from another source- that of my back-action or impingement of steam against the backs of the cavity next in succession. The following, therefore, is a description of the engine or apparatus by which I propose to effect these objects :

"It is a section of a case, with a wheel fitting within, about half an inch all round, and revolving within the case. This case, for a high-pressure engine, may be made of cast or sheet-iron, of suitable thickness to resist the pressure of the atmosphere caused by a partial vacuum from the rapid action of the steam wheel. The whole of the inside should be made perfectly smooth, to prevent friction of the used steam when it re-issues from the cavities. For a condensing engine the case must be made strong enough to resist the atmospheric pressure. At those parts where the jets of steam are admitted, several cavities are made with or attached to the inside circumference of the case, for the purpose of receiving the used steam, and returning the same into the cavities of the steam wheel; and where a double row of cavities are used round the steam wheel, the rows of cavities in the case must be reversed accordingly. These cavities in the case must be placed in such a position to the cavities of the steam wheel that, on the used steam issuing therefrom and rushing into the cavities of the case, it may be returned therefrom into the cavities on the steam wheel at the same angle as that at which the steam first issued from the jet-pipes, and the used steam will then, as the wheel revolves, be thrown out at the eduction passage.

"The periphery of the steam wheel is formed by a series of cavities all round of the same dimensions, for the reception of one or more jets of steam. The size of the cavities in the wheel is three quarters of an inch deep, by one and a half long, and half an inch wide."

The method adopted to determine the best velocity is this: The velocity with which steam at 60lbs. pressure above the atmosphere rushes through an orifice is known experimentally to be about 104,490 feet per minute. The difference between this and the velocity of the periphery of the wheel gives the velocity of impact of the steam on the cavities of the steam wheel, and the power or pressure exerted on the wheel, being proportional to the square of the velocity of impact, the power of the engine can be calculated for various velocities of the wheel. Mr. Pilbrow has done this, and finds that "the greatest results are obtained when the wheel revolves at precisely one third of the velocity of the steam, or 34,830 feet per minute."

We have now given sufficiently copious extracts from Mr. Pilbrow's specification to explain the principal features of his new engine. In practice Mr. Pilbrow has not been able to make his engine succeed, and we do not believe that, by the plan proposed, he will be able to make a good, durable, and economical engine. He has not yet, at least, proved that his engine will realise nearly the power he anticipates from it. To make a few experiments on the force of impact of a jet of steam on surfaces at rest, and from that to theorise on the mechanical effect of steam rushing into and rebounding from cavities in rapid motion; to calculate the velocity of impact on one surface, and of rebound from another, as if it were a solid and perfectly elastic body, whose laws of motion are exactly known, is a method of proceeding that can only lead to error, or at least can never bring us with certainty to the truth. We agree with Mr. Pilbrow, that the whole power of the steam is expended in giving velocity to its own particles, and, theoretically speaking, the same power ought to be extractible from them. In Mr. Pilbrow's engine, however, we have seen that the periphery of the wheel is to move with one third of the velocity of the steam, leaving only two thirds of the velocity to produce mechanical power. Now, Mr. Pilbrow himself says, that the mechanical power produced is proportional to the square of the velocity of impact, and the square of is, or scarcely one-half. So that here is fully onehalf of the whole mechanical effect dissipated at once: but this is not all; for when the steam strikes the periphery of the wheel, with two-thirds of its actual velocity, it must, by the laws of elastic bodies, rebound with the same velocity from the wheel; that is, it will rebound with an actual velocity of one-third, while the wheel moves in the opposite direction with an equal velocity. There is, therefore, this velocity of one-third of its original velocity still remaining in the steam, and unextracted after impact. One-third of the velocity corresponds to one-ninth of the mechanical effect. This, therefore, is also lost. We believe, indeed, that in the later modifications of the plan, it has been Mr. Pilbrow's object to have the steam reflected from the wheel, at a velocity equal to that of the wheel's rotation, so that it will, in effect, be brought to a state of rest; and professor Mosely has made a number of calculations to show that there is no loss of power if this condition be fulfilled. But the condition implies that the particles of steam are perfectly elastic, which is uncertain, and involves the necessity of the wheel moving with half the speed of the effluent steam, which is impossible in practice. A velocity of one-third of that of the steam, viz. 34,830 feet per minute, at the periphery is 3,166 revolutions per minute,—a velocity sufficient to make the engine fly to pieces. Mr. Pilbrow himself could not avoid seeing this objection, and accordingly he has attempted to prescribe a remedy. He proposes to make the wheel go slower, by which means the steam will rebound from its periphery with a correspondingly greater velocity, and is to be made to impinge on a cavity in the case similar to those in the wheel. From this it will again rebound, and strike the wheel, and again be retnrned to another cavity in the case, and thence

once more impinge on the wheel. These alternate saltations from the wheel to the case, and from the case to the wheel, are to be continued till the velocity of the steam has been entirely expended in imparting power to the wheel. We believe few persons will be able to refrain from a smile, when they contemplate the steam performing these fantastic gyrations. We suspect that it will be more apt to take one long jump to the exhausting port, and make its way off the stage as quickly as possible, instead of stopping to complete so many ingenious evolutions. It is but justice to Mr. Pilbrow to say, that he is very doubtful of the success of this part of his plan. His having ventured to propose it at all is a strong proof to us that he felt very strongly the force of our objection to the enormous velocity proposed to be given to his steam wheel.

Such, then, is a slender specimen of the rotatory engines which have at various times been projected for the supercession of the cylinder engine : many of them display much ingenuity, and, indeed, on no single subject perhaps has so much ingenuity been expended: nevertheless, up to the present time, no rotatory engine has been contrived which can be esteemed preferable to the common engine. We have already expressed our conviction that this will not be always so; but we believe an efficient rotatory engine must be sought rather among dynamical than statical resources. Pilbrow's engine, though it has weighty faults, is a movement in the right direction, and an engine on the impulse principle permits the benefit due to expansive action to be realised. From engines of the Elopile class a beneficial result is hardly to be expected, at least if made in the ordinary method; for by such engines there must be a loss of effect if the steam leaves the revolving arm with a greater velocity than that with which the arm moves, and this, unless by a combination of Elopiles, it must do at all ordinary pressures and speeds.

AMERICAN ENGINES.

The engines made in America are for the most part of a very rude and primitive description; yet the performance of the American steam vessels is such as to make them fully a match in point of speed to the best steam vessels of English construction. Their efficient performance is partly perhaps due to the high pressure of steam employed, and partly to the peculiarities of American river navigation, which are such as to enable a large sized vessel to subsist with a very moderate draft of water. On the Mississippi and its tributaries most of the engines employed are of the high pressure kind: a hundred pounds upon the square inch is esteemed a moderate pressure, and sometimes the pressure is raised as high as a hundred and fifty pounds on the inch. The engines employed in the vessels on the Hudson are for the most part condensing engines. Some of them have horizontal or inclined cylinders; in other cases the cylinder is placed above the shaft, with side-rods extending from the ends of a cylinder crosshead to cranks on the shaft beneath; while in a third variety a beam is employed as in the ordinary land rotative engine. Sometimes two engines are employed; but very often only one, the crank being carried over the centre by the momentum of the vessel. A few sets of engines for steam vessels have been made in America after English examples, and some of these specimens, which have come under our observation, are highly creditable performances; but most of the engines made in the country are of the quality and complexion of that of the North America. We subjoin a view of the machinery of that vessel; and we may here set down a few of the chief dimensions:- Diameter of cylinder, 43 inches; length of stroke, 11 ft.; length of keel, 200 ft.; breadth of beam, 25 ft.; diameter of paddle wheel, 27 ft.; length of float, 10 ft.; dip of float, 27 in.; pressure of steam, 50 lbs.

The framing of this engine is of timber: the working beam consists of a cast-iron skeleton frame trussed with wrought iron; and the crank and connecting-rod are both trussed with malleable iron rods. The beam is very short in proportion to the length of the stroke, and the place of the parallel motion is supplied by guides, the piston-rod being coupled to the beam by a long link, to enable the guides to operate efficiently. The several pieces composing the wooden framework which supports the crank shaft are keyed together with wooden keys, and bound with iron knees and plates of iron, to make the whole stiff and firm. The valves are double spindle valves, so that they are kept in equilibrium: this valve has the disadvantage that the valve-spindle expands more than the valve-casing, both from its higher temperature and the greater expansibility of the metal. Such valves, therefore, though tight when cold, will not be tight when hot. The cut off or expansion valve consists of a disc turning on a centre, like the throttle valve, and set in the steam-pipe, which it exactly fills when closed: in the rest of the engine there is little that is peculiar. The trussed beam might, it appears to us, be adopted with advantage in all large beam engines in this country, as it is not merely lighter than the cast-iron beams commonly used, but stronger and more safe. The trussing of the crank and connecting-rod we look upon as superfluous; and it must be difficult when they are trussed to keep those parts clean.

In many of the American steamers the engines and boilers are placed upon the deck, and beneath the deck a saloon extends the whole length of the vessel, for the accommodation of passengers. Most of these saloons are magnificently fitted up: many of them are upwards of 150 ft. long, 20 ft. wide, and 12 ft. high; and the accommodations are in every respect most

complete and commodious. The paddle-wheels are generally entirely constructed of wood, with the exception of the centres, to which the arms are bolted, and which are of cast-iron. The usual number of strokes per minute, with an 11 ft. stroke, is from 25 to 27, so that the piston travels at a great velocity. In the vessels on the Mississippi the paddles are made with a clutch or friction-strap, so that they may be thrown out of gear, and the engines may be turned so as to feed the boilers when the vessel is alongside a wharf, without moving the paddle-wheels. The steam vessels plying on the Mississippi are chiefly built at Pittsburg and Cincinnati: these vessels are much inferior to those plying from New York to Albany, and to Providence; and they are managed in the most reckless manner. Mr. Stevenson, speaking of one of these vessels, says, "She was steered close in-shore amongst stones and stumps of trees, where she lay for some hours to take in goods: the additional weight increased her draught of water, and caused her to heel a good deal; and when her engines were put in motion, she actually crawled into deep water on her paddle-wheels: the steam had been got up to an enormous pressure, to enable her to get off, and the volume of steam discharged from the escapement-pipe at every half stroke of the piston made a sharp sound, almost like the discharge of fire-arms, while every timber in the vessel seemed to tremble." The numerous explosions of boilers on the Mississippi cannot excite astonishment amid such provocations as that here narrated. The boilers are not calculated to withstand any very high pressure, even when new. Of the best form of American boiler we have given an example in page 56'; such a boiler would not be considered safe in this country with a greater pressure than 10 or 12 lbs. on the inch.

THE LOCOMOTIVE ENGINE.

The whole of the engines we have yet described at any length are condensing engines-the steam, after having urged the piston to the end of its travel, escaping into the condenser, where, by the abstraction of heat, it is reduced again to the form of water. The locomotive engine, however, is not of the condensing, but of the high pressure kind, the steam escaping, after having given motion to the piston, into the atmosphere. We have already explained that there is a loss of effect incidental to the use of the high pressure engine, but it would be most inconvenient to carry the large quantity of water a locomotive would require for condensation, and the rush of waste steam acts beneficially by blowing the fire; so that in this case to obviate the loss of effect incidental to the omission of condensation would involve the introduction of greater evils.

The locomotive engine of Messrs. Bury, Curtis, and Kennedy is a very complete and elegant piece of mechanism; and we may here give an explanation of the several parts of the merchandise engine of these eminent makers, which is represented in detail in Plate 23. Fig. 1. is a sectional elevation; fig. 2. is a sectional plan, xr being the principal line of section on which fig. 1. is shown; fig. 3. is a transverse section through the smokebox. The same letters of reference are used throughout all the figures; and to render them more explicit, several parts are shown which do not properly come into view on the lines of section. A is the fire-box: it is made of wrought iron of an inch thick, except the tube-plate, which is half an inch thick. The joints are welded wherever they are in contact with the burning fuel, as a rivetted joint, from its presenting a double thickness of metal, will not long resist the intense heat to which it is exposed. The fire-box is of a cylindrical form, with its back flattened to receive the ends of the tubes: the top is hemispherical, surmounted by a small dome, into which the upper end of the steam-pipe is carried, to obviate priming. B are the fire-tubes, of which there are 96, 2 in. diameter, and 9 ft. long. It will be seen, from the transverse section, that the tubes are so disposed as to concentrate the heat towards their centre, with the view of making a current from the outside of the tubes, where the water is colder, towards the bottom of the tubes, whence it will rise when heated up among the tubes. Partly for this reason, and also to prevent the tubes from being uncovered by the centrifugal recession of the water when the engine is travelling upon sharp curves, the upper row of tubes follows a circular sweep, the highest point being in the centre of the engine.

C. Is the smoke-box. D. The regulator. E. The steam-pipe, 3} inches diameter. F. The safety-valve and spring pressure-gauge, 2 inches diameter. G. The locked-up safety-valve, 2 inches diameter. H. The damper. I. The buffer-bar. J. The steam-whistle. L. The steam-cylinders, 13 inches diameter, 18 inches stroke. M. The force-pumps, plunger 2 inches diameter, 18 inches stroke. N. The cranked axle; the journals are 5 inches diameter, and 7 inches long: the bearing of each crank is 5 inches diameter and 3 inches long. O. The connecting-rods, ovalshaped, 2 inches by 21. P. The axle of the front wheels, 4 inches diameter. Q. The springs. The springs for the cranked axle are composed of 16 plates, together 4 inches deep at the centre; those for the front axle are composed of 10 plates, together 37 inches deep at the centre. aa. are the steam-pistons of gun-metal: the packing consists of two rings of castiron segments, forced outwards by brass wedges and steel springs. The piston-rods are 2 inches diameter. bb. are the inlet passages for the steam, 1x 6 inches, c. The outlet passages for the steam, 13 x 6 inches. d. The slide valves. d'. The slide-valve rods, 1 inch diameter. e. The pen

dulum-rods for carrying the ends of the eccentric rods, f. The shaft to which the eccentric levers are fixed, g. The shaft connecting the motion of the lever h, and the rod i, to the shaft. h. The guides for the piston-rods. i. Steadying pieces for the guides. j. Shaft carrying the steadying pieces. kl. The rods for moving the slide-valves. kl. The levers of the handgear. m. The shaft carrying the valve trappings. n. The lever for working the valves. n'. The lever worked by the eccentrics. p. The eccentrics for the retrograde motion. q. The eccentrics for the advancing motion. r. The pipes (2 inches diameter) connecting the force-pumps with the tender. s. The cock for letting the water out of the boilers. t. The rods (1 inch diameter) for coupling together the front wheels and the driving-wheels. u is a lead plug, placed at the culminant point of the dome-shaped top, and which will melt before any other part of the fire-box is left dry.

The framing of this engine is that which has been so long used by Messrs. Bury, and which is known as the inside framing. Mr. Bury claims for the inside framing a great superiority over the outside, on the ground that it forms a stronger connection between the cylinder, crank axle, and other moving parts, and bears all strains and concussions without throwing any of them upon the boiler. The following is Mr. Bury's comparison between the inside and outside framings:

:

"The advantages of the inside bearing are best described by comparing it with the ordinary outside framing when submitted to the principal strains which it has to resist. The most important is that caused by the whole power of the engine acting as a direct strain upon the crank as it passes over either centre. With the inside framing, the centre line of the connecting rod is only 10 inches distant from the centre line of the frame, and the total distance between the bearings is 43 inches; but when the framing is outside the wheels, these dimensions are necessarily 20 inches and 72 inches respectively, and the effect of the strain upon the crank in this case would be to its effect with the inside framing as 14 is to 8. For this reason, when the principal frame is placed outside the wheels, it becomes necessary to have an additional inside framing, to prevent the friction of the axle; these additional inside frames not only cause an increase of friction on the bearings of the cranked axle, but also throw a considerable strain on the boiler, which then becomes the medium of connection between the inside and outside frames, the inside frames being fixed at one end to the bottom of the smoke-box, and at the other end to the fire-box; while the principal frame is attached by long brackets to the body of the boiler. The fact that the use of four additional inside frames occasions six bearings on the axle (that axle being only 6 feet long), renders the system of principal outside framings so objectionable, that that circumstance alone should suffice to cause their rejection; for it is well known to practical men that it is impossible to key so many bearings perfectly true, and to maintain them so, when the engine is working; and even if this precision were attained, the aggregate friction on the four inside and the two outside bearings would be much greater than when it is all thrown upon two bearings; because, in the first place, all the friction due to the weight of the boiler is borne by the two outside bearings alone, and that which results from the pressure of the steam, through the medium of the connecting-rod, is thrown upon the four inside bearings: the pressure upon the outside bearings is vertical, and the mean pressure upon the inside bearings is nearly horizontal. So that if, instead of acting separately, these two amounts of pressure were thrown on the same bearings, the friction would only be due to the resultant of the pressures, and would consequently be much reduced. The friction on the cranked axle, having only two bearings, as, where a single inside frame is used, will be, under ordinary circumstances, that due to the resultant of the vertical and horizontal pressures, or,

In addition to the friction resulting from these forces, there is a considerable pressure on the bearings, arising from the tightness of the brasses; and it is evident that the friction arising from this cause will be three times greater with six than with two bearings. Another important feature is the strain to which locomotive engines are liable, from the pressing or striking of the flanges of the wheels against the rail when travelling in a curve. In engines with the bearing inside the wheels, the weight of the boiler has a tendency to bend the axle down in the centre; while the pressure of the flange against the rail acts upon it in a contrary direction, and thus one strain counteracts the effect of the other. If the bearing is outside the wheel, the weight of the boiler tends to bend the axle upwards, and a strain upon the flange of the wheel acts in the same direction, and in addition to it."

The position of the bearings inside the wheels is of great practical advantage in case of the fracture of the cranked axle, as the weight on the bearings presses the flange of the wheel against the rail, and assists the length of the journal in keeping it from being thrown off the rail. Instances have occurred on the London and Birmingham Railway, when an axle has broken, that not only have the wheels remained on the rails, but the driver has been enabled to proceed with the train to the nearest station. The stiffness of the single inside framing is not only a remedy against the excessive wear and tear which is consequent on a less perfect union between the parts of the engine, but its simplicity allows the whole machinery to be arranged in a more compact form, and constructed with greater solidity.

The best proof of the soundness of the doctrines here put forth is the fact that inside framings have now become nearly universal. Mr. Bury prefers four-wheeled to six-wheeled engines, and adduces many weighty reasons in justification of the preference: the circumstances of the case, however, may be altered by the exigencies arising out of the demand for a higher rate of speed; for engines of the power requisite to maintain that speed may be too heavy for the rails of existing lines, unless supported by a greater number of wheels: but in such a case it appears better to change the plan of engine altogether, to the end that the four wheels may be still retained.

The locomotive engine of Messrs. R. Stephenson and Co., represented in Plate 24., differs in many respects from that of Messrs. Bury and Co. It is furnished with six wheels, and the two middle wheels, which are coupled to the wheels behind them, are divested of their flanges, so that the engine may be better able to adapt itself to curves. The fire-box is square, and the water space between it and the outside shell is securely stayed, by means of bolts tapped into the plates, and rivetted over at the ends. The plate we have given will readily be understood by the aid of the following references:

-

A A is the steam dome. A' is the fire-box, which is square, with a flat top, according to Messrs. Stephenson's usual practice. A" is the foot-plate, for the engineer to stand upon when working the engine. BB is the smokebox. B' is the chimney, which is broken to save room. C is the boiler. E is the blast pipe. GG are the cylinders. His the piston. K is the crank axle. I is the eccentric rod for reversing one engine, and I' is the eccentric rod for reversing the other. M, N, are the jaws of the eccentric gab. O is the eccentric gab. Q is the starting and reversing lever. Q' is the fulcrum of the lever. S is a wide-mouthed pipe, carried up into the steam dome, to obviate priming. S' is the chest in which is placed the stop-valve or regulator. S" is the steam pipe carried through the boiler, and S"" is the short vertical steam pipe passing through the smoke-box to the cylinders. T is a cross stay between the piston guides, which serves as a guide for the valve spindles. U is the cock upon the feed-pipe where it enters the boiler. YY is the framing. aa are the water spaces round the fire-box, strengthened with stays. a'a' is the fire-grate. a" a" are the fire-tubes. dd are two safety-valves placed upon the steam dome, and loaded by spring balances. d' is the whistle. e is the eduction port. g is the crank of one engine, and g' that of the other. hh is the piston rod. h' is the connecting rod. h" h" are the guide blocks of piston cross-head. i is the eccentric rod of one engine for going ahead, and that of the other engine. 1 are the links connecting the drawing and the hind wheels. m, n, are the jaws of the eccentric gab for advancing. p is the feed-pump of which p' is the plunger, worked by p", the pump rod, which is attached to a lug on the eccentric hoop. qq is a long link, connecting the starting handle with lever q', fixed on the shaft r, on which is also fixed another lever q', and connected with the link q" to the end of one of the eccentric rods. The ends of the ex

centric rods of each engine are connected permanently by a short link, and have each a pin at their extremities fitting into the gabs, which are in this case fastened upon the valve rod. r is the regulator handle fixed upon the spindle which passes through a collar into the steam pipe s', and is then attached to the regulator r'. ss is the valve chest, in which the valves of the two cylinders are placed back to back. t is the slide valve. is the valve spindle. t' is the prolongation of the valve spindle through the front end of the cylinder, and also through the cross guide T. u' is the suction pipe, leading from the tender to the feed-pump. u"u" are the force-pipe and valve-chest of feed-pump. ""u" is the feed-pipe leading from the valve chest to the boiler. x is the shaft of reversing levers. yy are the axle bearings. zz is the smoke box door.

In this engine, it will be remarked, the boiler tubes are very long, which necessarily makes the engine long, and several inconveniences arise from the elongation. The hot air proceeding to the chimney experiences a greater resistance in passing through the longer tubes, which makes a greater intensity of draught necessary, and this in its turn involves a greater contraction of the blast pipe, whereby power is lost equivalent in amount perhaps to the steam gained by the extension of the tube surface. As the engine too is poised as it were upon the central driving wheels, it is subject to violent pitching and oscillatory movements should the driving wheels be screwed too tightly down upon the rail, and this there is a perpetual temptation on the part of the engine driver to do if the rails are wet or greasy. In most locomotive engines the bearings of the driving axle are too short, in consequence of which they speedily wear; and the engine will be subject to dangerous oscillations if there be much side play in the bearings of the axles. This would be effectually obviated by making the bearings spheroidal, and the next best alternative is to make them long and with very large fillets in the corners. We look upon it, however, as quite impossible that the present construction of locomotive engines can last very long, as it is full defects notwithstanding all the ingenuity expended upon it. The area of fire grate is much too small, the tubes are too long, and as a consequence of these defects the blast is in general too much contracted. The mass of matter partaking of the reciprocating motion of the piston is also too large: it would be better that the motion were communicated to the driving axle by means of two inverted oscillating engines set at an angle of 45 with one another and attaching themselves to a single crank in the middle of the driving axle: but in this case a provision would have to be made of such a nature that the movements of the engine would not affect the action of the springs, for the engine would otherwise work the springs up and down at every stroke. In cases where outside cylinders are employed, which, instead of attaching themselves to the crank axle, connect to pins in the driving wheels, the cylinders must be carefully felted to retain the heat, and they may then be covered with oil-cloth, which may also be employed to cover the wooden casing of the boiler with advantage.

PISTONS.

WE believe the variety that appears in the subjoined collection of pistons will excite some remark. Had it been our object to present as extensive a catalogue as possible, we should have added to the number; or had we intended to make merely a selection of the most approved, we might greatly have retrenched it. By presenting a somewhat heterogeneous and unselected enumeration of the principal varieties, just as they have occurred to us in practice, and noting their respective characteristics, whether merits or defects, we may, perhaps, make our account more instructive than if we were to give either a more methodical description or more restricted catalogue.

A consideration of the annexed sketches leads us to rank metallic packed pistons under two divisions (which, if not perfectly distinct, are sufficiently so for the purposes of classification), to one or other of which each may be referred: those in which the expansive force of the rings alone is used, and those in which it is either assisted or entirely superseded by springs. The former kind, by far the simplest and most obvious, seems to be of much the more recent adoption; and just adds one to the many instances which the history of engineering presents, of the shortest road being the last to be taken. Of this species we may reckon three varieties; first, those in which hemp packing requires to be compressed into the space between the rings and the piston, to aid the elasticity of the former; this is commonly the case in pumping engines, where hitherto the more complicated and expensive descriptions of packing have not been generally adopted.

Fig. 206. represents a piston of this variety, at work in a pumping-engine in Perthshire, manufactured by the Messrs. Maxton, of Leith; the oblique cut in the ring being designed to prevent the sharp edges of the break from grooving the cylinder. In this description of piston, two rings are generally Fig. 206.

Scale 1 inch=1 foot.

PISTON OF A PUMPING ENGINE, BY MESSRS. MAXTON, LEITH.

employed, one above the other, with the breaks about 90° apart, to prevent the steam escaping at the joint. This is the simplest kind of metallic packing, but when properly fitted up, it possesses nearly all the advantages of the more complicated descriptions; the only disadvantage being the necessity of occasional adjustment, when the internal hemp gasket loses its

elasticity, the frequency of which depends upon the accuracy with which the end joinings of the external rings are contrived to prevent the inlet of steam into the hemp, and also in some degree upon the temperature and pressure of the steam itself. Another form consists of a single ring with a tongue joint, as shown in the figure.

Closely allied to the above is what we may term the second variety of this species, an external packing ring like the former, but deriving its tightness from its own elasticity, and, of course, not dependent on the hempen packing behind. The simplest and perhaps the most common method of giving the requisite spring is to turn the packing rings a little larger than the diameter of the cylinder, and when sawn through to cut a tenon and mortice, or a half check in the abutting ends, and then to compress the ring by an iron hoop with screws, and to fix it temporarily with a pin put through the overlapping or morticed ends; in this state the rings are ground on the surface joints, and the piston made ready for its place; when, the hoop being unscrewed and the temporary pin withdrawn, the rings are suffered to expand in the cylinder by their own elasticity, which will generally continue to act till the rings and the cylinder are so much worn as to permit the rings to expand to their natural extent.

Sometimes the abutting ends are left plain, in which case a piece is merely cut off one end, to allow the ring to be compressed to a lesser diameter. Great diversity of opinion exists as to the merit of this species of packing: that it is a decided improvement upon the former is unquestionable, but it is alleged, that in accuracy of form, and facility of application to the cylinder, it is greatly inferior to the ordinary more complex varieties, with a number of segments and artificial springs to each. It is said, in the first place, that it tends to wear the cylinder off the truth, and no small degree of ingenuity has been put in requisition to remedy the evil; with what degree of success we may here briefly consider. Appeal to experience might seem the shortest way of settling the question, as many pistons both of the simple and the amended construction are in use, but in this, as in many other cases, recorded experience only serves to prevent any conclusion whatever, or rather tends equally to two contradictory results.

Let us then investigate the action of the simplest form. The ring, when first compressed, does not naturally assume the circular form, as by the two ends being brought together the tendency to expand to its original dimensions is mainly checked in only one direction. When confined in the cylinder, however, it will be seen at once that it is compressed equally in all directions, and must therefore exert a corresponding force equally in every direction, to recover its original dimensions. This appears so plain, as scarcely to be susceptible of illustration. Suppose, for instance, the original diameter of the ring to be 18 in., and to be compressed into an 18 in. cylinder, it will at once be seen that every diameter of the ring being compressed equally has an equal tendency, and must exert an equal force, to regain its original dimensions.

Instead of an exterior compressing cylinder to confine the ring, we may, to assist the true conception of the case, suppose the various diameters to be replaced by so many bowstrings, each of which plainly exerts an equal strain in confining its respective semicircumference, and thus the amount of pressure exerted around the entire circumference must every where be equal. We are told that this has been denied, and the reverse maintained by some judicious practical men, but chiefly, we believe, if not entirely, on theoretical grounds; and even though recorded experience be appealed to in support of their views, it cannot neutralise the opposing testimony, that in some cases the cylinder has been found to have worn perfectly true. The variety and ingenuity, and we might add, complexity of the various contrivances adopted, to correct this supposed defect in this description of packing, and the alleged success by which they have been attended, must be our apology for dwelling on the subject. But although the apprehended evil did exist in its full amount, and the rings really had a greater tendency to expand in the direction perpendicular to the diameter passing through the break, we do not see what bad effect would ensue; the cylinder would be worn almost imperceptibly oval, till at length the inequality of pressure

H H