Chapter V (1)

I have thus endeavored to trace the history of the building of a railway; and it must have been seen, from what has been said, that the evolution of the railway and of its rolling stock follows the same laws which govern the rest of the world: adaptation to circumstances decides what is fittest, and that alone survives. The scrap-heap of a great railway tells its own story.

Our railways have now reached a development which is wonderful. The railways of the United States, if placed continuously, would reach more than half-way to the moon. Their bridges alone would reach from New York to Liverpool. Notwithstanding the number of accidents that we read of in the daily papers, statistics show that less persons are killed annually on railways than are killed annually by falling out of windows.

Railways have so cheapened the cost of transportation that, while a load of wheat loses all of its value by being hauled one hundred miles on a common road, meat and flour enough to supply one man a year can, according to Mr. Edward Atkinson, be hauled 1,500 miles from the West to the East for one day's wages of that man, if he be a skilled mechanic. If freight charges are diminished in the future as in the past, this can soon be done for one day's wages of a common laborer.

The number of persons employed in constructing, equipping, and operating our railways is about two millions.

The combined armies and navies of the world, while on peace footing, will draw from gainful occupations 3,455,000 men.

Those create wealth--these destroy it. Is it any wonder that America is the richest country in the world?

The rapidity with which it is possible to build railways over the prairies of the West is extraordinary. It is true that the amount of earth necessary to be moved is much less than on the railways of the East. In Iowa and Wisconsin, the amount runs from 20,000 to 25,000 yards per mile, while in Dakota it is only 12,000 to 15,000 yards per mile. After making all due allowance for this, the result is still remarkable.

The Manitoba system was extended in 1887 through Dakota and Montana, a distance of 545 miles. A small army of 10,000 men, with about 3,500 teams, commanded by General D. C. Shepard, of St. Paul, a veteran engineer and contractor, did it all between April 2 and October 19. All materials and subsistence had to be hauled to the front, from the base of supplies. The army slept in its own tents, shanties, and cars. The grading was cast up from the side ditches, sometimes by carts, and sometimes by the digging machine.

Everything was done with military organization, except that what was left behind was a railway and not earth-work lines of defence. Assuming that this railway, ready for its equipment, cost $15,100 per mile, or $8,175,000, and if it be true, as statisticians tell us, that every dollar expended in building railways in a new country adds ten to the value of land and other property, then this six months' campaign shows a solid increase of the wealth of our country of over eighty millions of dollars. Had it been necessary for our Government to keep an army of observation of the same size on the Canadian frontier, there would have been a dead loss of over eight millions of dollars, and the only result would have been a slight reduction of the Treasury surplus.

It must be remembered that this railway was built after the American system: when the rails were laid, so as to carry trains, it was not much more than half finished; the track had to be ballasted, the temporary wooden structures replaced by stone and iron, and many buildings and miles of sidings were yet to be constructed. But it began to earn money from the very day the last rail was laid, and out of its earnings, and the credit thereby acquired, it will complete itself.

And this is only one instance out of many. The armies of peace are working all over our country, increasing our wealth, and binding all parts into a common whole. We have here the true answer to the Carlyles and the Ruskins who ask: "What is the use of all this? Is a man any better who goes sixty miles an hour than one who went five miles an hour?" "Were we not happier when our fields were covered with their golden harvests, than now, when our wheat is brought to us from Dakota?"

The grand function of the railway is to change the whole basis of civilization from military to industrial. The talent, the energy, the money, which is expended in maintaining the whole of Europe as an armed camp is here expended in building and maintaining railways, with their army of two millions of men. Without the help of railways the rebellion of the Southern States could never have been put down, and two great standing armies would have been necessary. By the railways, aided by telegraphs, it is easy to extend our Federal system over an entire continent, and thus dispense forever with standing armies.

The moral effect of this upon Europe is great, but its physical effect is still greater. American railways have nearly abolished landlordism in Ireland, and they will one day abolish it in England, and over the continent of Europe. So long as Europe was dependent for food upon its own fields, the owner of those fields could fix his own rental. This he can no longer do, owing to the cheapness of transportation from Australia and from the prairies of America, due to the inventions of Watt, the Stephensons, Bessemer, and Holley.

With the wealth of the landlord his political power will pass away. The government of European countries will pass out of the hands of the great landowners, but not into those of the rabble, as is feared. It will pass into the same hands that govern America to-day--the territorial democracy, the owners of small farms, and the manufacturers and merchants. When this comes to pass, attempts will be made to settle international disputes by arbitration instead of war, following the example of the Geneva arbitration between the two greatest industrial nations of the world. Whether our Federal system will ever extend to the rest of the world, no one knows, but we do know that without railways it would be impossible.

When we consider the effects of all these wonderful changes upon the sum of human happiness, we must admit that the engineer should justly take rank with statesmen and soldiers, and that no greater benefactors to the human race can be named than the Stephensons and their American disciples--Allen, Rogers, Jervis, Winans, Latrobe, and Holley.

FOOTNOTES:

[1] It is proper here to say that English engineers now appreciate the merits of the American swivelling truck or bogie. In the article on Railways in the last edition of the "Encyclopædia Britannica," speaking of locomotives, the author of the article, who is an English engineer of high authority, says: "American practice, many years since, arrived at two leading types of locomotive for passenger, and for goods traffic. The passenger locomotive has eight wheels, of which four in front are framed in a bogie, and the four wheels behind are coupled drivers. _This is the type to which English practice has been approximating._" The italics are ours.

[2] The statistics of ten leading English and ten leading American lines, given by Dorsey, show the following results: 1. The cost per year of the rations, wages, fuel of an American locomotive is $5,590; of an English locomotive, $3,080. 2. Average yearly number of train-miles run by American locomotive, 23,928; English locomotive, 17,539. 3. Yearly earnings: American locomotive, $14,860; English locomotive, $10,940, although the English freight charges are much greater than those of the United States.

[3] The writer has obtained many of the statistics used in this article from A. M. Wellington's "Economic Theory of Railway Location," a perfect mine of valuable information upon all such matters.

[4] The amount of permanent wood and iron truss bridges, and of temporary wooden trestles on the Chicago, Milwaukee, and St. Paul is as follows:

Truss bridges, 700 spans, average 93 feet, 12-4/5 miles.
Trestle " 7,196 " " 77 " 103-1/10 "
------ --------
Total, 7,896 115-9/10 "

The approximate total number of bridges in the United States was in 1888:

Iron and wood truss bridges, 61,562 spans, 1,086 miles.
Wooden trestles, 147,187 2,127 "
-------- ------
Total, 208,749 3,213 "

Probably three-fourths of the truss bridges are now of iron or steel, and may be considered perfectly safe so long as the trains remain upon the rails and do not strike the side trusses. The wooden trestles are a constant source of danger from decay or burning or from derailed trains, and should be replaced by permanent structures as fast as time and money will allow.

[5] See following article on "Feats of Railroad Engineering," page 86.

[6] For fuller description of work in a caisson see "Feats of Railway Engineering," page 69.

[7] See "Feats of Railway Engineering," page 55.

FEATS OF RAILWAY ENGINEERING.

BY JOHN BOGART.

Development of the Rail--Problems for the Engineer--How
Heights are Climbed--The Use of Trestles--Construction on a
Mountain Side--Engineering on Rope Ladders--Through the Portals
of a Cañon--Feats on the Oroya Railroad, Peru--Nochistongo
Cut--Rack Rails for Heavy Grades--Difficulties in Tunnel
Construction--Bridge Foundations--Cribs and Pneumatic
Caissons--How Men work under Water--The Construction of Stone
Arches--Wood and Iron in Bridge-building--Great Suspension
Bridges--The Niagara Cantilever and the enormous Forth
Bridge--Elevated and Underground Roads--Responsibilities of the
Civil Engineer.

There are one hundred and fifty thousand miles of railway in the United States: three hundred thousand miles of rails--in length enough to make twelve steel girdles for the earth's circumference. This enormous length of rail is wonderful--we do not really grasp its significance. But the rail itself, the little section of steel, is an engineering feat. The change of its form from the curious and clumsy iron pear-head of thirty years ago to the present refined section of steel is a scientific development. It is now a beam whose every dimension and curve and angle are exactly suited to the tremendous work it has to do. The loads it carries are enormous, the blows it receives are heavy and constant, but it carries the loads and bears the blows and does its duty. The locomotive and the modern passenger and freight cars are great achievements; and so is the little rail which carries them all.

The railway to-day is one of the matter-of-fact associations of our active life. We use it so constantly that it requires some little effort to think of it as a wonderful thing; a creation of man's ingenuity, which did not exist when our grandfathers were young. Its long bridges, high viaducts, and dark tunnels may be remarked and remembered by the traveller, but the narrow way of steel, the road itself, seems but a simple work. And yet the problem of location, the determination, foot by foot and mile by mile, of where the line must go, calls in its successful solution for the highest skill of the engineer, whose profession before the railway was created hardly existed at all. Locomotives now climb heights which a few years ago no vehicle on wheels could ascend. The writer, with some engineer friends, was in the mountains of Colorado during the summer of 1887, and saw a train of very intelligent donkeys loaded with ore from the mines, to which no access could be had but by those sure-footed beasts. Within a year one of that party of engineers had located and was building a railway to those very mines. No heights seem too great to-day, no valleys too deep, no cañons too forbidding, no streams too wide; if commerce demands, the engineer will respond and the railways will be built.

The location of the line of a railway through difficult country requires the trained judgment of an engineer of special experience, and the most difficult country is not by any means that which might at first be supposed. A line through a narrow pass almost locates itself. But the approach to a summit through rolling country is often a serious problem. The rate of grade must be kept as light as possible, and must never exceed the prescribed maximum. The cuttings and the embankments must be as shallow as they can be made--the quantities of material taken from the excavations should be just about enough to make adjacent embankments. The curves must be few and of light radius--never exceeding an arranged limit. The line must always be kept as direct as these considerations will allow--so that the final location will give the shortest practicable economical distance from point to point. Many a mile of railway over which we travel now at the highest speed has been a weary problem to the engineer of location, and he has often accomplished a really greater success by securing a line which seems to closely fit the country over which it runs without marking itself sharply upon nature's moulding, than if he had with apparent boldness cut deep into the hills and raised embankments and viaducts high over lowlands and valleys.

But roads must run through many regions where very different measures must be taken to secure a location practicable for traffic. For instance, a line at a high elevation approaches a wide valley which it must cross. The rate of descent is fixed by the established maximum grade, and the sides of the valley are much steeper than that rate. Then the engineer must gain distance--that is to say, he must make the line long enough to overcome the vertical height. This can often be accomplished by carrying it up the valley on one side and down on the other. Tributary valleys can be made use of if necessary, and the desired crossing thus accomplished. But at times even these expedients will not suffice. Then the line is made to bend upon itself and wind down the hillside upon benches cut into the earth, or rock, curving at points where nature affords any sort of opportunity, and reaching the valley at last in long convolutions like the path of a great serpent on the mountain side. These lines often show several tiers of railway, one directly above the other, as may be seen in the illustrations on pages 49 and 51.

The long trestle shown in the illustration opposite is an example of an expedient often of the greatest service in railway construction. These trestles are built of wood, simply but strongly framed together, and are entirely effective for the transport of traffic for a number of years. Then they must be renewed, or, what is better, be replaced by embankment, which can be gradually made by depositing the material from cars on the trestle itself. The trestle illustrated is interesting as conforming to the curve of the line, which in that country, the mountains of Colorado, was probably a necessity of location.

* * * * *

Where the direct turning of a line upon itself may not be necessary, there may and often must be bold work done in the construction of the road upon a mountain side. It must be supported where necessary by walls built up from suitable foundations, often only secured at a great depth below the grade of the road. Projecting points of rock must be cut through, and any practicable natural shelf or favorable formation must be made use of, as in the picture on page 61. In some of the mountain locations, galleries have been cut directly into the rock, the cliff overhanging the roadway, and the line being carried in a horizontal cut or niche in the solid wall.

The Oroya and the Chimbote railways in South America demanded constant locations of this character. At many points it was necessary to suspend the persons making the preliminary measurements from the cliff above. The engineer who made these locations told the writer that on the Oroya line the galleries were often from 100 to 400 feet above the base of the cliff, and were generally reached from above. Rope ladders were used to great advantage. One 64 feet long and one 106 feet long covered the usual practice, and were sometimes spliced together. The side ropes were ¾ and 1¼ inches in diameter, and the rounds of wood 1¼ inches in diameter, and 16 inches and 24 inches long. These were notched at the ends and passed through the ropes, to which they were afterward lashed. These ladders could be rolled up and carried about on donkeys or mules. When swung over the side of a cliff and secured at the top, and when practicable at the bottom, they formed a very useful instrument in location and construction. For simple examination of the cliff, and for rough or broken slopes not exceeding 70 to 80 degrees, an active fellow would, after some experience, walk up and down such a slope simply grasping the rope in his hands. If required to do any work he would secure the rope about his body, or wind it around his arm, leaving his hands comparatively free for light work.

The boatswain's chair--consisting of a wooden seat 6 inches wide and two feet long, through the ends of which pass the side ropes, looped at the top, and having their ends knotted--is a particularly convenient seat to use where cliffs overhang to a slight degree. The riggers were generally Portuguese sailors, who seemed to have more agility and less fear than any other men to be found. At Cuesta Blanca, on the Oroya, a prominent discoloration on the cliff served as a triangulation point for locating the chief gallery. Men were swung over the side of the cliff in a cage about 2½ feet by 6 feet, open at the top and on the side next the rock. This was a peculiar cliff about 1,000 feet high, rising from the river at a general slope of about 70 degrees. The grade line of the road was 420 feet above the river. The Chileno miners climbed up a rope ladder to a large seam near the grade, where they lived; provisions, water, etc., being hoisted up to them. The first men sent over the cliff to begin the preliminary work were lowered in a cage and took their dinners with them, for fear they would not return to the work, and that unless a genuine start was made others could not be induced to take their places. It is safe to say that 80 per cent. of the sixty odd tunnels on the Oroya and the seven tunnels on the Chimbote lines were located and constructed on lines determined by triangulation, and the results were so satisfactory that the method may be depended upon as the best system for determining topographical data or for locating and constructing the lines in any similar locality.

Where the rocks close in together, as in some of the cañons of our Southwest, the railway curves about them and finds its way often where one would hardly suppose a decent wagon road could be built. The portals of the Grand River Cañon, as here shown, present such a line, passing through narrow gateways of rock rising precipitously on either side to enormous heights.

When such a cañon or a narrow valley directly crosses the line of the road, it must be spanned by a bridge or viaduct. The Kentucky River Bridge, shown below, is an instance. The Verrugas Bridge, on the Lima and Oroya Railroad in Peru, is another. This bridge is at an elevation of 5,836 feet above sea-level. It crosses a ravine at the bottom of which is a small stream. The bridge is 575 feet long, in four spans, and is supported by iron towers, the central one of which is 252 feet in height. The construction was accomplished entirely from above, the material all having been delivered at the top of the ravine, and the erection was made by lowering each piece to its position. This was done by the use of two wire-rope cables, suspended across the ravine from temporary towers at each end of the bridge.

On the line of the same Oroya Railroad is a striking example of the difficulties encountered in such mountain country and of the method by which they have been overcome. A tunnel reaches a narrow gorge, a truss is thrown across, and the tunnel continued. Nature's wildest scenery, the deep ravine, the mountain cliffs, and the graceful truss carrying the locomotive and train safely over what would seem an impossible pass, here combine to give a vivid illustration of an engineering feat.

The location of a part of the Mexican Central Railway through the cut of Nochistongo is peculiarly interesting. Far underneath the level of this line of railway there was skilfully constructed, in 1608, a tunnel which at that period was a very bold piece of engineering. It was designed to drain the Valley of Mexico, which has no natural outlet. This tunnel was more than six miles long and ten feet wide. It was driven through the formation called _tepetate_, a peculiar earth with strata of sand and marl. It was finished in eleven months. At first excavated without a lining, it was afterward faced with masonry. It was not entirely protected when a great flood came, the dikes above gave way, and the tunnel became obstructed. The City of Mexico was flooded, and it was decided that, instead of repairing the tunnel an open cut should be made. The engineer who had constructed the tunnel, Enrico Martinez, was put in charge of this enormous undertaking, and others took his place after his death. The cut is believed to be the largest ever made in the world. For more than a century the work was continued. Its greatest depth is now 200 feet. It was cut deeper, but has partially filled with the washings from the slopes. The cost was enormous, more than 6,000,000 dollars in silver having been actually disbursed! Wages for workmen were then from 9 to 12 cents a day. All convicts sentenced to hard labor were put at work in the great cut. The loss of life was very great. Writers of the time state that more than 100,000 Indians perished while engaged in the work.

When a line of railway encountered a grade too steep for ascent by the traction of the locomotive, the earlier engineers adopted the inclined plane. Such planes were in use at important points during many years. Notable instances were those by which traffic was carried across the Alleghany Mountains, connecting on each side with the Pennsylvania railway lines. These old planes are still visible from the present Pennsylvania Railroad where it crosses the summit west of Altoona. The planes were operated by stationary engines acting upon cables attached to the cars. These cables passed around drums at the head of the planes, the weight of the cars on one track partially balancing those on the other. Similar planes were in use also at Albany, Schenectady, and other places.

Another effective expedient is the central rack rail. No better or more successful example of this method of construction can be given than the Mount Washington Railway, illustrated above. The road was completed in 1869. Its length is 3-1/3 miles and its total rise 3,625 feet. Its steepest grade is about 1 foot rise in every 3 feet in length; the average grade is 1 in 4. It is built of heavy timber, well bolted to the rock. Low places are spanned by substantial trestle work. The gauge of the road is 4 feet 7½ inches, and it is provided with the two ordinary rails and also the central rack rail, which is really like an iron ladder, the sides being of angle iron and the cross-pieces of round iron 1½ inches in diameter and 4 inches apart. Into these plays the central cog-wheel on the locomotive, which thus climbs this iron ladder with entire safety. Very complete arrangements are made to control the descent of the train in case of accident to the machinery. The locomotive is always below the train, and pushes it up the mountain. Many thousands of passengers have been transported every year without accident.

The rack railroad ascending the Righi, in Switzerland, was copied after the Mount Washington line. Some improvements in the construction of the rack rail and attachments have been introduced upon mountain roads in Germany, and this system seems very advantageous for use in exceptionally steep locations.

* * * * *

When a line of railway meets in its course a barrier of rock, it is often best to cut directly through. If the grade is not too far below the surface of the rock, the cut is made like a great trench with the sides as steep as the nature of the material will allow. Very deep cuts are, however, not desirable. The rains bring down upon their slopes the softer material from above, and the frost detaches pieces of rock which, falling, may result in serious accidents to trains. Snow lodges in these deep cuts, at times entirely stopping traffic, as in the blizzard near New York, in March, 1888. A tunnel, therefore, while perhaps greater in first cost than a moderately deep cut, is really often the more economical expedient.

(The glacier 8,200 feet above the Railway.)]

on the La Guayra and Carácas Railway, Venezuela.]

And here is as good a place, perhaps, as any other in this chapter, to say that true engineering is the economical adaptation of the means and opportunities existing, to the end desired. Civil engineering was defined, by one of the greatest of England's engineers, as "the art of directing the great sources of power in nature for the use and convenience of man," and that definition was adopted as a fundamental idea in the charter of the English Institution of Civil Engineers. But the development of engineering-works in America has been effected successfully by American engineers only because they have appreciated another side of the problem presented to them. A past president of the American Society of Civil Engineers, a man of rare judgment and remarkable executive ability, the late Ashbel Welch, said, in discussing a great undertaking proposed by an eminent Frenchman: "That is the best engineering, not which makes the most splendid, or even the most perfect, work, but that which makes a work that answers the purpose well, at the least cost." And it may be remarked, as to the project which he was then discussing, that after a very large expenditure and an experience of eight years since that discussion, the plans of the work were modified and the identical suggestions made by Mr. Welch of a radical economical change were adopted in 1888.[8] Another eminent American engineer, whose practical experience has been gained in the construction and engineering supervision of more than five thousand miles of railway, said, in his address as President of the American Society of Civil Engineers: "The high object of our profession is to consider and determine the most economic use of time, power, and matter."

That true economy, which finally secures in a completed work the best results from the investment of capital, in first cost and continued maintenance, is an essential element in the consideration of any really great engineering feat.

* * * * *

The difficulties involved in the construction of a tunnel, after the line and dimensions have been determined, depend generally upon the nature of the material found as the work advances. Solid rock presents really the fewest difficulties, but it is seldom that tunnels of considerable length occur without meeting material which requires special provision for successful treatment. In some cases great portions of the rock, where the roof of the tunnel is to be, press downward with enormous weight, being detached from the adjacent mass by the occurrence of natural seams.

At other places soft material may be encountered, and the passage then is attended with great difficulty. Temporary supports, generally of timber, and of great strength, have often to be used at every foot of progress to prevent the material from forcing its way into the excavation already made.

In long tunnels the ventilation is a difficult problem, although the use of compressed air drills has aided greatly in its solution.

Among the great tunnels which have been excavated, the St. Gothard is the most remarkable. It is 9¼ miles long, with a section 26¼ feet wide by 19-2/3 feet high. The work on this tunnel was continuous, and it required 9¼ years for its completion.

The Mont Cenis tunnel, 8-1/3 miles in length, was completed in 12 years.

The Hoosac Tunnel, 4¾ miles in length, 26 feet wide and 21½ feet high, was not prosecuted continuously; it was completed in 1876. These tunnels are notable chiefly on account of their great length; there are others of more moderate extent which have peculiar features; one, illustrated on the preceding page, is unique. This tunnel is a portion of the St. Gothard Railway, and not very far distant from the great tunnel referred to above. In the descent of the mountain it was absolutely necessary to secure a longer distance than a straight line or an ordinary curve would give; the line was therefore doubly curved upon itself. It enters the mountain at a high elevation, describes a circle through the rock and, constantly descending, reappears under itself at the side; still descending, it enters the mountain at another point and continues in another circular tunnel until it finally emerges again, under itself, but at a comparatively short horizontal distance from its first entry, having gained the required descent by a continued grade through the tunnels. The profile above shows the descent, upon a greatly reduced scale, the heavy lines marking where the line is in the tunnel.

The remarkable success achieved by engineers in securing suitable foundations at great depths is, of course, hardly known to the thousands who constantly see the structures supported on those foundations, but in any fair consideration of such engineering achievements this must not be omitted. The beautiful bridge built by Captain Eads over the Mississippi River at St. Louis, bold in its design and excellent in its execution, is an object of admiration to all who visit it, but the impression of its importance would be greatly magnified if the part below the surface of the water, which bears the massive towers, and which extends to a depth twice as great as the height of the pier above the water, could be visible.

The simplest and most effective foundation is, of course, on solid rock. In many localities reliable foundations are built upon earth, when it exists at a suitable depth and of such a character as properly to sustain the weight. Foundations under water, when rock or good material occurs at moderate depth, are constructed frequently by means of the coffer-dam, which is simply an enclosure made water-tight and properly connected with the bottom of the stream. The water is then pumped out and the foundation and masonry built within this temporary dam. When the material is not of a character to sustain the weight, the next expedient is the use of piles, which are driven into the ground, often to a very considerable depth, and sustain the load placed upon them by the friction upon the sides of the piles of the material in which they are driven. It is seldom that dependence is placed upon the load being transferred from the top to the point of the pile, even though the point may have penetrated to a comparatively solid material. Wood is generally used for piles, and where the ground is permanently saturated there seems to be hardly any known limit to its durability. The substructure of foundations, where it is certain that they will always be in contact with water, can be, and generally is, of wood, and the permanency of such foundations is well established. An exception to this, however, occurs in salt-water, particularly in warmer countries, where the ravages of the minute _Teredo Navalis_, and of the still more minute _Limnoria Terebrans_, destroy the wood in a very short period of time. These insects, however, do not work below the ground-line or bed of the water. In many special cases hollow iron piles are used successfully.

The ordinary method of forcing a pile into the ground is by repeated blows of a hammer of moderate weight; better success being obtained by frequent blows of the hammer, lifted to a slight elevation, than results from a greater fall, there being danger also in the latter case of injuring the material of the pile. The use of the water-jet for sinking piles, particularly in sand, is interesting. A tube, generally of ordinary gas-pipe, open at the lower end, is fastened to the pile; the upper end is connected by a hose to a powerful pump and, the pile being placed in position on the surface of the sand, water is forced through the tube and excavates a passage for the pile, which, by the application of very light pressure, descends rapidly to the desired depth. The stream of water must be continuous, as it rises along the side of the pile and keeps the sand in a mobile state. Immediately upon the cessation of pumping, the sand settles about the pile, and it is sometimes quite impossible to afterward move it. The water-jet is used in sinking iron piles by conducting the water through the interior of the hollow pile and out of a hole at its point. The piles of the great iron pier at Coney Island were sunk with great celerity in this way. The illustration opposite shows one of the piers of a bridge founded upon wooden piling.

In many cases it would be impossible to drive piling in such a way as to insure the durability of the structure above it. This is particularly true of the foundations of structures crossing many of our rivers, where the bottom is of material which, in time of flood, sometimes scours to very remarkable depths; the material often being replaced when the flood has subsided. The expedient adopted is the pneumatic tube, or the caisson. Both are merely applications of the well-known principle of the diving-bell. In the former case hollow iron tubes, open at the bottom, are sunk to considerable depths, the water being expelled by air pumped into the tubes at a pressure sufficient to resist the weight of the water. Entrance to the tubes is obtained by an air-lock at the top, the material is excavated from the inside, and sufficient weight placed upon the tube to force it gradually to the desired depth. When that depth is attained, the tubes are filled with concrete, and thus solid pillars of hydraulic concrete, surrounded by cast-iron tubing, are obtained.

The pneumatic caisson is an enlargement of this idea of the diving-bell. The caisson is simply a great chamber or box, open at the bottom; the outside bottom edges are shod and cased with iron so as to give a cutting surface; the roof and sides are made of timber, thoroughly bolted together, and of such strength as to resist the pressure of the structure to be finally founded upon it. The chamber in the open bottom is of sufficient height to enable the laborers to work comfortably in it. This caisson is generally constructed upon the shore in the vicinity of the structure and towed to the point where the foundation is to be sunk. Air is supplied by powerful pumps and is forced into the working chamber. The pressure of the air of course increases constantly as the caisson descends; it must always be sufficient to overbalance the weight of the water and thus prevent the water from entering the chamber.

Descent to the caisson is made through a tube, generally of wrought iron, and having, at a suitable point, an air-lock, which is substantially an enlargement of the tube, forming a chamber, and of sufficient size to accommodate a number of men. This air-lock is provided with doors or valves at the top and at the bottom, both opening downward, and also with small tubes connecting the air-lock with the chamber below and with the external air above. Entrance to the caisson is effected through this air-lock. The lower door, or valve, being at the bottom, closes and is kept closed by the pressure of the air in the caisson below. After the air-lock is entered the upper door or valve is shut, and held shut a few moments, and the tube connecting with the outer air is closed; a small valve in the tube connecting with the caisson is then opened gradually and the pressure in the air-lock becomes the same as that in the chamber below; as soon as this is effected the valve, or door, at the bottom of the air-lock falls open and the air-lock becomes really a part of the caisson.

A sufficient force of men is employed in the chamber to gradually excavate the material from its whole surface and from under the cutting edge, and the masonry structure is founded upon the top of the caisson and built gradually, so as to give constantly a sufficient weight to carry the whole construction down to its final location upon the stable foundation, which may be the bed-rock or may be some strata of permanent character.

The problem of lighting the chamber was until recently of considerable difficulty. The rapid combustion under great pressure made the use of lamps and candles very troublesome, particularly on account of the dense smoke and large production of lampblack.

The introduction of the electric light has greatly aided in the more comfortable prosecution of pneumatic foundation work.

The removal of rock, or any large mass, from the caisson is effected through the air-chamber; but the removal of finer material, as sand or earth, is accomplished by the sand pump or by the pressure of the air. A tube, extending from the top of the masonry and kept above the surface by additions, as may be required, enters the working chamber and is controlled by proper valves. Lines of tubing and hose extend to all portions of the chamber. A slight excavation is made and kept filled with water. The bottom of the tube, or the hose connected with it, is placed in this excavation, and, the material being agitated so as to be in suspension in the water, the valve is opened, and the pressure of the air throws the water and the material held in suspension to the surface, through the tube, from the end of which it is projected with great velocity and may be deposited at any desired adjacent point. This method, however, exhausts the air from the caisson too rapidly for continuous service. The Eads sand-pump is therefore generally used. This is an ingenious apparatus, somewhat the same in principle as the injector which forces water into steam-boilers. A stream of water is thrown by a powerful pump through a tube which, at a point near the inlet for the excavated material, is enlarged so as to surround another tube. The water is forced upward with great velocity into the second tube, through a conical annular opening, and, expelling the atmosphere, carries with it to the surface a continuous stream of sand and water from the bottom of the excavation.

This system has been used successfully in the foundations of piers and abutments of bridges in all parts of the world. The rapidity of the descent of the caisson varies with the material through which it has to pass. The speed with which such foundations are executed is remarkable, when one remembers with what delicacy and intelligent supervision they have to be balanced and controlled. In some instances it has been necessary to carry them to great depths, one at St. Louis being 107 feet below ordinary water level in the river.

The pressure of air in caissons at these depths is very great; at 110 feet below the surface of the water it would be 50 pounds to the square inch. Its effect upon the men entering and working in the caisson has been carefully noted in various works, and these effects are sometimes very serious; the frequency of respiration is increased, the action of the heart becomes excited, and many persons become affected by what is known as the "caisson disease," which is accompanied by extreme pain and in some cases results in more or less complete paralysis. The careful observations of eminent physicians who have given this disease special attention have resulted in the formulation of rules which have reduced the danger to a minimum.

The execution of work within a deep pneumatic caisson is worth a moment's consideration. Just above the surface of the water is a busy force engaged in laying the solid blocks of masonry which are to support the structure. Great derricks lift the stones and lay them in their proper position. Powerful pumps are forcing air, regularly and at uniform pressure, through tubes to the chamber below. Occasionally a stream of sand and water issues with such velocity from the discharge pipe that, in the night, the friction of the particles causes it to look like a stream of living fire. Far below is another busy force. Under the great pressure and abnormal supply of oxygen they work with an energy which makes it impossible to remain there more than a few hours. The water from without is only kept from entering by the steady action of the pumps far above and beyond their control. An irregular settlement might overturn the structure. Should the descent of the caisson be arrested by any solid under its edge, immediate and judicious action must be taken. If the obstruction be a log, it must be cut off outside the edge and pulled into the chamber. Boulders must be undermined and often must be broken up by blasting. The excavation must be systematic and regular. A constant danger menaces the lives of these workers, and the wonderful success with which they have accomplished what they have undertaken is entitled to notice and admiration.

Another process, which has succeeded in carrying a foundation to greater depths than is possible with compressed air, is by building a crib or caisson, with chambers entirely open at the top, but having the alternate ones closed at the bottom and furnished with cutting edges. These closed chambers are weighted with stone or gravel until the structure rests upon the bottom of the river; the material is then excavated from the bottom through the open chambers, by means of dredges, thus permitting the structure to sink by its weight to the desired depth. When that depth is reached, the chambers which have been used for dredging are filled with concrete, and the masonry is constructed upon the top of this structure. The use of this system has enabled the engineer to place foundations deeper than has been accomplished by any other device, one recently built in Australia being 175 feet below the surface of the water. The illustrations above and on page 76 show this method of construction.

Even more remarkable than the pneumatic caisson is this method of sinking these great foundations. The removal of material must be made with such systematic regularity that the structure shall descend evenly and always maintain its upright position. The dredge is handled and operated entirely from the surface. The very idea is startling, of managing an excavation more than a hundred feet below the operator, entirely by means of the ropes which connect with the dredge, and doing it with such delicacy that the movement of an enormous structure, weighing many tons, is absolutely controlled. This is one of the latest and most interesting advances of engineering skill.

* * * * *

While it is true that the avoidance of large expenditure, when possible, is a mark of the best engineering, yet great structures often become absolutely necessary in the development of railway communication. Wide rivers must be crossed, deep valleys must be spanned, and much study has been given to the best methods of accomplishing these results. In the early history of railways in Europe substantial viaducts of brick and stone masonry were generally built; and in this country there are notable instances of such constructions. The approach to the depot of the Pennsylvania Railroad, in the city of Philadelphia, is an excellent example. Each street crossed by the viaduct is spanned by a bold arch of brick. Upon a number of our railways there are heavy masonry arches and culverts, and at some places these are of a very interesting character. The arches in the approach to the bridge over the Harlem Valley (recently completed) are shown above. They are of granite, having a span of 60 feet. The illustration shows also the method of supporting the stone work of such arches during construction. Braced timbers form what is called the centre, and support the curved frame of plank upon which the masonry is built, which, of course, cannot be self-supporting until the keystone is in place; then the centre is lowered by a loosening of the wedges which support it, and the stone work of the arch is permitted to assume its final bearing. It is generally considered that where it is practicable to construct masonry arches under railways there is a fair assurance of their permanency, but some engineers of great experience in railway construction advance the theory that the constant jar and tremor produced by passing railway trains is really more destructive to masonry work than has been supposed, and that it may be true that the elements of the best economy will be found in metal structures rather than in masonry. It is a fact that repairs and renewals of metal bridges are much more easily accomplished than of masonry constructions.

In this country the wooden bridge has been an important, in fact an essential element in the successful building of our railways.

Timber is also used extensively in railroad construction in the form of trestles; one example of which has been alluded to on page 50. There were also constructed, years ago, some very bold viaducts in wood. One of the most interesting is shown above, being the viaduct at Portage, N. Y. This construction was over 800 feet long, and 234 feet high from the bed of the river to the rail. The masonry foundations were 30 feet high, the trestles 190 feet, and the truss 14 feet; it contained more than a million and a half feet, board measure, of timber. The timber piers, which were 50 feet apart, are formed by three trestles, grouped together. It was framed so that defective pieces could be taken out and replaced at any time. This bridge was finished in 1852 and was completely destroyed by fire in 1875. The new metal structure which took its place is shown on the opposite page, and is an interesting example of the American method of metal viaduct construction, an essential feature of that construction being the concentration of the material into the least possible number of parts. This bridge has ten spans of 50 feet, two of 100 feet, and one of 118 feet. The trusses are of what is called the Pratt pattern, and are supported by wrought-iron columns, two pairs of columns forming a skeleton tower 20 feet wide and 50 feet long on the top. There are six of these towers, one of which has a total height from the masonry to the rail of 203 feet 8 inches. There are over 1,300,000 pounds of iron in this structure.

The fundamental idea of a bridge is a simple beam of wood. If metal is substituted it is still a beam with all superfluous parts cut away. This results in what is called an I beam. When greater loads have to be carried, the I beam is enlarged and built up of metal plates riveted together and thus becomes a plate girder. These are used for all short railway spans. For greater spans the truss must be employed.

Before referring, however, to examples of truss bridges, a description should be given of the Britannia Bridge, built by Robert Stephenson in 1850, over the Menai Straits. This great construction carries two lines of rails and is built of two square tubes, side by side, each being continuous, 1,511 feet long, supported at each extremity and at three intermediate points, and having two spans of 460 feet each and two spans of 230 feet each. The towers which support this structure are of very massive masonry, and rise considerably above the top of the tubes. These tubes are each 27 feet high and 14 feet 8 inches wide; they are built up of plate iron, the top and bottom being cellular in construction, and the sides of a single thickness of iron. The tubes for the long spans were built on shore and floated to the side of the bridge and then lifted by hydraulic presses to their final position. The rapid current, and other considerations, made the erection of false works for these spans impracticable. The beautiful suspension bridge, built by Telford in 1820, over the Menai Straits, is only a mile away from this Britannia Bridge, but, at the time of the construction of the latter, it was not deemed possible by English engineers to erect a suspension bridge of sufficient strength and stability to accommodate railway traffic.

The Victoria Bridge at Montreal is of the same general character of construction as the Britannia Bridge, but is built only for a single line of rails; this bridge also was built by Mr. Stephenson, in 1859. These two structures were enormous works; their strength is undoubted, but they lack that element of permanent economy which has been spoken of in this article; their cost was very great, and the expense of maintenance is also very great. A very large amount of rust is taken from these tubes every year; they require very frequent painting, and there are on the Victoria Bridge 30 acres of iron surface to be thus painted.

A remarkable and interesting contrast to these heavy tubes of iron is the Niagara Falls railway suspension bridge, completed in March, 1855. The span of this bridge is 821 feet, and the track is 245 feet above the water surface. It is supported by 4 cables which rested on the tops of two masonry towers at each end of the central span, the ends of the cables being carried to and anchored in the solid rock. The suspended superstructure has two floors, one above the other, connected together at each side by posts and truss rods, inclined in such a manner as to form an open trussed tube, not intended to support the load, but to prevent excessive undulations. The floors are suspended from the cables by wire ropes, the upper floor carrying the railroad track, and the lower forming a foot and carriage way. Each cable has 3,640 iron wires. This bridge carried successfully a heavy traffic for 26 years; it was then found that some repairs to the cable were required at the anchorage, the portions of the cables exposed to the air being in excellent condition. These repairs were made, and the anchorage was substantially reinforced. At the same time it was found that the wooden suspended superstructure was in bad condition, and this was entirely removed and replaced by a structure of iron, built and adjusted in such a manner as to secure the best possible results. For some time it had been noticed that the stone towers which supported the great cables of the bridge showed evidences of disintegration at the surface, and a careful engineering examination in 1885 showed that these towers were in a really dangerous condition. The reason for this was that the saddles over which the cables pass on the top of the towers had not the freedom of motion which was required for the action of the cables, caused by differences of temperature and by passing loads. These saddles had been placed upon rollers but, at some period, cement had been allowed to be put between these rollers, thus preventing their free motion. The result was a bending strain upon the towers which was too great for the strength and cohesion of the stone. A most interesting and successful feat was accomplished in the substitution of iron towers for these stone towers, without interrupting the traffic across the bridge. This was accomplished within a year or two by building a skeleton iron tower outside of the stone tower, and transferring the cables from the stone to the iron tower by a most ingenious arrangement of hydraulic jacks. The stone towers were then removed. Thus, by the renewal of its suspended structure and the replacing of its towers, the bridge has been given a new lease of life and is in excellent condition to-day.