inches thick near the crown and 23 4 inches thick at the haunches; (2) the road platform, with a thickness of 4 to 5.5 inches; and (3) the connection between the two former parts, consisting of three vertical longitudinal webs 6.24 inches thick, which are themselves stiffened by a few cross ribs.

The working-load specified was 50 lbs. per square foot evenly distributed, or a wagon weighing 7.5 tons. In order to avoid vibration the bridge was actually designed for a distributed load of 61.5 lbs. per square foot, or a wagon of 12 tons; and considering these as dead loads the parts were so designed as to have ample factors of safety. The illustrations in the original show the method of construction, which is very simple, the ironwork consisting merely of interlaced bars. Wood centering was built on a temporary framework fixed across the bed of the river. The concrete consisted of 660 lbs. of Portland cement by weight to 28.25 cubic feet of gravel and 17.65 cubic feet of sand by volume.

The total weight of the reinforcement was 16 tons and the bars were of round section 0.39 inch to 0.6 inch in diameter. The total cost of the bridge, exclusive of the metalling of the road, was £1,328, which is about the same as the cost of a purely metallic structure; but reinforced concrete was preferred on the score of lower cost of maintenance.

E. R. D.

First Connection by Precise Levelling between the Atlantic and Pacific Oceans. J. F. HAYFORD.

(Engineering News, New York, 16 March, 1905, p. 279.)

On the 4th October, 1904, a Coast and Geodetic Survey party working eastward from Seattle, Washington, met a small party working westward, at Hunts Junction, in the south-eastern part of Washington; which completed the first connection by precise levelling between the Atlantic and Pacific. The discrepancy developed at the junction was 0.615 foot, the Pacific being apparently higher than the Atlantic. This discrepancy must be due either to errors in the tidal observations which furnished the connection with mean sea-level, to errors in levelling, or to a real difference in the elevation of the mean sea-surface at the points at which the tidal observations were made. As the tidal observations extended over at least 5 years and the maximum annual variation at Seattle and Biloxi amounted to 0.100 foot and 0.204 foot respectively, these variations cannot account for the discrepancy. The shortest line of levelling from Seattle to Biloxi is 3,500 miles, and if it be assumed that the discrepancy of 0.615 foot is simply an accumulated error in levelling, and the rate be assumed to be uniform, it amounts to 1 foot in 5,700 miles, or 0.00018 foot per

mile, which is well within the possible limits of error of precise levelling. There are therefore no grounds for inferring that an actual difference in level of the two oceans really exists. The average cost during the last 4 years per completed mile was 428., and 64 miles were completed per month.

A. W. B.

Canal from the North Sea to the Mediterranean.
ARDOUIN-DUMAZET.

(La Nature, Paris, 10 June, 1905, pp. 20–22.)

By the end of 1906, or early in 1907, vessels starting from Dunkerque or Lille will be able to proceed without breaking cargo to Lyons, with a saving in distance of upwards of 170 miles, and with enormous advantages in the matter of transportation. This saving is calculated by means of a formula, each lock passed through being reckoned as 0.621 mile. The last link in this waterway has been the construction of the canal between the Marne and the Saône, a prolongation of the Haute-Marne Canal. One of the most notable facts connected with this undertaking is that the requisite funds for its completion have been provided by a small community of 15,000 inhabitants, without State grants or contributions from the provinces. St. Dizier, in one of the most remarkable metalliferous districts in France, has achieved this result unaided, in order to open out her products by the piercing of this "French isthmus." This town offered to the State a subvention of £240,000 in return for the right to impose small dues upon each ton of merchandize. Some of the works which before the passing of this Act in 1900 had already been accomplished are described, namely, the construction of a tunnel 5,271 yards in length under the plateau of Langres, and the formation of the two vast reservoirs on the Mouche and the Liez, impounding 1,412 million cubic feet of water, but the works still remaining to be completed were by no means inconsiderable. A fall of 134 feet had to be overcome in a rocky district abounding with fissures, which threatened to lay the canal dry, and provision had to be made for an additional supply of 706 million cubic feet of water in a country where the watercourses become mere rivulets during certain seasons. The engineers, however, have overcome all these difficulties; new reservoirs have been constructed at Villegusien and near Charmes, and the descent of 134 feet has been arranged for by means of a chain of six locks, which have entailed a total deviation of only 3,280 yards. One of the other formidable obstacles to be surmounted was the passage of the upper valley of the Vingeanne, which involved deep rocky cuttings, and presented more problems for solution in 5 miles than in all the

rest of this part of the canal. Some of the industrial advantages which will be afforded by the completion of this important undertaking are discussed.

G. R. R.

Stolze Gas-Turbine. A. GRADENWITZ.

(L'Électricien, Paris, 1905, vol. xxix. pp. 273-76.)

This article describes the principle and the general construction of a 200-HP. gas-turbine, devised by Stolze, which is now being constructed at Berlin. The cycle of operations is as follows:Atmospheric air is compressed to about 2.5 atmospheres, it is then heated at constant pressure till its volume reaches two to twoand-a-half times the original value before heating; it is then allowed to expand to atmospheric pressure. The efficiency of the cycle depends on the nature of the compression, the quantity of heat supplied and the manner in which it is supplied, and finally on the nature of the expansion. In practice a turbine draws in the air and compresses it to a pressure depending on the angular velocity of the turbine. The air then passes through a heating vessel, heated by the exhaust gases of the apparatus, and is used partly in a generator to convert coke or anthracite into gas, but mostly to burn this gas in special burners, thus producing the desired increase of temperature and of volume. The heated gases finally pass into another turbine where they are expanded to atmospheric pressure, and on exhaust are made to assist in heating the air entering the apparatus. The theoretical thermal efficiency of the cycle for moderate heating to about 400° C. is more than 30 per cent., and could be very considerably increased for higher temperatures, which might be readily realized with an appropriate system of turbines. The Author gives some diagrams showing the arrangement of the apparatus.

W. C. H.

Testing the Efficiency of a 500-HP. Coke-Oven-Gas Engine.

E. MEYER.

(Zeitschrift des Vereines deutscher Ingenieure, Berlin, 1905, vol. xlix. pp. 324–30.) The engine tested was of the Oechelhäuser type, built for use with blast-furnace gas, and fitted with a double-acting charging pump, delivering on the one side air and on the other gas, to the working cylinder. To fit the engine for working with the stronger gas from coke-ovens, both sides of the charging pump were arranged to deliver air, the gas being supplied by a separate pump; but as this arrangement gave an excess of air, a blow-off

valve was provided between the pump and the cylinder. For adjusting the relative apertures of the air- and gas-ports, the latter were fitted with a sliding collar.

The principal dimensions of the engine were as follows:— diameter of working double-piston cylinder 27 inches, stroke of front piston 38 inches, stroke of back piston 37 inches. Diameter of air-pump cylinder 45 inches, stroke 20 inches; diameter of gaspump cylinder 23 inches, stroke 20 inches. The engine was employed to drive a 66-inch blast-furnace blower, coupled direct on to the tail piston.

The coke-oven-gas ranged in calorific value between 3,112 and 3,860 heat-units per cubic metre. The best results in the tests were obtained with the gas-valve half open and the air blow-off valve about one-quarter open. Under these conditions, and running at 110 revolutions per minute, the engine developed a useful effect of 628 HP., with a consumption of 15,220 cubic feet of gas per hour, and a heat-consumption of 1,290,000 units. The indicated horse-power was 765, of which 137 HP. were consumed in overcoming the internal friction of the machine. The mechanical efficiency of the engine was 82.13 per cent.

A series of other tests made, with the gas- and air-valves adjusted in different ways, led to the following general conclusions:-For a given engine-speed the consumption of power per revolution by the air-pump varies directly with the load; and conversely with a given load the consumption per revolution varies with the speed. The ratio between the consumption of power by the pump and the indicated power of the engine varies inversely with the load. For a given load this ratio diminishes with the speed, but not to any considerable extent. Under full load and at normal speed, about 15 per cent. of the indicated work is absorbed by the charging pump. Tests made at highly divergent speeds show that the consumption of heat per indicated horse-power-hour is only about 1,660 heat-units, so that about 38 per cent. of the calorific value of the gas is converted into indicated power, though when the load is reduced to about 0 42, the consumption of heat rises to 1,820 units. Owing to the diminution of the mechanical efficiency, however, the consumption of heat per horse-power-hour of useful effect increases rapidly as the load decreases.

C. S.

Three-Cylinder Diesel Engine by Sulzer Bros., of Winterthur. (Schweizerische Bauzeitung, vol. xliv., p. 253. 8 Figs.)

After pointing out the advantages of the Diesel engine in the possibility of using various qualities of oil, the Author describes engines built by Messrs. Sulzer Bros., who are the sole licensees for the manufacture of the engines in Switzerland. Engines of

20, 40, 80 and 120 HP. are illustrated, the last having three cylinders; and the construction and working of the installations are described.

The consumption of oil is 0.396 to 0.55 lb. of oil per brake horse-power-hour. The Author compares the efficiency of the Diesel engines with that of other types and states that in the best steam plants with heated feed-water to the boilers and superheated steam, only about 15 to 16 per cent. of the thermal value of the fuel is used. In the best modern power-gas plants only about 22 to 24 per cent. of the thermal value of the fuel is used; while in the Diesel engine more than 35 per cent. of that value is usefully employed. A large Diesel motor will give only about 25 per cent. lower consumption of oil than small engines, while in the case of steam-engines of 10 to 20 HP. the fuel-consumption is two to three times as much as that of a steam-engine developing several hundred horse-power.

E. R. D.

Recording Work done in Stopes in Mines. C. F. Morrison.

(School of Mines Quarterly, New York, 1905, vol. xxvi. p. 120.)

In order to obtain a systematic record of all work done in the copper mines of the Butte district, Montana, at every level a survey is made from the shaft to the face of each cross-cut and drift as the work advances. Along the drifts, which are timbered with square sets, the points of the traverse are established on the caps, and, when possible, the line of sight is the centre line of the drift. Notes and sketches are made showing the timberingangles, side sets, stations, and other details; and from these an accurate plan is plotted of the level with the timbers as they actually stand in the mine. The field-book is about 10 inches long by 4 inches wide, and consists of sheets of squared paper. A scale of 20 feet to the inch is used. Each square represents a square set, and a dot at the line of intersection represents a post. With such a scale, the book opened will represent about 400 feet of the vein and will be wide enough, where the veins are narrow, to permit of two floors being plotted on the same sheet. Specimen pages of the stope-book are given by the Author. The field stopebook furnishes a record of the work in the mine, shows the manner in which the development was conducted, and supplies data for office plans and estimates of amount of ore extracted and of possible ore reserves.

B. II. B.