did not extend beyond the antipathy of thirty-two feet of water. The speculations of Galileo were prosecuted after his death by his pupil Torricelli, who, in 1643, discovered that the effort of any fluid open to the atmosphere to enter a vacuous space is owing to the pressure of the air upon bodies on the earth's surface. This discovery gave a great impulse to mechanical ingenuity, and many schemes were contrived to make this new agent available as a motive power; but the first of these projects that appears to have been of any avail was the fire engine of Captain Thomas Savery, who produced a vacuum by condensing steam in close vessels, and then applied the vacuum so obtained to the elevation of water. Savery, however, also made use of the elastic force of the steam after the manner proposed by the Marquis of Worcester; but he made the pressure of the atmosphere carry the water up the first stage, which the Marquis of Worcester does not appear to have done, although there must have been a vacuum created in his receivers as effectually as in those of Savery, occasioned by the condensation of the steam on the entrance of the cold water. Savery obtained a patent from William III. "for raising water, and occasioning motion to all sorts of mill-work by the impellent force of fire" in 1698; and for several years thereafter he appears to have been actively engaged in getting his engine introduced into practice. In 1702 he published a small work, called the "Miner's Friend," in which he gives the following account of the structure of his engine, and of the mode of its operation:

"AA the furnaces which contain the boilers; B 1, B2 the two fire-places; C the funnel, or chimney, which is common to both furnaces. In these two furnaces are placed two vessels of copper, which I call boilers, the one large, as L, the other small, as D; D the small boiler contained in the furnace, which is heated by the fire at B 2; E the pipe and cock to admit cold water into the small boiler to fill it; F the screw that covers and confines the cock E to the top of the small boiler; G a small gauge cock at the top of a pipe, going within eight inches of the bottom of the small boiler; Ha

larger pipe, which goes the same depth into the small boiler; I a clack or valve at the top of the pipe H (opening upwards); K a pipe going from the box above the said clack or valve, in the great boiler, and passing about an inch into it; LL the great boiler contained in the other furnace, which is heated by the fire at B1; M, the screw with the regulator, which is moved by the handle Z, and opens or shuts the apertures at which the steam passes out of the great boiler into the steam-pipes OO; N a small gauge cock at the top of a pipe which goes half way down into the great boiler; 01, 02, steam-pipes, one end of each screwed to the regulator, the other ends to the receivers PP, to convey the steam from the great boiler into those receivers ; P1, P2, copper vessels called receivers, which are to receive the water which is to be raised; Q screw joints, by which the branches of the water-pipes are connected with the lower parts of the receivers; R 1, 2, 3, and 4, valves or clacks of brass in the water-pipes, two above the branches Q, and two below them; they allow the water to pass upwards through the pipes, but prevent its descent; there are screw plugs to take out on occasion, to get at the valves R; S the forcing-pipe which conveys the water upwards to its place of delivery, when it is forced out from the receivers by the impellent steam; T the sucking-pipe, which conveys the water up from the bottom of the pit, to fill the receivers by suction; V a square frame of wood, or a box, with holes round its bottom in the water, to enclose the lower end of the sucking-pipe, to keep away dirt and obstructions; X a cistern with a buoy-cock coming from the force-pipe, so as it shall always be kept filled with cold water; YY a cock and pipe coming from the bottom of the said cistern, with a spout to let the cold water run down on the outside of either of the receivers P P; Z the handle of the regulator, to move it by, either open or shut, so as to let the steam out of the great boiler into either of the receivers.

The Manner of working the Engine.

"The first thing is to fix the two boilers of the engine in a good double furnace, so contrived that the flame of the fire may circulate round, and encompass the boilers to the best advantage, as you do coppers for brewing. Before you make any fire, unscrew the two small gauge-pipes, and cocks, G and N, belonging to the two boilers, and at the holes, fill the great boiler L two-thirds full of water, and the small boiler D quite full; then screw in the said pipes again as fast and tight as possible, and light the fire under the large boiler at B 1, to make the water therein boil, and the steam of it being quite confined must become wonderfully compressed, and therefore will, on the opening of a way for it to issue out (which is done by pushing the handle Z of the regulator as far as it will go from you), rush with a great force through the steam pipe O 1, into the receiver P 1, driving out all the air before it, and forcing it up through the clack R 1, into the forcepipe, as you will perceive by the noise and rattling of that clack; and, when all the air is thus driven out, the receiver P1 will be very much heated by the steam. When you find it is thoroughly emptied, and is grown very hot, as you may both see and feel. then pull the handle Z of the regulator towards you, by which means you will stop the steam pipe O 1, so that no more steam can come into the receiver P 1, but you will open a way for it to pass through the other steam pipe O 2, and by that means fill the other receiver P 2 with the hot steam, until that vessel has discharged its air through the clack R 2 up the force pipe, as the other vessel did before.

"While this is doing, let some cold water be poured on the first-mentioned receiver P 1, from the spout Y, by which means the steam in it being cooled and condensed, and contracted into a very little room, a vacuum or emptiness is created, and consequently the steam pressing but very little (if at all) on the clack R 3 at the bottom of the receiver P 1, there is nothing there to counterbalance the pressure of the atmosphere on the surface of the water at the lower part V of the sucking-pipe T, wherefore the water will be pressed np, and ascend into and fill the receiver P 1, by what is commonly called suction: the water as it rises lifts up the clack or valve R 3, which afterwards falling down again and shutting close, hinders the descent of the water that way.

"The receiver P 2 being by this time emptied of its air, push the handle of the regulator from you again, and the force of the steam coming from the great boiler will be again admitted through Q 1, and will act upon the surface of the water contained in the receiver P1; which surface only being heated by the steam, it does not condense it, but the steam gravitates or presses with an elastic quality like air, and still increasing its elasticity or spring until it counterpoises, or rather exceeds the weight of the column of water in the receiver and pipe S, which it will then necessarily drive up through the passage Q R 1 into the force-pipe S. The steam takes up some time to recover its power, but it will at last discharge the water out at the top of the force-pipe S, as it is represented in fig. 3. After the same maner, though alternately, the receiver P2 is filled with water by means of the suction, and then emptied by the impellent force of the steam, whereby a regular stream is kept continually running out at top of the force-pipe S, and so the water is raised very easily from the bottom of the mine, &c. to the place where it is designed to be discharged. I should add, that after the engine begins to work, and the water is risen into and hath filled the force-pipe S, then it also fills the little cistern X, and by that means supplies the spout or pipe Y Y, which I call the condensing pipe, and which by its handle can be turned sideways over either of the

receivers, and is then open; by this spout cold water is conveyed down from the force-pipe to fall upon the outside of either of the receivers when thoroughly heated by the steam, in order to cool and condense the steam within, and make it suck (as it is usually called) the water out of the well up into that receiver.

"It is easy for any one, that never saw the engine, after half an hour's experience, to keep a constant stream; for on the outside of the receiver you may see how the water goes out as well as if the receiver were transparent for as far as the steam continues within the receiver, so far is that vessel dry without, and so very hot as scarce to endure the least touch of the hand; but as far as the water is withinside of the said vessel, it will be cold and wet on the outside where any water has fallen on it; which cold and moisture vanish as fast as the steam in its descent takes place of the water. But if you force all the water out of the receiver, the steam, or a small part thereof, will go through the clack R 1 or R 2, and will rattle that clack so as to give notice to move the handle of the regulator, and then the steam begins to force out the water from the other receiver P, without the least alteration of the stream, only sometimes the stream will be rather stronger than before, if you pull the handle before any considerable quantity of steam be got up the clack R: but it is much better to let none of that steam go off, for that is but losing so much strength, and it is easily prevented by pulling the regulator some little time before that receiver which is forcing is quite emptied.

"This being done, turn the cock, or condensing pipe Y of the cistern X, over the empty receiver, so that the cold water proceeding from X may run down through Y, which is never opened but when turned over one of the receivers, but when it stands between them is tight and stanch. This cold water falling on the outside of the receiver, by its coolness causes that steam which had such great force just before, to condense and become an empty space, so that the receiver is immediately refilled by the external pressure of the atmosphere, or what is vulgarly called suction, whilst the other receiver is emptying by the impellent force of the steam, which being done, you push the handle of the regulator from you, and thus throw the force into the other receiver, pulling the condensing pipe over the receiver P 2, causing the steam in that vessel to condense, so that it fills while the other empties-the labour of turning these two parts of the engine, viz. the regulator and condensing water-cock, and tending the fire, being no more than what a boy's strength can perform for a day together, and is as easily learned as their driving of a horse in a tubgin. Yet after all, I would have men employed in working of the engine, and those too the most apprehensive, supposing them more careful than boys: the difference of this charge is not to be mentioned when we consider the vast profit which those who use this engine will reap by it.

"The ingenuous reader will here probably object, that the steam being the cause of this motion and force, and that as steam is but water rarefied, the boiler L must in some certain time be emptied, so as the work of the engine must stop to replenish the boiler, or endanger the burning out or melting the bottom of the boiler. To answer which, please to observe the use of the small boiler D; it is supplied with water from the force-pipe by a small pipe and cock E; when it is thought fit by the person tending the engine to replenish the great boiler (which requires about an hour and a half or two hours' time to the sinking one foot of water), he turns the cock E, so that there can be no communication between the force pipe S, and the small boiler D, and putting in a little fire under the small boiler at B 2 the water will there grow presently hot, and when it boils, its own steam, which hath no vent out, will gain more strength than the steam in the great boiler. For the force of the great boiler being perpetually spending and going out, and the other confined and increasing, it is not long before the force in the small boiler exceeds that in the great one; so that the water in the small boiler being depressed by its own steam pressing on its surface, will force the water up the pipe H, through K, into the great boiler L ; and so long will it run till the surface of the water in the small boiler D gets to be as low as the bottom of the pipe H, and then the steam and water will run together, and by its noise and rattling of the clack I, will give sufficient assurance to him that works the engine that the small boiler hath emptied and discharged itself into the greater one L, and carried in as much water as is then necessary; after which, by turning the cock E again, you may let fresh cold water out of the force-pipe S into the lesser boiler D, as before, and thus there will be a constant motion and a continual supply of the engine, without fear of decay or disorder. And inasmuch as from the top of the small boiler D to the bottom of its pipe H (which is within eight inches of the bottom of the boiler) there is contained about as much water as will replenish the great boiler L one foot, so you may be certain it is replenished one foot of course.

46

Also, to know when the great boiler wants replenishing or not, you need only turn the gauge-cock N, and if water come out there is no need to replenish it, but if steam only come, you may conclude there is want of water; and the like will the cock G do in reference to the small boiler D, showing when it is necessary to supply that with fresh water from S, so that in working the engine there is very little skill or labour required; it is only to be injured by either a stupid or wilful neglect.

"And if a master is suspicious of the design of a servant to do mischief, it is easily discovered by those gauge-pipes; for if he come when the engine is at work, and find the surface of the water in the great boiler L below the bottom of the gauge-pipe N, or the water in the small boiler D below the

bottom of the gauge-pipe G, such a servant deserves correction, though three hours after that, the working on, would not damage or exhaust the boilers. In a word, the clacks being, in all water-works, always found the better the longer they are used, and all the moving parts of our engine being of like nature, the furnace being made of Sturbridge or Windsor brick or firestone, I do not see it possible for the engine to decay in many years; for the clacks, buckets, and mitre-pipes, regulator and cocks, are all brass; and the vessels made of the best hammered copper, of sufficient thickness to sustain the force of the working of the engine. In short, the engine is so naturally adapted to perform what is required, that even those of the most ordinary and meanest capacity may work it for some years without any injury."

A good many of these engines appear to have been constructed and set to work. They were not employed, however, in any case in which water had to be lifted from a great depth, as the great pressure of steam requisite to overcome the gravity of a high column of water was reckoned inconvenient and dangerous in those days of fragile boilers and imperfect workmanship. These disadvantages appear to have been the chief cause of the relinquishment of Savery's engine in favour of that of Newcomen; for in the early career of Newcomen's contrivance the useful effect under certain circumstances was inferior to that of the engine it superseded, and the first expense was greater: but then Newcomen's plan made the column of water that could be lifted independent of the pressure of the steam, an increase in the height of the column only requiring an increase in the diameter of the cylinder. The details of Savery's contrivance are 'extremely judicious, and the scheme altogether speaks very favourably of his ingenuity and perseverance. It has been a matter of controversy whether he invented the engine altogether himself, or borrowed the idea from the Marquis of Worcester, and merely matured the plan of that noble mechanic. Desaguliers retails an idle story of Savery buying up the Marquis of Worcester's book in Paternoster Row and destroying it, so that the priority of the Marquis might remain unknown; but we give very little credit to the tale. There were probably at the time some of the Marquis's engines in being, which would be a much more conclusive testimony than any book could be, and at least there must have been many persons then living who recollected the engine set up by the Marquis at Vauxhall. Besides, the very act of buying up a book which had been many years in circulation, or rather of attempting to buy it up, would be the surest way to attract attention towards it, and thus produce the very disclosure it was Savery's imputed object to avert. We think it likely enough that Savery may have invented this engine entirely himself, without knowing anything of what the Marquis of Worcester had done forty years before, although in the pursuit of the subject he must of course have become acquainted with the achievements of his predecessor. The same wants generate so naturally the same expedients for their relief, that simultaneous discoveries and inventions become inevitable, and identical projects start up at different epochs without imitation, under the force of similar circumstances. The ingenuity, too, displayed by Savery in the details of his engine encourages the idea that it may have been altogether of his own contrivance, for he does not appear to have been unequal to such a performance, which indeed it did not require any very brilliant genius to accomplish. But whatever conclusion may be come to upon this head, we are at least certain that it was Savery who first introduced the fire-engine into extended use; and he prosecuted his undertaking with great assiduity and success; though, as he says, he was obliged to encounter the oddest and almost insuperable difficulties," and at a great expenditure of energy and means. The merit of this achievement is in our apprehension much greater than that of the mere invention of the machine; for such a work is one of equal difficulty, and of infinite toil and discouragement, and is little lightened by the bright dreams of fancy or the consolations of applause.

66

FROM THE APPLICATION OF THE PRINCIPLE OF ATMOSPHERIC PRESSURE TO THE INTRODUCTION OF THE CYLINDER.

About seven years after the discovery of the pressure of the atmosphere by Torricelli, Otto Guericke, a magistrate of Magdeburgh, without any knowledge of what Torricelli had done, succeeded in obtaining a vacuum in a cask by means of an air pump, and contrived a great variety of ingenious pneumatic apparatus; some for raising water in a pipe by means of an exhausted receiver screwed on the top, and others for lifting a great weight by means of a cylinder fitted with a piston, beneath which was a vacuum, while the atmosphere pressed on its superior surface. The nature of the arrangement he adopted in this latter case will be at once apprehended by a reference to the annexed figure, where A is a cylinder fitted with a piston, to the shank of which a rope is attached, passing over the pullies B and C with a scale D hung at its extremity loaded heavily with weights. A small pump was screwed into the lower part of the cylinder, by means of which the cylinder was exhausted, and the piston descending raised the weights.

We have in these ingenious devices the same mechanical arrangements as those afterwards employed by Savery and Papin in their projects; and indeed the only material difference between the cylinder apparatus here represented and the atmospheric steam engine is that in the one case the vacuum is produced by means of a pump, and in the other by the condensation of steam. It is in the production of a vacuum, therefore, by the

agency of steam, that any merit in the contrivance of the atmospheric steam engine must be held to consist, for the efficacy of a vacuum when produced in raising heavy weights, or in the production of other powerful mechanical effects, could not fail after Guericke's experiments to be obvious to every one. Savery says he discovered the efficacy of steam in the production of a vacuum by plunging the neck of a flask he had accidentally thrown upon the fire with a little wine in it into a basin of water, when the steam raised from the wine was condensed, and the water rose into the flask and filled; but Desaguliers throws discredit upon this account, by saying that in such an experiment the flask would have been forced from the hand of the operator by the impulse of the entrant water, a circumstance of which Savery makes no mention. Mr. Scott Russell, however, says that he has tried the experiment many times, and that the catastrophe indicated by Desaguliers is conditional upon the diameter of the neck of the flask, the temperature of the water, and other circumstances, so that we really see very little reason to doubt the truth of Savery's statement. The date of this experiment of Savery's does not very clearly appear. It must of course have been antecedent to his patent of 1698; but whether it was before the year 1690, at which time Papin published a scheme for creating a vacuum in a cylinder by the condensation of steam, is uncertain. The question, however, is one of very little moment, for both of these projectors were long preceded in their discoveries; the method of producing a vacuum by steam agency having been known even before any idea had been formed of the atmospheric pressure. The manner of re-filling the Æolipile was to plunge it, while a jet of steam was issuing from it, into a vessel of water, whereby the steam within it was condensed, and the water was drawn with great rapidity through the small orifices through which the steam escaped, by means of the internal vacuum. In the Marquis of Worcester's engine, again, as, we believe, we have already indicated, a vacuum must have been formed by the condensation of the steam so soon as the cold water began to enter the receivers, though from what Cosmo de Medici tells us, we should be disposed to conclude that the power thus generated was wasted in drawing the water through contracted passages, rather than in raising it up to any considerable elevation. We believe, however, that both Savery and Papin re-discovered the means of producing a vacuum by the condensation of steam, or more properly perhaps saw, independently of their predecessors, and of each other, the feasibility of producing a convenient and effectual vacuum by this agency, for it is hardly conceivable that in the times of Newton, Hooke, and Boyle, the effect of the condensation of steam in producing a vacuum should not have been a matter of extended notoriety. In this, indeed, as in most other cases, a reference to the current intelligence of the time will show that even the most noted projectors advanced very little beyond it; and many a genius, with whose name the world has rung, will, when measured by this standard, collapse to the dimensions of a pigmy. Inventions sown in one age are ripened in another by the progress of events, but a multitude of agencies are necessary to the final effect, and of these the most important is Time. Individual projectors are merely like bubbles dancing upon the waves of that majestic stream: though higher than the other liquid particles around them, yet their altitude is not to be measured from the position of the lowest surface, and they are not the cause of the total exaltation.

As we have several times mentioned the name of Papin, we must, we suppose, say who he was. Denys Papin, then, was born at Blois, in France, and was educated to the profession of medicine. Being a Protestant, he was driven from France by the revocation of the edict of Nantes, and settled in London, where he assisted Mr. Boyle in his experiments with the air-pump, and was elected a member of the Royal Society in 1680. During this period he invented the culinary utensil called Papin's digester, of which he published an account in 1682. In 1687 he was appointed professor of mathematics at Marburg, in Germany, and in 1690 he proposed a scheme in the Acta Eruditorum of Leipsic, for producing the vacuum required in Otto Guericke's cylinder arrangement, by the agency of steam. A A (fig. 3.) is a thin metal cylinder, fitted with a piston B, and a pistonrod H with a notch in its side, into which a latch, E, falls to prevent the descent of the piston until the latch is withdrawn. A rope, L, is fastened to the piston-rod, and passes over pullies T T, precisely as in Guericke's contrivance, and may be employed to raise weights in the descent of the piston. Water is introduced underneath the piston through the hole closed by the screwed rod M, and a fire is applied beneath the cylinder bottom, which raises the water into steam, and forces the piston up, the latch falling into the notch in the piston-rod so soon as the piston has

3

T

T

M

H

A

risen sufficiently. The fire is now to be withdrawn from the cylinder bottom, and in a certain time the steam in the cylinder will become condensed, when the latch is to be drawn backwards, and the piston will be forced down by the pressure of the atmosphere. It requires little penetration to see that this scheme would be quite useless in practice; for the continual removals of the fire, the injurious effect of fire on the cylinder, and the absence of any expedient to facilitate refrigeration, are most effectual barriers to its successful application. Indeed, the project is greatly inferior in efficacy to the plan of the Marquis of Worcester, which was provided with a boiler, and which proved itself susceptible of a practical use. With Savery's engine, projected about the same time, the crude project of Papin is not to be put in comparison; and indeed Papin appears to have been so thoroughly convinced of the superiority of Savery's contrivance, that he gave up his own plan and adopted that of Savery. Papin, however, was a person of much ingenuity, but most of his schemes were visionary; and whatever might be his other merits, he certainly contributed nothing to the improvement of the steam engine, either by his performances or suggestions. He neither was the first to suggest the formation of a vacuum in a cylinder fitted with a piston, nor the first to discover the method of producing a vacuum by the agency of steam, and he certainly was not the person who first carried this combination to any result of moment or utility. The first successful engine which operated by means of a cylinder and piston was contrived by Thomas Newcomen, an ironmonger of Dartmouth, and John Cawley, a glazier of the same place, about the year 1710. The object of the innovation appears to have been not so much to obtain a more effectual and economical engine than that of Savery, as to dispense with the necessity of employing steam of a very high elasticity where water had to be raised to a considerable height, and which, in those days, was the occasion of considerable inconvenience and danger. We think it very probable that Newcomen was acquainted with the project of Papin at the time he introduced his own engine into practice, as well as his relinquishment of it in favour of Savery's scheme, but this rather adds to Newcomen's merit than diminishes it, as it shows him to be no imitator of Papin, and to be in no wise influenced by his authority. It is to Newcomen, therefore, that we owe the cylinder steam engine, for Papin not only never projected any cylinder steam engine that was practicable, and never made one, but had gone back before Newcomen's time to Savery's scheme; and although Sir Samuel Morland, so far back as 1682, had contrived cylinders for raising water by steam power, yet we have no reason to believe that he put his plan in practice; and the outline he has left of it in his "Principles of the new Force of Fire" is too meagre to give any certain intimation of the nature of the contrivance, though the probability is that it more nearly resembled the Marquis of Worcester's engine than that of Newcomen. It does not appear, however, that Newcomen was himself conscious of the advantages of his own method of construction beyond that of the greater convenience of low pressure steam, and, indeed, his engine would have remained very little better than that of Savery, at least for moderate heights, but for the improvements effected by subsequent inventors. Of these improvements we must now give some account.

FROM THE APPLICATION OF THE CYLINDER TO THE DISCOVERY OF THE MODE OF CONDENSING BY JET, THE USE OF SELF-ACTING VALVES, AND THE COMPLETION OF THE IMPROVEMENTS OF SMEATON.

Newcomen's engine, as at first constructed by him, was provided with a casing around the cylinder for holding cold water, to accelerate the condensation of the steam, and the surface of the piston was covered with water to prevent the passage of air past it. The steam was admitted from the boiler by a slide valve or regulator, as in Savery's engine, which was moved by hand, and after the cylinder was filled with steam, water was introduced into the external casing, by which, in process of time, the steam was condensed, and the piston then made a stroke. This, however, it is evident, must have been very tedious work, and the engine indeed was very untractable until the method of condensing the steam by a jet of cold water was introduced, and the movement of the valves was effected by the engine. These improvements were the effect of accident: the mode of their development is thus related by Desaguliers :-"In the latter part of the year 1711 Newcomen and Cawley made proposals to drain the water of a colliery at Griff, in Warwickshire, where the proprietors employed 500 horses, at an expense of 9001. a year; but their invention not meeting with the reception they expected, in March following, through the acquaintance of Mr. Potter of Bromsgrove, in Worcestershire, they bargained to draw water for Mr. Back of Wolverhampton; where, after a great many laborious attempts, they did make the engine work: but not being either phi

Losophers to understand the reason, or mathematicians enough to calculate the powers and proportions of the parts, they very luckily, by accident, found what they sought for.

66

They were at a loss about the pumps, but being so near Birmingham, and having the assistance of so many admirable and ingenious workmen, they came about 1712, to the method of making the pump-valves, clacks, and buckets, whereas they had but an imperfect notion of them before. One thing very remarkable: as they at first were working, they were surprised to see the engine go several strokes, and very quick together, when, after a search, they found a hole in the piston, which let the cold water in to condense the steam in the inside of the cylinder, whereas before they had always done it on the outside. They used before to work with a buoy to the cylinder, enclosed in a pipe, which buoy rose when the steam was strong and opened the injection, and made a stroke; thereby they were capable of only giving six, eight, or ten strokes in a minute, till a boy, named Humphrey Potter, in 1713, who attended the engine, added (what he called scoygan) a catch, that the beam always opened, and then it would go 15 or 16 strokes a minute. But this being perplexed with catches and strings, Mr. Henry Beighton, in an engine he had built at Newcastle-uponTyne, in 1718, took them all away but the beam itself, and supplied them in a much better manner.'

The boilers of the engines made by Savery do not appear to have been provided with safety valves, nor do they appear to have required such an application, as the regulators were in all probability so constructed that the passages of both receivers could not be closed at the same time. About the year 1717, Desaguliers constructed an engine on Savery's plan, but with only one receiver, and, as the only passage through which the steam could find vent would sometimes in such an engine be closed, and might be so left by a careless attendant, it became necessary to apply a safety-valve to the boiler. This species of valve had been applied by Papin to his digester, and its application to the steam-engine had been suggested by him; but the expedient was superfluous so long as the pressure of the steam was balanced by a column of water, which would be overcome before any dangerous elasticity, could be attained.

4

In 1720 a very elegant high pressure engine was contrived by Leupold, and described in his Theatrum Machinarum; the plan of which will at once be understood by reference to the annexed diagram, where A is the boiler, E and C the cylinders, F the waste steam proceeding from the four-way cock. In this engine, the four-way cock first occurs, and the scheme altogether is marked by a refined ingenuity not unworthy of Watt and his successors. The effect of the two cylinders placed in juxtaposition is identical with that of the double-acting engine; and the machine is altogether the simplest and most artistic that, up to this time, had been projected. The employment, however, of high pressure steam could not fail to be looked upon as very objectionable at a period when boilers were so rude and unsubstantial; and we suppose it was owing to this objection that the plan of working by high pressure steam did not, until very recently, come into practice. The domes of the boilers employed by Newcomen were generally of lead, the inferior portions being composed of copper; and it was not until many years after steam engines were applied to practical uses, that boilers composed altogether of iron plates rivetted together, were introduced, or that the art of making boilers had reached such perfection, that a high pressure might be put upon them with safety.

From the improvement of the valve gearing by Beighton, in 1718, very little alteration appears to have been made in the atmospheric engine until the time of Smeaton. Brindley, indeed, added the contrivance of a float in the boiler, to regulate the admission of the feed-water, and also proposed that the boiler should be constructed of wood or stone, furnished with an internal furnace and internal flues: but his attention was withdrawn from the steam-engine to other subjects, and very little real improvement resulted from his ingenuity. In 1767, Smeaton was employed to construct an engine by the New River Company, into which he resolved to introduce several alterations. Conceiving that a good deal of power was dissipated in the rapid reciprocation of the beam, and in giving a needless Inomentum to the water in the pumps, he determined to work the engine at a less velocity, to employ larger pumps, and to load the piston with 10 lbs upon the inch instead of 7 lbs., as had been the usual practice. To carry into effect these intentions, the main centre was removed from the middle of the beam to a point nearer the end from which the pumps were hung, and the cylinder was made tall and of small diameter, by which peculiarity it was conceived the steam would be more readily condensed, and a less amount of injection water would suffice. "I thought myself," says Smeaton, "quite secure under those advantages; but how great was my surprise and mortification, to find that, instead of requiring less injection water than common, although the injection pump was calculated to afford as much injection water as usual, in proportion to the area of the cylinder, with a sufficient overplus to answer all imaginable wants, it was unable to support the engine with injection, and that two men were obliged to assist to raise the injection water quicker, by hand, to keep the engine in motion; at the same time that the cylinder was so cold, I could keep my hand upon any part

of it, and bear it for a length of time in the hot well. By good fortune, the engine performed the work it was appointed to do, as to the raising of water, but the coals by no means answered my calculation. The injection pump being enlarged, the engine was in a state for doing business, and I tried many smaller experiments, but without any good effect, till I altered the fulcrum of the beam so much, as reduced the load upon the piston from 10 lbs. to 8 lbs. per inch. Under this load, though it shortened the stroke at the pump end, the engine went so much quicker as not only to raise more water, but consume less coals; took less injection water; the cylinder became hot, and the injection water came out at 180° of Fahrenheit; and the engine in every respect not only did its work better, but went more pleasantly. This at once convinced me that a considerable degree of condensation of the steam took place in entering the cylinder, and that I had lost more by this way, by the coldness of the cylinder, than I had gained by the increase of load. In short, this single alteration seemed to have unfettered the engine; but in what degree this condensation took place, under different circumstances of heat, and where to strike the medium, so as upon the whole to do best, was still unknown to me. But resolving, if possible, to make myself master of the subject, I immediately began to build a small fire engine at home, that I could easily convert into different shapes for experiments, and which engine was set to work in the winter of 1769."

This engine was set up at Austhorpe, and by its assistance Smeaton investigated the operation of the steam-engine after the same practical fashion that he had previously adopted in investigating the operation of water-wheels and windmills. The diameter of its cylinder was 9.9 inches, length of stroke 3 feet, number of strokes per minute 17, and the load upon the piston 7.89 lbs. per square inch. This is not quite one-horse power, for (9-9)=77 square inches × 7·89=607 lbs. and × by 52.5, the effective motion per minute=31,867 lbs. raised one foot high per minute, or+by 33,000= 966 HP. The consumption of coals per hour was 55 lbs.: a bushel, therefore, or 84 lbs. would suffice for 91.6 minutes, which is equivalent to 2.92 millions of pounds raised one foot high by a bushel of coals. The quantity of injection water required was found by measurement to be 95 cubic inches per stroke, and the quantity of water evaporated from the boiler was about 89 cubic inches per stroke, so that the quantity of water required for injection was 10.66 times greater than that requisite for the supply of the boiler, though this proportion would, of course, vary somewhat with the temperature of the atmosphere. This engine must have lost about half of its efficacy by condensation in the cylinder, for 8.9 cubic inches of water will make 8.9 cubic feet of steam to be consumed per stroke, which is about double the capacity of the cylinder and its waste spaces. A good deal of the heat thus lost, however, appears to be due to radiation from the surface of the cylinder, for not above two thirds of the whole heat is to be found in the hot well, the temperature of which was only 134° when the temperature of the injection water was 68° or 70°. The load on the piston was varied in the experiments by taking away or adding a length of pipe to the superior part of the pump barrel, and it was found that the load of 7.81 lbs. per square inch on the piston was the most beneficial one, both as regards the quantity of work done and the consumption of fuel. When the load was diminished to 6.6 lbs. the efficacy, or, in other words, the duty of the engine, estimated at 100, was reduced to 94, and, by diminishing the load to 55 lbs. the efficiency was reduced as low as 82, while by increasing the load from 7.81 to 8.8 lbs. the work done was only increased from 100 to 107, although the consumption of fuel proceeded in a much higher ratio, and by a further increase of the load to 9.1 lbs. the efficiency was reduced from 100 to 96, though at an increased expenditure of fuel. In diminishing the load below 7.81 lbs. to 6.6 lbs. the efficacy of a pound of coal was reduced from 100 to 94, and the further diminution to 5.5 lbs. reduced the efficacy of a pound of coal from 94 to 80, while the increments in the load reduced the efficacy of the fuel from 100 to 97 in the first case, and from 97 to 93 in the second. The relative efficacy of different kinds of coal was also tested in these experiments. Rubbly coal was found superior to slack in the proportion of 100 to 80. Coke was of about the same efficacy as slack, and 100 lbs. of coal were found to produce about 66 lbs. of coke. The Newcastle coal called Team Top was found more effectual than the Halton or common Yorkshire coal in the proportion of 120 to 100, and Cannel coal more effectual than Halton in the proportion of 133 to 100. Smeaton continued his experiments with this engine for four years, and recorded the results in a book of tables which he employed to regulate his future practice. In 1772 he was employed to construct an engine for Long Benton colliery near Newcastle, in which he embodied the several improvements his experiments had suggested. In all its main features this engine was identical with the previous constructions of Beighton, but it was distinguished by juster proportions and greater nicety of detail than had yet been realised, and the innovations thus introduced were found to be highly beneficial in practice. The engine at Long Benton Mr. Smeaton looked upon as his standard work, but his most magnificent performance in this way was the Chase Water engine erected in 1775 at one of the mines of Cornwall. Of these engines we must contrive to give a brief description. The diameter of cylinder in the Long Benton engine was 52 inches; length of stroke 7 feet; load on the piston 7 lbs. per square inch; number of strokes per minute 12; and 522 x 7854 x 7 x 84÷33,000 = about 40 horsepower. The consumption of fuel was 8 bushels of Newcastle coals per hour, which at 84 lbs. per bushel is 17.63 lbs. per horse-power per hour, or 9:45 mil

lions of pounds, raised one foot high by a bushel of coal. The evaporation of water by the boilers was 90 cubic feet per hour, by 8 bushels, 10:58 cubic feet, evaporated by a bushel or 84 lbs., or 7.88 lbs. evaporated by each pound of coal. If the steam within the boiler be considered to be 1700 times less dense than water, which would be near the truth, then 2550 cubic feet of steam would be generated per minute; but of this not so much as one-half could be spent in producing a useful effect, for the space within the cylinder occupied by the motion of the piston would only be 103.25 cubic feet × 12 1239 cubic feet per minute, which is not half of the steam generated. The residue, therefore, must have been lost by leakage and condensation, and by the space left between the piston and the bottom of the cylinder. This space would cause a loss of about 360 cubic feet per minute, leaving a loss of above 950 cubic feet per minute, to be accounted for by leakage and condensation, or indeed by condensation alone, as the leakage was inappreciable. The internal surface was 168 square feet, so that each foot of cylinder surface must have condensed about 5-66 cubic feet of steam per minute. There were three boilers, and they were so proportioned as to afford 3.5 square feet of furnace surface, 7·83 of flue surface, and 867 feet of fire-grate per horse-power.

The structure and arrangement of the Chase Water engine were almost identical with those of the engine at Long Benton: it was, however, of greatly superior power. The cylinder of the Chase Water engine was 72 inches in diameter; length of stroke 9 feet; number of strokes per minute 9-81 feet of effective motion, and load upon the piston 73 lbs. This gives a power of about 76 horses for 722 x 7854 x 81 x 7}÷33,000=76} HP. The configuration of these engines will be seen at once by a reference to the following wood-cut, where A B is the cylinder, C the piston made of iron and covered on the bottom with elm planking, secured with screwbolts to the piston. Between this planking and the iron of the piston, tarred flannel is interposed to retard the transmission of heat, and thereby

render the condensation within the cylinder less serious. D is the great lever, 27 feet 4 inches long, consisting of twenty pieces of fir, the four nearest the middle being 12 inches square, and the other sixteen being 6 inches by 12 inches. These pieces of timber are prevented from rubbing upon one another by means of oak keys, driven into mortices prepared for their reception, and are bolted together by 32 long iron bolts, 1 inch in diameter, passing through the whole depth of the lever. The axis of the lever, where it passes through the wood, is 30 inches wide by 5 inches thick. The bearings are 8 inches in diameter, resting in brass bushes, let into blocks of wood built into the lever wall. Breadth between bearings 3 feet; breadth of beam 2 feet; depth of beam at centre 74 inches; depth at extremities 60 inches. E is the lever wall; F F the beams which support the cylinder, or rather which keep it down, and which are retained in their places by the walls of the house; G is the boiler, H the furnace, I the steam pipe, J the injection pipe, terminating in a wooden jet piece jammed

between the steam pipe and the side of the cylinder, the use of wood in this situation being to diminish the condensation by a non-conducting substance. K is the cistern for holding the injection water, and which is placed in an elevated situation to cause the more perfect dispersion of the injection water by the force with which it enters the cylinder. L is a pump for raising the injection water into the cistern. M is the eduction pipe, terminating in the hot well, and fitted with a valve at its inferior extremity, to prevent the regurgitation of the water. N pump rod of the pump of the mine; O a band or strap extending to the cataract; PP the plug rod, by which the valve gearing is wrought. The steam pipe is 12 inches in diameter, the eduction pipe 8 inches; and the two short necks for the injection cock and snifting valve 6 inches diameter. These are all cast in the same piece with the hemispherical bottom. The snifting valve consists of a short plug of boxwood driven tight into its seat, fitted with a circular flap valve applied to the end of the plug; over this a conical pipe is fixed with a cock at its extremity, so as to regulate the amount of the snift at pleasure. "The upright feed pipe, Q," says Smeaton, "is designed to answer the double purpose of a lower gauge cock and a safety pipe. Two or three small holes are to be pierced through it at the proper depth for a lower gange cock, that is, at the lowest level which is proper for the water in the boiler, and three inches below these holes the pipe is to be pierced with a hole half an inch in diameter, and again at three inches lower the pipe must terminate altogether. By this means, if the water in the boiler sinks too low, the small holes in the feeding pipe will emit steam and give notice by a rattling noise; but if the water is suffered to subside still lower, the steam will blow out by the half inch hole, and the noise will produce a greater alarm; and when the water sinks so low as to be at intervals below the bottom of the pipe, the water and steam will issue in such a manner as to make a very great noise, and call the engine-man to his duty even if he have fallen asleep."

Such then are the principal details of this celebrated engine; and we must now proceed to explain in a few words the manner in which the atmospheric engine operates. When the engine is at rest, the end of the lever from which the pump is hung will always proponderate, that end being purposely made heavier than the other, to overcome the friction and inertia of the lever, as it cannot be elevated by the force of the steam. The first step in starting the engine is to open the regulator or slide valve which interposes between the boiler and cylinder. The cylinder will of course immediately become charged with steam, but the steam will at first be very rapidly condensed by the cold metal, and will continue to condense until the cylinder has become nearly as hot as the steam itself. As soon as this takes place, the steam will open the snifting valve, and drive out the air with which the cylinder had been filled, and the sign of the thorough expulsion of the air is that the steam will then issue in a transparent, and not in a cloudy state. While the steam is thus issuing from the snifting valve the injection valve situated in the injection cistern at the top of the enginehouse must be drawn open by the cord attached to it, and the handle which had been depressed to open the regulator must be raised. This will close the connection between the boiler and the cylinder, and will at the same time open the injection cock, and an instant vacuum being produced by the jet of cold water, the engine will make a stroke. The plug-rod in its descent strikes with a projection or tappet the handle which had just been raised, which shuts the injection cock and opens the regulator, and as the pressure of the steam is about equal to the pressure of the atmosphere, the piston ascends, being drawn up by the preponderance of weight at the other end of the lever. The plug-rod in its ascent, when near the end of the stroke, strikes the handle upwards by means of another peg or tappet, by which the steam passage is again closed, and the injection passage opened. The piston thus makes another stroke, and the same action is continued so long as the engine is kept supplied with steam and cold water. The hot water produced by the condensation of the steam flows out of the cylinder by the eduction pipe into the hot well, whence a part of it flows into the boiler to restore the quantity evaporated, and the remainder runs to waste. A small branch leading from the injection pipe distributes water over the upper surface of the piston to prevent the leakage of air, and any air that may find admission either in the steam or otherwise, and which, if suffered to accumulate in the cylinder, would destroy the action of the engine, is driven out every stroke by the snifting valve. A certain leakage of air, however is beneficial, as it diminishes greatly the condensation of the steam; for the air, being diffused equably among the steam, will accumulate against the sides of the cylinder as against a sieve, which is pervious to heat but not to air; and this film of air will prevent the access of the steam to the naked iron to suffer a destructive refrigeration. After a better quality of workmanship had been introduced by Smeaton, it was found that the engines were actually too tight; and an air-cock was therefore appended to give a regular admission to such a quantity of air as might be found beneficial.

Before Smeaton commenced his improvements in the steam engine, he ascertained with much care the performance of a number of the engines then in existence; and of 15 engines at Newcastle he found that the best raised about 7 millions of pounds one foot high with a bushel of coals, and the worst only 3 millions. The engine he afterwards erected at Long Benton, and in which his improvements were fully carried out, raised, as we believe we have already stated, about 94 millions; so that there was much less difference between the performance of the best of the common engines and Smeaton's engines than between the common engines them