overcome, except that of the air; and such roads will allow the velocity to be increased almost without limit."

Iron Railways are of two descriptions. The flat rail or tram road consists of cast iron plates about three feet long, four inches broad, and half an inch or an inch thick, with a flaunch, or turned-up edge on the inside, to guide the wheels of the carriage. These plates rest at each end on stone sleepers of three or four hundred weight, sunk into the earth, and they are joined to each other so as to form a continuous horizontal pathway. They are of course double, and the distance between the opposite rails is from 3 to 4 feet, according to the breadth of the car or waggon to be employed. The edge rail, which is found to be superior to the tramrail,is made either of wrought or cast iron. If the latter is used, the rails are about 3 feet long, 3 or 4 inches broad, and from 1 to 2 inches thick; and they are joined at their ends by cast metal sockets attached to the sleepers. The upper edge of the rail is generally made with a convex surface, to which the wheel of the car is adapted by a groove made somewhat wider. When wrought iron is used (which is found to be almost equally cheap with cast metal, and greatly preferable in many respects, 1) the bars are made of a smaller size, of a wedge shape, and 12 or 18 feet long, but they are supported by sleepers at the distance of every three feet. The waggons generally used run on four wheels of from two to three feet diameter, and carry from 20 to 50 cwt. Four or five of these are drawn by one horse. On the dead-level Railway constructed by Mr. John Grieve for Sir John Hope, near Musselburgh, which is one of the most perfect in Britain, a single horse draws five loaded waggons, each containing 30 cwt. of coals, at the rate of four miles an hour-in all 7 tons exclusive of the waggons, which weigh 3 tons more. Reducing the velocity to two miles an hour, by Professor Leslie's rule the horse should drag 12 tons, or 15 tons including the waggons. Mr. Stevenson observes, that "an ordinary horse, on a well-constructed edge Railway, on a level line of draught, will work with about ten tons of goods. 3 Mr. Palmer, an English engineer, gives the following as the effect

2.

It is proper to mention, that I find the superiority of the wrought-iron rail, when formed by rollers, questioned in the report of an English engineer, published in the Newcastle Courant. Its defect, according to his statement, is, that under the incessant pressure of the wheels, it ultimately splits into laminæ.

2 See the article Railways by Mr. Buchanan, in the Supplement to the Encyclopedia Britannica.

3 Memorial relative to opening the great valleys of Strathmore and Strathearn by means of a railway or canal, 1821. See also the article Railways, in Dr. Brewster's Encyclopedia, by the same writer.

of a single horse's draught, on different railways, at 23 miles an

This table shows the great superiority of the edge rail. The engineer observes too, that the vehicles are made in a very rude manner, and that were more care employed in their construction, the load might be much increased.

Railways are generally made double, one for going and the other for returning. The breadth of ground required for a single railway, is from 9 to 12 feet; for a double one, from 15 to 25. The expense of a double road, including the price of the ground, may be estimated generally at from 3000l. to 5000l. per mile, or from one-half to one-third of the expense of a canal. Mr. Stevenson says, "the first expense of a canal will be found to be double if not treble the expense of a railway; such are the difficulties of passing through a well-cultivated country, and especially of procuring a sufficient supply of water in manufacturing districts, that four times the expense will in most cases be nearer the mark." (Memorial, p. 12.) We speak here of railways of the ordinary kind for the transportation of goods; but it is probable, that one destined to serve the purpose of a great national thoroughfare for vehicles of all kinds, quick and slow, would cost at least twice as much. Even in this case, however, the original outlay would certainly not amount to more than a half or a third of what would be required for a canal of such a magnitude as to afford the same amount of commercial accommodation. The Union Canal has cost altogether about 12,000l. per mile; the Forth and Clyde, if executed at this day, would cost twice as much; the Caledonian Canal, if we exclude the lochs, and reckon only what has been cut, will ultimately cost almost 50,000l. per mile. Even deducting what has been expended on the lochs and on the harbors at its

1 Description of a railway on a new principle, &c. 1823. For farther details respecting the construction of railways, and an account of various contrivances for raising the waggon from a lower level to a higher, the reader may consult the Essays edited by Mr. Stevenson, in the 6th vol. of the Highland Society's Transactions.

extremities, the expense will still be nearly 40,000l. per mile. The projected canal from Bridgewater-bay to Lime, for sea vessels, is estimated to cost more than 40,000l. per mile.

A railway from Glasgow to Berwick, 125 miles long, projected in 1810, was surveyed by Mr. Telford, and estimated to cost 365,700l., or 2,9261. per mile. The estimated expense of a railway from Birmingham to Liverpool, distance 104 miles, surveyed within these few months, is 350,000l., or 3,365l. per mile. That of one from the Cromford Canal to the Peak Forest Canal in Derbyshire, 32 miles long, is 150,000l., or 4,700l. per mile. A recent Carlisle paper states, that the expense of a railway between that city and Newcastle was estimated at 252,000l., or 4,000l. per mile; and that of a canal at 888,000l., or 14,000l. per mile. A railway projected to run from Manchester to Liverpool, 33 miles, has been estimated to cost 400,000/., which is no less than 13,000/. per mile; but this includes a large amount for warehouses and locomotive engines, besides a sum set apart for the probable expenses of litigation. Lastly, a railway from Dalkeith to Edinburgh, including a branch to Fisherrow Harbour, 9 miles long altogether, will cost, according to the recent estimate of Mr. John Grieve, 36,862., or 3,9837. per mile, including the expense of five locomotive and one stationary steam engines.

Mr. Palmer, the engineer already mentioned, has proposed a new and ingenious species of railway, which deserves notice. It consists of a single rail, or continuous rod, of the usual form, but raised about three feet above the surface of the ground, and supported by cast metal pillars every ten feet. Two wheels with grooved edges, and 24 or 30 inches diameter, run, the one before the other, on this railway; and from the iron frame to which they are attached by their axles, two chests or receptacles made of iron are suspended by stiff rods, exactly like panniers from the back of a horse. The chests hang down very near to the surface of the ground; the load which is placed in these chests being so low, that the centre of gravity is always beneath the level of the rail, the machine, unless very unequally loaded, has no tendency to overset. The principal advantages of this contrivance are the following: A moderate fall of snow would produce no obstruction; it could be carried over uneven ground, and over small hollows or ravines, without cutting, embanking, or casting bridges, by merely lengthening or shortening the pillars; the lateral friction from the want of perfect parallelism in the two opposite rails of the ordinary railways is avoided; and in many cases the rail might even be carried along the side of a common cart-road with a very small additional expense. Mr. Palmer, who made some trials with a portion of railway formed in this manner, states, that the effect

produced by the draught of a single horse was nearly double of that produced on the common double railway, or 45,000 pounds, including the vehicle. There is nothing in the nature of this machine to render steam-power inapplicable to it.

In calculations respecting the power of a horse exerted in different modes, errors often arise from considering this power as a constant quantity, which it is not. At a dead pull an ordinary horse exerts a force of traction equal to 150 pounds; this is reduced to less than one-half when he travels four miles an hour; to one-ninth part when he travels eight miles an hour, and at twelve miles an hour his whole strength is expended in carrying forward his own body, and his power of traction ceases. It is supposed here that the horse performs pretty long journeys. When travelling very short stages, he may exert a force considerably greater; and his power of traction may perhaps cease only at a velocity of 14 or 15 miles an hour. But in common cases a velocity of 12 miles may be taken as the maximum, and for the convenience of calculation, the dead-pull may be taken at 144 pounds. Adopting then Professor Leslie's formula, the force of traction at any degree of velocity (v) will be (12). Thus, the force exerted, at two miles an hour, will be 100 pounds; at four miles, 64 pounds; at six miles, 36 pounds; at eight miles, 16 pounds; and at ten miles, only 4 pounds. For such horses as are employed in stage coaches, and for light cart horses, we have no doubt that this estimate is substantially correct. But there is a very great difference in the natural strength and speed of horses, and for those of a very heavy description, the force of traction may probably be assumed as one-half greater at low velocities, and the maximum speed as somewhat less. Mr. Stevenson found, by the dynamometer, that a very heavy draught horse (weighing 10 cwt. or 70 stone), in tracking a boat, exerted a force of 160 pounds. ' He unluckily does not mention the velocity; but this may be assumed as rather under two miles an hour. Estimating the force of traction at two miles an hour to be 150 pounds, Professor Leslie's formula will become (12-). The horse moving at one mile an hour would pull with a force of 181 pounds; at two miles, with 150 pounds; at three miles, with 120 pounds; at four miles, with 96 pounds; and at a dead pull he would exert a force of 225 pounds. It is proper to add, that the dead-pull here mentioned is a continued-not a momentary effort. It is the last term in the series of descending velocities, in all of which constant action is supposed, and means properly the effort which the horse could make if released from the exertion required to carry forward

'Article Railways. Dr. Brewster's Encyclopedia.

his own body. When the action is continued only for a few minutes, the horse may undoubtedly pull with twice the force which the formula exhibits, or even more. The maximum velocity also, for a heavy horse, should have been assumed a little lower; but not to embarrass ourselves with a multiplicity of formulæ, I have kept it unchanged. On these grounds, therefore, and to avoid every thing like exaggeration, I shall consider the force of traction exerted by a horse, moving at two miles an hour, as equivalent to about 150 pounds, or one-half greater than Professor Leslie's formula indicates. Steam-engine makers assume a horse-power to be equal to a weight of 180 or 200 pounds; but this is to be considered merely as an arbitrary and conventional standard, adopted for a particular purpose.

The resistance to the motion of a vessel in the sea or a canal is of an extremely different kind from that which a carriage of any kind experiences on a common road or a railway. In the former case it arises from the pressure of the water on the bow and sides of the vessel; in the latter, from the friction of the axle in its box, and that of the rim of the wheel on the gravel or iron rail. The motion of the body in both cases is resisted also by the air; but this resistance, which is small in amount, generally speaking, I shall throw entirely out of view in the first instance, in order to simplify the calculations.

On a well-made road a horse will draw a load of one ton, in a cart weighing 7 cwt., at the rate of two miles an hour. (Leslie's Elements, p. 253.) The whole strength of the horse is exerted in overcoming the friction. On such a road, therefore, a force of traction of 150 pounds moves a weight of 3000 pounds, or the friction is 1-20th part of the load (the cart included).

On a railway of the best construction, it has been already shown, that a horse travelling at the same rate of two miles an hour, draws 15 tons, including the vehicles. In this case, then, a power of traction of 150 pounds moves a weight of 33,600 pounds; the friction of course is 1-224th part-or in round numbers, 1-200th part of the load. Mr. Wood in his Practical Treatise on Railroads, published in May 1825, gives the results of thirteen experiments made by himself, from which it appears that the average amount of the friction is 1-204th part of the load. The rail was an edge rail; the diameter of the wheel was 34 inches, and that of the axle 1-12th of the wheel.

On a canal, a horse travelling at two miles an hour, draws 30 tons in a boat weighing probably 15 tons. Reducing the ton

I

Boats in some cases carry only 15 or 20 tons; in others 35 (as the coal boats on the Union Canal), but in the one case they travel quicker, and others slower, than the rate mentioned. The boats drawing 35 tons travel 27 miles in 15 hours, and work by relays.