CHAPTER XXIX.

LOCKS, PLANES, SLUICE-GATES AND LIFTS. The main difference between rivers and canals, is that the former are usually capable of being navigated without any artificial provision for overcoming differences of level, whereas canals are so constructed that differences of level are overcome by locks or lifts. There are, of course, many cases in which the navigation of a river is suddenly and effectually obstructed by differences of level which are unsurmountable. This is notably the case on the Niagara river, where the falls of that name interpose a bar to the further navigation of a stream which would otherwise be the natural connection between lakes Erie and Ontario. The same sort of obstruction is interposed to the navigation of the Gotha river in Sweden, by the Falls of Trolhätta. There are many cataracts on the Mississippi river and its tributaries which render navigation all but impossible. These natural barriers have in many cases been got over by the risky and difficult operation of “shooting the rapids”—a feat in which the red Indian navigators have long excelled. But while it may be possible with a canoe to overcome such obstructions without absolute disaster, it is manifest that such risks could never be run in the everyday business of commercial transport. For these reasons, it has, in not a few cases, been found expedient to overcome the obstructions to river navigation that are interposed by rapids or cataracts, by constructing an artificial waterway parallel to the falls, on which the rise or fall of the natural waterway is surmounted by locks or lifts. The Welland Canal performs this function in Canada, and the Gotha Canal in Sweden. There is practically no limit to the differences of level that may be met by this arrangement, always assuming that water supply can be commanded at the summit. [Illustration: VIEW FROM THE NIAGARA ESCARPMENT, LOOKING DOWN THE WELLAND CANAL TOWARDS LAKE ONTARIO.] Obviously, however, on canals, as on rivers, the fewer locks or lifts the better. The process of passing through a canal lock is tedious, and while it involves a considerable expenditure of time, it involves also a corresponding amount of cost. For ship canals, it is much better to have no locks or lifts whatever. This aim was kept in view in the laying out of the Suez and Panama Canals, but on the latter canal, the cost of cutting through the Culebra mountain was found to be so very considerable, and the financial position of the company was so unsatisfactory, that M. Eiffel submitted a proposal to make use of locks, which was adopted by M. de Lesseps and his colleagues as a _dernier ressort_.[272] The Panama Canal has proved that the avoidance of locks can only be purchased in an uneven country at an immense cost, and the canal engineer has therefore to consider whether the resources at his disposal will enable him to pay the price involved. The proposed Nicaraguan Canal is another enterprise that presents some interesting problems of this description. This canal will make use for a great part of the total distance between ocean and ocean—169 miles—of Lake Nicaragua, and as that lake is 600 feet above sea level, it is manifestly necessary to make use of locks. In other words, as the lake cannot be brought to the tide level of the canal, the canal must be carried up to the level of the lake. A lock chamber, enclosed by a double pair of gates is said to have been employed for the first time in Italy in 1481, the designers and builders having been two clock-makers in Viterbo of the name of Domenico. The State of Venice was the first to adopt the system, but before the end of the fifteenth century, Leonardo da Vinci had united the two chief canals of Milan by six such locks, having a fall of 17 braces.[273] On one of the canals constructed in Italy, between Padua and Vicenza, about the fifteenth century, there are several sluice-gates, or _pertuis_, which are said by Cresy[274] to have been thus contrived:— “The lower beam of each gate was framed with the head and heel posts, so as to allow a space of 6 inches between it and the sill. From the middle beam to the top, the gates were planked over in the ordinary way; the lower part was left open, or in skeleton framing, and was closed by paddles or sluices, which were moved up and down by a rack and pinion. When the paddles were let down, they descended 3 or 4 inches lower than the surface of the floor on the lower side, which acted as a rebate, against which they pressed, and effectually shut the lock. They also had a bearing against the lower cross-beam of the gate, and the head and heel posts rested on square stones made fast in the sill. “To make use of gates upon this construction, it was necessary first to raise the paddle as high as the lower cross-beam, which permitted the water to pass through at the foot of the gate. The paddles were then elevated to the height of the middle beam, which was placed at the ordinary level of the water, usually 4 or 5 feet deep upon the sill. “These gates were easily opened, as the boarded part was entirely out of the water, and a deposit on the floor of the chamber of the lock could form but little obstruction, as from the scour of the water the greater part would be washed away. The only serious objection to this early contrivance in aid of internal navigation is the injury that vessels might sustain at the time they were passing through, when one half of their length would be out of the water, producing a considerable strain upon them. The water passing through a space, walled in on both sides, would, to a certain extent, allow the barge or vessel to slide down an apparent plane; but, before it could again resume its level position, it would be subjected to another strain. These side walls were, however, made of considerable length, a foot being usually allowed for every inch of fall; a timber floor was laid throughout, to prevent the force of the water from deepening and undermining the foundations.” The Chinese, who have early distinguished themselves in many inventions that have been worked out and improved upon under our Western civilisation, introduced piers into their rivers and canals, in order to overcome the difficulties incidental to falls or shoal water. These piers have been termed by De la Lande half locks, and it has been remarked by Chapman that the casual position of two pairs of piers near to each other has no doubt suggested the invention of locks, as it would be seen, when the gates of the lower piers were closed, and of sufficient height, that the water would be nearly still between the upper pair of piers, and afford an easy passage, so that, in place of a single pair of piers, two pairs would be erected sufficiently near to each other for the purpose, and capacious enough to hold a fleet of boats. It would soon afterwards be found that in dry seasons the waste of water was greater than could be conveniently afforded, and the operation was tedious for single boats. Thus would progressively arise the invention of locks with walled chambers and sluices through their gates and walls. There are, or were recently, on some rivers, locks of the first construction, composed simply of two pairs of piers, without any connection of walls or pavement between them. The Kennet and the Lea have unwalled locks. Thomas Telford, when projecting the Inverness and Fort William Canal, on account of the great plenty of water, and size of the vessels to be used, proposed not to wall the locks the whole length, but to have earthen banks between the two pair of piers of masonry that support the upper and lower gates of the locks. It appears from M. De la Lande’s ‘Traité des canaux de Navigation,’ that the first lock was supposed to be erected in the year 1488, upon the Brenta, near Padua; and that shortly after, the two canals of Milan, between which there was a fall of nearly 34 feet were joined by means of six locks, similar in principle to those at present in use. The first lock that James Brindley erected appears to have been at Compton, on the Stafford and Worcester Canal; but they were not at that time uncommon in England, on several of the rivers, and on the Sankey Canal. PLANES. William Reynolds, of Ketley, in Shropshire was the first who contrived and executed an inclined plane (which was completed in 1788) for the passage of boats and their cargoes. It was found to answer the purpose, and continued in practical use. Thomas Telford has thus described the plane, in ‘Plymley’s Agricultural Report of Shropshire’ (p. 291): “Mr. Reynolds having occasion to improve the mode of conveying iron, stone and coals, from the neighbourhood of the Oaken-gates to the ironworks of Ketley, these materials lying generally at the distance of about a mile and a half from the ironworks, and at 73 feet above their level, he made a navigable canal,” called the Ketley Canal, “and instead of descending in the usual way, by locks, continued to bring the canal forward to an abrupt part of the bank, the skirts of which terminated on a level with the ironworks. At the top of this bank he built a small lock, and from the bottom of the lock, and down the face of the bank, he constructed an inclined plane, with a double iron railway. He then erected an upright frame of timber, in which, across the lock, was fixed a large wooden barrel; round this barrel a rope was passed and was fixed to a movable frame; this last frame was formed of a size sufficient to receive a canal boat. These boats were 20 feet in length, 6 feet 4 inches wide, 3 feet 10 inches deep, and each carrying 8 tons, and the bottom upon which the boat rested was preserved in nearly a horizontal position, by having two large wheels before and two small ones behind, varying as much in the diameters as the inclined plane varied from a horizontal plane. This frame was placed in the lock, the loaded boat was also brought from the upper canal into the lock, the lock gates were shut, and on the water being drawn from the lock into a side pond, the boat settled on a horizontal wooden frame, and as the bottom of the lock was formed with nearly the same declivity as the inclined plane, upon the lower gates being opened, the frame with the boat passed down the iron railway on the inclined plane on to the lower canal, which had been formed on a level with the Ketley iron works, being a fall of 73 feet. Very little water was required to perform this operation, because the lock was formed of no greater depth than the upper canal, except the addition of such a declivity as was sufficient for the loaded boat to move out of the lock; and in dry seasons, by the assistance of a small steam engine, the whole of the water drawn off from the lock was returned into the upper canal by means of a short pump. A double railway having been laid upon the inclined plane, the loaded boat in passing down brought up another boat containing a load nearly equal to one-third of that which passed down. The velocities of the boats were regulated by a brake acting upon a large wheel placed upon the axis, on which the ropes connected with the carriage were coiled. It appears that this plane has an inclination of about 22°, except near the extremities, where it diminishes to about 111°; and that about 400 tons of coals usually descend thereon daily.” In 1789 a copper medal, or halfpenny, having a representation of this plane on one side, and of the cast-iron bridge at Coalbrookdale on the other, was struck and issued by the Coalbrookdale Company. After the practicability of inclined planes had been established, by the success of the Ketley plane, few Acts were passed for a new canal, without a clause authorising the company to erect inclined planes, instead of locks, if they should be found most advisable. At Walkden Moor, an underground plane was completed in October 1797, upon the Bridgwater Canal, similar to the Ketley plane above described. Reynolds introduced another form of inclined plane on the Shropshire Canal, where there were three planes employed of 120, 126, and 207 feet rises. The Act for this canal was obtained in 1788, and it was completed and opened in 1792. These planes were of the same construction as those at Ketley, except that there were no locks at the top of the descending planes, but the latter were continued above the surface of the water in the upper canal and terminated in a cross beam, from which another plane and railway descended into the upper canal, this being intended to avoid the waste of water which locks at the top of the planes occasion. The first incline up which barges were conveyed in a large caisson containing water was at Blackhill, on the Monkland Canal, near Glasgow, the system having been previously introduced, on a small scale, on the Chard Canal in Somersetshire. The Blackhill incline, with a rise of 96 feet, and a gradient of 1 in 10, replaced two flights of four locks each. The wrought-iron caisson, 70 feet long, and 13⅓ feet wide, runs on twenty wheels, and carries barges of 60 tons on the incline. An incline with larger caissons was constructed at Georgetown in 1876, in substitution for two locks connecting the Chesapeake and Ohio Canal with the Potomac. This incline rises 39 feet, with a gradient of 1 in 12; and barges of 115 tons are transferred from the lower to the upper reach in 8 or 10 minutes. The caisson is 112 feet long, 16¾ feet wide, and 7⅚ feet high. It is carried on three trucks, with twelve wheels each, and is drawn up by wire cables worked by a turbine. A canal incline, as described by Mr. Vernon-Harcourt,[275] consists of two lines of way, laid on a steep uniform gradient, on which barges are drawn up or let down, by wire cables, from one reach to the next, either resting on a cradle, or water-borne, in a caisson, running on wheels on the incline. The cables wind round a drum at the top of the incline; and the ascending barge is generally more or less counterbalanced by another descending, whereby the tractive force required to pull the barge up is considerably reduced. Primitive inclines exist on the Bude Canal in Cornwall. Inclines are often used, as alternatives to locks, as at Hampton Court, on the Thames. Inclines, up which barges are drawn in cradles, were carried out on the most extensive scale on the Morris Canal in America, where there are twenty-three inclines with gradients of 1 in 10, and an average lift of 58 feet. The largest of these is 1100 feet long, and rises 100 feet; and barges of 70 tons are drawn up the inclines. LOCKS. In the great majority of cases, however, locks are the means adopted for overcoming differences of level. In Great Britain it is calculated that on the existing canal system of 2240 miles, there are 1901 locks, being at the rate of one lock to every 1·37 miles of canal, of which 931, or nearly one-half, are 80 feet long or more.[276] This, of course, means very slow transport and great loss of time. On the canal system between Birmingham and London there are about 130 locks in all. The loss of time due to the passage of locks arises from two causes, one of which, as Mr. Conder points out,[277] it is easy to calculate, while the other varies extremely according to the management of the line, and the nature and volume of the traffic. The rise or fall of the water in the lock occupies an ascertainable time, ranging from three to six minutes; but the time lost in entering and leaving the locks is less easy to calculate. With perfect arrangement the loss is very small; frequently it is, in fact, very considerable. “In the event of a heavy traffic being thrown on our canals, it will probably be advisable to double the locks, a communication being made practicable between the pair, in order to save half a lock full of water at each passage. With this arrangement much time as well as much water may be saved. The average retardation due to the hydraulic requirements alone of the locks on the English canals is from 1¾ to 2 minutes per mile, the average rise to be overcome being under 6 feet per mile of canal.” In France, on all the more important canal routes, the locks are designed to accommodate the large _péniches_ or boats of 270 tons burden, 116 feet long, and 16 feet beam, which are the usual craft employed. In cases of exceptional traffic, the locks are made 130 yards long by 13 yards wide, in order to allow of several vessels passing through together. These arrangements are very favourable to the transport of large quantities of freight, so much so, that it is no unusual thing to see 25 to 30 barges, each laden with 270 tons of coal, towed by a small tug of 20 horse-power, working a submerged chain or wire rope, which the tug raises from the bottom as it progresses, the rope being nipped between revolving pulleys. The time occupied in passing through a lock on the French canals used to amount to from 16 to 20 minutes at least. The time is spent in filling or emptying the lock, in closing and opening the lock-gates, and in passing the barge into and out of the lock-chamber. The adjustment of the water-level in the lock-chamber may be hastened by large sluices in the side walls of the lock. The moving of the lock-gates can be rapidly effected by hydraulic machinery. Delays have been experienced in dragging a barge into or out of the lock when it is nearly the width of the lock-chamber, owing to its acting like a piston, and preventing the flowing back of the water along the sides; but this inconvenience can be obviated by carrying the culverts for the sluiceways all along the side walls, and providing lateral openings through which the water finds an exit. The dimensions of the canal locks resolved upon in France, under the extension scheme of 1878, was 126·2 feet in length by 17 feet in clear width, and 6·56 feet of water on the sill of the lock-gate. Boats of 120 tons burden can make use of such locks without difficulty. No canal in England, except the Gloucester and Berkeley, has locks of this size. The nearest approach to such dimensions is that made by the Grand Junction Canal, with locks 87 feet 6 inches long, 15 feet in the clear, and a depth of 5 feet, allowing, however, for the passage of an 80-ton boat only. The Grand Union Canal, which is connected with the Grand Junction, has only locks of 78 feet by 7 feet 2 inches. It has been computed that the difference between the cost of locks for a 120-ton boat and that of locks suited for an 80-ton boat is not more than 3000_l._ per mile.[278] Assuming the accuracy of this figure, the cost of enlarging the dimensions of the principal British canals ought not to be a serious item. Locks provided with sluiceways running the whole length of the side walls have been constructed on the Aire and Calder navigation, on the Scheldt and Meuse Canal, and the Canal du Centre of France. These large sluiceways ensure the rapid filling or emptying of the lock; and by making several side openings along the side walls into the lock-chamber, the inflowing or outflowing currents are distributed so as to have no injurious effect on the vessel inside. [Illustration: _DOMINION OF CANADA._ GENERAL PLAN OF ENLARGED LOCK _ON THE_ _S^T LAWRENCE AND WELLAND CANALS_.] [Illustration: _E & F N Spon. London & New York_ “INK-PHOTO” SPRAGUE & CO. LONDON. VIEW FROM THE NIAGARA ESCARPMENT, LOOKING UP THE WELLAND CANAL TOWARDS LAKE ERIE.] Balanced cylindrical sluice-gates, rising and falling vertically in a circular well communicating with the sluiceways, have been adopted at the new locks of the Scheldt and Meuse Canal, and the Canal du Centre, for opening and closing the sluiceways easily and rapidly. The enlarged locks on the Canal du Centre can be filled or emptied in two minutes; and the passage of vessels through the locks takes less than half the original time. Mr. E. J. Lloyd, speaking from experience of canals in the Midlands, is of opinion that a multiple of the present size of lock which prevails throughout the Midland district would be best. This would enable the existing craft on those canals, and also most of the barges on larger navigations, to be used in the most economical way possible. It would greatly simplify the conduct and management of low-class mineral traffic, which does not require any care, and could be treated in a similar manner to traffic of the like description on railways—no crews being attached to the boats, which could be detached from the trains, and left at any roadside wharf, until they could be unladen at the convenience of the owners. There is, no doubt, as Mr. Lloyd points out, a distinct advantage in small craft for such traffic, as it is obvious that a coal merchant could purchase small boat-loads of different classes of coals, to suit his customers, who could not find capital and wharf space for large cargoes of one class of coal only, and this would apply in equal degree to traffic in road-stone, bricks, drain-pipes, building materials, and many other classes of undamageable goods, and these small craft might also ply successfully on short branch canals, in districts which would not produce a sufficient traffic to warrant a large expenditure in improvement. Such locks would also, of course, accommodate craft sufficiently large to cross the estuaries of rivers, and to approach any docks with safety, and if sufficient depth of waterway is provided in the improved main lines, say 8 feet, or thereabouts, short coasting voyages might also be undertaken by craft specially constructed to do so, and also to navigate the canals.[279] Mr. Lloyd thinks that the heavy cost involved in constructing canals of sufficient size to pass craft suitable for coasting and short continental voyages would be fatal to cheap conveyance. The largest locks hitherto constructed are those on the St. Mary’s Falls Canal, in the United States, and the Welland Canal in Canada. On the former canal the lock opened in 1851 is 515 feet in length and 80 feet wide. The great tidal lock at Eastham, on the Manchester Ship Canal, will be 600 feet long and 80 feet wide. _The Welland Canal_, which is in some respects the most important in Canada, was begun by a private company in 1824 and opened in 1829. The original locks were of wood, 110 feet by 22 feet by 8 feet, and they bulged out on each side of the chamber to such an extent that they had to be hewn down from time to time to let vessels pass through. The canal was enlarged in 1841, and again in 1871, the depth of the canal having, on the occasion of the last enlargement, been increased to 14 feet. The drawing (p. 415) shows the general plan of the enlarged lock on this canal. It is 270 feet between the gates, 45 feet between the side walls, and has 12 to 14 feet of water upon the mitre sill. The entire system of locks on the _Manchester Ship Canal_, now under construction, will be as under:— Three locks at Eastham, namely, one 600 feet long by 80 feet wide; one 450 feet long by 50 feet wide; one 150 feet long by 30 feet wide. Two locks at Latchford, namely, one 600 feet long by 65 feet wide; one 450 feet long by 45 feet wide. Two locks at Irlam: similar to Latchford. Two locks at Barton: similar to Latchford. Two entrance locks to docks: similar to Latchford. Small lock (Weston Marsh Lock), 229 feet long by 42 feet 8 inches wide, to connect the ship canal with the Weston canal. Weston Mersey Lock, opposite Weston Point; 600 feet long and 45 feet wide. Bridgwater Lock, opposite Bridgwater Dock, Runcorn, 300 feet long and 45 feet wide. Runcorn Old Quay Lock below Old Quay Docks, Runcorn, 300 feet long and 45 feet wide. The three locks above mentioned connect the ship canal with the Mersey Estuary at their several points. Two small barge locks at Walton and Stockton Heath, Warrington. [Illustration: MANCHESTER SHIP CANAL ENTRANCE LOCKS AT EASTHAM.] The largest canal locks hitherto contemplated are those designed by Col. Blackman for the proposed _Nicaragua Canal_. They are 700 feet long, 100 feet wide, and with 30 feet depth of water over the sill; whilst the lift proposed is from 50 to 120 feet. The filling of the lock-chamber was to be rapidly effected through 18 feet cast-iron pipes, built into the side walls along the whole length on each side, connected across the bottom of the lock-chamber by a series of 3-feet pipes, perforated by a number of 2-inch holes, from which the water was to be distributed over the whole area of the chamber in numerous small streams, so as to avoid any prejudicial agitation of the water. The emptying was to be similarly effected; and where a saving of water is important, the pipes were to discharge at the lower end into a series of long ponds, formed in terraces, so that most of the water might be used again for filling the lock. These arrangements are a large extension of the system of sluiceways, all along the side walls, with lateral openings, and of side ponds, already referred to. The most novel feature was the form of the caisson-gate, proposed to be constructed of wrought iron and steel, which, being increased in width towards the bottom, would become stronger in proportion to the depth, as the water-pressure increases. HYDRAULIC LIFTS. The adoption of hydraulic lifts, in the place of locks and inclines, has recently come prominently to the front. The first lift of this description[280] was erected on the Weaver Navigation in 1875, for the purpose of connecting that river with the Trent and Mersey Canal. In this case the difference of level between the canal and the river is rather over 50 feet. Two wrought-iron troughs are employed to raise and lower the barges. The troughs are 75 feet by 15½ feet, and have 5 feet depth of water. They rest on a central hydraulic ram, three feet in diameter, working in two hydraulic presses underground, which can be connected at pleasure, making the troughs counterbalance one another. One trough ascends as the other descends. Hydraulic power is only required when the descending trough reaches the water in the lift-pit, the motion of the troughs being effected by removing about six inches from the lower trough. This arrangement is very economical of time, inasmuch as a 100-ton barge can be transferred from the river to the canal, and another from the canal to the river, in eight minutes. Although the difference of level, as already stated, is 50 feet, only the final lift of 4½ feet requires the expenditure of hydraulic power. La Louviére hydraulic lift, which was only completed during the summer of 1888, is the largest hydraulic canal lift in the world. It was constructed for the Belgian Government by the Société Cockerill, of Seraing, from the designs and under the superintendence of Messrs. Clark, Stanfield and Clark, of Westminster, the consulting engineers of the Government, and the patentees of the system. The difference between the levels of the upper and lower canals—that is, the height the boats are raised—is 50 feet 6¼ inches. The lift consists of two pontoons or troughs, each 141 feet long by 19 feet broad, with 8 feet draught of water, and are capable of holding the largest size of barge that navigates the Belgian broad-gauge canal system. Such barges are capable of taking 400 tons of coal or other cargo, so that the total weight of the trough, water, and barge is not much under 1000 tons. This immense weight is supported on the top of a single colossal hydraulic ram of 6 feet 6¾ inches diameter, and 63 feet 9½ inches long, working in a press of cast iron, hooped continuously for greater security with weldless steel coils. The working pressure in this press is about 470 lb. to the square inch. The time actually occupied in the operation of lifting or lowering is only two and a half minutes. (See illustration at p. 141.) It is probable that there is a greater liability to accidents, with lifts than with either locks or inclined planes. Where a dead weight of some hundreds of tons has to be moved bodily, it must, of course, be necessary to provide correspondingly strong machinery, and this is not to be done without considerable cost. Several accidents have, indeed, recently occurred in hydraulic presses for lifts. One of these occurred with a steel press which burst under a pressure of 70 atmospheres. Another happened with a riveted steel-plate press, which leaked and rent under pressure. On the Brussels and Charleroi Canal it was recently proposed to apply an hydraulic lift, for which the Cockerill Company made a cylinder, in which tightness was obtained by the use of cast iron, and strength by the use of steel, the cylinder being cut with projecting rings, turned to receive steel hoops, which were bored to a slightly smaller diameter, put on hot, and allowed to contract. A cylinder 2·06 metres (6 feet 9 inches) in diameter and 2 metres (6 feet 7 inches) high, was subjected, by means of force-pumps, to an internal pressure of 131 atmospheres, or four times that which would be required in practice. The French engineers, following the Seraing system, formed their cylinders of a series of steel hoops fitting one into the other, and with the flanges of the two outside hoops drawn together by tie-rods, the inside being lined with brass 0·0025 metre thick, applied with the mallet, so that the water may not come into contact with the joints. A trial cylinder was tested up to 170 atmospheres without yielding. On the Brussels and Charleroi Canal, however, it was decided to substitute a tunnel for the intended lift, so that the Cockerill cylinders were not applied. Inclines, or lifts, are said by some authorities to effect a great economy, both in time and water, as compared with flights of locks. Much, however, must depend upon local circumstances. One problem that is likely to press for solution in the immediate future is that of constructing locks or lifts that will enable ship canals to be worked with facility and economy. The proposal of the late Mr. Eads to construct a ship railway across the Isthmus of Tehuantepec was intended to overcome the necessity of such a canal, and was, indeed, a form of lift, of the practicability of which, however, we still await a conclusive demonstration. There is, of course, a natural limit to the size of locks that it is possible to work. That limit, however, does not appear to have been reached in any locks hitherto constructed. The Eastham locks on the Manchester Ship Canal will be the largest hitherto made; but we have seen that locks, even 100 feet longer, have been proposed for the Nicaraguan Canal. Of course, by the application of steam power, canal locks may be made of larger dimensions, and there are some instances in which such power has been attended with much advantage. In 1868, steam-power was applied to the locks of the Delaware and Raritan Canal, and is said to have increased their capacity for traffic, and therefore that of the canal, by 50 per cent. The engine has two cylinders, 6 inches diameter, 12 inches stroke, and works a 3-feet drum, actuating a 1-inch wire rope, which passes over rollers, along the face of the lock, and round sheaves above and below. To this rope the boats are attached and hauled in and out, two at a time. The engine also raises and lowers the valves, opens and shuts the gates, and in one case works a swing bridge. For large docks (e. g. 600 feet by 800 feet) it has been proposed[281] to admit and take off the water by channels the full width of the length of the lock, the water entering and leaving the lock by a number of small sluiceways, through the walls at right angles to the axis. This water should, if possible, be supplied, not from the reach above, but from separate reservoirs. There will thus be two waterways at right angles to reach each other, one longitudinal for the passage of vessels only, and one transverse, for the passage of water only. This avoids the expense of maintaining the paddles in the lock gates, and all the risks attending longitudinal currents. The canal should be widened above and below by floating pontoons, not by fixed walls. The height of the lift may be varied as required up to 30 feet; a lift of 33 feet has been worked for twenty-five years with ease and safety. FOOTNOTES: [272] These locks will be found described and illustrated in a previous chapter. [273] Zendrini’s ‘Della Acque Correnti,’ c. 12. [274] ‘Encyclopædia of Civil Engineering.’ [275] Report of the Conference on Canals and Inland Navigation, held at the Hall of the Society of Arts in 1888, p. 5. [276] Report of the Select Committee on Canals, 1883, p. 125. [277] Paper on “Inland Transport in the Nineteenth Century by Land and Water.” [278] Mr. Conder on ‘Inland Transport in the Nineteenth Century by Land and by Water.’ [279] Report of the Conference on Canals and Inland Navigation, ‘Journal of the Society of Arts,’ for 1888. [280] On the Great Western Canal, a simple lift, with two counterbalancing troughs, was used in the early part of the century, but it was not found successful. [281] Min. Proc. I. C. E., vol. lxiii. p. 350.