been constructed that the waterway was insufficient, and that in both cases an extra span of 65 6 feet was needed.

Very few difficulties were encountered in the construction of the masonry for the abutments, as this work could generally be executed in dry ravines. In crossing certain rivers and marshes, however, the water present involved rather heavy hand-pumping. The breast-walls of the metallic culverts were constructed in 18-inch rubble masonry, built up flush with the ends of the tubes. In some instances, when the volume of water to be dealt with was large, three or more of the metal tubes were laid side by side and even one above the other, so as to give a sufficient area of outlet, both in width and depth. In these cases the breast-wall was of course made to enclose the ends of the entire group of culverts. In order to provide means for building the abutments of the bridges the rows of piles were so arranged as not to interfere with the work, and the temporary bridges were so constructed as to leave a free space between the piling for the stonework. As soon as the abutments were finished all that remained was to bring

Fig. 2.

forward and fix the iron superstructure. Only cast steel and mild steel were employed in the bridge-work. In the first 180.8 miles, there are in all 110 bridges, which vary in span from 328 feet down to 13.1 feet. The bridges first erected were riveted together in the usual way, but this process proved so defective, so costly and so exceedingly slow, when executed by negro riveters, who were novices at the work, that in all the later ironwork recourse was had to turned bolts, which succeeded extremely well. The bolts were made in a special way by machine, and the Author states that the results in actual practice by the use of bolts were better than those obtained by means of riveting.

Many of the bridges are on very steep gradients, and in one case not only was the gradient 1 in 35, but the line was on a curve of 2 chains radius. The span of this bridge was 131.23 feet. The Author describes and illustrates by means of a diagram, Fig. 2, the special method of construction adopted in the case of this unusual bridge. The lattice-work on the outer side of the curve is straight, but on the concave face the girder is spread out at an angle towards each end and thus bears on to the abutments obliquely, the centre part of the lattice-work remaining parallel to the outer girder. In order to secure the gradient of 1 in 351⁄2,

the top flange, or roadway, finishes at this inclination, while the bottom of the girder is made level to avoid the danger of shifting downwards on the abutments-the girder is thus trapezoid in form. Certain special features in the construction of other bridges are described in detail by the Author. In all cases the design of the ironwork was made as graceful as was possible without entailing any pecuniary sacrifices. Certain of the cast-steel bed-plates for the bridges were too large to be made in one single piece, and had to be joined together in sitú by key-pieces bolted on. All the superstructures of the bridges up to 40 feet in length were put together in Matadi and loaded there on trucks. The larger bridges were erected in close proximity to the line, and were rolled into position on to the abutments on Sunday mornings when the traffic was stopped for the purpose. Some of the largest spans were put together on their own abutments, the traffic being only interrupted during the time needed to place the transverse joists and rails. Special precautions were taken to guard against the expansion due to the high temperatures encountered, both in the case of the bridge-work and in the method of jointing the rails.

The placing of the ballast was executed as speedily as possible after the rails were laid, though this work was generally 15 to 18 miles behind the rail-head. Materials suitable for ballast were found in more or less abundance all along the line. White quartz pebbles, which are plentiful in the upper layers of the soil in certain districts traversed, were chiefly used. It was found unnecessary to sift the soil containing the pebbles, as it was soon cleansed by the first heavy rains. Beyond the Inkissi the soil is mainly sandy and it was thought unsuited for metalling, but on trial the sand as used for ballast has turned out very well. It is found to be fairly cohesive, it is not washed away nor blown away by the wind and by passing trains, and it keeps well to the trimmed surface.

Having so far described the construction of the line, the Author proceeds to deal with the proposed mode of working the traffic and with the accessory details. First as to the means of transport, which vary in accordance with the particular phase reached in the progress of the work. In the section of the line actually completed for traffic, as far, namely, as the Inkissi, or 164 miles, as also in the section which is completely ballasted, the transport is provided for by the management staff (service de l'exploitation). Beyond the section where the ballast ends and where the metals only are laid, the traffic is still under the charge of the ballasting department. The trucks, laden with the materials and with the [THE INST. C.E. VOL. CXXXV.]

food for the workpeople, are all conveyed to the temporary depôt in the proximity of the rail-head. From this depót all the stores, tools, materials used in the camps, &c., have to be carried forward by the ballasting department. They are consigned to a special agent to whom they are entrusted, and who causes them to be transported to the temporary stores situated where they are actually required for use. Small boxes, cases of dried fish, sacks of rice, &c., not exceeding 66 lbs. in weight are carried by porters; barrels, and iron tubes for aqueducts, are rolled along the temporary way. The depôt stores are very lightly constructed in galvanized corrugated iron. They are each of them 39 feet long by 13 feet wide.

For purposes of communication along the line and in order to regulate the despatch of trains, the company has established a telephone service, the terminus of which is the most advanced point reached by the earthworks. The telephone wire is of oxydized phosphor-bronze, in order not to excite the cupidity of the natives, who have a great admiration for copper. This wire is supported by means of metal standards 16.4 feet in height, with a spiked end from 2 to 3 feet deep in the ground. The posts at first used were of drawn barrel-tubing; but subsequently T-section posts in soft steel, of a weight of 22 lbs. per yard, have been substituted for them with advantage. These posts are fixed from 110 to 165 yards apart. The telephone stations are usually from 12 to 15 miles apart. Some curious interruptions in the service, supposed to be due to the induction of momentary terrestrial currents, which occur during the hottest part of the rainy season, are noticed. The telephone posts are also used to carry a telegraph wire joining Boma and Matadi with the furthest point reached by the railway.

Passing on to the question of the method in which the traffic of the line will be dealt with when the works are finally completed, the Author states that it is proposed to divide the entire length into three sections about equal in length, say each of 80 miles. These will be worked separately, and at the end of each section there will be a change of engines. Stores and workshops will be constructed at each stopping place. As long as no night trains are run the goods trains will cover only one section in the day and the entire journey will thus need three days. Certain passenger trains will complete the journey in two days, and for these there will be a change of engines midway, where engine-sheds and workshops have also been constructed. The stoppage for the night is arranged for at Tumba, at the 116th mile, and here hotel

The

accommodation has been provided by private enterprise. traffic will be worked on the block system. A train on reaching a siding will have to await a telephonic message from the signal station beyond giving "line clear." In order to avoid as far as possible the danger caused by the fact that telephonic communication leaves no trace in case of accidents, and with the view to fix responsibility, a very perfect system of booking all the messages has been devised.

Since the railway was projected the progress of trade has been far beyond expectation. In the first instance the tariff rates were fixed very high. All merchandise carried upwards was charged for at the rate of 38. per ton per mile. The only exception made was in the case of salt, which was to be conveyed at half rates. Since these rates were fixed the company has agreed to carry all vessels, steam-engines, mechanical appliances used in agriculture or in industry, as also all electrical and telephonic apparatus, at a reduction of 40 per cent. By a still later arrangement it has been agreed to convey all railway plant for new lines communicating with the Upper Congo above Stanley Pool at half rates.

In the case of the downward traffic differential rates dependent upon the value of the products have been agreed upon, with a view of promoting the export trade. Palm-kernels, ground-nuts and timber pay the lowest rates, about 4d. per ton per mile, and ivory the highest, 38. per ton per mile. White travellers pay 18. 6d. per mile, and negroes travel at one-tenth this rate. Return fares are calculated at one and a half times the rate for a single journey. First-class passengers are allowed 220 lbs. of luggage free, and second-class passengers, 44 lbs. Thirty or more negroes travelling together in parties, in the service of a white employer, are carried at half fares.

During the construction of the line the maintenance and repairs of the finished section are entrusted to the service which corresponds to the permanent-way department (voies et travaux) on European lines; but the section still incomplete is in charge of the construction department. The whole line is divided into sections and sub-sections for the purpose of repairs, each of which is entrusted to a group of native workmen under a black foreman. Each section of about 62 miles has a works engine and a stock of trucks for ballasting purposes. There will eventually be four sets of repairing-sheds, the two chief ones at either end of the line, and two smaller ones at the sectional stations where the engines are changed.

A detailed account is given of the repairing shops at Matadi,

and of the supply of machine-tools, &c. Two types of goodsengines are in use, and after careful study by experts in Belgium a locomotive was specially designed to meet all the exigencies of the case, and adapted for the sharp curves and heavy gradients. The engine in question has eight wheels, the three front pairs coupled with a wheel-base of 7.32 feet, and the axle of the trailingwheels placed far back, 6.56 feet behind the front group. The total wheel-base of 13.94 feet being quite out of the question on curves of 2 chains, the trailing-axle is given a transverse movement by the employment of radial axle-boxes, and in addition to this provision is made for a slight amount of play in the leadingaxle. This locomotive, like all the later ones, was furnished with a tender. The total weight when empty is 24 tons, and 31 tons when in running order. The boiler is in mild steel, tested to 180 lbs. per square inch; it has a copper fire-box and brass tubes. This engine can propel three 10-ton trucks on rising gradients of 1 in 22. It has great stability, but has not proved well adapted for the sharp curves. This type has, therefore, been given up in favour of a locomotive having six coupled wheels on a wheel-base of 11.8 feet. This engine weighs 21 tons empty, and 26 tons in running order, and can take four 10-ton trucks up the steepest gradients on the line. A much lighter engine has been designed for the passenger traffic, having only four wheels coupled on a wheel-base of 6.56 feet, and capable of drawing two carriages, or a carriage and a goods-truck. The passenger engine weighs 16 tons when empty, or 18 tons in running order. A still smaller engine is used for works purposes. Illustrations are given of each form of engine, as also of the 10-ton trucks on bogies. Some account follows of the coaches and rolling-stock and of the staff in charge of the trains.

Mention has from time to time been made of the rivers and torrents traversed, and it will naturally appear probable that some attempt should be made to utilize the water-power for the generation of electricity. The river Congo at Matadi could easily be made to furnish 250,000 HP., and if three trains were despatched daily over the whole line each way, it would only involve the expenditure of about 2,000 HP. No doubt the practical difficulties in the way of obtaining the power from the Congo would be enormous. It is shown, however, that even if the power could only be utilized within a radius of, say, 37 to 43 miles, the rivers M'Poso, Lufu, Kwilu, Inkissi, and Lukaya would furnish ample power for the sections of the line in their vicinity. For this purpose the rivers in question would have to be dammed, and