purposes, and certain new pieces of mechanism applicable to the same.'

99. The specification, dated 3rd of July, 1782, contains, first the expansive steam engine, with six different contrivances for equalising the power; second, the double power steam engine, in which the steam is alternately applied to press on each side of the piston, while a vacuum is formed on the other; third, a new compound engine, or method of connecting together the cylinders and condensers of two or more distinct engines, so as to make the steam which has been employed to press on the piston of the first, act expansively upon the piston of the second, &c., and thus derive an additional power to act either alternately or conjointly with that of the first cylinder; fourth, the application of toothed racks and sectors to the ends of the piston or pump-rods, and to the arches of the workingbeams, instead of chains; fifth, a new reciprocating semi-rotative engine, and a new rotative engine or steam-wheel.

100. Patent, 28th of April, 1784.- For certain new improvements upon fire and steamengines, and upon machines worked or moved

by the same.'

101. The specification, dated 24th of August, 1784, describes, first, a new rotative engine, in which the steam vessel turns upon a pivot, and is placed in a dense fluid, the resistance of which to the action of the steam causes the rotative motion; second, methods of causing the piston-rods, pump-rods, and other parts of engines, to move in perpendicular or other straight lines, and to enable the engine to act upon the working-beams both in pushing and pulling; this is now called the parallel motion, and three varieties are described; third, improved methods of applying the steam engine to work pumps, or other alternating machinery, by making the rods balance each other; fourth, a new method of applying the power of steam engines to move mills which have many wheels required to move round in concert; fifth, a simplified method of applying the power of steam engines to the working of heavy hammers or stampers; sixth, a new construction and mode of opening the valves, and an improved working gear; seventh, a portable steam engine and machinery for moving wheel-carriages.

102. Patent, in 1785. For certain newly improved methods of constructing furnaces or fire-places for heating, boiling, or evaporating of water and other liquids which are applicable to steam engines and other purposes; and also for heating, melting, and smelting of metals and their ores, whereby greater effects are produced from the fuel, and the smoke is in a great measure prevented or consumed.'

103. The specification is dated 14th of June, 1785.

104. The act of parliament, extending Mr. Watt's exclusive privilege for the improvements secured to him by his first patent, expired in 1800, at which period he retired from business, having, for some years before, ceased to take an active part.

105. Mr. Watt's first great improvement in the engine of Newcomen may be best under

nal cavity of the tube will be filled with condensable vapor. On closing the steam-cock, and opening that connected with the vessel G, a portion of the vapor will immediately expand itself, and, coming in contact with the cold sides of the vessel, a portion of its heat must be absorbed by the water at E. A new supply of steam then descends, and is also condensed; and indeed the same process continues till the whole of the steam is drawn from the tube. A vacuum being thus formed, the pressure of the atmosphere will preponderate, and the pistonrod be depressed to the bottom of the tube. On closing the stop-cock F a new portion of steam may be admitted by the other pipe, and, after raising the piston, the process of condensation may be readily repeated.

106. The advantages that arise from this mode of forming a vacuum are very considerable, not the least important of which is a saving of nearly half the fuel. In the old engine the condensing water must reduce the temperature of the internal surface of the cylinder to that of the atmosphere, before a vacuum could be produced; and, when the condensing-water was applied more sparingly, the elastic vapor remaining in the cylinder was found to materially reduce the pressure of the air operating above. The great advantage, then, of Mr. Watt's apparatus consists in performing the condensation in a separate vessel, so that the cylinder is always preserved at the temperature of boiling water.

107. Having thus produced a vacuum, without the intervention of condensing-water beneath the piston, Mr. Watt's next improvement consisted in closing the top of the cylinder, so that the piston-rod worked through an air-tight hole in the centre of the cap; and, to ensure the necessary pressure within the cylinder, steam with an elastic force greater than that of the atmosphere was admitted above the piston. The atmospheric engine of Newcomen was thus

machine.

der with its piston, and the piston rod moving through a stuffing-box, so as to be quite air-tight; is the steam-pipe for bringing steam from a boiler, and this terminates in a four-way cock hi, which together with the pipes m, n, and o, are drawn much too large for the cylinder Y Z, in order that their parts may be distinctly seen: p is the end of a lever or handle for the purpose of turning the cock when required. This cock is constructed precisely in the same manner as other cocks for fluids, and consists of a conical plug or pin ground very truly into the body of the cock itself; but instead of having a single straight passage through the plug, as is commonly the case, there are two curved ones h and i, which give the plug the appearance of having four openings, each appearing in the same plane, and at one quarter of the circumference apart, so that, as the cock stands in the figure, steam coming from g would pass through the opening i in the plug, and would be delivered into the pipe n at right angles to its first direction, instead of passing directly onwards to o, as would be the case in a common cock. From the pipe n the steam would pass immediately into the lower part Z of the cylinder, and consequently would drive the piston upwards. At the same time it

will be seen that there is an immediate connexion between the upper part Y of the cylinder through the pipe m, and opening of the cock to the pipe o o, leading down to the condenser; consequently, so long as the steam is acting against the under side of the piston, there will be a vacuum at Y to permit it to rise. So soon as the piston has arrived near the top of the cylinder, the cock must be turned one quarter round by depressing its handle p from its present position to that shown by the dotted lines at q, by which all the openings in the plug of the cock will be changed: steam, therefore, which now enters at g, will be turned upwards by the passage h, and by entering the pipe m will be conveyed above instead of below the piston, while the passage i forms a connexion between the pipe n, leading to the lower part of the cylinder, and the pipe o leading to the condenser; consequently, in this position, a vacuum will be formed below the piston, while the steam is operating above it. The piston will therefore descend, and, on coming near the bottom, the cock must be again turned into its first position, when the piston will ascend, and so on; thus producing an equality of force, both in the up and down strokes, by simply turning a small cock.

110. Having in the preceding pages strictly adhered to a chronological arrangement of the improvements in the steam engine, it may now be advisable to furnish our readers with a sketch of its earliest application to the purposes of navigation :

111. The possibility of employing steam as a moving power in the navigation of vessels was known early in the last century; its practical application, however, on a large scale, has not been fully established above twenty years.

112. In 1698 Savery recommended the use of paddle-wheels, similar to those now so generally employed in steam vessels, though without in the remotest degree alluding to his engine as a prime mover; and it is probable that he intended to employ the force of men or animals working at a winch for that purpose. About forty years after the publication of this mode of propelling vessels, Mr. Jonathan Hulls obtained a patent for a vessel in which the paddle wheels were to be driven by an atmospheric engine of considerable power.

113. In describing his mode of producing a force sufficient for towing of vessels, and other purposes, the ingenious patentee says, 'In some convenient part of the tow-boat there is placed a vessel about two-thirds full of water, with the top close shut; this vessel, being kept boiling, rarefies the water into steam; this steam being conveyed through a large pipe into a cylindrical vessel, and there condensed, makes a vacuum, which causes the weight of the atmosphere to press on this vessel, and so presses down a piston that is fitted into this cylindrical vessel, in the same manner as in Mr. Newcomen's engine, with which he raises water by fire.

114. 'It has been already demonstrated that when the air is driven out of a vessel of thirty inches diameter (which is but two feet and a half) the atmosphere will press on it to the weight of four tons 16 cwt. and upwards; when

proper instruments for this work are applied to it, it must drive a vessel with great force.'

115. Mr. Hulls's patent is dated 1736, and he suggested the use of a crank to produce the rotatory motion of his paddle-wheels; this ingenious mode of converting a reciprocating into a rotatory motion was afterwards recommended by the abbé Arnal, canon of Alais in Languedoc, who, in 1781, proposed the crank for the purpose of turning paddle-wheels in the navigation of lighters.

116. It is probable that Mr. Hulls anticipated some objection to his new mode of propelling vessels; and it appears from captain Savery's statement, to which we have already alluded, that a strong prejudice had been raised against the use of propelling wheels in vessels. Mr. Secretary Trenchard, who was at that time at the head of the admiralty, had also given a decided negative to the proposition. In answer therefore to the objections which might have been anticipated, Mr. Hulls proposed the following queries, which he afterwards solved in the most satisfactory way.

117.

Query 1.-Is it possible to fix instruments of sufficient strength to move so prodigious a weight as may be contained in a very large vessel?

118. Answer.-All mechanics will allow it is possible to make a machine to move an immense weight, if there is force enough to drive the same; for every member must be made in a proportionable strength to the intended work, and properly braced with laces of iron, so that no part can give way, or break.

119. Query 2.-Will not the force of the waves break any instrument to pieces that is placed to move in the water?

120. Answer. First, It cannot be supposed that this machine will be used in a storm or tempest at sea, when the waves are very raging; for if a merchant lieth in a harbour, &c., he would not choose to put out to sea in a storm, if it were possible to get out, but rather stay until it were abated. Secondly, when the wind comes a-head of the tow-boat, the fans will be protected by it from the violence of the waves, and, when the wind comes side-ways, the waves will come edge-ways of the fans, and therefore strike them with the less force. Thirdly, there may be pieces of timber laid to swim on the surface of the water on each side of the fans, and so contrived as they shall not touch them, which will protect them from the force of the waves.

121. Up inland rivers, where the bottom can possibly be reached, the fans may be taken out, and cranks placed at the hindmost axis to strike a shaft to the bottom of the river, which will drive the vessel forward with the greater force. 122. Query 3.-It being a continual expense to keep this machine at work, will the expense be answered?

6

123. Answer. The work to be done by this machine will be upon particular occasions, when all other means yet found out are wholly insufficient. How often does a merchant wish that his ship were on the ocean, when, if he were there, the wind would serve tolerably well to carry him on his intended voyage, but does

not serve at the same time to carry him out of the river, &c., he happens to be in, which a few hours' work at this machine would do. Besides, I know engines that are driven by the same power as this is, where materials for the purpose are dearer than in any navigable river in England. Experience, therefore, demonstrates that the expense will be but a trifle to the value of the work performed by those sort of machines which any person who knows the nature of those things may easily calculate.'

124. M. Duquet appears to have tried revolving oars as early as the year 1699, and experiments were made with them on a large scale both at Marseilles and at Havre: this mode, however, of impelling vessels was soon given up as impracticable; and after our countryman, Hulls, the marquis de Jouffroy unquestionably holds the most distinguished rank in the list of practical engineers, who have added to the value of this invention.

125. It is evident, from an article published in the Journal des Debats, that in 1781 the marquis constructed a steam boat at Lyons, of 140 feet in length. With this he made several successful experiments on the Saone, near that city. The events of the revolution, which broke out a few years afterwards, prevented M. de Jouffroy from prosecuting this undertaking, or reaping any advantage from it. On his return to France after a long exile, in 1796, he learned from the newspapers that M. de Blanc, an artist of Trevoux, had obtained a patent for the construction of a steam boat, built probably from such information as he could procure relative to the experiments of the marquis. The latter appealed to the government, which was then too much occupied with public affairs to attend to those of individuals. Meanwhile Fulton, who had gained the same information, and was making similar experiments near the Isle des Cygnes, alarmed M. de Blanc, who knew that he had much more to fear from the influence and mechanical skill of an Anglo-American than from that of an emigrant. He accordingly alleged his patent right, and requested the stoppage of Mr. Fulton's works, who returned for answer that his essays could not affect France, as he had no intention to set up a practical competition upon the rivers of that country, but should soon return to America, which he actually did, and commenced the erection of those engines to which he has since laid claim as exclusive inventor.

126. Shortly after the first experiments were made, by the marquis de Jouffroy, a gentleman of the name of Miller, who resided at Dalswinton, published a work, in which he described the application of wheels to the working of triple vessels on canals; and in 1794 he completed a model of a boat on this construction, impelled by a steam engine.

127. From this period till 1801, but little progress appears to have been made in this species of navigation: in that year Mr. Symington, who had been employed in the construction of Miller's vessel, tried a boat propelled by steam on the Forth-and-Clyde inland navigation; this, however, was shortly laid aside, on account of the injury with which it threatened the banks of

the canal, from the violent agitation produced by the paddle-wheels.

128. Mr. Symington's mode of connecting the piston and paddle-wheel was by placing the cylinder nearly in a horizontal position, so that by this means the necessity of employing a working beam was avoided. The piston was also supported in its position by friction-wheels, and communicated, by means of a rod, with a crank connected with the wheel, which imparts a motion to a paddle somewhat slower than its own. The paddle wheel was placed in the middle of the boat towards the stern, and on this account it became necessary to have a double rudder, connected by rods, which were moved by a winch placed at the head of the boat.

129. Mr. Symington also employed stampers placed at the head of the boat, for the purpose of breaking the ice on canals; and this plan, we believe, was also adopted in the original construction of the vessels intended for the Arctic expedition.

130. In 1795 a very ingenious apparatus was invented by lord Stanhope, and tried by that nobleman in Greenland Dock. In this experiment the paddles were made to resemble the feet of a duck, and were placed under the quarters of the vessel. This plan was also tried in America, but it does not appear in either case to have answered the expectations of its projector. 131. A plan has also been tried which in some measure resembles the endless chain of a pump. This was, we believe, first employed in the duke of Bridgewater's canal, and consists in the use of a chain, with a number of paddles attached to it, going over two wheels placed level with the water line. A steam engine, acting on the foremost roller, gave motion to the chain, and a continuous parallel motion was thus effected.

132. Having already described the general arrangement of a steam engine, as improved by Mr. Watt, it may now be advisable more particularly to point out the arrangement of its mechanism in detail. For this purpose we may select the single acting engine as employed for pumping.

133. Plate III. ABC represents the beam, which is made of cast-iron instead of wood, and is composed of two large plates, of the shape represented in the figure, put together at twelve inches distance from each other, leaving a space between them, the centre or axis B passing through the middle of both plates. The axis lies on the floor D, which is sustained by the wall E, built beneath the centre. Q is the cylinder, contained within a steam jacket, composed of segments screwed together. FF is the steampipe coming from the boiler G. ab is the piston-rod, connected with the end C of the beam, by links C, b; and whilst the upper ends of these links move in the arc of a circle, with the end of the beam, the lower ends, b, are made to accommodate themselves to the vertical motion of the piston-rod a, by means of the rods c, extending to the smaller links, d, which form a parallelogram. The motion of the parallelogram is governed by the bridle-rods, which move about a fixed centre m; the lower ends b, of the links b, C, will ascend and descend in a perpen

dicular right line. A similar motion, but of half the quantity, is given to the rod R, which works the air-pump, N, of the engine at the lower end, and the middle part of the rod has the plug-beam, R, attached to it, which has pins, or chocks, screwed on it to actuate the handles, x, y, and z, of the mechanism for the valves, which mechanism is very different from that employed in the old engines, and even from that of the first engines of Mr. Watt.

134. The pump-rod p is suspended at the end A of the beam by another parallel motion, and the upper part S of the rod is made of castiron, and very massive, to have a sufficient weight in itself to draw up the piston, and make the returning stroke. The real pump-rod p is jointed to the heavy counter-weight S, and is polished, like the piston-rod a, that it may slide through a collar of leathers in the head of the pump Y, because the pump is of that kind called lifting force-pumps; its bucket raises the water in ascending, but it forces it through the air-vessel T, and pipe X, which leads to a reservoir two miles distant, in Hyde Park, and elevated 150 feet above the level of the water in the well where the pump draws from. This well has a communication with the river.

135. The cylinder Q is kept down by the weight of a pier of masonry, on which it is placed, and large iron bolts, n, descend from the lower stanch to the groundsills, upon which the masonry is built. Immediately before the pier is the condensing cistern M, which contains the air-pump N, the condenser L, partly concealed, and hot-well g, and is kept supplied with cold water by the cold-water pump I, worked by the beam at the outer end, and the waste runs off again into the well, so as to keep the water in the cistern always cold.

136. The valves, which must be opened and shut to produce the action of the engine, are four in number; viz. the upper steam-valve at F, the lower steam-valve O, and the exhaustion-valve K, fig. 2, and a small valve 1, beneath the water in the cistern M, to admit the injection into the condenser: but these parts are better explained in fig. 2, which is a section of the cylinder, airpump, and condenser on a double scale.

137. A, the section of the cylinder, in which the piston X moves; F, the steam-pipe coming from the boiler; L, the condenser; N, the air or discharging-pump; m, a passage or pipe from the pump L to the condenser N, in which passage is an occasional communication by a hanging-valve at m, which shuts towards the condenser; is the injection-valve, to be lifted by the engine at every stroke, for the purpose of condensing the steam in the condenser L; w is the snifting or blowing-valve, placed outside the condensing cistern (of which M M is a section, on purpose to show the contents); the sniftingvalve w communicates with the condenser by a pipe passing through the side of the cistern M, and is inserted at the side of the condenser; K is the exhaustion-valve, to be lifted by the engine, and open a, communication between the cylinder A and the condenser L; O is the steam-valve, to be lifted by the engine, and open a communication between the lower part of the

cylinder, and upper part thereof, through the steam-pipe r; and F is the upper steam valve of the same kind, opening a passage from the boiler to the top of the cylinder; and thence by the pipe r, and valves O, K, to all parts of the engine.

138. We must now attend to the mechanism by which the engine is made to feed itself, and perform its reciprocations. The valves are lifted by means of a lever applied to each, within the iron box in which it is contained, entering into an opening in the stem of the valve; and a second lever is fixed on the axis of the lever, on the outside of the box, to be connected with the levers and handles x, y, z, which open and shut the valves. There are three separate axles, or spindles, placed parallel and above one another, and each has a handle or spanner x, y, z, by which it is moved, either by the hand, to start the engine, or by the chocks on the plug-beam R, when the engine is in action. The two upper spindles, a andy, have short levers projecting from them towards the cylinder; and from each of these levers a rod is suspended, with a sufficient weight, o, at the lower end to turn round the spindle, each upon its axis, in that direction which will cause the handles, a and y, to fly upwards. Also the lower spindle has a lever projecting from it, away from the cylinder, with a heavy weight, n, fixed at the end; but this being applied, on the opposite side, to the weights of the two upper handies x, y, the weight, n, causes the handle, z, to descend. Both the axles of the lower handles, y, z, have small levers, or catches, 1 and 2, which act in the hooks of a double latch, or detent, tv, which is moveable upon a centre-pin situated between the two axles. The hooks of this detent are to detain the catches of the spindles, and prevent the handles y z from moving by the action of their respective weights, until the detent is moved on its centre, so as to relieve the catches of the levers from its hooks t, v. But it is evident, from fig. A, that when only one catch, as 2, is hooked by the lower hook v of the detent, and consequently the weight of its spindle is held up, if the other catch, 1, is moved by depressing its handle y, so as to raise its weight in the act of entering the hook t of the detent, it will press the end t of the detent forwards upon its centre, and this at the same time pressing back the hook v, at the opposite end of the detent, releases the catch, z, of the lower handle, z, therefrom, and the weight n on that spindle immediately falls.

139. The spindle of the upper handle, r, is devoted to opening and shutting the upper steam-valve F, having a lever which communicates by a rod, 2, with the lever, 3, of that valve; so that by pressing down the handle r, it will shut the valve F. The weight o, which is applied to the upper spindle, tends to lift up the handler, and open the valve F; and, when the upper handler is depressed, the valve will be shut; or, when the handle is suffered to fly up by the action of its weight, it will open the valve.

140. The second spindle, y, has a lever communicating with the lever of the exhaustingvalve K, by a rod 4. The weight o, applied to this like the former, tends to lift up the handle y, and draw open the valve; but, when the handle

y is depressed, the valve is shut, and in this position the catch 1 is held down by the hook of the detent, before explained, and retains the valve shut.

141. Lastly, the lower spindle, z, is for the lower steam-valve O, which is opened by the rod 14, when the handle is suffered to fall down, and shut when the same is up, being held by the catch 2, and hook v. In all these the weight tends to open the valve; but, when the valve is to be kept shut, the detent holds the weight up. Now, by removing the detent, the weight falls and opens the valve in an instant.

142. The upper spindle has no detent to detain it; but what is equivalent is a rod, 5, jointed to that lever of the middle axis which has its weight and rod, o, suspended from it. The upper end of the rod, 5, is made with a loop, or long slit, in which works a pin at the end of a lever, 6, projecting from the upper axis towards the cylinder. The consequence of this is, that while the middle axis is detained by its catch, and detent t, to keep the exhausting-valve K shut, the lever 6 of the upper spindle will be borne up by its pin resting in the bottom of the loop of the rod 5, so as to keep the weight from opening the upper steam-valve F, as long as the exhausting-valve is kept shut; but when the catch, 1 t, of the middle axis is discharged, and its weight has opened the exhausting-valve, the looped rod, 5, will no longer support the lever, 6, of the upper axis, but allows its weight to descend and open the upper steam-valve; but at the same time the upper steam-valve, F, is not confined to be always open when the exhausting-valve, K, is open; for the upper steam-valve may be shut by depressing the upper handler, without affecting the exhausting-valve at all, because the slit, or loop, in the top of the rod 5, allows that motion. This property must be attended to, because the action of the engine, by expansion, depends upon it. We have not before noticed the injection-valve, from which a long wire ascends, and is attached to a strap, 9, which winds upon the middle axis; therefore, when the middle handle, y, flies up by its weight, it winds the strap, and opens the injection-valve at the same instant that the exhausting-valve is opened.

143. The injection-valve, l, is placed to close the orifice at the end of a short curved pipe, which enters into the condenser and turns up; and the pipe has a cock in it, between the valve and the condenser, to cut off the communication, or to regulate the supply of injection when the valve is opened. This cock must be always shut when the engine is not at work, to prevent the condenser filling with water.

144. Suppose the fire lighted beneath the boiler G; all the valves are kept shut by pressing down the two upper handles r and y, and lifting up the lower one, their respective catches detaining them in those positions, until the steam is sufficiently heated, and the engine is ready to work. In the quiescent position of the engine, when it is at rest, the counter-weight always draws the piston fully at the top of its cylinder, as in the figure; the air-pump bucket will also be at the top of its barrel.