RAILWAYS. Railways had their origin in the tramways (q.v.) or wagon-ways which at least as early as the middle of the 16th century were used in the mineral districts of England round Newcastle for the conveyance of coal from the pits to the river Tyne for shipment. It may be supposed that originally the public roads, when worn by the cartage of the coal, were repaired by laying planks of timber at the bottom of the ruts, and that then the planks were laid on the surface of special roads or ways formed between the Collieries and the river. “The manner of the carriage,” says Lord Keeper North in 1676, “is by laying rails of timber . . . exactly straight and parallel, and bulky carts are made with four rowlets fitting these rails, whereby the carriage is so easy that one horse will draw down four or five chaldrons of coals” (from 10.6 to 13.2 tons). The planks were of wood, often beech, a few inches wide, and were fastened down, end to end, on logs of wood, or “sleepers,” placed crosswise at intervals of two or three feet. In time it became a common practice to cover them with a thin sheathing or plating of iron, in order to add to their life; this expedient caused more wear on the wooden rollers of the wagons, and, apparently towards the middle of the 18th century, led to the introduction of iron wheels, the use of which is recorded on a wooden railway near Bath in 1734. But the iron sheathing was not strong enough to resist buckling under the passage of the loaded wagons, and to remedy this defect the plan was tried of making the rails wholly of iron. In 1767 the Colebrookdale Iron Works cast a batch of iron rails or plates, each 3 ft. long and 4 in. broad, having at the inner side an upright ledge or flange, 3 in. high at the centre and tapering to a height of 21 in. at the ends, for the purpose of keeping the flat wheels on the track. Subsequently, to increase the strength, a similar flange was added below the rail. Wooden sleepers continued to be used, the rails being secured by spikes passing through the extremities, but about 1793 stone blocks also began to be employed—an innovation associated with the name of Benjamin Outram, who, however, apparently was not actually the first to make it. This type of rail (fig. 1) was known as the plate-rail, tramway-plate or barrow way-plate—names which are preserved in the modern term “plate layer” applied to the men who lay and maintain the permanent way of a railway.
Another form of rail, distinguished as the edgerail, was first used on a line which was opened between Loughborough and Nanpantan in 1789. This line was originally designed as a “plateway” on the Outram system, but objections were raised to rails with upstanding ledges or flanges being laid on the turnpike road which was crossed at Loughborough on the level. In other cases this difficulty was overcome by paving or “causewaying” the road up to the level of the top of the flanges, but on this occasion William Jessop, of the Butterley Iron Works, near Derby, proposed to get over it by laying down two plates of iron, perfectly flat and level with the road but each having on its outside a groove 3 in. wide and 3 in. deep to control extra guiding wheels which were to be of somewhat larger diameter than the bearing wheels and to be affixed to them. The rest of the line was laid with what were substantially plate-rails placed on their edge instead of flat. These were cast in 3 ft. lengths, of a double-flanged section, and for the sake of strength they were “fish-bellied” or deeper in the middle than at the ends. At one end of each rail the flange spread out to form a foot which rested on a cross sleeper, being secured to the latter by a spike passing through a central hole, and above this foot the rail was so shaped as to form a socket into which was fitted the end of the next rail. Each length was thus fastened to a sleeper at one end, while at the other it was socketed into the end of its fellow. This method, however, was not found satisfactory: the projecting feet were liable to be broken off, and in 1799 or 1800 Jessop abandoned them, using instead separate cast-iron sockets or chairs, which were fastened to the sleepers and in which the rails were supported in an upright position. In the first instance he proposed to place the guiding wheels outside the bearing wheels, and the Nanpantan line was laid on this plan with a width of 5 ft. between the guide wheels; but before it was opened he decided not only to cast the guiding wheels and bearing wheels in one piece but also to put the former inside the rails, arguing that with this arrangement the edge-rails themselves would keep the wheels in position on the axles, whereas with that first contemplated fastenings would have been required for them (fig. 2). Jessop thus produced what was virtually the flanged wheel of to-day, having the flanges inside the rails, and further, it is said, established what has become the standard gauge of the world, 4 ft. 81 in., or 5 ft. minus the width of two of his rails.
These two systems of constructing railways the plate-rail and the edge-rail-continued to exist side by side until well on in the 19th century. In most parts of England the plate-rail was preferred, and it was used on the Surrey iron railway, from Wandsworth to Croydon, which, sanctioned by parliament in 1801, was finished in 1803, and was the the first railway available to the public on payment of of tolls, previous lines having all been private and reserved exclusively for the use of their owners. In South Wales again, where in 1811 the railways in connexion with canals, Collieries and iron and copper works had a total length of nearly 150 miles, the plate-way was almost universal. But in the north of England and in Scotland the edge~rail was held in greater favour, and by the third decade of the century its superiority was generally established. The manufacture of the rails themselves was gradually improved. By making them in longer lengths a reduction was effected in the number of joints—always the weakest part of the line; and another advance consisted in the substitution of wrought iron for cast iron, though that material did not gain wide adoption until after the patent for an improved method of rolling rails granted in 1820 to John Birkinshaw, of the Bedlington Ironworks, Durham. His rails were wedge-shaped in section, much wider at the top than at the bottom, with the intermediate portion or web thinner still, and he recommended that they should be made 18 ft. long, even suggesting that several of them might be welded together end to end to form considerable lengths. They were supported on sleepers by chairs at intervals of 3 ft., and were fish-bellied between the points of support. As used by George Stephenson on the Stockton & Darlington and Whitstable & Canterbury lines they weighed 28 ℔ per yard. On the Liverpool & Manchester railway they were usually 12 ft. or 15 ft. long and weighed 35 ℔ to the yard, and they were fastened by iron wedges to chairs weighing 15 or 17 ℔ each. The chairs were in turn fixed to the sleepers by two iron spikes, half-round wooden cross sleepers being employed on embankments and stone blocks 20 in. square by 10 in. deep in cuttings. The fish bellied rails, however, were found to break near the chairs, and from 1834 they began to be replaced with parallel rails weighing 50 ℔ to the yard.
The next important development in rail design originated in America, which, for the few lines that had been laid up to 1830, remained content with wooden bars faced with iron. In that year Robert Livingston Stevens (1787–1856), devised for the Camden & Amboy railway a rail similar as to its top to those in use in England, but having a flat base or foot by which it was secured to the sleepers by hook-headed spikes, without chairs (fig. 3); he had to get the first lot of these rails, which were 15 ft. long and weighed 36 ℔ to the yard, manufactured in England, since there were then no mills in America able to roll them. This type, which is often known as the Vignoles rail, after Charles Blacker Vignoles (1793–1875), who re-invented it in England in 1836, is in general use in America and on the continent of Europe. The bridge-rail (fig. 4)–so called because it was first laid on bridges—was supported on continuous longitudinal sleepers and held down by bolts passing through the flanges, and was employed by I. K. Brunel on the Great Western railway, where, however, it was abandoned after the line was converted from broad to standard gauge in 1892. In the double-headed rail (fig. 5), originated by Joseph Locke in 1837, and first laid on the Grand Junction railway, the two tables were equal. This rail was more easily rolled than others, and, being reversible, was in fact two rails in one. But as it was laid in cast-iron chairs the lower table was exposed to damage under the hammering of the traffic, and thus was liable to be rendered useless as a running surface. In consequence the bull-headed rail (fig. 6) was evolved, in which the lower table was made of smaller size and was intended merely as a support, not as a surface to be used by the wheels. There was a waste of metal in these early rails owing to the excessive thickness of the vertical web, and subsequent improvements have consisted in adjusting the dimensions so as to combine strength with economy of metal, as well as in the substitution of steel for wrought iron (after the introduction of the Bessemer process) and in minute attention to the composition of the steel employed.
It was found, naturally, that the rails would not rest in their chairs at the joints, but were loosened and bruised at the ends by the blows of the traffic. The fish-joint was therefore devised in 1847 by W. Bridges Adams, the intention being by “fishing” the joints to convert the rails into continuous beams. In the original design two chairs were placed, one under each rail, a few inches apart, as in fig. 7. The joint was thus suspended between the two chairs, and two keys of iron, called “fishes,” fitting the side channels of the rails, were driven in on each side between the chairs and the rails. In subsequent modifications the fishes were, as they continue to be, bolted to and through the rails, the sleepers being placed rather further apart and the joint being generally suspended between them.
The iron tramway or railway had been known for half a century and had come into considerable use in connexion with collieries and quarries before it was realized that for the carriage of general merchandise it might prove a serious competitor to the canals, of which a large mileage had been constructed in Great Britain during that period. In the article on “Railways” in the Supplement to the Encyclopaedia Britannica, published in 1824, it is said: “It will appear that this species of inland carriage [railways] is principally applicable where trade is considerable and the length of conveyance short; and is chiefly useful, therefore, in transporting the mineral produce of the kingdom from the mines to the nearest land or water communication, whether sea, river or canal. Attempts have been made to bring it into more general use, but without success; and it is only in particular circumstances that navigation, with the aid either of locks or inclined planes to surmount the elevations, will not present a more convenient medium for an extended trade.” It must be remembered, however, that at this time the railways were nearly all worked by horse-traction, and that the use of steam had made but little progress. Richard Trevithick, indeed, had in 1804 tried a high-pressure steam locomotive, with smooth wheels, on a plate-way near Merthyr Tydvil, but it was found more expensive than horses; John Blenkinsop in 1811 patented an engine with cogged wheel and rack-rail which was used, with commercial success, to convey coal from his Middleton Colliery to Leeds; William Hedley in 1813 built two locomotives—Puffing Billy and Wylam Dilly—for hauling coal from Wylam Colliery, near Newcastle; and in the following year George Stephenson’s first engine, the Blucher, drew a train of eight loaded wagons, weighing 30 tons, at a speed of 4 m. an hour up a gradient of 1 in 450. But, in the words of the same article, “This application of steam has not yet arrived at such perfection as to have brought it into general use.”
The steam locomotive, however, and with it the railways, soon began to make rapid progress. On the Stockton & Darlington railway, which was authorized by parliament in 1821, animal power was at first proposed, but on the advice of Stephenson, its engineer, steam-engines were adopted. This line, with three branches, was over 38 m. in length, and was in the first instance laid with a single track, passing-places being provided at intervals of a quarter of a mile. At its opening, on the 27th of September 1825, a train of thirty four vehicles, making a gross load of about 90 tons, was drawn by one engine driven by Stephenson, with a signalman on horseback in advance. The train moved off at the rate of from 10 to 12 m. an hour, and attained a speed of 15 m. an hour on favourable parts of the line. A train weighing 92 tons could be drawn by one engine at the rate of 5 m. an hour. The principal business of the new railway was the conveyance of minerals and goods, but from the first passengers insisted upon being carried, and on the 10th of October 1825 the company began to run a daily coach, called the “Experiment,” to carry six inside, and from fifteen to twenty outside, making the journey from Darlington to Stockton and back in two hours. The fare was 1s., and each passenger was allowed to take baggage not exceeding 14 ℔ weight. The rate for carriage of merchandise was reduced from 5d. to one-fifth of a penny per ton per mile, and that of minerals from 7d. to 1½d. per ton per mile. The price of coals at Darlington fell from 18s. to 8s. 6d. a ton.
The example of the Stockton & Darlington line was followed by the Monklands railway in Scotland, opened in 1826, and several other small lines—including the Canterbury & Whitstable, worked partly by fixed engines and partly by locomotives—quickly adopted steam traction. But the Liverpool & Manchester railway, opened in 1830, first impressed the national mind with the fact that a revolution in the methods of travelling had really taken place; and further, it was for it that the first high-speed locomotive of the modern type was invented and constructed. The directors having offered a prize of, £500 for the best engine, trials were held on a finished portion of the line at Rainhill in October 1829, and three engines took part—the Rocket of George and Robert Stephenson, the Novelty of John Braithwaite and John Ericsson, and the Sanspareil of Timothy Hackworth. The last two of these engines broke down under trial, but the Rocket fulfilled the conditions and won the prize. Its two steam cylinders were 8 in. in diameter, with a stroke of 16½ in., and the driving wheels, which were placed in front under the funnel, were 4 ft. 8½ in. in diameter. The engine weighed 4¼ tons; the tender following it, 3 tons 4cwt.; and the two loaded carriages drawn by it on the trial, 9 tons 11 cwt.: thus the weight drawn was 12¾ tons, and the gross total of the train 17 tons. The boiler evaporated 18¼ cub. ft., or 114 gals., of water an hour, and the steam pressure was 50 ℔ per square inch. The engine drew a train weighing 13 tons 35 m. in 48 minutes, the rate being thus nearly 44 m. an hour; subsequently it drew an average gross load of 40 tons behind the tender at 13·3 m. an hour. The Rocket possessed the three elements of efficiency of the modern locomotive—the internal water-surrounded fire-box and the multitubular flue in the boiler; the blast-pipe, by which the steam after doing its work in the cylinders was exhausted up the chimney, and thus served to increase the draught and promote the rapid combustion of the fuel; and the direct connexion of the steam cylinders, one on each side of the engine, with the two driving wheels mounted on one axle. Of these features, the blast-pipe had been employed by Trevithick on his engine of 1804, and direct driving, without intermediate gearing, had been adopted in several previous engines; but the use of a number (25) of small tubes in place of one or two large flues was an innovation which in conjunction with the blast-pipe contributed greatly to the efficiency of the engine. After the success of the Rocket, the Stephensons received orders to build seven more engines, which were of very similar design, though rather larger, being four-wheeled engines, with the two driving wheels in front and the cylinders behind; and in October 1830 they constructed a ninth engine, the Planet, also for the Liverpool & Manchester railway, which still more closely resembled the modern type, since the driving wheels were placed at the fire-box end, while the two cylinders were arranged under the smoke-box, inside the frames. The main features of the steam locomotive were thus established, and its subsequent development is chiefly a history of gradual increase in size and power, and of improvements in design, in material and in mechanical construction, tending to increased efficiency and economy of operation.
In America the development of the locomotive dates from almost the same time as in England. The earliest examples used in that country, apart from a small experimental model constructed by Peter Cooper, came from England. In 1828, on behalf of the Delaware & Hudson Canal Company, which had determined to build a line, 16 m. long, from Carbondale to Honesdale, Pennsylvania, Horatio Allen ordered three locomotives from Messrs Foster & Rastrick, of Stourbridge, and one from George Stephenson. The latter, named the America, was the first to be delivered, reaching New York in January 1829, but one of the others, the Stourbridge Lion, was actually the first practical steam locomotive to run in America, which it did on the 9th of August 1829. The first American-built locomotive, the Best Friend, of Charleston, was made at the West Point Foundry, New York, in 1830, and was put to work on the South Carolina railroad in that year. It had a vertical boiler, and was carried on four wheels all coupled, the two cylinders being placed in an inclined position and having a bore of about 6 in. with a stroke of 16 in. It is reported to have hauled 40 or 50 passengers in 4 or 5 cars at a speed of 16–21 m. an hour. After a few months of life it was blown up, its attendant, annoyed by the sound of the escaping steam, having fastened down the safety-valve. A second engine, the West Point, also built at West Point Foundry for the South Carolina railroad, differed from the Best Friend in having a horizontal boiler with 6 or 8 tubes, though in other respects it was similar. In 1831 the Baltimore & Ohio Company offered a prize of $4000 for an American engine weighing 3½ tons, able to draw 15 tons at 15 m. an hour on the level: it was won by the York of Messrs Davis & Gartner in the following year. Matthias W. Baldwin, the founder of the famous Baldwin Locomotive Works in Philadelphia, built his first engine, Old Ironsides, for the Philadelphia, Germantown & Morristown railroad; first tried in November 1832, it was modelled on Stephenson’s Planet, and had a single pair of driving wheels at the firebox end and a pair of carrying wheels under the smoke-box. His second engine, the E. L. Miller, delivered to the South Carolina railroad in 1834, presented a feature which has remained characteristic of American locomotives—the front part was supported on a four-wheeled swivelling bogie-truck, a device, however, which had been applied to Puffing Billy in England when it was rebuilt in 1815.
The Liverpool & Manchester line achieved a success which surpassed the anticipations even of its promoters, and in consequence numerous projects were started for the construction of railways in various parts of Great Britain. In the decade following its opening nearly 2000 m. of railway were sanctioned by parliament, including the beginnings of most of the existing trunk-lines, and in 1840 the actual mileage reached 1331 m. The next decade saw the “railway mania.” The amount of capital which parliament authorized railway companies to raise was about 4½ millions on the average of the two years 1842–1843, 17¾ millions in 1844, 60 millions in 1845, and 132 millions in 1846, though this last sum was less than a quarter of the capital proposed in the schemes submitted to the Board of Trade; and the wild speculation which occurred in railway shares in 1845 contributed largely to the financial crisis of 1847. In 1850 the mileage was 6635, in 1860 it was 10,410, and in 1870 it was 15,310. The increase in the decade 1860–1870 was thus nearly 50%, but subsequently the rate of increase slackened, and the mileages in 1880, 1890 and 1900 were 17,935, 20,073 and 21,855. In the United States progress was more rapid, for, beginning at 2816 in 1840, the mileage reached 9015 in 1850, 30,600 in 1860, 87,801 in 1880, and 198,964 in 1900. Canada had no railway till 1853, and in South America construction did not begin till about the same time. France and Austria opened their first lines in 1828; Belgium, Germany, Russia, Italy and Holland in the succeeding decade; Switzerland and Denmark in 1844, Spain in 1848, Sweden in 1851, Norway in 1853, and Portugal in 1854; while Turkey and Greece delayed till 1860 and 1869. In Africa Egypt opened her first line (between Alexandria and Cairo) in 1856, and Cape Colony followed in 1860. In Asia the first line was that between Bombay and Tannah, opened in 1853, and in Australia Victoria began her railway system in 1854 (see also the articles on the various countries for further details about their railways).
Transcontinental Railways.—A railway line across North America was first completed in 1869, when the Union Pacific, building from the Missouri river at Omaha (1400 m. west of New York), met the Central Pacific, which built from San Francisco eastwards, making a line 1848 m. long through a country then for the most part uninhabited. This was followed by the Southern Pacific in 1881, from San Francisco to New Orleans, 2489 miles; the Northern Pacific, from St Paul to Portland, Ore., in 1883; the Atchison, Topeka & Santa Fé, from Kansas City to San Diego; and the Great Northern from St Paul to Seattle and New Westminster in 1893. Meanwhile the Canadian Pacific, a true transcontinental line, was built from Montreal, on Atlantic tide-water, to the Pacific at Vancouver, 2906 m. But these lines have been dwarfed since 1891 by the Siberian railway, built by the Russian government entirely across the continent of Asia from Cheliabinsk (1769 m. by rail east of St Petersburg) to Vladivostok, a distance of 4073 m., with a branch from Kharbin about 500 m. long to Dalny and Port Arthur. The main line was finished in 1902, except for a length of about 170 m. in very difficult country around the south end of Lake Baikal; this was constructed in 1904, communication being maintained in the interval by ferry-boats, which conveyed all the carriages of a train across the lake, more than 40 m., when the ice permitted. A transcontinental line was long ago undertaken across South America from Buenos Aires to Valparaiso, where the continent is only about 900 m. wide. The last section through the Andes was finished in 1910. (H. M. R.)
- “Another thing that is remarkable is their way-leaves; for, when men have pieces of ground between the colliery and the river, they sell leave to lead coals over their ground” (Roger North).