PLATE DI.

Fig. 1. a A b B

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

Fig. 6. Savary's Steam Engine.

Fig. 7. Newcomen's Steam Engine. BEIGHTON'S STEAM ENGINE.

PLATE XII.

Fig. 8.

Fig. 1.

Fig. 2.

Fig. 9.

Fig. 10.

Engraved by W.B. Paterson, Edinburgh. WATT'S STEAM ENGINE.

PLATE DIII.

Fig. 12.

Fig. II.

Fig. 13.

Engd by W. & D. Litcarr Edin'c. Hornblower's Steam Engine.

PLATE DIV.

Fig. 16. Fig. 15. Fig. 14.

 the link, the inner half (that is, the half next the cylinder) of the wheel U will work on the inner half of the wheel W, so that at the end of the revolution of the fly the wheel W must have got to the top of the wheel U, and the outer end of the beam must be raised to its highest position. The next revolution of the fly will bring the wheel W and the beam connected with it to their first positions; and thus every two revolutions of the fly will make a complete period of the beam's reciprocating movements. Now, instead of supposing the fly to drive the beam, let the beam drive the fly. The motions must be perfectly the same, and the ascent or descent of the piston will produce one revolution of the fly.

A side view of this apparatus is given in fig. 12, marked by the same letters of reference. This shows the situation of parts which were fore-shortened in fig. 11, particularly the descending branch C of the steam-pipe, and the situation and communications of the two pumps K and I. 8, 8 is the horizontal part of the steam-pipe. 9 is a part of it whose box is represented by the dark circle of fig. 11. D is the box of the steam-clack; and the little circle at its corner represents the end of the axis which turns it, as will be described afterwards. N is the place of the upper eduction valve. A part only of the upper eduction-pipe G is represented, the rest being cut off, because it would have covered the descending steam-pipe CC. When continued down, it comes between the eye and the box E of the lower steam-valve, and the box F of the lower eduction-valve.

Let us now trace the operation of this machine through all its steps. Recurring to fig. 11, let us suppose that the lower part of the cylinder BB is exhausted of all elastic fluids; that the upper steam-valve D and the lower eduction-valve F are open, and that the lower steam-valve E and upper eduction-valve N are shut. It is evident that the piston must be pressed toward the bottom of the cylinder, and must pull down the end of the working beam by means of the toothed rack OO and sector QQ, causing the other end of the beam to urge forward the machinery with which it is connected. When the piston arrives at the bottom of the cylinder, the valves D and F are shut by the plug frame, and E and N are opened. By this last passage the steam gets into the eduction-pipe, where it meets with the injection water, and is rapidly condensed. The steam from the boiler enters at the same time by E, and pressing on the lower side of the piston, forces it upwards, and by means of the toothed rack OO and toothed sector QQ forces up that end of the working beam, and causes the other end to urge forward the machinery with which it is connected; and in this manner the operation of the engine may be continued for ever.

The injection water is continually running into the eduction-pipe, because condensation is continually going on, and therefore there is a continual atmospheric pressure to produce a jet. The air which is disengaged from the water, or enters by leaks, is evacuated only during the rise of the piston of the air-pump K. When this is very copious, it renders a very large air-pump necessary; and in some situations Mr Watt has been obliged to employ two air-pumps, one worked by each arm of the beam. This in every case expends a very considerable portion of the power, for the air-pump is always working against the whole pressure of the atmosphere.

It is evident that this form of the engine, by maintaining an almost constant and uninterrupted impulsion, is much fitter for driving any machinery of continued motion than any of the former engines, which were inactive during half of their motion. It does not, however, seem to have this superiority when employed to draw water: But it is equally fitted for this task. Let the engine be loaded with twice as much as would be proper for it if a single-stroke engine, and let a fly be connected with it. Then it is plain that the power of the engine during the rise of the steam-piston will be accumulated in the fly; and this, in conjunction with the power of the engine during the descent of the steam-piston, will be equal to the whole load of water.

In speaking of the steam and eduction-valves, we said that they were all puppet-valves. Mr Watt employed cocks, and also sliding-valves, such as the regulator or steam-valves in the old engines. But he found them always lose their tightness after a short time. This is not surprising, when we consider that they are always perfectly dry, and almost burning hot. He was therefore obliged to change them all for puppet-clacks, which, when truly ground and nicely fitted in their motions at first, are not found to go out of order by any length of time. Other engineers now universally use them in the old form of the steam-engine, without the same reasons, and merely by servile and ignorant imitation.

The way in which Mr Watt opens and shuts these valves is as follows. Fig. 13 represents a clack with its seat and box. Suppose it one of the eduction-valves, HH is part of the pipe which introduces the steam, and GG is the upper part of the pipe which communicates with the condenser. At EE may be observed a piece more faintly shaded than the surrounding parts. This is the seat of the valve, and is a brass or bell-metal ring turned conical on the outside, so as to fit exactly into a conical part of the pipe GG. These two pieces are fitted by grinding; and the cone being of a long taper, the ring sticks firmly in it, especially after having been there for some time and united by rust. The clack itself is a strong brass plate D, turned conical on the edge, so as to fit the conical or sloping inner edge of the seat. These are very nicely ground on each other with emery. This conical joining is much more obtuse than the outer side of the ring; so that although the joint is air-tight, the two pieces do not stick strongly together. The clack has a round tail DG, which is freely moveable up and down in the hole of a cross piece FF. On the upper side of the valve is a strong piece of metal DC firmly joined to it, one side of which is formed into a toothed rack. A is the section of an iron axle which turns in holes in the opposite sides of the valve-box, where it is nicely fitted by grinding, so as to be air-tight. Collets of thick leather, well soaked in melted tallow and rosin, are sewed on the outside of these holes to prevent all ingress of air. One end of this axis projects a good way without the box, and carries a spanner or handle, which is moved by the plug-frame. To this axis is fixed a strong piece of metal B, the edge of which is formed into an arch of a circle having the axis A in its centre, and is cut into teeth, which work in the teeth of the rack DC. K is a cover which is fixed by forewos to the top of the box HJH, and may be taken off in order to get at the valve when it needs repairs.

From this description it is easy to see that by turning the handle which is on the axis A, the sector B must lift up the valve by means of its toothed rack DC, till the upper end of the rack touch the knob or button K. Turning the handle in the opposite direction brings the valve down again to its seat.

This valve is extremely tight. But in order to open it for the passage of the steam, we must exert a force equal to the pressure of the atmosphere. This in a large engine is a very great weight. A valve of fix inches diameter sustains a pressure not less that 400 pounds. But this force is quite momentary, and hardly impedes the motion of the engine; for the instant the valve is detached from its seat, although it has not moved the 100th part of an inch, the pressure is over. Even this little inconvenience has been removed by a delicate thought of Mr Watt. He has put the spanner in such a position when it begins to raise the valve, that its mechanical energy is almost infinitely great. Let QR (fig. 14.) be part of the plug-frame descending, and P one of its pins just going to lay hold of the spanner NO moveable round the axis N. On the same axis is another arm NM connected by a joint with the leader ML, which is connected also by a joint with the spanner LA that is on the axis A of the sector within the valve-box. Therefore when the pin P puts down the spanner NO, the arm NM moves sidewise and pulls down the spanner AL by means of the connecting rod. Things are so disposed, that when the cock is shut, LM and MN are in one straight line. The intelligent mechanic will perceive that, in this position, the force of the lever ONM is insuperable. It has this further advantage, that if any thing should tend to force open the valve, it would be ineffectual; for no force exerted at A, and transmitted by the rod LM, can possibly push the joint M out of its position. Of such importance is it to practical mechanics, that its professors should be persons of penetration as well as knowledge. Yet this circumstance is unheeded by hundreds who have servilely copied from Mr Watt, as may be seen in every engine that is puffed on the public as a discovery and an improvement. When these puppet-valves have been introduced into the common engine, we have not seen one instance where this has been attended to; certainly because its utility has not been observed: and there is one situation where it is of more consequence than in Mr Watt's engine, viz. in the injection cock. Here the valve is drawn back into a box, where the water is so awkwardly disposed round it that it can hardly get out of its way, and where the pressure even exceeds that of the atmosphere. Indeed this particular substitution of the button valve for the cock is most injudicious.

We postponed any account of the office of the fly XX (fig. 11.), as it is not of use in an engine regulated by the fly VV. The fly XX is only for regulating the reciprocating motion of the beam when the steam is not admitted during the whole descent of the piston. This it evidently must render more uniform, accumulating a momentum equal to the whole pressure of the full supply of steam, and then sharing it with the beam during the rest of the descent of the piston.

When a person properly skilled in mechanics and chemistry reviews these different forms of Mr Watt's steam-engine, he will easily perceive them susceptible of many intermediate forms, in which any one or more of the distinguishing improvements may be employed. The first great improvement was the condensation in a separate vessel. This increased the original powers of the three great engine, giving to the atmospheric pressure and to the counter-weight their full energy; at the same time the waste of steam is greatly diminished. The next improvement, by employing the prelude of the steam instead of that of the atmosphere, aimed only at a still farther diminution of the waste; but was fertile in advantages, rendering the machine more manageable, and particularly enabled us at all times, and without trouble, to suit the power of the engine to its load of work, however variable and increasing; and brought into view a very interesting proposition in the mechanical theory of the engine, viz. that the whole performance of a given quantity of steam may be augmented by admitting it into the cylinder only during a part of the piston's motion. Mr Watt has varied the application of this proposition in a thousand ways; and there is nothing about the machine which gives more employment to the sagacity and judgment of the engineer. The third improvement of the double impulse may be considered as the finishing touch given to the engine, and renders it as uniform in its action as any water-wheel. In the engine in its most perfect form there does not seem to be above one-fourth of the steam wasted by warming the apparatus; so that it is not possible to make it one-fourth part more powerful than it is at present. The only thing that seems susceptible of considerable improvement is the great beam. The enormous strains exerted now on its arms require a proportional strength. This requires a vast mass of matter, not less indeed in an engine with a cylinder of 54 inches than three tons and a half, moving with the velocity of three feet in a second, which must be communicated in about half a second. This mass must be brought into motion from a state of rest, must again be brought to rest, again into motion, and again to rest, to complete the period of a stroke. This consumes much power; and Mr Watt has not been able to load an engine with more than 10 or 11 pounds on the inch and preserve a sufficient quantity of motion, so as to make 12 or 15 fix-feet strokes in a second. Many attempts have been made to lessen this mass by using a light framed wheel, or a light frame of carpentry, in place of a solid beam. These have generally been constructed by persons ignorant of the true scientific principles of carpentry, and have fared accordingly. Mr Watt has made similar attempts; but found, that although at first they were abundantly strong, yet after a short time's employment the straps and bolts with which the wooden parts were connected cut their way into the wood, and the framing grew loose in the joints, and, without giving any warning, went to pieces in an instant. A solid malleable beam, of sufficient strength, bends, and sensibly complains (as the carpenters express it), before it breaks. In all great engines, therefore, such only are employed, and in smaller engines he sometimes uses cast-iron wheels or pulleys; nay, he frequently uses no beam or equivalent whatever, but employs the steam-piston rod to drive the machinery to which the engine is applied.

We presume that our thinking readers will not be disappointed. displeased with this rational history of the progress of this engine in the hands of its ingenious and worthy inventor. We owe it to the communications of a friend, well acquainted with him, and able to judge of his merits. The public see him always associated with the no less celebrated mechanic and philosopher Mr Boulton of Soho near Birmingham (see Soho). They have shared the royal patent from the beginning; and the alliance is equally honourable to both.

The advantages derived from the patent right show both the superiority of the engine and the liberal minds of the proprietors. They erect the engines at the expense of the employers, or give working drafts of all the parts, with instructions, by which any resident engineer may execute the work. The employers select the best engine of the ordinary kind in the kingdom, compare the quantities of fuel expended by each, and pay to Messrs Watt and Boulton one-third of the annual savings for a certain term of years. By this the patentees are excited to do their utmost to make the engine perfect; and the employer pays in proportion to the advantage he derives from it.

It may not be here improper to state the actual performance of some of these engines, as they have been ascertained by experiment.

An engine having a cylinder of 31 inches in diameter, and making 17 double strokes per minute, performs the work of forty horses working night and day (for which three relays of 120 horses must be kept), and burns 11,000 pounds of Staffordshire coal per day. A cylinder of 19 inches, making 25 strokes of 4 feet each per minute, performs the work of 12 horses working constantly, and burns 3700 pounds of coals per day. A cylinder of 24 inches, making 22 strokes of 5 feet, burns 5500 pounds of coals, and is equivalent to the constant work of 20 horses. And the patentees think themselves authorized by experience to say in general, that these engines will raise more than 20,000 cubic feet of water 24 feet high for every hundred weight of good pit-coal consumed by them.

In consequence of the great superiority of Mr Watt's engines, both with respect to economy and manageableness, they have become of most extensive use; and in every demand of manufacture on a great scale they offer us an indefatigable servant, whose strength has no bounds. The greatest mechanical project that ever engaged the attention of man was on the point of being executed by this machine. The States of Holland were treating with Messrs Watt and Boulton for draining the Haerlem Meer, and even reducing the Zuyder Zee; and we doubt not but that it will be accomplished whenever that unhappy nation has sufficiently felt the difference between liberty and foreign tyranny. Indeed such unlimited powers are afforded by this engine, that the engineer now thinks that no task can be proposed to him which he cannot execute with profit to his employer.

No wonder then that all classes of engineers have turned much of their attention to this engine; and seeing that it has done so much, that they try to make it do still more. Numberless attempts have been made to improve Mr Watt's engine; and it would occupy a volume to give an account of them; whilst that account would do no more than indulge curiosity. Our engineers by profession are in general miserably deficient in that accurate knowledge of mechanics and of chemistry which is necessary for understanding this machine; and we have not heard of one in this kingdom who can be put on a par with the present patentees in this respect.

Most of the attempts of engineers have been made with the humbler view of availing themselves of Mr Watt's discoveries, so as to construct a steam-engine superior to Newcomen's, and yet of a form sufficiently different from Watt's to keep it without the reach of his patent. This they have in general accomplished by performing the condensation in a place which, with a little stretch of fancy, not unfrequent in a court of law, may be called part of the cylinder.

The success of most of these attempts has interfered and the fo little with the interest of the patentees, that they succeed or have not hindered the erection of many engines which the law would have deemed encroachments. We think it our duty to give our opinion on this subject without reserve. These are most expensive undertakings, and few employers are able to judge accurately of the merits of a project presented to them by an ingenious artificer. They may see the practicability of the scheme, by having a general notion of the expansion and condensation of steam, and they may be misled by the ingenuity apparent in the construction. The engineer himself is frequently the dupe of his own ingenuity; and it is not always dishonesty, but frequently ignorance, which makes him prefer his own invention or (as he thinks it) improvement. It is a most delicate engine, and requires much knowledge to see what does and what does not improve its performance. We have gone into the preceding minute investigation of Mr Watt's progress with the express purpose of making our readers fully masters of its principles, and have more than once pointed out the real improvements, that they may be firmly fixed and always ready in the mind. By having recourse to them, the reader may pronounce with confidence on the merits of any new construction, and will not be deceived by the puffs of an ignorant or dishonest engineer.

We must except from this general criticism a construction by Mr Jonathan Hornblower near Bristol, on account of its singularity, and the ingenuity and real skill which appears in some particulars of its construction. The following short description will sufficiently explain its principle, and enable our readers to appreciate its merit.

A and B (fig. 15.) represent two cylinders, of which A is the largest. A piston moves in each, having their rods C and D moving through collars at E and F. These cylinders may be supplied with steam from the boiler by means of the square pipe G, which has a flanch steam-end to connect it with the rest of the steam-pipe. This fine square part is represented as branching off to both cylinders. c and d are two cocks, which have handles and tumblers as usual, worked by the plug-beam W. On the fore-side (that is, the side next the eye) of the cylinders is represented another communicating pipe, whose section is also square or rectangular, having also two cocks a, b. The pipe X, immediately under the cock b, establishes a communication between the upper and lower parts of the small cylinder B, by opening the cock b. There is a similar pipe on the other side of the cylinder A, immediately under the cock d. When the cocks c and a are open, and the cocks b and d are shut, the steam from the boiler has free admission into the upper part of the cylinder B, and the steam from the lower part of B has free admission into the upper part of A; but the upper part of each cylinder has no communication with its lower part.

From the bottom of the great cylinder proceeds the eduction-pipe K, having a valve at its opening into the cylinder, which bends downwards, and is connected with the conical condenser L (c). The condenser is fixed on a hollow box M, on which stand the pumps N and O, for extracting the air and water; which last runs along the trough T into a cistern U, from which it is raised by the pump V for recruiting the boiler, being already nearly boiling hot. Immediately under the condenser there is a spigot-valve at S, over which is a small jet-pipe, reaching to the bend of the eduction-pipe. The whole of the condensing apparatus is contained in a cistern R of cold water. A small pipe P comes from the side of the condenser, and terminates on the bottom of the trough T, and is there covered with a valve Q, which is kept tight by the water that is always running over it. Lastly, the pump-rods X cause the outer end of the beam to preponderate, so that the quiescent position of the beam is that represented in the figure, the pistons being at the top of the cylinders.

Suppose all the cocks open, and steam coming in copiously from the boiler, and no condensation going on in L; the steam must drive out all the air, and at last follow it through the valve Q. Now shut the valves b and d, and open the valve S of the condenser. The condensation will immediately commence. There is now no pressure on the under side of the piston of A, and it immediately descends. The communication between the lower part of B and the upper part of A being open, the steam will go from B into the space left by the piston of A. It must therefore expand, and its elasticity must diminish, and will no longer balance the pressure of the steam above the piston of B. This piston therefore, if not withheld by the beam, would descend till it is in equilibrium, having steam of equal density above and below it. But it cannot descend so far; for the cylinder A is wider than B, and the arm of the beam at which its piston hangs is longer than the arm which supports the piston of B; therefore when the piston of B has descended as far as the beam will permit it, the steam between the two pistons occupies a larger space than it did when both pistons were at the tops of their cylinders. Its density, therefore, and its elasticity, diminish as its bulk increases. It is therefore not a balance; for the steam on the upper side of B, and the piston B, pulls at the beam with all the difference of these pressures. The slightest view of the subject must show the reader, that as the pistons descend, the steam that is between them will grow continually rarer and less elastic, and that both pistons will pull the beam downwards.

Suppose now that each has reached the bottom of its cylinder. Shut the cock a and the eduction-cock at the bottom of A, and open the cocks b and d. The communication being now established between the upper and lower part of each cylinder, nothing hinders the counter weight from raising the pistons to the top. Let them arrive there. The cylinder B is at this time filled with steam of the ordinary density, and the cylinder A with an equal absolute quantity of steam, but expanded into a larger space.

Shut the cocks b and d, and open the cock a, and the eduction-cock at the bottom of A; the condensation will again operate, and the pistons descend. And thus the operation may be repeated as long as steam is supplied; and one full of the cylinder B of ordinary steam is expended during each working stroke.

Let us now examine the power of this engine. It is evident, that when both pistons are at the top of their respective cylinders, the active pressure (that is, the difference of the pressure on its two sides) on the piston of B is nothing, while that on the piston of A is equal to the full pressure of the atmosphere on its area. This, multiplied by the length of the arm by which it is supported, gives its mechanical energy. As the pistons descend, the pressure on the piston of B increases, while that on the piston of A diminishes. When both are at the bottom, the pressure on the piston of B is at its maximum, and that on the piston of A at its minimum.

Mr Hornblower saw that this must be a beneficial employment of steam, and preferable to the practice of condensing it while its full elasticity remained; but he has not considered it with the attention necessary for ascertaining the advantage with precision.

Let a and b represent the areas of the pistons of A and B, and let α and β be the lengths of the arms by which they are supported. It is evident, that when both pistons have arrived at the bottoms of their cylinders, the capacities of the cylinders are as aa and bb. Let this be the ratio of m to 1. Let g h i k (fig. 16.) and l m n o be two cylinders of equal length, communicating with each other, and fitted with a piston-rod p q, on which are fixed two pistons aa and bb, whose areas are as m and 1. Let the distance between the pistons be precisely equal to the height of each cylinder, which height we shall call h. Let x be the space g b or b a, through which the pistons have descended. Let the upper cylinder communicate with the boiler, and the lower cylinder with the condenser or vacuum V.

Any person in the least conversant in mechanics and pneumatics will clearly see that the strain or pressure on the piston-rod pq is precisely the same with the united energies of the two piston rods of Mr Hornblower's engine, by which they tend to turn the working beam round its axis.

The base of the upper cylinder being 1, and its height h, its capacity or bulk is 1 h or h; and this expresses the natural bulk of the steam which formerly filled it, and is now expanded into the space b h l a a m i b. The part b h i b is plainly = h - x, and the part l a a m i s = m x. The whole space, therefore, is m x + h - x, = h + m x - x, or h + m - 1 x. Therefore the density of the steam between the pistons is

\[ \frac{h}{h + m - 1 x} \]

Let p be the downward pressure of the steam from the the boiler on the upper piston \( b b \). This piston is also pressed up with a force \( = p \frac{h}{h+m-1 \times} \) by the steam between the pistons. It is therefore, on the whole, pressed downward with a force \( = p \left( 1 - \frac{h}{h+m-1 \times} \right) \).

The lower piston \( a a \), having a vacuum below it, is pressed downwards with a force \( = p \frac{m h}{h+m-1 \times} \). Therefore the whole pressure on the piston rod downwards is \( = p \left( 1 + \frac{m h}{h+m-1 \times} - \frac{h}{h+m-1 \times} \right) = p \left( 1 + \frac{m-1 h}{h+m-1 \times} \right) = p + \frac{p h m-1}{h+m-1 \times} = p + \frac{p h}{m-1 + x} \).

This then is the momentary pressure on the piston rod corresponding to its descent \( x \) from its highest position. When the pistons are in their highest position, this pressure is equal to \( m p \). When they are in their lowest position, it is \( = p \frac{2 m-1}{m} \). Here therefore is an accession of power. In the beginning the pressure is greater than on a single piston in the proportion of \( m \) to \( 1 \); and at the end of the stroke, where the pressure is weakest, it is still much greater than the pressure on a single piston. Thus, if \( m \) be 4, the pressure at the beginning of the stroke is \( 4 p \), and at the end it is \( \frac{7}{4} p \), almost double, and in all intermediate positions it is greater. It is worth while to obtain the sum total of all the accumulated pressures, that we may compare it with the constant pressure on a single piston.

We may do this by considering the momentary pressure \( p + \frac{p h}{m-1 + x} \), as equal to the ordinate GF, H b,

or M c, of a curve F b c (fig. 10.), which has for its axis the line GM equal to \( h \) the height of our cylinder. Call this ordinate y. We have \( y = p + \frac{p h}{m-1 \times} \), and \( y - p = \frac{p h}{m-1 + x} \). Now it is plain that

\[ \frac{p h}{m-1 + x} \]

is the ordinate of an equilateral hyperbola,

of which \( p h \) is the power or rectangle of the ordinate and absciss, and of which the absciss reckoned from the centre is \( \frac{h}{m-1} + x \). Therefore make GE \( = p \), and draw DEA parallel to MG, and make EA \( = \frac{GM}{m-1} \),

\( = \frac{h}{m-1} \). The curve F b c is an equilateral hyperbola having A for its centre and AD for its asymptote. Draw the other asymptote AB, and its ordinate FB. Since the power of the hyperbola is \( = p h \), \( = GEDM \) (for GE \( = p \), and GM \( = h \)); and since all the inscribed rectangles, such as AEFB, are equal to \( p h \), it follows that AEFB is equal to GEDM, and that the area ABFC DA is equal to the area GF c MG, which expresses the accumulated pressure in Hornblower's engine.

We can now compute the accumulated pressure very easily. It is evidently \( = p h \times \left( 1 + L, \frac{\Delta D}{\Delta E} \right) \).

The intelligent reader cannot but observe that this is precisely the same with the accumulated pressure of a quantity of steam admitted in the beginning, and stopped in Mr Watt's method, when the piston has descended through the \( m \)th part of the cylinder. In considering Mr Hornblower's engine, the thing was presented in so different a form that we did not perceive the analogy at first, and we were surprised at the result. We could not help even regretting it, because it had the appearance of a new principle and an improvement; and we doubt not but that it appeared so to the ingenious author; for we have had such proofs of his liberality of mind as permit us not to suppose that he saw it from the beginning, and availed himself of the difficulty of tracing the analogy. And as the thing may mislead others in the same way, we have done a service to the public by showing that this engine, so costly and so difficult in its construction, is no way superior in power to Mr Watt's simple method of stopping the steam. It is even inferior, because there must be a condensation in the communicating passages. We may add, that if the condensation is performed in the cylinder A, which it must be unless with the permission of Watt and Boulton, the engine cannot be much superior to a common engine; for much of the steam from below B will be condensed between the pistons by the coldness of the cylinder A; and this diminishes the downward pressure on A more than it increases the downward pressure on B. We learn however that, by confining the condensation to a small part of the cylinder A, Mr Hornblower has erected engines clear of Mr Watt's patent, which are considerably superior to Newcomen's: so has Mr Symington.

We said that there was much ingenuity and real skill still observable in many particulars of this engine. The disposition and connection of the cylinders, and the whole condensing apparatus, are contrived with peculiar neatness. The cocks are very ingenious; they are and skill composed of two flat circular plates ground very true to each other, and one of them turns round on a pin through their centres; each is pierced with three sectional apertures, exactly corresponding with each other, and occupying a little less than one-half of their surfaces. By turning the moveable plate so that the apertures coincide, a large passage is opened for the steam; and by turning it so that the solid of the one covers the aperture of the other, the cock is shut. Such regulators are now very common in the cast iron stoves for warming rooms.

Mr Hornblower's contrivance for making the collars for the piston rods air-tight is also uncommonly ingenious. This collar is in fact two, at a small distance from each other. A small pipe, branching off from the main steam-pipe, communicates with the space between the collars. This steam, being a little stronger than the pressure of the atmosphere, effectually hinders the air from penetrating by the upper collar; and though a little steam should get through the lower collar into the cylinder A, it can do no harm. We see many cases in which this pretty contrivance may be of signal service. But it is in the framing of the great working beam that Mr Hornblower's scientific knowledge is most conspicuous; and we have no hesitation in affirming that it is stronger than a beam of the common form, and containing twenty times its quantity of timber. There is hardly a part of it exposed to a transverse strain, if we except the strain of the pump V on the strut by which it is worked. Every piece is either pushed or pulled in the direction of its length. We only fear that the bolts which connect the upper beam with the two iron bars under its ends will work loose in their holes, and tear out the wood which lies between them. We would propose to substitute an iron bar for the whole of this upper beam. This working beam highly deserves the attention of all carpenters and engineers. We have that opinion of Mr Hornblower's knowledge and talents, that we are confident that he will see the fairness of our examination of his engine, and we trust to his candour for an excuse for our criticism.

The reciprocating motion of the steam-engine has always been considered as a great defect; for though it be now obviated by connecting it with a fly, yet, unless it is an engine of double stroke, this fly must be an enormous mass of matter moving with great velocity. Any accident happening to it would produce dreadful effects: A part of the rim detaching itself would have the force of a bomb, and no building could withstand it. Many attempts have been made to produce a circular motion at once by the steam. It has been made to blow on the vanes of a wheel of various forms. But the rarity of steam is such, that even if none is condensed by the cold of the vanes, the impulse is exceedingly feeble, and the expence of steam, so as to produce any serviceable impulse, is enormous. Mr Watt, among his first speculations on the steam engine, made some attempts of this kind. One in particular was uncommonly ingenious. It consisted of a drum turning air-tight within another, with cavities so disposed that there was a constant and great pressure urging it in one direction. But no packing of the common kind could preserve it air-tight with sufficient mobility. He succeeded by immersing it in mercury, or in an amalgam which remained fluid in the heat of boiling water; but the continual trituration soon calcined the fluid and rendered it useless. He then tried Parent's or Dr Barker's mill, inclosing the arms in a metal drum, which was immersed in cold water. The steam rushed rapidly along the pipe which was the axis, and it was hoped that a great reaction would have been exerted at the ends of the arms; but it was almost nothing. The reason seems to be, that the greatest part of the steam was condensed in the cold arms. It was then tried in a drum kept boiling hot; but the impulse was now very small in comparison with the expence of steam. This must be the case.

Mr Watt has described in his specification to the patent office some contrivances for producing a circular motion by the immediate action of the steam. Some of these produce alternate motions, and are perfectly analogous to his double-stroke engine. Others produce a continued motion. But he has not given such a description of his valves for this purpose as can enable an engineer to construct one of them. From any guess that we can form, we think the machine very imperfect; and we do not find that Mr Watt has ever erected a continuous circular engine. He has doubtless found all his attempts inferior to the reciprocating engine with a fly. A very crude scheme of this kind may be seen in the Transactions of the Royal Society of Dublin 1787. But although our attempts have hitherto failed, we hope that the case is not yet desperate: we see different principles which have not yet been employed.

We shall conclude our account of this noble engine with observing, that Mr Watt's form suggests the construction of an excellent air-pump. A large vessel may ciples may be made to communicate with a boiler at one side, and be employed with the pump-receiver on the other, and also with a condenser. Suppose this vessel of ten times the capacity of the receiver; fill it with steam from the boiler, engine and drive out the air from it; then open its communication with the receiver and the condenser. This will rarely the air of the receiver ten times. Repeating the operation will rarely it 100 times; the third operation will rarely it 1000 times; the fourth 10,000 times, &c. All this may be done in half a minute.

STEAM-Kitchen. Ever since Dr Papin contrived his digester (about the year 1690), schemes have been proposed for dressing viands by the steam of boiling water. A philosophical club used to dine at Salters' coffee-house, Chelsea, about 40 years ago, and had their victuals dressed by hanging them in the boiler of the steam-engine which raised water for the supply of Piccadilly and its neighbourhood. They were completely dressed, and both expeditiously and with high flavour.

A patent was obtained for an apparatus for this purpose by a tin-man in London; we think of the name of Tate. They were afterwards made on a much more effective plan by Mr Gregory, an ingenious tradesman in Edinburgh, and are coming into very general use.

It is well known to the philosopher that the steam of boiling water contains a prodigious quantity of heat, which it retains in a latent state ready to be faithfully accounted for, and communicated to any colder body. Every cook knows the great scalding power of steam, and is disposed to think that it is much hotter than boiling water. This, however, is a mistake; for it will raise the thermometer no higher than the water from which it comes. But we can assure the cook, that if he make the steam from the spout of a tea-kettle pass through a great body of cold water, it will be condensed or changed into water; and when one pound of water has in this manner been boiled off, it will have heated the mass of cold water as much as if we had thrown into it seven or eight hundred pounds of boiling hot water.

If, therefore, a boiler be properly fitted up in a furnace, and if the steam of the water boiling in it be conveyed by a pipe into a pan containing viands to be dressed, every thing can be cooked that requires no higher degree of heat than that of boiling water: And this will be done without any risk of scorching, or any kind of overheating, which frequently spoils our dishes, and proceeds from the burning heat of air coming to those parts of the pot or pan which is not filled with liquor, and is covered only with a film, which quickly burns and taints the whole dish. Nor will the cook be scorched by the great heat of the open fire that is necessary for dressing at once a number of dishes, nor have his person and clothes soiled by the smoke and foot unavoidable in the cooking on an open fire. Indeed the whole whole process is so neat, so manageable, so open to inspection, and so cleanly, that it need neither fatigue nor offend the delicacy of the nicest lady.

We had great doubts, when we first heard of this as a general mode of cookery, as to its economy; we had none as to its efficacy. We thought that the steam, and consequently the fuel expended, must be vastly greater than by the immediate use of an open fire; but we have seen a large tavern dinner expeditiously dressed in this manner, seemingly with much less fuel than in the common method. The following simple narration of facts will show the superiority. In a paper manufactory in this neighbourhood, the vats containing the pulp into which the frames are dipped are about six feet diameter, and contain above 200 gallons. This is brought to a proper heat by means of a small cockle or furnace in the middle of the liquor. This is heated by putting in about one hundred weight of coals about eight o'clock in the evening, and continuing this till four next morning, renewing the fuel as it burns away. This method was lately changed for a steam heater. A furnace, having a boiler of five or fix feet diameter and three feet deep, is heated about one o'clock in the morning with two hundred weight of coals, and the water kept in brisk ebullition. Pipes go off from this boiler to fix vats, some of which are at 90 feet distance. It is conveyed into a flat box or vessel in the midst of the pulp, where it condenses, imparting its heat to the sides of the box, and thus heats the surrounding pulp. These fix vats are as completely heated in three hours, expending about three hundred weight of coals, as they were formerly in eight hours, expending near eighteen hundred weight of coals. Mr Gregory, the inventor of this steam-heater, has obtained (in company with Mr Scott, plumber, Edinburgh) a patent for the invention; and we are persuaded that it will come into very general use for many similar purposes. The dyers, hatmakers, and many other manufacturers, have occasion for large vats kept in a continual heat; and there seems no way so effectual.

Indeed when we reflect seriously on the subject, we see that this method has immense advantages considered merely as a mode of applying heat. The steam may be applied to the vessel containing the viands in every part of its surface; it may even be made to enter the vessel, and apply itself immediately to the piece of meat that is to be dressed, and this without any risk of scorching or overdoing.—And it will give out about \( \frac{7}{8} \) of the heat which it contains, and will do this only if it be wanted; so that no heat whatever is wasted except what is required for heating the apparatus. Experience shows that this is a mere trifle in comparison of what was supposed necessary. But with an open fire we only apply the flame and hot air to the bottom and part of the sides of our boiling vessels; and this application is hurried in the extreme; for to make a great heat, we must have a great fire, which requires a prodigious and most rapid current of air. This air touches our pans but for a moment, imparts to them but a small portion of its heat; and we are persuaded that three-fourths of the heat is carried up the chimney, and escapes in pure waste, while another great portion beams out into the kitchen to the great annoyance of the scorched cook. We think, therefore, that a page or two of this work will not be thrown away in the description of a contrivance by which a saving may be made to the entertainer, and the providing the pleasures of his table prove a less fatiguing task to this valuable corps of practical chemists.

Let A (fig. 1.) represent a kitchen-boiler, either properly fitted up in a furnace, with its proper fire-place, ash-pit, and flue, or set on a tripod on the open fire, or built up in the general fire-place. The steam-pipe BC rises from the cover of this boiler, and then is led away with a gentle ascent in any convenient direction. C represents the section of this conducting steam-pipe. Branches are taken off from the side at proper distances. One of these is represented at CDE, furnished with a cock D, and having a taper nozzle E, fitted by grinding into a conical piece F, which communicates with an upright pipe GH, which is foldered to the side of the stewing vessel PQRS, communicating with it by the short pipe I. The vessel is fitted with a cover OT', having a staple handle V. The piece of meat M is laid on a tin-plate grate KL, pierced with holes like a cul-tender, and standing on three short feet n n n.

The steam from the boiler comes in by the pipe I, and is condensed by the meat and by the sides of the vessel, communicating to them all its heat. What is not so condensed escapes between the vessel and its cover. The condensed water lies on the bottom of the vessel, mixed with a very small quantity of gravy and fatty matter from the viands. Frequently, instead of a cover, another stew-vessel with a cul-tender bottom is set on this one, the bottom of the one fitting the mouth of the other: and it is observed, that when this is done, the dish in the under vessel is more expeditiously and better dressed, and the upper dish is more slowly, but as completely, stewed.

This description of one stewing vessel may serve to give a notion of the whole; only we must observe, that when broths, soups, and dishes with made sauces or containing liquids, are to be dressed, they must be put into a smaller vessel, which is set into the vessel PQRS, and is supported on three short feet, so that there may be a space all round it of about an inch or three-quarters of an inch. It is observed, that dishes of this kind are not so expeditiously cooked as on an open fire, but as completely in the end, only requiring to be turned up now and then to mix the ingredients; because as the liquids in the inner vessel can never come into ebullition, unless the steam from the boiler be made of a dangerous heat, and every thing be close confined, there cannot be any of that tumbling motion that we observe in a boiling pot.

The performance of this apparatus is far beyond any expectation we had formed of it. In one which we examined, fix pans were stewing together by means of a boiler 10\(\frac{1}{2}\) inches in diameter, standing on a brisk open fire. It boiled very briskly, and the steam puffed frequently through the chinks between the stew-pans and their covers. In one of them was a piece of meat considerably above 30 pounds weight. This required above four hours stewing, and was then very thoroughly and equally cooked; the outside being no more done than the heart, and it was near two pounds heavier than when put in, and greatly swelled. In the mean time, several dishes had been dressed in the other pans. As far as we could judge, this cooking did not consume one-third part of the fuel which an open fire would have required for the same effect.

When we consider this apparatus with a little more knowledge of the mode of operation of fire than falls to the share of the cooks (we speak with deference), and consider the very injudicious manner in which the steam is applied, we think that it may be improved so as to surpass any thing that the cook can have a notion of.

When the steam enters the stew-pan, it is condensed on the meat and on the vessel; but we do not want it to be condensed on the vessel. And the surface of the vessel is much greater than that of the meat, and continues much colder; for the meat grows hot, and continues so, while the vessel, made of metal, which is a very perfect conductor of heat, is continually robbed of its heat by the air of the kitchen, and carried off by it. If the meat touch the side of the pan in any part, no steam can be applied to that part of the meat, while it is continually imparting heat to the air by the intermediate of the vessel. Nay, the meat can hardly be dressed unless there be a current of steam through it; and we think this confirmed by what is observed above, that when another stew-pan is set over the first, and thus gives occasion to a current of steam through its cullender bottom to be condensed by its sides and contents, the lower dish is more expeditiously dressed. We imagine, therefore, that not less than half of the steam is wasted on the sides of the different stew-pans. Our first attention is therefore called to this circumstance, and we wish to apply the steam more economically and effectually.

We would therefore construct the steam-kitchen in the following manner:

We would make a wooden chest (which we shall call the STEW-CHEST) ABCD (fig. 2.). This should be made of deal, in very narrow slips, not exceeding an inch, that it may not shrink. This should be lined with very thin copper, lead, or even strong tinfoil. This will prevent it from becoming a conductor of heat by soaking with steam. For further security it might be set in another chest, with a space of an inch or two all round, and this space filled with a composition of powdered charcoal and clay. This should be made by first making a mixture of fine potter's clay and water about as thick as poor cream: then as much powdered charcoal must be beat up with this as can be made to stick together. When this is rammed in and dry, it may be hot enough on one side to melt glass, and will not discolour white paper on the other.

This chest must have a cover LMNO, also of wood, having holes in it to receive the stew-pans P, Q, R. Between each pan is a wooden partition, covered on both sides with milled lead or tinfoil. The whole top must be covered with very spongy leather or felt, and made very flat. Each stew-pan must have a bearing or shoulder all round it, by which it is supported, resting on the felt, and lying so true and close that no steam can escape. Some of the pans should be simple, like the pan F, for dressing broths and other liquid dishes. Others should be like E and G, having in the bottom a pretty wide hole H, K, which has a pipe in its upper side, rising about an inch or an inch and half into the stew-pan. The meat is laid on a cullender plate, as in the common way; only there must be no holes in the cullender immediately above the pipe.—These stew-pans must be fitted with covers, or they may have others fitted to their mouths, for warming sauces or other dishes, or stewing greens, and many other subordinate purposes for which they may be fitted.

The main-pipe from the boiler must have branches, (each furnished with a cock), which admit the steam into these divisions. At its first entry none will be condensed on the bottom and sides; but we imagine that these will in two minutes be heated so as to condense no more, or almost nothing. The steam will also quickly condense on the stew-pan, and in half a minute make it boiling hot, so that it will condense no more; all the rest will now apply itself to the meat and to the cover. It may perhaps be advisable to allow the cover to condense steam, and even to waste it. This may be promoted by laying on it flannel soaked in water. Our view in this is to create a demand for steam, and thus produce a current through the stew-pan, which will be applied in its passage to the viands. But we are not certain of the necessity of this. Steam is not like common air of the same temperature, which would glide along the surfaces of bodies, and impart to them a small portion of its heat, and escape with the rest. To produce this effect there must be a current; for air hot enough to melt lead, will not boil water, if it be kept stagnant round the vessel. But steam imparts the whole of its latent heat to any body colder than boiling water, and goes no farther till this body be made boiling hot. It is a most faithful carrier of heat, and will deliver its whole charge to any body that can take it. Therefore, although there were no partitions in the stew-chest, and the steam were admitted at the end next the boiler, if the pan at the farther end be colder than the rest, it will all go thither; and will, in short, communicate to everything impartially according to the demand. If any person has not the confidence in the steam which we express, he may still be certain that there must be a prodigious saving of heat by confining the whole in the stew-chest; and he may make the pans with entire bottoms, and admit the steam into them in the common way, by pipes which come through the sides of the chest and then go into the pan. There will be none lost by condensation on the sides of the chest; and the pans will soon be heated up to the boiling temperature; and hardly any of their heat will be wasted, because the air in the chest will be stagnant. The chief reason for recommending our method is the much greater ease with which the stew-pans can be shifted and cleaned. There will be little difference in the performance.

Nay, even the common steam-kitchen may be prodigiously improved by merely wrapping each pan in three or four folds of coarse dry flannel, or making flannel bags of three or four folds fitted to their shape, which can be put on or removed in a minute. It will also greatly conduce to the good performance to wrap the main steam pipe in the same manner in flannel.

We said that this main-pipe is conducted from the boiler with a gentle ascent. The intention of this is, that the water produced by the unavoidable condensation of the steam may run back into the boiler. But the rapid motion of the steam generally sweeps it up hill, and it runs into the branch-pipes and descends into the stew-pans. Perhaps it would be as well to give the main-