CHAPTER XXVIII.

SYSTEMS OF TRANSPORT AND HAULAGE. The cost of transport, whether by land or by water, is necessarily largely affected by the method of propulsion or traction employed. On the ocean, on lakes, and, for the most part, on rivers as well, steam and wind are the systems available. On canals, however, the wind is practically impossible as a motive power, and steam is not always convenient. It has, therefore, become necessary and customary to employ other methods. Of these, the most common in Great Britain is horse traction, which, however, is often varied by manual labour on the towing path. In either of these forms, traction is slow, tedious, and costly, but there are many cases in which it is not possible to make use of any other system. Much depends upon the width of the canal, the number of locks that have to be passed through, and other conditions that affect the problem. It has, however, been placed beyond all doubt that where steam traction can be introduced, it is much more economical than either horse or manual labour. Steam may, of course, be employed in either of two ways—either in the form of a tug-boat, with a number of barges in tow, as on the great lakes of the United States; or, where the locks are not long and wide enough to permit of this system, in the form of a locomotive, instead of a horse, on the towing-path. The former system is, of course, much more general, and, so far as it is possible to judge from recorded experiments, much more satisfactory than the latter. But there are few towing paths that could not be adapted for a narrow-gauge railway, and a small locomotive engine might, therefore, be frequently employed where a steam-tug was out of the question. Besides the systems of traction already named, there are various systems of chain towage that have been employed, especially on the Continent, with more or less satisfactory results. These usually take the form of ordinary chain towage, by an endless chain or rope, laid along the bottom of the canal in lengths of two or three miles, the tug being drawn along by the engine pulleys engaging with the rope or chain; or endless chain towage, by which, as practised on the Rhone, the tug carries two independent engines, each of which puts in motion an endless chain drawn along by the tug. This chain, on the Rhone, receives a motion like that of the bucket chain of a dredge, but the upper part remains horizontal, while the lower follows the bottom of the canal, the length and weight of the chain being determined by the adhesion necessary to draw the tug. Another system which is practised in France to some extent, and especially on the Rhone, is that of a keel carrying at the stem or prow a large wheel with cams, which draws the boat along by pushing against the bottom, the initial motion being given by a steam engine. The moving of boats upon canals or narrow rivers, where sailing is impracticable, has always been attended with difficulties. Where the width and depth of water will admit, long oars have been used, worked by one or two men on each side of the vessel, as is done on the coal barges or lighters on the Thames. On the Tyne, at Newcastle, these keels are said to have been in use ever since 1378, and are rowed by an immense oar on one side, another being used at the stem to steer by, and so to counteract the tendency of this strange mode of rowing. It is said that the large oar is hung by an iron ring, so as to admit of its being laid on the gunwale of the keel, when not in use, but not of its being removed. Owing to the want of any regular and proper path on which horses could travel by the sides of rivers, the first hauling or towing of boats was performed by men. This still continues to be the case on the canals of China and some other countries; and in this country most of our navigable rivers were without horse towing-paths until the early part of the present century. Formerly ten or fifteen men were seen tugging at the hauling line of a barge on the Thames in the meadows of Twickenham. A good horse-path now begins at Putney bridge, on the south side, and continues uninterruptedly on one side or other of the river to the extreme points of the navigation. These essential appendages to navigation were even more recently adopted on the Severn river. The towing path on many of our old navigations is continually interrupted and broken off by mills and other obstacles without any bridges for the crossing of the towing horses and boys. On the Ouse river, below Bedford, the towing-path used to be interrupted at the end of almost every field by high and dangerous stiles, over which the ill-fated navigation horses had to leap, encumbered by their harness and the heavy rope. The records of the machines approved by the Academy at Paris, and the Cabinet of M. de Servier, printed in 1719, contain plates and descriptions of many different contrivances, designed for the propelling or rowing of boats on canals and rivers. One of these systems depends upon gaining an impulse or hold against the ground at the bottom of the river or canal, in one of which a small boat moved by oars was proposed to be employed in successively carrying forwards and dropping anchors whose ropes were to be attached to a horse-gin, on board of a barge, which was designed to tow or drag a great number of others. In another, a spiked wheel was proposed to roll on the bottom of the canal, attached by a frame, movable on hinges, at the stern of a barge, where a roller, turned by a winch, was to give motion to the spiked wheel, and propel the barge by means of an endless rope or chain. A second kind depended upon the same principles as an oar, except in the construction and mode of applying the power. On the 20th of July, 1796, one Thomas Potts took out a patent for the use of a large flap or oar moving upon a horizontal hinge, attached to a framed lever at the stern of a barge, intended, when the handle of this lever was lifted up by several men, to turn on its hinge and present but little resistance; but on the descent of the lever, its whole surface was, by the action of the men at the lever, to be exerted on the water for propelling the barge. In the year 1801, one Edward Steers took out a patent which seems to have differed but little from the above, except in having two paddles or oars. Robert Beatson took out a patent for applying the principle of luffer boards or Venetian blinds to several purposes, which he has explained at length in an essay printed in 1798; and he proposed to propel ships by large oars or fins of this kind to be hung on the sides thereof by hinges, and worked by a lever, as a rudder is by its tiller-poles, with square frames fixed on their ends, to push against the water behind the vessel. A third kind, depending on the reverse of the action of an undershot water-wheel, has had many advocates. Thomas Savery, in 1698, proposed the use of six or eight paddles, like those of a water-wheel, on each side of the vessel, fixed on an axis across the same, by the force of a capstan to be turned by men. In the year 1781, the Abbé Arnal proposed to apply the power of a steam engine on board of a vessel for working paddles. Soon after this period, there was employed on the Thames, at Westminster, a small barge with a water-wheel in a cavity in its stern, with a steam engine for working it, which was said to be the contrivance of Earl Stanhope, and had been tried with success against the tide in the river. In the year 1797 a vessel having rowers by its side, that made 18 strokes per minute, from the action of a steam engine on board, was tried on the Sankey Canal near Liverpool, by which it was propelled 10 miles and back again to the same place.[266] About the year 1800, Messrs. Hunter and Dickenson, took out a patent for a propeller for ships, which was tried in January 1801, on board of a Government sloop off Deptford on the Thames, and the sloop thereby made way against the tide at the rate of three knots an hour.[267] In the Journal of the Royal Institution, about the year 1802, there is a description of an improved application of the steam engine to the turning of a wheel for propelling boats; the cylinder of this engine was horizontal, and the wheels with paddles were in a cavity in the stem of the boat, which, therefore, had two rudders, one on each side of the wheel, connected together by cross rods. A vessel of this kind was constructed for the Forth and Clyde Company under the direction of Mr. Symington, the inventor, and, in a trial made in December 1801, drew three vessels of 60 and 70 tons burthen each, at the rate of 2½ miles per hour on their canal.[268] Robert Fulton exhibited a vessel on the Seine at Paris, in August 1803, having two wheels with paddles, worked by a steam engine, and it was reported that two other vessels were towed by it against the stream at the rate of three miles per hour. A fourth kind of boat propellors, depended upon the rotary motion of a screw or fliers, like those of a jack. Daniel Bushnel, in his attempts to navigate submarine vessels,[269] used oars, placed near the sides and top of the vessel, formed upon the principle of a screw, the axles of which entered the vessel, and by turning the same one way, the vessel was made to advance or descend by a contrary motion of the screw. John Vidler contrived a vessel—which was tried in the Thames at Westminster, about 1810—that had a boom hung by a universal joint (hooks) at the stern to a rotative axis, turned by a capstan upon the deck of the vessel. At the end of this boom was fixed a circle of strong flyers, just like those of a jack, which, by striking the water obliquely as the boom was turned round, propelled the vessel forward. Near to the flyers there was a collar on the boom that turned easily therein; to this collar ropes were attached, which were carried to different parts of the stern of the vessel, and by means of which the boom could be stopped when in motion, if it was desired to stop its propelling action on any temporary occasion, or the flies thereof could be let down into the water to any depth required, or be turned aside from the direct line of the vessel to steer her on any course, without expending so much of the propelling power upon the rudder as was usually done in steering. These are but a few of the many services that have either been proposed or applied to the propulsion of boats on rivers and canals. Most of them, it need hardly be added, were found to be failures, although in some cases they contained the germs of the remarkable progress that has since taken place in the matter of propulsion generally. The number of patents that have been taken out with a view to overcoming the difficulties incidental to canal haulage have been legion. The real gist of the matter is that no two waterways present exactly the same conditions, and no system of transport will be found to answer equally well in all cases, unless the circumstances under which it is applied are identical and parallel. Hence, it becomes important to show what has been done on different waterways to meet the special conditions that have existed, and the results of these different applications. In the earliest traction experiments made on the Elbe in 1720 a hempen rope was fastened on shore, the other end being wound up on board, and vessels were thus propelled. Nothing better than this rough system obtained for a hundred years, when, in 1820, Messrs. Tourasse and Courteaut designed special flat-bottomed tugs, 75 feet long and 17 feet wide, with a horse capstan for winding up the rope; and subsequently, on the Seine, a 6 horse-power steam-engine was substituted for the horse capstan. Chains next took the place of hempen ropes, and between 1820 and 1830 many chain-tugs were employed on French rivers; but the first systematic service was carried out in 1846 between Paris and Montereau (65 miles) with tugs designed by Mr. Dietz, which in their essential features are similar to those in use at the present day. These tugs drew 18 inches of water, and were fitted with engines of from 35 to 40 horse-power, actuating the drum on which the chain was wound, two sets of gear being provided for going up and down stream, respectively. The boiler pressure was 5½ atmospheres, and the expenditure of fuel 5½ lbs. per horse-power per hour. Subsequently the chain was laid further up the Seine, and it was also applied to some rivers in France. In Germany, in 1866, chain-tugs were running on 200 miles of the Elbe, and in the next ten or twelve years this system was in use on the Saale, the Brahe, and the Neckar. The Elbe tugs are 138 to 150 feet long and 24 feet wide, with 18 inches draught. On the other rivers of Germany they are somewhat smaller. The sides are of ¼-inch iron plate, and formerly the bottoms were of ½-inch iron, but now they are built of 4-inch pine planks, as suffering less from abrasion on dragging over a rough bed. There is a rudder at each end, the wheel being amidships. The engines are from 60 to 70 horse-power, and work with a pressure of from 5 to 7 atmospheres. In slight currents a single drum is sufficient, the chain being kept pressed against it by rollers, and the drum is nicked to prevent the slip of the chain, but ordinarily there are two drums, to which the engine power is transmitted by two sets of gearing with different rates of speed—one for working up stream, with great power and small speed; the other for down stream, with less power and greater speed. Projecting over each end of the tug are booms furnished with guide-rollers for the chain, which give increased steering facilities. The chains are from ¾ to 1 inch thick. When fractures occur, which is seldom, it is generally at the moment of the chain being first wound round the drum. Each drum is fitted with a brake, and at the ends of the booms there are clips, designed to prevent a running out of the chain in case of the brake failing to hold. Chain-towing has so increased on the Elbe that in 1874 there were twenty-eight tugs running regularly between Hamburg and Aussig (420 miles). On the Neckar, at the same date, five tugs were employed on 56 miles of chain, and this was to be extended for 30 miles more, from Heilbronn to Cannstatt. Experience has shown that chain-tugs have great advantages over paddle-tugs, even in smooth water, for in the latter 60 to 70 per cent. of the power is lost in slips. Another advantage of chain-towing is that it produces no wash or swell. The charge for transport by this system is said to average about ¼_d._ per ton per mile. In 1865 Mr. de Meseil, a Belgian, introduced a system of transport where a wire rope was substituted for the chain. The same system was taken up and improved by Max Eith of Wurtemburg, and worked with success on a 40-mile section of the Maas (from Namur to Liége). It was subsequently employed on canals in Holland and Belgium, and also on the Rhine. Extensive trials were also made on the Danube with satisfactory results. A wire-rope tug company in 1873 laid down the line from Bingen to Rotterdam, but worked the upper section only themselves, viz. from Bingen to Ruhrort (155 miles). From Ruhrort downwards a concession was granted to a Dutch company, who employed a special kind of tug, in which the rope passed over drums inside the vessel, similar to the chain-tug system; but the usual arrangement of having the rope outside the tug has been found most convenient, as it enables it to be easily cast off and taken up again when two tugs meet. The wire rope generally used on the Rhine is formed of forty-nine wires 0·189 inch thick, is 1·7 inch in diameter, and weighs 4¾ lb. per yard. It usually costs 10 _d._ per foot, which is about one-third the weight and cost per foot of an iron chain of equal strength. The first wire-rope tugs at work in Holland and Belgium had a 20 horse-power engine for the driving wheels, and another 10 horse-power engine to work a screw when going down stream clear of the rope. At each end, outside the tug, there are guide-wheels to keep the rope clear of the vessel, and at the centre are two large wheels which lead the rope on to a Fowler’s clip-drum, against which it is kept pressed by small rollers. To pick up the rope and pass it over the wheels and drum takes a quarter of an hour. The Danube Company’s tug _Nyitra_, which resembles the Rhine tugs, is 140 feet long, 24½ feet wide, and draws 3½ feet of water; the clip-drum is 10½ feet, and the adjoining wheels about 9 feet, in diameter. Against a current of 4¼ feet per second, it can draw eight barges, with a total load of over 2000 tons, at a speed of 3 miles an hour, with useful effect of 75 per cent. In chain-tugs this percentage is higher on account of the greater flexibility of the chain. Fractures of the rope seldom occur, in spite of the rocky bottom in certain sections of the river. The life of a wire rope may be taken at from four to six years. It has been found that wire-rope tugs cannot work in less than 3 feet of water, or only with difficulty, whereas chain tugs can work in one-half of that depth. As regards steering facility, they are much alike. The delay caused by fractures is an important item in the comparison. Repairs to chains usually occupy considerably less time than repairs to wire ropes. Chain tugs in any depth under 3 feet, and in sharp curves, are said to be preferable to rope tugs; in moderately strong currents, and in larger curves, they are about equal; but in canals, and in large deep rivers, rope tugs are the best, and both are superior, in ordinary circumstances, to paddle tugs. In canal tunnels, as in the 4-mile section between Mons and Paris, where steam cannot be used on account of the smoke, chain tugs, worked by a horse capstan, tow a barge through in one-third the time, and at one-fourth the cost, of the former system, when men were employed for towing. Where strong rapids are met with, special appliances called “grapins” are sometimes employed. This consists of an iron wheel of about 20 feet in diameter and 17½ tons weight, furnished with projections or picks, fixed in a well-hole at midships, and worked by a chain attached to the paddle-shaft. On ascending a river the “grapin” is lowered till the picks grip the bed, on which the wheel slowly turns, and the paddles, working at the same time, in this way tow barges over the strongest rapids. Busquet’s tug, which is used in France, works on a chain, though it is similar to a wire-rope tug. The _Baxter_ steamboat, used on the Erie canal, was the outcome of a competition invited by the State of New York for a prize of 20,000_l._ for the steamer which best fulfilled the following, viz. a mean speed of 3 miles per hour with a load of 200 tons, small cost, and no wash or swell. This steamboat is 100 feet long, 17½ feet wide, and about 9 feet deep, with a flat bottom and vertical sides, and, including engines and coal, weighs 52 tons. It carries a load of 200 tons, with a draught of 6 feet of water, and has an average speed of about 4 miles, but can work up to 7½ miles an hour. On the Saar coal canal Jacquel’s steam-tug system is in use, where the screw is within the body of the vessel, and surrounded by a cylinder, and is fed with water by two large channels leading from the sides of the vessel to the front of the screw.[270] The tugs of the Rhine are large, very tapering vessels; some of them have engines of from 600 to 700 horse-power, and they are provided with all the latest improvement for economising fuel. Vessels with two screws are preferred, as combining adequate power with small draught; nevertheless, when the river is very low, paddle-wheel tugs of the old type have to be resorted to. Towing by aid of a submerged cable was started some years ago, but it has since been abandoned, except in the most difficult part of the river between St. Goar and Bingen, where it has proved serviceable, especially when the water is low. A serious disadvantage of this system is that in descending the river the tug has to let go the cable, and act simply as a tug, for which it is not well suited. Improvements have been introduced in the vessels as well as in the tugs. Narrow iron vessels have been substituted for the broad wooden barges in order to reduce the tractive force. Some of these vessels are 1000 tons register; but vessels from 400 to 500 tons are the most common. On the Rhine, vessels forming one convoy are not connected together in trains, as in France, but each is provided with its tug, which is a great advantage where the navigation is difficult. Human labour is still employed for towage on some of the Dutch, Belgian, and German canals. Boats of from 15 to 26 tons are towed by men at a speed of 1 to 1⅓ miles per hour. Dr. Mitzen, a German authority, allows for this system of transport a duty of 11 miles a day, including all stoppages. Steam-tug boats on the Belgian canals are restricted to a speed of 2⅔ miles per hour, and on the wider rivers to 4½ miles per hour. On the canal joining the Tiege to the Vistula, steam-tugs draw trains of barges 410 feet long, the speed being restricted to three miles per hour. The steam-tugs put by Mr. Beardmore on the river Lea towed from 50 to 60 tons, at from two to two and a half miles per hour, in the cuts, three to three and a half miles per hour in the larger sections, and five miles per hour in the Thames. On the Grand Junction Canal the speed of a steamer towing one vessel is put from three to three and a half miles per hour. On the Rotterdam Canal, four boats, of 130 tons each, are towed by a screw steamer. Several attempts have been made on the Leeds and Liverpool Canal to introduce steam towage, and in the year 1879 the company tried a screw steamer with compound condensing engines, to tow six 40-ton barges on a river or deep canal. It was very quickly discovered that the vessel was next to useless on a shallow canal—the section of that particular waterway only averages from 40 feet to 50 feet in width at the surface, with flat sloping sides under water, tapering down to a mid-channel or gutter with an average depth of only 4½ feet—inasmuch as with that depth (in mid-channel only) a screw propeller of sufficient diameter could not be used to utilise the power of the engines without a very great amount of “slip” and churning of the water instead of doing useful work. It was also found that when the least obstruction took place by meeting other barges near bridges or sharp curves, causing the slowing up or stoppage entirely of the tug, the barges in tow would, so to speak, insist on running pell-mell into one another, for the simple reason that they could not apply a brake, and besides they used to get zig-zagged across the canal in every direction, which often caused a delay of fifteen or twenty minutes before all could be marshalled and got under weigh again. Another attempt has since been made, which utilised the power of the engines with more success. Two narrow boats of about five feet beam were braced side by side under one deck, with a longitudinal space of about three feet between each, and in this space was one paddle-wheel with a long-stroke horizontal engine on deck over each boat (two engines) driving a crank on each end of the paddle shaft, set at right-angles, and across the deck stood a locomotive boiler, each boat carrying its own proportion of the weight of the boiler. The funnel had to be placed at an angle of 45 degrees, so as to get under the very low bridges. This steamer towed fairly well five barges of coal, but caused a great waste in the canal, to the injury of the banks, and was subject to the steering difficulties whenever any obstruction took place, which in this canal are frequent, owing to its very tortuous character. The ordinary barges on the Leeds and Liverpool Canal have been utilised as tugs by putting in small engines of just sufficient power to drive a screw propeller as large as could be made available without a large percentage of positive “slip,” each tug carrying a paying cargo. When the first barge was fitted up in this way, it was found that it would tow two others very well at two miles an hour. In some parts of the canal where the depth is a little greater the speed would rise to 2½ and 2¾ miles an hour; and under similar conditions, with only one barge in tow, as high as 3¼ to 3½ miles an hour. At the latter speed, however, the displacement sets up a rolling wave along banks, which does injury, whereas at 2 to 2½ miles an hour there is no perceptible disturbance of the water at the sides, and only a very slight disturbance in the centre. A number of these steam barges are now employed on this canal, in addition to one for towing through Foulridge tunnel, one mile in length. This tug has both ends alike, with two propellers, one at the bow and one at the stern, as well as a rudder at bow and stern, so that the boat does not require to be turned about at each journey. Prior to the adoption of this tug, all barges had to be worked through the tunnel by men, who lay on their side on the gunwale of the boat, pushing it along with their feet against the tunnel wall, and taking 2 to 2¼ hours to travel the mile, whereas the tug tows two and three loaded barges at a time the same distance (one mile) in twenty to twenty-five minutes, the only hands required being the engineer and helmsman. The engine and boiler are placed as far aft as possible. The form of propeller is the result of a very exhaustive and costly series of experiments. With full-size ones in actual work, it gives the best results in shallow waters. It would not, however, be well adapted for deep-water towage. The helmsman can perform the following duties without leaving his helm, viz., start, stop, or reverse the engines, lower the funnel at bridges, blow the whistle and use the auxiliary steam jet for funnel. He can also observe the conditions of his boiler, for he has the water-gauge and steam-gauge in full view before him. Mr. Ald. Bailey, of Salford, has given the following interesting details of the cost of a steamer for twenty-four hours’ work, towing two barges fully loaded, on the Leeds and Liverpool Canal:—[271] COST OF STEAMER. £ _s. d_. One captain 0 4 8 One mate 0 4 8 Two ordinary hands 0 8 0 Gas coke for engines: 24 cwt. at 6_s_. 8_d_. per ton 0 8 0 Tallow (2 lb.) at 5_d_. 0 0 10 Oil (2 quarts) at 10_d_. 0 1 8 Stores, waste and lights 0 1 0 COST OF TWO BARGES. Two captains at 4_s_. 4_d_. 0 8 8 Two ordinary hands at 4_s_. 0 8 0 Five per cent. interest, and 10 per cent. depreciation, on first cost of steamer and barges (£1000) for one day 0 8 3 Fifteen per cent. of steamer and barges for repairs per day 0 8 3 ────────── £3 1 8 The distance averaged in twenty-four hours (including locks) was 40 miles. The weight carried was—steamer, 35 tons; barges, each 40 tons; total 115 tons. The cost was about one-sixth of a penny per ton per mile. Mr. Bartholomew, of the Aire and Calder Navigation, has introduced a system of a train of boats about ten or twelve in number, each carrying about 40 tons, 20 feet long, 16 feet wide, and 7 feet 6 inches deep, propelled by a steam tug. By having a tug behind the train of boats, greater control of the steaming power is obtained. The boats are threaded together by means of wire rope controlled by two cylinders which are self-acting, and are under the charge of the man who is steering. By lengthening and shortening the wire ropes on each side of the train, it can be guided to go to any curve by making it convex or concave, the train being left to rise and fall vertically according to any little variation of headline. Buffers are attached to the ends of the boats, which have a tendency to bring them back again into line in case of any slight disorganisation caused by wind or water, the full control of the train and its direction being under the guidance of the steerer. This system, however, could not be introduced on many of the canals in England, unless larger locks were made, or inclined planes to get from one level to another. The system has been well described as a train of waggons on water without wheels. On the Gloucester and Berkeley Canal, Mr. Clegram found that, after allowing 15 per cent. for interest and depreciation, the cost of steam haulage amounted to 1/11th of a penny per ton per mile, being a saving of two-thirds as compared with horse power. With a heavier trade, however, which allowed the barges to be more generally employed, the work was done for 1/16th of a penny per ton per mile. In a number of cases both chain and wire rope haulage has been tried unsuccessfully on English canals, but that, no doubt, has been owing to their peculiar local circumstances. The wire rope system has been tried on the Bridgwater Canal and found unworkable owing to the large number of bends and turns and the difficulty of working the traffic in different directions. The chain system of haulage was tried on the Grand Junction Canal of Ireland as far back as 1860, but it was soon abandoned as impracticable, and steam power was substituted. On the canals of Deûle and Neufossés locomotive haulage is employed for a total length of about 50 miles. The line is of metre gauge, and the locomotives, of which there are twenty-two, weigh from six to ten tons each. The speed employed, however, is only about 1¼ miles per hour, at which rate each locomotive can draw about 1000 tons. In some interesting experiments lately made on French canals, a railway was laid down on the towing-path, about a yard from the brink of the canal, and a small locomotive of about four tons weight was placed upon it. The wheels were coupled and geared, with a driving wheel making 140 revolutions per minute, and allowing a maximum speed of 7 miles per hour. The engine, which was worked by one man, was attached to a cable about 80 yards long, and then drew a team of barges with complete success. It was found capable of drawing a net load of 100 tons of goods for each ton of its own weight. The actual speed was 2·4 miles per hour, and the average speed, allowing for stoppages, 1·8 miles per hour. With horses the average speed on the same canals was only 0·9 mile per hour, so that an important saving in time, as well as of expense, was obtained. The system has since been tried on a larger scale upon the canals between Dunkirk and Paris. It seems, on a survey of the various systems heretofore applied to canal towage, that they may be divided into two categories. In the most important of these, the fulcrum lies out of the water, as in chain and wire-rope towage, in the employment of grapplers, in locomotive towage, and in the use of horses and men. In the other category, we find paddle-wheels and screw-propellers, which have their fulcrum in the water. In the former category, the amount of power utilised is much greater than in the latter, and, for that reason, chain, wire-rope, or locomotive towage would appear to be preferable, more especially so, as the use of screw propellers or paddle-wheels has a tendency to damage the embankments of the canal, and thereby to increase the expense of maintenance. [Illustration: CABLE TRACTION ON THE ST. MAURICE CANAL.] [Illustration: PLAN OF THE ST. MAURICE CANAL, SHOWING CABLE TRACTION.] During the year 1888, experiments were carried out on the Saint Maurice canal with a system of cable haulage introduced by M. Levy, which seems to be of some value. An endless cable, supported by pulleys on posts along the banks of the canal, is set in motion by a hauling engine situated at some convenient point, and the barges which are attached to this cable are thus drawn along. On one side of the canal the cable runs in one direction, and on the other side it runs in the opposite direction, so as to accommodate both up and down traffic. Notwithstanding the extreme simplicity of the idea, there occur considerable difficulties in its practical application, the most formidable of these being the danger that, by the oblique pull from the barges, the cable may be thrown off its supporting pulleys into the water, especially where there occurs a bend in the canal. To prevent the cable from leaving the pulleys, the latter are provided with deep flanges; but as these would prevent the easy passage of the oblique hauling rope, some special provision had to be made for this purpose. The flange on the water side of each pulley has two gaps, as shown in the drawings (pp. 405-406), and as the cable with its hauling rope passes into the groove, one or the other of these gaps engages the oblique rope, but not the cable which passes on in a straight line. The rope passing through the gap is thus shunted out of the groove, and passes clear of the pulley. The attachment of the rope to the cable is shown at 3. At certain intervals along the cable are attached ferrules, between which is a shackle A, which can freely revolve. Through this shackle is passed the hauling rope, made fast upon itself by an easily detachable clamp D, from which a line is taken on board. By a pull at this line the clamp is unfastened, and the hauling rope is slipped through the shackle, so that the man in charge of the barge can at any moment disconnect the latter from the cable. The speed of the cable is from 2¼ to 2½ miles per hour, and with this speed no difficulty was experienced in making the attachment. The difficulty, however, was to impart motion to the barge without unnecessarily straining the cable. It will be easily understood that when a weight of 200 tons to 300 tons has to be set in motion, even at a comparatively slow speed, the acceleration must not be too great, otherwise the strain on the cable and hauling rope would be excessive. The attachment must therefore not be an absolutely rigid one, and, to give time for the gradual starting of the barge, the hauling rope is taken round a brake drum, and allowed to slip at first, so that the barge may be gradually set in motion; the brake is then locked, and the only further attention required is the steering. At the end of the length of canal served by the rope, the bargeman simply pulls the line, and the momentum of the barge is sufficient to carry it on to the next section, where it would be similarly attached to a running cable. [Illustration: CABLE TRACTION ON THE ST. MAURICE CANAL.] The illustration on p. 404, reproduced from _Industries_, shows the engine house by the side of the canal bank: and a plan of the experimental installation as at present carried out is shown on p. 405. The results have been so encouraging, that it is intended to equip about 6½ miles of canal with this system. Compared with horse haulage, there is said to be a considerable gain in speed; and, as far as can be judged at present, the cost of haulage is reduced from 10 to 30 per cent. FOOTNOTES: [266] ‘Monthly Magazine,’ vol. iv. p. 75. [267] Ibid., vol. xi. p. 195. [268] ‘Agricultural Magazine,’ vol. vii. p. 152. [269] ‘Transactions of the American Philosophical Society,’ vol. iv. p. 303. [270] These particulars are abstracted, through the “Minutes of Proceedings of the Institution of Civil Engineers,” from the ‘Zeitschrift für technische Hochschulen’ for 1881. [271] Paper read before the Manchester Association of Engineers.