full speed have been carried round and round for hours

together. It is quite an amusing sight to watch the very mixed collection of floatables of different description, which is here kept dancing round in large and small circles. Native canoes are often torn away from their fastenings by the sudden rise of the river and carried away, but they never get beyond this place as long as the freshet lasts. From Nadurulolo the river has a comparatively straight course to Naitasiri; on either side the flat land is found in narrow strips only, and in several places bushcovered hill-sides descend to the water's edge. At Naitasiri a range more than 200 feet high commences on the right bank, and takes a westerly direction; on the opposite side lies an extensive flat between the river and the foot of the hills to the eastward. Owing to the tortuous course between here and Bau Levu, combined with a narrower bed and rocky bottom and banks, with generally a swifter current, the navigation requires considerable care and local knowledge. From Ban Levu to Vuni-cibicibi another large flat is formed between the river course and the foot of the hills on the right bank. Opposite Muanaweni the river for a short distance gets broader and very shallow; this is probably caused by the river water from above meeting for the first time the checking influence of a contrary tide: under certain circumstances the water comes to a stand-still, whereon many of the particles carried by it in suspension are precipitated. Above Wai-Dina, at a place called Viria, is the last flat of any importance: it is used, like most of the others on the lower river, for the cultivation of sugar-cane. A couple of days' steady rain in the mountains will often cause the river at Viria to rise 30 or 40 feet, and it has been as high as 50 feet above the ordinary level. With the exception of a few patches of land on which the natives grow bananas, there is no cultivation on the Rewa above Viria. No steamers go beyond this place, and canoes only are used for traffic further into the interior. The river from here upwards keeps a pretty uniform width of less than a hundred yards; on both sides the banks are formed by the slopes of bush-covered hills, seldom more than 100 feet high; low islands formed of coarse gravel are numerous, and about half-way between the Wai-inmala and Viria the first rounded pebbles and stones are met with in the river-bed. The land of the surrounding district has not yet been surveyed, and the exact courses of the numerous creeks are therefore still unknown this must be borne in mind when looking at the chart, where only their junction with the main river is given. Most of them drain hills and valleys many miles away, but, as I could not give their precise course, I have not shown any.

The distances to which under ordinary circumstances the different materials are carried away in suspension or pushed along the bottom by the river water are rather distinctly marked on the Rewa. Stones do not reach Viria; coarse gravel settles between there and Nadurulolo, finer gravel and sand are found in the Lower Rewa, while the mud is carried down to the sea. I do not mean to say that there is no mud in the riverbed, or no sand in the Middle or Upper Rewa, but that no final settlement of these finer substances takes place until after the water has lost power to move stones or pebbles and carry gravel in suspension. Where mud is found in the river-bed it is still, though perhaps slowly, in

motion towards the coast; the ebb-tide brings a large quantity with it to the sea, but the flood-tide also carries a considerable quantity of mud and sand, which had nearly reached the coast, up-river again. Thus many of the finest particles held in suspension by the water probably have to pass down and up the part of the river affected by tides a couple of times before they ultimately reach the ocean.

It is worth notice that on the flats the land close to the river banks is generally higher than that further away; when, therefore, floods occur here, they are not caused by the river overflowing its banks, but by the swelling of creeks near the foot of the hills, or by the water flowing from the river into the back-land through the same creeks, which under ordinary circumstances bring water to the main stream. As an average result of a good many trials at Nansori during a flood, which caused a rise of about 18 feet over ordinary water, I found that the speed of the current in the middle of the river did not exceed 4 miles per hour. At the same time the matter held in suspension was 1000 by weight; this is about ten times more than under ordinary circumstances. These figures were obtained by weighing a quantity of river water and finding the weight of the residue after evaporation.

The only evidences of active volcanic action in the Rewa basin are some hot springs in the Wai-Dina, too far to the westward to be shown on the chart. I have not seen them myself, or been able to obtain a reliable detailed description of them.

The scenery on the Rewa is totally different from that on the Siga Toka there primeval Nature displays a grand variety of forms; here almost everything has been interfered with by the hand of man. Signs of civilisation are met with everywhere in the form of well-built cottages, with pretty gardens, and large plains planted with sugar-cane. Steamers, cutters, boats, canoes, and bamboo rafts are continually on the move; permanent tramways, worked with locomotives, have been constructed at Nansori, Naitasiro, Bau Levu, and Viria; and at Nansori for six months of the year the electric light turns night into day. Even the hills have not been spared, and a tunnel through a range at Viria allows sugar-cane grown along the Wai-Dina to be transported by rail to the Viria Mill. Above this place Nature holds the sway; the scenery, however, is very monotonous; nothing but bush-covered hills, where the trees are so hidden by creepers that it is almost impossible to determine their species; here and there fern-trees vary the general effect by thrusting their graceful fronds high above the other vegetation. Coco-nut palms, so conspicuous and numerous on the lower river, are very scarce up here, for they love sea air and salt water, and do not thrive far from the coast.

From the above description of the Rewa the following main points may be selected :-(1) The headwaters of its principal affluents have their springs in a very mountainous district of decided volcanic origin. (2) The basin generally consists of a series of irregular ranges or isolated hills of argillaceous composition, with alluvial flats between them or along the river banks. The coast is protected by coral reefs, with openings corresponding to the different mouths of the river.

A few words about these coral reefs may not be out of place here.

It is generally supposed, as stated by Darwin, that Viti Levu is one of those islands which are gradually being submerged, while their barrier reefs are being built up as fast as the dry land disappears. I trust, however, it will not be looked upon as presumption on my part if I state that I am of a different opinion. To support the conclusion I have arrived at, I would point out

(1.) At Ovalau, an island about fifteen miles to the north of Kaba, the reefs are exactly of the same description as those at the mouth of the Rewa; in fact, they may be considered a continuation of the latter, interrupted in two places only by narrow channels. Opposite the port of Levuka, on the east coast of Ovalau, there is a passage through the reef which admits a considerable sea during strong easterly winds the coast opposite this opening is in some places formed of hard conglomerate rock, which for a considerable distance along the shore has been thoroughly undermined and eroded by the action of the sea, at a height of two or three feet above the line at present reached by the water during spring-tides.

I consider this a decided proof of emergence, and, if Ovalau is rising, it is highly improbable that Viti Levu is sinking.

(2.) The coral reefs on the south-east coast of Viti Levu are exposed too long above water during ebb-tide to be "growing." Some parts of the reefs are exposed at half-tide.

Such are my principal reasons for believing that the island has been slightly rising in recent times, though it is hard to say whether it is still rising or stationary.

For Rewa district this is an important question. If the river navigation should become considerably more difficult than at present, a railway between this district and Suva would become a matter of necessity for the transport of sugar to the mills.

The accompanying chart cannot boast of artistic execution, but I think it will be found quite correct.

Note.-In pronouncing Fijian words the following rules should be noted :

Letter a is pronounced as in the word lamb.

A SPECIAL MEETING of the Society was held in Edinburgh on the afternoon of Friday the 17th July. Mr. A. L. Bruce presided. A paper on "Matabele-land and Mashona-land" was read by Mr. Denis Doyle, who was accompanied to the platform by Gungunhana's envoys, Huluhulu and Umfeti. Dr. George Smith moved the vote of thanks to the lecturer.

GEOGRAPHICAL NOTES.

BY W. A. TAYLOR, M.A.

EUROPE.

The Population of Norway.-On January 1st, the total population of Norway amounted to 1,999,176 (Forelöbige Resultater af Folketællingen i Kongeriget Norge). In 1876 the number was 1,813,424, so that the increase in the last fifteen years has been 185,752, or 102 per cent. The numbers of inhabitants in the larger towns are as follows, the figures in brackets indicating the population in 1886 :Kristiania, 150,444 (130,027); Bergen, 53,686 (47,075); Trondhjem, 25,051 (23,979); Stavanger, 23,930 (23,911); and Drammen, 20,684 (19,601).

Migrations of the Lapps.-During the last two centuries the Lapps of Norway have been moving gradually southwards, preserving their uncivilised and nomadic mode of life in their new environment. Dr. Yngar Nielsen of Christiania has recently studied this interesting ethnological question (Vide Le Tour du Monde, Nouvelles Géog., p. 137). According to him the southern limit of this people is now marked by the railway from Trondhjem to Östersund, nearly along the 63d parallel of N. lat. To the north of this line are found ancient tombs, places of worship, and names of Lappish origin. Here the Lapps of the present day, though nominally converted to Christianity, retain in secret some of their pagan customs, whereas further south they are good Christians, and have changed even in type. About the year 1600 the southern limit of the Lapps was on the parallel of the northern extremity of the fiord of Trondhjem; since then they have made several excursions southward, and have been repeatedly checked by the Norwegian Government. In 1890 they advanced as far as the plateaus of the Hardanger Fjeld. The Norwegians do not resort to violence, but defend their property by legal processes. The question of the Lapp invasion is, however, one that demands the serious attention of the Government.

Ascent of Mt. Elbruz.-In July 1890, G. A. Pastukhoff, of Tiflis, ascended this mountain with the object of making topographical surveys. Both summits are caldron-shaped, and on the edge of each rise three distinct peaks. The highest of these on the eastern summit is 18,353 feet above the sea-level. (The height of the western summit is 18,470 feet. See Scottish Geog. Mag. vol. vi. p. 319). Both summits are undoubtedly extinct craters. On the descent M. Pastukhoff and his men were nearly lost in a violent snow-storm.-Deutsche Rundschau, Jahrg. xiii. Heft 9.

ASIA.

Trans-Asiatic Journeys.-M. Paul Barré contributes to the Revue Française (April 15th, 1891) a short paper on this subject, from which we extract the following. The Dutchman, Ruysbroeck, visited Mongolia between 1246 and 1273, but, though he advanced far towards the east, he did not succeed in reaching the Chinese coast. The first European to traverse the whole continent was Marco Polo (1271-1295), who, passing through Turkestan and China, entered Pekin, and extended his journey even to Japan. Irmak Timofeef, a Cossack brigand, opened Siberia to Muscovite influence (1530); and Elisée Bouza (1635), Kopylof (1639), and Sladukhim and Ignatief (1644) succeeded in reaching the north-western limits of this country. Dejnef, in 1648, reached the Gulf of Anadir, and ascertained the existence of a strait between Siberia and America before Behring sailed to that