Among the readers of this volume there will be some who have reached the summit of the "divide" which separates the spring and summer of life from its autumn and winter, and whose first information about railroads was received from Peter Parley's "First Book of History," which was used as a schoolbook forty or fifty years ago. In his chapter on Maryland, he says:

The picture reproduced below (Fig. 2) of a car drawn by horses was given with the above description of the Baltimore & Ohio Railroad. The mutilated copy of the book from which the engraving and extract were copied does not give the date when it was written or published. It was probably some time between the years 1830 and 1835. That the car shown in the engraving was evolved from the Conestoga wagon is obvious from the illustrations.

This engraving and description, made for children, more than fifty years ago, will give some idea of the state of the art of railroading at that time; and it is a remarkable fact that the present wonderful development and the improvements in railroads and their equipments in this country have been made during the lives of persons still living.

In the latter part of 1827, the Delaware & Hudson Canal Company put the Carbondale Railroad under construction. The road extends from the head of the Delaware & Hudson Canal at Honesdale, Pa., to the coal mines belonging to the Delaware & Hudson Canal Company at Carbondale, a distance of about sixteen miles. This line was opened, probably in 1829, and was operated partly by stationary engines, and partly by horses. The road is noted chiefly for being the one on which a locomotive was first used in this country. This was the "Stourbridge Lion," which was built in England under the direction of Mr. Horatio Allen, who afterward was president of the Novelty Works in New York, and who is still (1889) living near New York at the ripe age of eighty-seven. Before the road was opened, he had been a civil engineer on the Carbondale line. In 1828 Mr. Allen went to England, the only place where a locomotive was then in daily operation, to study the subject in all its practical details. Before leaving this country he was intrusted by the Delaware & Hudson Canal Company with the commission to have rails made for that line, and to have three locomotives built on plans to be decided by him when in England. This, it must be remembered, was before the celebrated trial of the "Rocket" on the Liverpool & Manchester Railway, which was not made until 1829. Previous to that trial, it had not been decided what type of boiler was the best for locomotives. The result of Mr. Allen's investigations was to produce in his mind a decided confidence in the multitubular boiler which is now universally used for locomotives. Other persons of experience recommended a boiler with small riveted flues of as small diameter as could be riveted. An order was therefore given to Messrs. Foster, Rastrick & Co., at Stourbridge, for one engine whose boiler was to have riveted flues of comparatively large size, and another order was given to Messrs. Stephenson & Co., of Newcastle-on-Tyne, for two locomotives with boilers having small tubes. The engine built by Foster, Rastrick & Co. was named the "Stourbridge Lion." It was sent to this country and was tried at Honesdale, Pa., on August 9, 1829. On its trial trip it was managed by Mr. Allen, to whom belongs the distinction of having run the first locomotive that was ever used in this country. In 1884 he wrote the following account of this trip:

The two engines contracted for with Messrs. Stephenson & Co. were made by them, and Mr. Allen has informed the writer that they were built on substantially the same plans that were afterward embodied in the famous "Rocket." They were shipped to New York and for a time were stored in an iron warehouse on the east side of the city, where they were exhibited to the public. They were never sent to the Delaware & Hudson Canal Company's road, and it is not now known whatever became of them. If they had been put to work on their arrival here the use of engines of the "Rocket" type would have been anticipated on this side the Atlantic.

The first railroad which was undertaken for the transportation of freight and passengers in this country, on a comprehensive scale, was the Baltimore & Ohio. Its construction was begun in 1828. The laying of rails was commenced in 1829, and in May, 1830, the first section of fifteen miles from Baltimore to Ellicott's Mills was opened. It was probably about this time that the animated sketch of the car given by Peter Parley was made. From 1830 to 1835 many lines were projected, and at the end of that year there were over a thousand miles of road in use.

Whether the motive power on these roads should be horses or steam was for a long time an open question. The celebrated trial of locomotives on the Liverpool & Manchester Railway, in England, was made in 1829. Reports of these trials, and of the use of locomotive engines on the Stockton & Darlington line, were published in this country, and, as Mr. Charles Francis Adams says, "The country, therefore, was not only ripe to accept the results of the Rainhill contest, but it was anticipating them with eager hope." In 1829 Mr. Horatio Allen, who had been in England the year before to learn all that could then be learned about steam locomotion, reported to the South Carolina Railway Company in favor of steam instead of horse power for that line. The basis of that report, he says, "Was on the broad ground that in the future there was no reason to expect any material improvement in the breed of horses, while, in my judgment, the man was not living who knew what the breed of locomotives was to place at command."

As early as 1829 and 1830, Peter Cooper experimented with a little locomotive on the Baltimore & Ohio Railroad (Fig. 4). At a meeting of the Master Mechanics' Association in New York, in 1875—at the Institute which bears his name—he related with great glee how on the trial trip he had beaten a gray horse, attached to another car. The coincidence that one of Peter Parley's horses is a gray one might lead to the inference that it was the same horse that Peter Cooper beat, a deduction which perhaps has as sound a basis to rest on as many historical conclusions of more importance.

The undeveloped condition of the art of machine construction at that time is indicated by the fact that the flues of the boiler of this engine were made of gun-barrels, which were the only tubes that could then be obtained for the purpose. The boiler itself is described as about the size of a flour-barrel. The whole machine was no larger than a hand-car of the present day.

In the same year that Peter Cooper built his engine, the South Carolina Railway Company had a locomotive, called the "Best Friend," built at the West Point Foundry for its line. In 1831 this company had another engine, the "South Carolina" (Fig. 5), which was designed by Mr. Horatio Allen, built at the same shop. It was remarkable in having eight wheels, which were arranged in two trucks. One pair of driving-wheels, D D and D' D', and a pair of leading-wheels, L L and L' L', were attached to frames, c d e f and g h i j, which were connected to the boiler by kingbolts, K K', about which the trucks could turn. Each pair of driving-wheels had one cylinder, C C'. These were in the middle of the engine and were connected to cranks on the axles A and B.

The "De Witt Clinton" (Fig. 6) was built for the Mohawk & Hudson Railroad, and was the third locomotive made by the West Point Foundry Association. The first excursion trip was made with passengers from Albany to Schenectady, August 9, 1831. This is the engine shown in the silhouette engraving of the "first[10] railroad train in America" which in recent years has been so widely distributed as an advertisement.

In 1831 the Baltimore & Ohio Railroad Company offered a premium of $4,000 "for the most approved engine which shall be delivered for trial upon the road on or before the 1st of June, 1831; and $3,500 for the engine which shall be adjudged the next best." The requirements were as follows:

In pursuance of this call upon American genius, three locomotives were produced, but only one of these was made to answer any useful purpose. This engine, the "York," was built at York, Pa., and was brought to Baltimore over the turnpike on wagons. It was built by Davis & Gartner, and was designed by Phineas Davis, of that firm, whose trade and business was that of a watch and clock maker. After undergoing certain modifications, it was found capable of performing what was required by the company. After thoroughly testing this engine, Mr. Davis built others, which were the progenitors of the "grasshopper" engines (Fig. 7) which were used for so many years on the Baltimore & Ohio Railroad. It is a remarkable fact that three of these are still in use on that road, and have been in continuous service for over fifty years. Probably there is no locomotive in existence which has had so long an active life.

In August, 1831, the locomotive "John Bull," which was built by George & Robert Stephenson & Company, of Newcastle-upon-Tyne, was received in Philadelphia, for the Camden & Amboy Railroad & Transportation Company. This is the old engine which was exhibited by the Pennsylvania Railroad Company at the Centennial Exhibition in 1876. After the arrival of the "John Bull" a very considerable number of locomotives which were built by the Stephensons were imported from England. Most of them were probably of what was known as the "Planet" class (Fig. 8), which was a form of engine that succeeded the famous "Rocket."

The following quotation is from "The Early History of Locomotives in this Country," issued by the Rogers Locomotive & Machine Works:

In all of the locomotives which have been illustrated, excepting the "South Carolina," the axles were held by the frames so that the former were always parallel to each other. In going around curves, therefore, there was somewhat the same difficulty that there would be in turning a corner with an ordinary wagon if both its axles were held parallel, and the front one could not turn on the kingbolt. The plan of the wheels and running gear of the "South Carolina" shows the position that they assumed on a curved track (Fig. 5). It will be seen that, by reason of their connection to the boiler by kingbolts, K K', the two pairs of wheels could adjust themselves to the curvature of the rails. This principle was afterward applied to cars, and nearly all the rolling-stock in this country is now constructed on this plan, which was proposed by Mr. Allen in a report dated May 16, 1831, made to the South Carolina Canal & Railroad Company; and an engine constructed on this principle was completed the same year.

In the latter part of the year 1831 the late John B. Jervis invented what he called "a new plan of frame, with a bearing-carriage for a locomotive engine," for the use of the Mohawk & Hudson Railroad. Jervis's engine is shown by Figure 9. In a letter published in the American Railroad Journal of July 27, 1833, he described the objects aimed at in the use of the truck as follows:

In Jervis's locomotive the main driving-axle, A, shown in the plan of the wheels and running gear, was rigidly attached to the engine-frame, a b c d, and only one truck, or "bearing-carriage," e f g h, consisting of the two pairs of small wheels attached to a frame, was used. This was connected to the main engine-frame by a kingbolt, K, as in Allen's engine.

The position of its wheels on a curve, and the capacity of the truck, or "bearing-carriage," to adapt itself to the sinuosities of the track are shown in the plan. The effectiveness of the single truck for locomotives, in accomplishing what Mr. Jervis intended it for, was at once recognized, and its almost general adoption on American locomotives followed.

In 1834, Ross Winans, of Baltimore, patented the application of the principle which Mr. Allen had proposed and adopted for locomotives "to passenger and other cars." He afterward brought a number of actions at law against railroads for infringement of his patent, which was a subject of legal controversy for twenty years. Winans claimed that his invention originated as far back as 1831, and was completed and reduced to practice in 1834. The dispute was finally carried to the Supreme Court of the United States, and was decided against the plaintiff, after an expenditure of as much as $200,000 by both sides. It involved the principle on which nearly all cars in this country are now and were then built; and, as one of the counsel for the defendants has said, "It was at one time a question of millions, to be assured by a verdict of a jury."

In 1836, Henry R. Campbell, of Philadelphia, patented the use of two pairs of driving-wheels and a truck, as shown in Figure 10. The driving-wheels were coupled by rods, as may be seen below. This plan has since been so generally adopted in this country that it is now known as the "American type" of locomotive, and is the one almost universally used here for passenger, and to a considerable extent for freight, service. An example of a modern locomotive of this type is represented by Figure 11.

From these comparatively small beginnings, the magnificent equipment of our railroads has grown. From Peter Cooper's locomotive, which weighed less than a ton, with a boiler the size of a flour-barrel, and which had difficulty in beating a gray horse, we now have locomotives which will easily run sixty and can exceed seventy miles an hour, and others which weigh seventy-five tons and over. A comparison of the engraving of Peter Cooper's engine with that of the modern standard express passenger locomotive (Fig. 11) shows vividly the progress which has been made since that first experiment was tried—little more than half a century ago. In that period there have been many modifications in the design of locomotives to adapt them to the changed conditions of the various kinds of traffic of to-day. An express train travelling at a high rate of speed requires a locomotive very different from one which is designed for handling heavy freight trains up steep mountain grades. A special class of engines is built for light trains making frequent stops, as on the elevated railroads in New York, and those provided for suburban traffic (Fig. 12)—and still others for street railroads (Fig. 13), for switching cars at stations (Fig. 14), etc. [Pp. 110 and 113]. The process of differentiation has gone on until there are now as many different kinds of these machines as there are breeds of dogs or horses.

Nearly all the early locomotives had only four wheels. In some cases one pair alone was used to drive the engine, and in others the two pairs were coupled together, so that the adhesion of all four could be utilized to draw loads. The four-wheeled type is still used a great deal for moving cars at stations, and other purposes where the speed is comparatively slow. But to run around sharp curves the wheels of such engines must be placed near together, just as they are under an ordinary street-car. This makes the wheel-base very short, and such engines are therefore very unsteady at high speeds, so that they are unsuited for any excepting slow service. They have the advantage, though, that the whole weight of the machine may be carried on the driving-wheels, and can thus be useful for increasing their friction, or adhesion to the rails. This gives such engines an advantage for starting and moving heavy trains, at stations or elsewhere, which is the kind of service in which they are usually employed.

If the front end of the engine is carried on a truck, as in Campbell's plan (Fig. 10)—which is the one that has been very generally adopted in this country—the wheel-base can be extended and at the same time the front wheels can adjust themselves to the curvature of the track. This gives the running-gear lateral flexibility. But as the tractive power of a locomotive is dependent upon the friction, or adhesion of the wheels to the rails, it is of the utmost importance that the pressure of the wheels on the rails should be uniform. For this reason the wheels must be able to adjust themselves to the vertical as well as the horizontal inequalities of the track.

Figure 15 shows the driving-wheels, axles, journal-boxes, and part of the frame and springs of an American type of engine—the circumference of the wheels only being shown. The axles A A each have journal-boxes or bearings, B B, in which they turn. These boxes are held between the jaws J J J J of the frames, and can slide vertically in the spaces c c c c between the jaws. The frames are suspended on springs, S S, which bear on the boxes B B. The vertical motion of the boxes and the flexibility of the springs allow the wheels to adjust themselves to some extent to the unevenness of the track. But, in order to distribute the weight equally on the two wheels, the springs S S on each side of the engine are connected together by an equalizing lever, E E. These levers each have a fulcrum, F, in the middle, and are connected by iron straps or hangers, h h, to the springs. It is evident that any strain or tension on one spring is transferred by the equalizing lever to the other spring, and thus the weight is equalized on both wheels.

But to give perfect vertical adjustment of such an engine to the track, still another provision must be made. Everyone has observed that a three-legged stool will always stand firm on any surface, no matter how irregular, but one with four legs will not. Now if the back end of a locomotive should rest on the fulcrums of the equalizing levers, as shown in Figure 15, and the front end should rest on the two sides of the truck, it would be in the condition of the four legged stool. Therefore, instead of resting on the two sides of the truck, locomotives are made to bear on the centre of it, so that they are carried on it and on the two fulcrums of the equalizing levers, which gives the machine the adjustability due to the three-legged principle. When more than four driving-wheels are used the springs are connected together by equalizing levers, as shown in Figure 29 (p. 124), which represents a consolidation engine as it appears before the wheels are put under it.

Having a vehicle which is adapted to running on a railroad track, it remains to supply the motive power. This, in all but some very few exceptional cases, is the expansive power of steam. What the infant electricity has in store for us it would be rash to predict, but for locomotives its steps have been thus far weak and uncertain, and when we want a giant of steel or a race-horse of iron our only sure reliance is steam. This is the breath of life to the locomotive, which is inhaled and exhaled to and from the cylinders, which act as lungs, while the boiler fulfils functions analogous to the digestive organs of an animal. A locomotive is as dependent on the action of its boiler for its capacity for doing work as a human being on that of his stomach. The mechanical appliances of the one and the mental and physical equipment of the other are nugatory without a good digestive apparatus.

A locomotive boiler consists of a rectangular fireplace or fire-box, as shown at A, in Figure 16, which is a longitudinal section, and Figure 17 a transverse section through the fire-box. The fire-box is connected with the smoke-box B by a large number of small tubes, a a, through which the smoke and products of combustion pass from the fire-box to the smoke-box, and from the latter they escape up the chimney D. The fire-box and tubes are all surrounded with water, so that as much surface as possible is exposed to the action of the fire. This is essential on account of the large amount of water which must be evaporated in such boilers. To create a strong draught, the steam which is exhausted from the cylinders is discharged up the chimney through pipes, and escapes at e. This produces a partial vacuum in the smoke-box, which causes a current of air to flow through the fire on the grate, into the fire-box, through the tubes, and thence to the smoke-box and up the chimney. Probably many readers have noticed, that of late years the smoke-boxes of locomotives have been extended forward in front of the chimneys. This has been done to give room for deflectors and wire netting inside to arrest sparks and cinders, which are collected in the extended front and are removed by a door or spout, L, below.

To get the water into the boiler against the pressure of steam a very curious instrument, called an injector, has been devised. Formerly force-pumps were used, but these are now being abandoned. The illustration (Fig. 18) shows what may be called a rudimentary injector. B is a boiler and E a conical tube open at its lower end—and connected to a water-supply tank by a pipe, C. A pipe, A, is connected with the steam-space of the boiler and terminates in a contracted mouth, F, inside of the cone E. If steam is admitted to A, it flows through the pipe and escapes at F. In doing so it produces a partial vacuum in E, and water is consequently drawn up the pipe C from the tank. The current of steam now carries with it the water, and they escape at G. After flowing for a few seconds the water has a high velocity and the steam, mingling with the water, is condensed. The momentum of the water soon becomes sufficient to force the valve H down against the pressure below it, and the jet of water then flows continuously into the boiler. A very curious phenomenon of this somewhat mysterious instrument is that if steam of a low pressure is taken from one boiler it will force water into another against a higher pressure. Figure 19 is a section of an actual injector used on locomotives.

Having explained how the steam is generated, it remains to show how it propels a locomotive. It does this very much as a person on a bicycle propels it—that is, by means of two cranks the wheels are made to revolve, and the latter must then either slip or the vehicle will move. In a locomotive the driving-wheels are turned by means of two cylinders and pistons, which are connected by rods to the cranks attached to the driving-wheels or axles. These cranks are placed at right angles to each other, so that when one of them is at the "dead-point" the piston connected with the other can exert its maximum power to rotate the wheels. This enables the locomotive to start with the pistons in any position; whereas, if one cylinder only was used it would be impossible to turn the wheels if the crank should stop at one of its dead-points.

It will probably interest a good many readers to know how the steam gets into the cylinders and moves the pistons and then gets out again, and how a locomotive is made to run either backward or forward at pleasure.

Figure 20 (p. 118) shows a section of a cylinder, A A', with the piston B and piston rod R. The cylinder has two passages, c c and d d, which connect its ends with a box, U, called a steam-chest, to which steam is admitted from the boiler by a pipe, J. The two passages c and d have another one, g, between them, which is connected with the chimney. These passages are covered by a slide-valve, V, which moves back and forth in the steam-chest, alternately uncovering the openings c and d. When the valve is in the position shown in Figure 20, obviously steam can flow into the front end A of the cylinder through the passage c, as indicated by the darts. The valve has a cavity, H, underneath it. When this cavity is over the passage d and g, it is plain that the steam in the back end A' of the cylinder can flow through d and g and then escape up the chimney. Under these circumstances the steam in the front end A of the cylinder will force the piston B to the back end. When it reaches the back end of the cylinder the valve is moved into the position shown in Figure 21, and steam can then enter d and will fill the back end A' while that in the front end escapes through c and g. The piston is then forced to the front end by the pressure of the steam behind it. It will thus be seen that the steam enters and escapes to and from the cylinder through the same openings.

From what has been said it is obvious, too, that every time the piston moves from one end of the cylinder to the other the valve must also be moved back and forth in the steam-chest. This is done by what is called an eccentric.

An "eccentric" is a disk or wheel (Fig. 22) with a hole, S, the size of the axle of the locomotive to which it is attached. The centre n of the outside periphery of the eccentric is some distance from S, the centre of the shaft. A metal ring, K K (Fig. 23), made in two halves, embraces the eccentric, and the latter revolves inside of this ring. A rod, L, is attached to the strap, and is connected with the valve so that the motion of the eccentric is communicated to it. It is obvious that if the eccentric revolves it will impart a reciprocating motion to the rod L, which is communicated to the valve.

If properly adjusted on the axle the eccentric will run the engine in one direction. To run the opposite way another eccentric must be provided. Therefore locomotives always have two eccentrics for each cylinder. These, J and K, are shown in Figure 24, which represents the "valve-gear" of a locomotive. S is a section of the main driving-axle, to which the eccentrics are attached by keys or screws. C is the eccentric rod of the forward-motion eccentric and D that of the one for running backward. As a locomotive must be run either backward or forward, and, as the one eccentric moves the valve to run forward and the other to run backward, we must be able to connect or disconnect the rods to and from the valve at will. The eccentric rods of the early locomotives had hooks on the ends by which they were attached to or detached from suitable pins connected with the valves. But these hooks were very uncertain in their action and therefore were abandoned, and now what is known as the "link-motion" is almost universally used for the valve-gear of locomotives. It consists of a "link" (a b, Fig. 24) which has a curved opening or slot, k, in it in which a block, B, fits accurately, so that it can slide from end to end of the link. This block has a hole bored in the middle which receives a pin, c, which is attached to the end of the arm N of the "rocker" M O N. The rocker has a shaft, O, which can turn in a suitable bearing, and two arms, M and N; the latter, as explained, is connected to the link by the pin c and block B. The upper arm M has another pin, V, on its end, which is connected by a rod, v V, to the main slide-valve V. The rocker-arms, as will be seen, can vibrate about the shaft O.

It is plain that, by moving the upper end of the reverse lever, the link a b can be raised up or lowered at will. When the link is down, or in the position represented in the engraving, the forward eccentric rod imparts its motion to the block B, pin c, and thence to the rocker and valve, and the engine will run forward. If, however, the reverse lever is thrown back into the position indicated by the dotted line J I, the link would then be raised up so that the end e of the backward-motion rod would be opposite to the block B and pin c and would communicate its motion to the rocker and valve, and the wheels would then be turned backward instead of forward. It will thus be seen how the movement of the reverse lever effects the reversal of the engine.

A locomotive is started by admitting steam to the cylinders by means of what is called the "throttle-valve." This is usually placed in the upper part of the boiler at T (Fig. 16). The valve is worked by a lever at l, which is also shown at 14, 14' (Fig. 36). The steam is conveyed to the cylinders by a pipe (s, Fig. 16, p. 115).

If steam is admitted to the cylinders and the wheels are turned, one of two results must follow: either the locomotive will move backward or forward according to the direction of revolution, or the wheels will slip, as they often do, on the rails. That is, if the resistance of the cars or train is less than the friction or "adhesion" of the wheels on the rails, the engine and train will be moved; if the adhesion is less than the resistance the wheels will turn without moving the train.

The capacity of a locomotive to draw loads is therefore dependent on the adhesion, and this is in proportion to the weight or pressure of the driving-wheels on the rails. The adhesion also varies somewhat with the weather and the condition of the wheels and rails. In ordinary weather it is equal to about one-fifth of the weight which bears on the track; when perfectly dry, if the rails are clean, it is about one-fourth, and with the rails sanded about one-third. In damp or frosty weather the adhesion is often considerably less than a fifth.

It would, then, seem as though all that is needed to increase the capacity of a locomotive to draw loads would be to add to the weight on its driving-wheels, and provide engine-power sufficient to turn them—which is true. But it has been found that if the weight on the wheels is excessive both the wheels and rails will be injured. Even when they are all made of steel, they are crushed out of shape or are rapidly worn if the loads are too great. The weight which rails will carry without being injured depends somewhat on their size or weight, but ordinarily from 12,000 to 16,000 pounds per wheel is about the greatest load which they should carry.

For these reasons, when the capacity of a locomotive must be increased beyond a limit indicated by these data, one or more additional pairs of driving-wheels must be used. Thus, if a more powerful engine was required than that shown in Figure 14 (p. 113), another pair of wheels would be added, as shown in Figures 26, 27, and 28. Or, if you wanted a more powerful engine than these, still another pair of driving-wheels would be provided, as shown in Figure 30. In this way the Mogul, ten-wheeled and consolidation engines have been developed from that shown in Figure 14. The Mogul locomotive (Fig. 27) has three pairs of driving-wheels, but only one pair of truck-wheels. The engravings shown in Figures 30 and 31 represent consolidation and decapod types of engines which have four and five pairs of driving-wheels.

From the illustrations, Figures 28, 30, and 31, it will be seen that when so many wheels are used, even if they are of small diameter, the wheel-base must necessarily be long, so that a limit is very soon reached beyond which the number of driving-wheels cannot be increased.

Improvements in the processes of manufacturing steel, which resulted in the general use of that material for rails and tires, have made it possible to nearly double the weight which was carried on each wheel when they were made of iron. The weight of rails has also been very much increased since they were first made of steel. Twenty or twenty-five years ago iron rails weighing 56 pounds per yard were about the heaviest that were laid in this country. Now steel rails weighing 72 pounds are commonly used, and some weighing 85 pounds have been laid on American roads, and others weighing 100 pounds have been laid on the Continent of Europe.

Of late years urban and suburban traffic has created a demand for a class of locomotives especially adapted to that kind of service. One of the conditions of that traffic is that trains must stop and start often, and therefore, to "make fast time," it is essential to start quickly. Few persons realize the great amount of force which must be exerted to start any object suddenly. A cannon-ball, for example, will fall through 16 feet in a second with no other resistance than the atmosphere. The impelling force in that case is the weight of the ball. If we want it to fall 32 feet during the first second, the force exerted on it must be equal to double its weight, and for higher speeds the increase of force must be in the same proportion. This law applies to the movement of trains. To start in half the time, double the force must be exerted. For this reason, trains which start and stop often require engines with a great deal of weight on the driving-wheels. In accordance with these conditions a class of engines has been designed which carry all, or nearly all, the weight of the boiler and machinery, and sometimes the water and fuel, on the driving-wheels. For suburban traffic, the speed between stops must often be quite rapid, and consequently the engine must have a long wheel-base for steadiness, as well as considerable weight on the wheels for adhesion. Four-wheeled engines (Fig. 14) have all their weight on the driving-wheels, but the wheel-base is short.

To combine the two features, engines have been built with the driving-wheels and axles arranged as in Figure 32. The frames are then extended backward, and the water-tank and fuel are placed on top of the frames, and their weight is carried by a truck underneath. This arrangement leaves the whole weight of the boiler and machinery on the driving-wheels, and at the same time gives a long wheel-base for steadiness. This plan of engine was patented by the author of this article in 1866, and has come into very general use—since the expiration of the patent. In some cases a two-wheeled truck is added at the opposite end, as shown in Figure 33. For street railroads, in which the speed is necessarily slow, engines such as Figure 13 (p. 110) are used. To hide the machine from view, and also to give sufficient room inside, they are enclosed in a cab large enough to cover the whole machine.

The size and weight of locomotives have steadily been increased ever since they were first used, and there is little reason for thinking that they have yet reached a limit, although it seems probable that some material change of design is impending which will permit of better proportions of the parts or organs of the larger sizes. The decapod engines built at the Baldwin Locomotive Works, in Philadelphia, for the Northern Pacific Railroad, weigh in working order 148,000 pounds. This gives a weight of 13,300 pounds on each driving-wheel. Some ten-wheeled passenger engines, built at the Schenectady Locomotive Works for the Michigan Central Railroad, weigh 118,000 pounds, and have 15,666 pounds on each driving-wheel. Some recent eight-wheeled passenger locomotives for the New York, Lake Erie & Western Railroad weigh 115,000 pounds, and have 19,500 pounds on each driving-wheel. At the Baldwin Works, some "consolidation" engines have recently been built which are still heavier than the decapod engines.

The following table gives dimensions, weight, price, and price per pound of locomotives at the present time. If we were to quote them at 8 to 8¼ cents per pound for heavy engines and 9 to 22¼ for smaller sizes, it would not be much out of the way.

Dimensions, Weights, and Approximate Prices of Locomotives.

The speed of locomotives, however, has not increased with their weight and size. There is a natural law which stands in the way of this. If we double the weight on the driving-wheels, the adhesion, and consequent capacity for drawing loads, is also doubled. Reasoning in an analogous way, it might be said that if we double the circumference of the wheels the distance that they will travel in one revolution, and consequently the speed of the engine, will be in like proportion. But, if this be done, it will require twice as much power to turn the large wheels as was needed for the small ones; and we then encounter the natural law that the resistance increases as the square of the speed, and probably at even a greater ratio at very high velocities. At 60 miles an hour the resistance of a train is four times as great as it is at 30 miles. That is, the pull on the draw-bar of the engine must be four times as great in the one case as it is in the other. But at 60 miles an hour this pull must be exerted for a given distance in half the time that it is at 30 miles, so that the amount of power exerted and steam generated in a given period of time must be eight times as great in the one case as in the other. This means that the capacity of the boiler, cylinders, and the other parts must be greater, with a corresponding addition to the weight of the machine. Obviously, if the weight per wheel is limited, we soon reach a point at which the size of the driving-wheels and other parts cannot be enlarged; which means that there is a certain proportion of wheels, cylinders, and boiler which will give a maximum speed.

The relative speed of trains here and in Europe has been the subject of a good deal of discussion and controversy. There appears to be very little difference in the speed of the fastest trains here and there; but there are more of them there than we have. From 48 to 53 miles an hour, including stops, is about the fastest time made by our regular trains on the summer time-tables.

When this rate of speed is compared with that of sixty or seventy miles an hour, which is not infrequent for short distances, there seems to be a great discrepancy. It must be kept in mind, though, that these high rates of speed are attained under very favorable conditions. That is, the track is straight and level, or perhaps descending, and unobstructed. In ordinary traffic it is never certain that the line is clear. A locomotive-runner must always be on the look-out for obstructions. Trains, ordinary vehicles, a fallen tree or rock, cows, and people may be in the way at any moment. Let anyone imagine himself in responsible charge of a locomotive and he will readily understand that, with the slightest suspicion that the line is not clear, he would slacken the speed as a precautionary measure. For this reason fast time on a railroad depends as much on having a good signal system to assure the locomotive-runners that the line is clear, as it does on the locomotives. If he is always liable to encounter, and must be on the look-out for, obstructions at frequent grade-crossings of common roads, or if he is not certain whether the train in front of him is out of his way or not, the locomotive-runner will be nervous and be almost sure to lose time. If the speed is to be increased on American railroads, the first steps should be to carry all streets and common roads either over or under the lines, have the lines well fenced, provide abundant side-tracks for trains, and adopt efficient systems of signals so that locomotive-runners can know whether the line is clear or not.

In what may be called the period of adolescence of railroads there was a very decided predilection on the part of locomotive engineers for large driving-wheels. Figure 34 represents one of the engines built as early as 1848 for the Camden & Amboy Railroad, with driving wheels 8 feet in diameter. Other engines with 6 and 7 feet wheels were not uncommon. In Europe many engines with very large wheels were made and are still in use. Here, as well as there, excessively large wheels have, however, been abandoned, and six feet in diameter is now about the limit of their size in this country.

So far as locomotives are concerned, fast time, especially with heavy trains, is generally dependent more upon the supply of steam than it is on the size of the wheels. Without steam to turn them, big wheels are useless; but with an abundant supply there is no difficulty in turning small wheels at a lively rate. Speed, therefore, is to a great extent a question of boiler capacity, and the general maxim has been formulated that "within the limits of weight and space to which a locomotive boiler must be confined, it cannot be made too big." But the maximum speed at which a locomotive can run when an adequate supply of steam is provided also depends on the perfection of the machinery. At 60 miles an hour a driving-wheel 5½ feet in diameter revolves five times every second. The reciprocating parts of each cylinder of a Pennsylvania Railroad passenger engine, including one piston, piston-rod, cross-head, and connecting rod, weigh about 650 pounds. These parts must move back and forth a distance equal to the stroke, usually two feet, every time the wheel revolves, or in a fifth of a second. It starts from a state of rest at each end of the stroke of the piston and must acquire a velocity of 32 feet per second, in one-twentieth of a second, and must be brought to a state of rest in the same period of time. A piston 18 inches in diameter has an area of 254½ square inches. Steam of 150 pounds pressure per square inch would therefore exert a force on the piston equal to 38,175 pounds. This force is applied alternately on each side of the piston, ten times in a second. The control of such forces requires mechanism which works with the utmost precision and with absolute certainty, and it is for this reason that the speed and the economical working of a locomotive depend so much on the proportions of the valves and the "valve-gear" by which the "distribution" of steam in the cylinders is controlled.

The engraving (Fig. 36) on p. 133 represents the cab end of a locomotive of the New York Central & Hudson River Railroad, looking forward from the tender, and shows the attachments by which the engineer works the engine.[12] This gives an idea of the number of keys on which he has to play in running such a machine. There is room here for little more than an enumeration of the parts which are numbered:

Besides being impressive as a triumph of human ingenuity, there is much about the construction and working of locomotives which is picturesque. A shop where they are constructed or repaired is always of interest. An engine-house (Fig. 35) especially at night, is full of weird suggestions and food for the imagination.

Figure 37 (p. 135) is an illustration from a photograph taken in the erecting shops of the Baldwin Locomotive Works in Philadelphia; and Figure 38 (p. 137) is a view of a similar shop of the Pennsylvania Railroad at Altoona, which suggests at a glance many of the processes of construction which go on in these great works. At Altoona are immense travelling cranes resting on brick arches and spanning the shop from side to side. These are powerful enough to take hold of the largest locomotive and lift it bodily from the rails and transfer it laterally or longitudinally at will. A large consolidation engine is shown in Figure 38, swung clear of the rails, and in the act of being moved laterally. The hooks of the crane are attached to heavy iron beams, from which the locomotive is suspended by strong bars. Figure 39 (p. 138) is a view in the blacksmiths' shop of the Baldwin Works, showing a steam hammer and the operation of forging a locomotive frame.

It is quite natural that the engineers, or "runners," as they generally call themselves, who have the care of locomotives should take a deep interest in and acquire a sort of attachment for them. In the earlier days of railroading this was much more the case than it is now. Then each locomotive had an individuality of its own. It was rare that two engines were exactly alike. Nearly always there was some difference in their proportions, or one engine had some device in it which the other had not. Now, many locomotives are made exactly alike, or as nearly so as the most improved machinery will permit. There is nothing to distinguish the one from the other. Therefore Bony Smith can claim no superiority for his machine which Windy Brown has not the advantage of. In the old days, too, each engine had its own runner and fireman, and it seldom fell into the hands of anyone else, and those in charge of it took as much pride in keeping it bright as the character in "Pinafore" did "in polishing up the handle of the big front door." On many roads—particularly the larger ones—engines are not assigned to special men. The system of "first in first out" has been adopted; that is, the engines are sent out in the order in which they come in, and the men take whichever machine happens to fall to their lot. This naturally results in a loss of personal attachment to special engines.

Every change in the construction, alteration in the proportions, or addition to the attachments of locomotives is a subject of intense interest to the men and a topic of endless discussion at all times and places. The theories which are propounded, and the yarns which are spun while sitting around hot stoves in round-houses, or waiting for passing trains on side-tracks, would fill many books. Jack never tires of telling what his engine did when "she was going up Rattlesnake Grade," and Smoky Bill grows excited when he describes how Ninety-six turned her wheels in making up forty-nine minutes time in the down run with the "electric express."

Locomotive engineers and firemen read with avidity everything which is explanatory of the construction or working of locomotives, but generally have a contempt for things which have no practical bearing. They demand "lucidity" in what they read with as much vehemence as Matthew Arnold did, and some editors and college professors, whose writing and thinking are foggy, would be greatly benefited by the criticisms of the Locomotive Brotherhood.

Much might be written about the duties of locomotive-runners and firemen, and the qualifications required. It is the general opinion of locomotive superintendents that it is not essential that the men who run locomotives should be good mechanics. The best runners or engineers are those who have been trained while young as firemen on locomotives. Brunel, the distinguished civil engineer, said that he never would trust himself to run a locomotive because he was sure to think of some problem relating to his profession which would distract his attention from the engine. It is probably a similar reason which sometimes unfits good mechanics for being good locomotive-runners.

It will perhaps interest some readers to know how much fuel a locomotive burns. This, of course, depends upon the quality of fuel, work done, speed, and character of the road. With freight trains consisting of as many cars as a heavy locomotive can draw without difficulty, the consumption of coal will not exceed from 1 to 1½ pounds of coal per car per mile if the engine is carefully managed. It takes from 15 to 20 pounds of coal per mile to move an engine and tender alone, the consumption being dependent upon the size of the engine, speed, grades, and number of stops. If this amount of coal is allowed for the engine and tender, and the balance that is consumed is divided among the cars, it will reduce the quantity for hauling the cars alone to even less amounts than those given above. In ordinary average practice the consumption is from 3 to 5 pounds per freight-car per mile, without making any allowance for the engine and tender. With passenger trains, the cars of which are heavier and the speed higher, the coal consumption is from 10 to 15 pounds per car per mile. A freight locomotive with a train of 40 cars will burn 40 to 200 pounds of coal per mile, the amount depending on the care with which it is managed, quality of the coal, grades, speed, weather, and other circumstances.

AMERICAN CARS.

Peter Parley's illustration (p. 101) of the Baltimore & Ohio Railroad represents one of the earliest passenger-cars used in this country. The accuracy of the illustration may, however, be questioned. Probably the artist depended upon his imagination and memory somewhat when he drew it. The engraving below (Fig. 40) is from a drawing made by the resident engineer of the Mohawk & Hudson Railroad, and from which six coaches were made by James Goold for the Mohawk & Hudson Railroad in 1831. It is an authentic representation of the cars as made at that time. Other old prints of railroad cars represent them as substantially stage-coach bodies mounted on four car-wheels, as shown by Figure 41. The next step in the development of cars was that of joining together several coach-bodies. This form was continued after the double-truck system was adopted, as shown by Figure 42, which represents an early Baltimore & Ohio Railroad car, having three sections, united. It was soon displaced by the rectangular body, as shown in Figure 43, which is a reproduction from an old print.

Figure 44 is an illustration of a car used for the transportation of flour on the Baltimore & Ohio Railroad, while horses were still used as the motive power. To show how nearly all progress is a process of evolution, it was asserted, in one of the trials of the validity of Winans' patent on eight-wheeled cars with two trucks, that before the date of his patent it was a practice to load firewood by connecting two such cars with long timbers, which rested on bolsters attached by kingbolts to the cars. The wood was loaded on top of these timbers, as shown in Figure 45. An old car (Fig. 46), which antedated Winans' patent and was used at the Quincy granite quarries for carrying large blocks of stone, was also introduced as evidence for the defendants in that suit. Although Winans was not able to establish the validity of his patent on eight-wheeled cars with two trucks, he was undoubtedly one of the first to put it into practical form, and did a great deal to introduce the system.

The progress in the construction of cars has been fully as great as in that of locomotives. If the old stage-coach bodies on wheels are compared with a vestibule train of to-day the difference will be very striking. Most of us who are no longer young can recall the days when sleeping-cars were unknown, when a journey from an Eastern city to Chicago meant forty-eight hours or more of sitting erect in a car with thirty or more passengers, and an atmosphere which was fetid. Happily those days are past, although the improvement in the ventilation of cars has been very slow, and is still very imperfect.

Improvement has also lagged in the matter of coupling cars. It has been shown by statistics and calculations that some hundreds of persons are killed and some thousands injured in this country annually in coupling cars. The use of automatic coupling, by which cars could be connected together without going between them, it has been supposed, would greatly lessen, if it would not entirely prevent, this fearful sacrifice of life and limb. To accomplish this end, though, it is essential that some one form of coupler shall be generally adopted by all railroads. One of the obstacles in the way of this has been the mechanical difficulty of finding a mechanism which will satisfactorily accomplish the purpose for which it was intended. After thirty or forty years of invention and experiment, no automatic coupler has been produced, which has been approved by competent judges with a sufficient degree of unanimity to justify its general adoption. The patents on that class of inventions are numbered by thousands, so that it is no light task to select the best one or even the best kind. Besides this difficulty, there is the other equally formidable one of inducing railroad men, of various degrees of knowledge, ignorance, and prejudice regarding this subject, and who are scattered all over the continent, to agree in adopting some one form or kind of automatic coupler. Various cliques had also been organized on different roads in the interest of some patents, and in such cases argument and reason addressed to them were generally wasted. Public indignation was, however, aroused; and the stimulus of legislation in different States compelled railroad officers to give serious attention to the subject. After devoting some years to the investigation, the Master Car-Builders' Association—which is composed of officers of railroad companies, who are in charge of the construction and repair of cars on the different lines—has recommended the adoption of a coupler of the type represented by Figures 47 to 49, which has been already applied to many cars and the indications are that it will be very generally adopted for freight and probably for passenger cars. If it should be, it will relieve railroad employees of the dangerous duty of going between cars to couple them. Figure 47 shows a plan looking down on the couplers with one of the latches, A, open; Figure 48 shows it with the two couplers partly engaged; and Figure 49 shows them when the coupling is completed.

One of the first problems which presented itself in the infancy of railroads was how to keep the cars on the rails.

Anyone who will stand close to a line of railroad when a train is rushing by at a speed of forty, fifty, or sixty miles an hour must wonder how the engine and cars are kept on the track; and even those familiar with the construction of railroad machinery often express astonishment that the flanges of the wheels, which are merely projecting ribs about 1¹/₈ inches deep and 1¼ inches thick, are sufficient to resist the impetus and swaying of a locomotive or car at full speed. The problem of the manufacture of wheels which will resist this wear, and will not break, has occupied a great deal of the attention of railroad managers and manufacturers.

Locomotive driving-wheels in this country are always made of cast-iron, with steel tires which are heated and put on the wheels and then cooled. They are thus contracted and "shrunk" on the wheel. The tread, that is, the surface which bears on the rail, and the flange of the tire are then turned off in a lathe, shown in Figure 25, on p. 121, made especially for the purpose. For engine-truck, tender, and car-wheels, until within a few years, "chilled" cast-iron wheels have been used almost exclusively on American railroads. If the tread and flange of a wheel were made of ordinary cast-iron they would soon be worn out in service, as such iron has ordinarily little capacity for resisting the wear to which wheels are subjected. Some cast-iron, however, has the singular property which causes it to assume a peculiar, hard crystalline form if, when it is melted, it is allowed to cool and solidify in contact with a cold iron mould. The iron which is thus cooled quickly, or "chilled," becomes very hard, and resists wear very much better than iron which is not chilled. Car-wheels which are made of this material are therefore cast in what is called a chill-mould. Figure 50 represents a section of such a mould and flask in which wheels are cast.

A A is the wheel, which is moulded in sand in the usual way. The part B B of the mould, which forms the rim or tread of the wheel, consists of a heavy cast-iron ring. The melted iron is poured into this mould and comes in contact with B B. This has the effect of chilling the hot iron, as has been explained. In cooling, the wheel contracts; and for that reason the part between the rim C and the hub D is made of a curved form, as shown in the section, so that if one part should cool more rapidly than another these parts can yield sufficiently to permit contraction without straining any portion of the wheels injuriously. For the same reason the ribs on the back of the wheels, as shown in Figure 51, are also curved. As an additional safeguard to the unequal contraction in cooling, the wheels are taken out of the mould while they are red-hot, and placed in ovens where they are allowed to remain several days so as to cool very slowly.

Figure 52, on p. 145, represents a section of the tread and flange of a chilled wheel, showing the peculiar crystalline appearance of the chilled iron.

In making cast-iron wheels the quality of the iron used is of the utmost importance. The difficulty in making good wheels lies in the fact that most iron which is ductile and tough will not chill, whereas hard white iron, which has the chilling property in a very high degree, is brittle, and wheels which are made of it are liable to break. There are some kinds of cast-iron produced in this country which have the two qualities combined, in a very remarkable degree; that is, they are ductile and tough, and will also chill. Wheel-founders also mix different qualities of irons to produce wheels with the required strength, and which will resist wear; that is, they use a certain amount of hard white iron which will chill, with that which is ductile and soft. By changing the proportions, any required amount of chill can be produced. The danger is that iron which has little strength or ductility will be fortified with hard chilling iron, and a very weak wheel will thus be the result. Thousands of such wheels have been bought and used because they are cheap, and many lamentable accidents are undoubtedly due to this cause. To guard against this, car-wheels should always be subjected to rigid tests and inspection.

In Europe wheels are made of wrought-iron, with tires which were also made of the same material before the discovery of the improved processes of manufacturing steel, but since then they have been made of the latter material. Owing to the breakage of a great many cast-iron wheels of poor quality, steel-tired wheels are now coming into very general use on American roads under passenger-cars and engines. A great variety of such wheels is now made. The "centres" or parts inside the tires of some of them are cast-iron, and others are wrought-iron constructed in various ways.

What is known as the Allen paper wheel is used a great deal in this country, especially under sleeping-cars. A section and front view of one of these wheels is shown by Figure 53. It consists of a cast-iron hub, A, which is bored out to fit the axle. An annular disk, B B, is made of layers of paper-board glued together and then subjected to an enormous pressure. The disk is then bored out to fit the hub, and its circumference is turned off, and the tire C C is fitted to it. Two wrought-iron plates, P P, are then placed on either side of it, and the disk, plates, tire, and hub are all bolted together. The paper, it will be seen, bears the weight which rests on the hub of the axle and the hub of the wheel.

Steel tires have the advantage that when they become worn their treads and flanges may be turned off anew, whereas chilled cast-iron wheels are so hard that it is almost impossible to cut them with any turning tool. For this reason machines have been constructed for grinding the tread with a rapidly revolving emery-wheel. In these the cast-iron wheel is made to turn slowly, whereas the emery-wheel revolves very rapidly. The emery-wheel is then brought close to the cast-iron wheel, so that as they revolve the projections on the latter are cut away, and the tread is thus reduced to a true circular form. These machines are much used for "truing-up" wheels which have been made flat by sliding, owing to the brakes being set too hard.

It would require a separate article to give even a brief description of the different kinds of cars which are now used. The following list could be increased considerably if all the different varieties were included.

The following table gives the size, weight, and price of cars at the present time. The length given is the length over the bodies not including the platforms.

Some years ago the master car-builders of the different railroads experienced great difficulty in the transaction of their business from the fact that there were no common names to designate the parts of cars in different places in the country. What was known by one name in Chicago had quite a different name in Pittsburg or Boston. A committee was therefore appointed by the Master Car-Builders' Association to make a dictionary of terms used in car-construction and repairs. Such a dictionary has been prepared, and is a book of 560 pages, and has over two thousand illustrations. It has some peculiar features, one of which is described as follows in the preface: "To supply the want which demanded such a vocabulary, what might be called a double dictionary is needed. Thus, supposing that a car-builder in Chicago received an order for a 'journal-box'; by looking in an alphabetical list of words he could readily find that term and a description and definition of it. But suppose that he wanted to order such castings from the shop in Albany, and did not know their name; it would be impracticable for him to commence at A and look through to Z, or until he found the proper term to designate that part." To meet this difficulty the dictionary has very copious illustrations in which the different parts of cars are represented and numbered, and the names of the parts designated by the numbers are then given in a list accompanying the engraving. An alphabetical list of names and definitions is also given, as in an ordinary dictionary. The definition usually contains a reference to a number and a figure in which the object described is illustrated. In making the dictionary the compilers selected terms from those in use, where appropriate ones could be found. In other cases new names were devised. The book is a curious illustration of a more rapid growth of an art than of the language by which it is described.

The following table, compiled from "Poor's Manual of Railroads," gives the number of locomotives and of different kinds of cars in this country, beginning with 1876, and for each year thereafter. If the average length of locomotives and tenders is taken at 50 feet, those now owned by the railroads would make a continuous train 280 miles long; and the 1,033,368 cars, if they average 35 feet in length, would form a train which would be more than 6,800 miles long.

Statement of the Rolling Stock of Railroads in the United States; from "Poor's Manual" for 1889.

The number of cars, it will be seen, has more than doubled in ten years, so that if the same rate of increase continues for the next decade there will be over two millions of them on the railroads of this country alone. Beyond a certain point, numbers convey little idea of magnitude. Our railroad system and its equipment seem to be rapidly outgrowing the capacity of the human imagination to realize their extent. What it will be with another half-century of development it is impossible even to imagine.

[9] An engraving of a team and of a "Conestoga" wagon—which was used in this traffic—taken from a photograph of one which has survived to the present day, is given opposite (Fig. 1).