pump, inside the frame, at 7. The feed-pump and the bilge, pumps are driven from the cross-head of the air-pump.

The successful introduction of the double-cylinder en. gine was finally accomplished by the exertions of a few engineers, who were at once intelligent enough to understand its advantages, and energetic and enterprising enough to push it forward in spite of active opposition, and powerful enough, pecuniarily and in influence, to succeed.

The most active and earnest of these eminent men was John Elder, of the firm of Randolph, Elder & Co., subsequently John Elder & Co., of Glasgow.'

Elder was of Scotch descent.

His ancestors had, for

1 Vide "Memoir of John Elder," W. J: M. Rankine, Glasgow, 1871.

generations, shown great skill and talent in construction, and had always been known as successful millwrights. John Elder was born at Glasgow, March 8, 1824, and died in London, September 17, 1869. He was educated at the Glasgow High-School and in the College of Engineering at the University of Glasgow, where, however, his attendance was but for a short time. He learned the trade under his father in the workshops of the Messrs. Napier, and became an unusually expert draughtsman. After spending three years in charge of the drawing-office at the engine-building works of Robert Napier, where his father had been manager, Elder became a partner in the firm which had previously been known as Randolph, Elliott & Co., in the year 1852. The firm commenced building iron vessels in 1860.

In the mean time, the experiments of Hornblower and Wolff, of Allaire and Smith, and of McNaught, Craddock, and Nicholson, together with the theoretical investigations of Thompson, Rankine, Clausius, and others, had shown plainly in what direction to look for improvement upon then standard engines, and what direction practice was taking with all types. The practical deductions which were becoming evident were recognized very early by Elder, and he promptly began to put in practice the principles which his knowledge of thermo-dynamics and of mechanics enabled him to appreciate. He adopted the compound engine, and coupled his cranks at angles of 180°, in order to avoid losses due to the friction of the crank-shaft in its bearings, by effecting a partial counterbalancing of pressures on the journals. Elder was one of the first to point out the fact that the compound engine had proved itself more efficient than the single-cylinder engine, only when the pressure of steam carried and the extent to which expansion was adopted exceeded the customary practice of his time. His own practice was, from the first, successful, and from 1853 to 1867 he and his partners were continually engaged in the construction of steamers and fitting them with compound engines.

The engines of their first vessel, the Brandon, required but 3 pounds of coal per hour and per horse-power, in 1854, when the usual consumption was a third more. Five years later, they had built engines which consumed a third less than those of the Brandon; and thenceforward, for many years, their engines, when of large size, exhibited what was then thought remarkable economy, running on a consumption of from 24 to 24 pounds.

In the year 1865 the British Government ordered a competitive trial of three naval vessels, which only differed in the form of their engines. The Arethusa was fitted with trunk-engines of the ordinary kind; the Octavia had three steam-cylinders, coupled to three cranks placed at angles of 120° with each other; and the Constance was fitted with compound engines, two sets of three cylinders each, and each taking steam from the boiler into one cylinder, passing it through the other two with continuous expansion, and finally exhausting from the third into the condenser. These vessels, during one week's steaming at sea, averaged, respectively, 3.64, 3.17, and 2.51 pounds of coal per hour and per horse-power, and the Constance showed a marked superiority in the efficiency of the mechanism of her engines, when the losses by friction were compared.

The change from the side-lever single-cylinder engine, with jet-condenser and paddle-wheels, to the direct-acting compound engine, with surface-condenser and screw-propellers, has occurred within the memory and under the observation of even young engineers, and it may be considered that the revolution has not been completely effected. This change in the design of engine is not as great as it at first seemed likely to become. Builders have but slowly learned the principles stated above in reference to expansion in one or more cylinders, and the earlier engines were made with a high and low pressure cylinder working on the same connecting-rod, and each machine consisted of four steam-cylinders. It was at last discovered that a high-pressure single

cylinder engine exhausting into a separate larger low-press ure engine might give good results, and the compound engine became as simple as the type of engine which it displaced. This independence of high and low pressure engines is not in itself novel, for the plan of using the exhaust of a high-pressure engine to drive a low-pressure condensing engine was one of the earliest of known combinations.

The advantage of introducing double engines at sea is considerably greater than on land. The coal carried by a steam-vessel is not only an item of great importance in consequence of its first cost, but, displacing its weight or bulk of freight which might otherwise be carried, it represents so much non-paying cargo, and is to be charged with the full cost of transportation in addition to first cost. The best of steam-coal is therefore usually chosen for steamers making long voyages, and the necessity of obtaining the most economical engines is at once seen, and is fully appreciated by steamship proprietors. Again, an economy of one-fourth of a pound per horse-power per hour gives, on a large transatlantic steamer, a saving of about 100 tons of coal for a single voyage. To this saving of cost is to be added the gain in wages and sustenance of the labor required to handle that coal, and the gain by 100 tons of freight carried in place of the coal.

For many years the change which has here been outlined, in the forms of engine and the working of steam expansively, was retarded by the inefficiency of methods and tools used in construction. With gradual improvement in tools and in methods of doing work, it became possible to control higher steam and to work it successfully; and the change in this direction has been steadily going on up to the present time with all types of steam-engine. At sea this rise of pressure was for a considerable time retarded by the serious difficulty encountered in the tendency of the sulphate of lime to deposit in the boiler. When steampressure had risen to 25 pounds per square inch, it was

found that no amount of "blowing out" would prevent the deposition of seriously large quantities of this salt, while at the lower pressures at first carried at sea no troublesome precipitation occurred, and the only precaution necessary was to blow out sufficient brine to prevent the precipitation of common salt from a supersaturated solution. The introduction of surface-condensation was promptly attempted as the remedy for this evil, but for many years it was extremely doubtful whether its disadvantages were not greater than its advantages. It was found very difficult to keep the condensers tight, and boilers were injured by some singular process of corrosion, evidently due to the presence of the surface-condenser. The simple expedient of permitting a very thin scale to form in the boiler was, after a time, hit upon as a means of overcoming this difficulty, and thenceforward the greatest obstacle to the general introduction was the conservative disposition found among those who had charge of marine machinery, which conservatism regarded with suspicion every innovation. Another trouble arose from the difficulty of finding men neither too indolent nor too ignorant to take charge of the new condenser, which, more complicated and more readily disarranged than the old, demanded a higher class of attendants. Once introduced, however, the surface-condenser removed the obstacle to further elevation of steam-pressure, and the rise from 20 to 60 pounds pressure soon occurred. Elder and his competitors on the Clyde were the first to take advantage of the fact when these higher pressures became practicable.

The lightness of engine and the smaller weight of boiler secured when the simpler type of "compound" engine is used are great advantages, and, when coupled with the fact that by no other satisfactory device can great expansion and consequent economy of fuel be obtained at sea, the advantages are such as to make the adoption of this style of engine imperative for ship-propulsion.