cylindrical boiler, with hemispherical ends. The Staffordshire plate is considered the best for the purpose; and they are generally bent to the circle in the cold state, by the aid of a machine. The joinings are all effected by riveting with short bolts, the plates being previously punched by a machine for the reception of the rivets. The rivet-bolts are inserted and riveted down in the red-hot state, so that, besides the effect of the riveting to draw the contiguous surfaces of the plates into close contact, there is the powerful contraction of the iron while cooling, to produce still more perfect contact. To insure perfect tightness in the joinings or landings, technically so called, the whole of the joints, after they are all riveted up, undergo a process of caulking, which is simply the stamping up of the edges of the plates into intimate contact with the adjoining surface, by means of flat-edged chisels struck with a hammer.

1709. Various rules are given for the dimensions of boilers, corresponding with any given power of engine; but the most natural are those founded upon the extent of surface exposed to the flame of the fire, and the flame and heated air of the flues, taken along with the cubical contents of the boiler. From researches into the relations of the fire-surface and contents, we can form a very correct value of the power of boilers from their capacity alone. According to these, a cylindrical boiler for high-pressure steam should have, in its entire capacity, 12 cubic feet of space for each cubic foot of water boiled off per hour. Now it has just been shown (1703) that a 6-horse power cylinder of 10 inches diameter will consume the steam of 9 cubic feet of water per hour, and 9×12=108 cubic feet for the capacity of boiler; and taking its diameter at 3 feet 4 inches, which gives a sectional area of 8.6 feet; and dividing the cubical capacity of the boiler by this

the length of a cylinder that will yield the capacity required. But as it will have hemispherical ends, the extreme length will be 13 feet nearly, which agrees very well with practice.

is to be set down, should be as near as possible to the engine, so that the pipe conveying the steam should be as short as possible, to prevent all unnecessary condensation of steam.

1711. The furnace of the boiler has undergone many changes, both in form and dimensions, and in the means of supplying it with air for the consumption of the smoke. The latter property, though always desirable, is not of weighty importance in farm-engines, but it is essential that, in every case, the furnace be so constructed as to produce the requisite supply of steam with certainty and despatch. To effect these objects, it is required to expose as much as possible of the bottom of the boiler to the action of the fire, without reducing that portion of the surface which falls to receive the heat from the flame and smoke passing along the flues. In cylindrical boilers, this is effected by making the width of the furnace equal to three-fifths of the diameter of the boiler, the covering of the side flues rising to within 2 inches of the level of the centre of the boiler, that being the line to which the sinking of the water should be restricted, so that the fire or heated air shall never impinge on any part of the surface that is not covered internally with water. The area of firegrate commonly allowed is one superficial foot to each horse power; or, if the furnace is 2 feet wide, the grate-bars should be 3 feet in length. This is too small, unless coal of the very best quality is burned; and, to insure abundance of fire, the bars should be of such length as to give 14 square foot to the horse power nearly, and should be laid with a slope of 1 inch on the foot of length. In front of the bars, a breadth of from 15 to 18 inches is occupied by the dumb-plate or dead-plate, upon which the fresh fuel should always be laid down on its first introduction to the furnace. The mouth of the furnace is closed by a door and frame of cast-iron, leaving the opening of the ash-pit about 3 feet in height.

these by engineers. The following points may, however, be taken as data not easily controverted. The height should not be less than 50 feet; and, if it is desirable to avoid the nuisance arising from smoke, the height should greatly exceed this. The internal sectional area should be as large as may be consistent with economy in the expense, but should never be under 80 square inches for each horse power. Thus a chimney for a 6-horse engine should have its area at the top 80×6= 480 square inches, and ✓ 480-22 inches nearly, the side of the square of the chimney internally; and if circular, the diameter should be 25 inches nearly. The height of the chimney being determined, and also the side of its square externally, the square of its base is found by adding to the length of the side at top the amount of increase arising from the slope given to the sides. The usual slope or battre is inch to a foot; with a height, therefore, of 50 feet, the increase at bottom will be 18 inches on each side; and the walls being one brick, or 5 inches thick, the side of the square at top will be 22+ (5×2)=32 inches, and this added to 18 x2=394+32=5 feet 11 inches, the sides of the square at the bottom.

1713. The regular supply of water to the boiler of a steam-engine is a matter of great importance, as is also the proper method of water-gauge; and it is also of great importance that the water sent into the boiler should have its temperature raised as high as possible before entering. This is effected in a very simple manner, and to a temperature of about 140° by the apparatus; but as the description of all the apparatus by which all these ends are best effected would involve the relation of much minutiæ, it is unnecessary for me to enter into them.

1714. The fuel for the steam-engine is always coal, where it can be procured; but either wood or peat may be used. In the neighbourhood of coal-works, the refuse or culm is always procurable at a low price, and is quite sufficient in point of quality for an engine furnace; 1 cwt. of this culm will, on an average, be required hour that the furnace is burning, for a 6-horse engine. Where the locality involves a distant carriage, it is then

per

better to use the large coal, of which, with due care, cwt. will suffice for an hour. In stoking, the coal should always be laid down on the dead-plate, pushing it forward before putting in a fresh supply. The fire should be kept clear, and always free of the clinkers that may be formed in the bars. On all occasions of stopping, the damper should be let down and the furnace-doors opened, to prevent unnecessary waste of fuel; but at all times, while working, the furnacedoors must be kept shut, unless the supply of steam is found to be too great.

1715. The Horse-wheel.-Until of late years the thrashing-machine was in most cases impelled by horses moving in a circular course; and as this power continues to be employed on the smaller class of farms, it is still of that importance to demand being here brought under notice. Horse-wheels are of various construction, as under-foot and over-head; the former being chiefly used where small powers are required, and the latter where four horses and upwards are employed. In general, in the under-foot wheel, the horses draw by means of trace-chains and swing-tree. In the over-head wheel, of old construction, we also find occasionally the same method of yoking practised; but in all modern over-head wheels the horses draw by a yoke descending over their back, from a horizontal beam placed over-head. Custom seems, as usual, to have produced a preference for this mode of yoking, though there appears good reason for calling its propriety in question, especially if the course has a diameter of 22 feet or upwards. The argument in support of the over-head draught is, that the horse exerts his force in the direction of a tangent, or very nearly so, to the curve in which he walks, or at right angles to the beam by which he draws; while, in the swing-tree draught, his shoulders being considerably more in advance of the point of attachment, his exertions must necessarily tend in a direction that will form an angle more acute than a right angle, but which will vary with the radius of the course. It is quite true that this is the case, and that the horse will draw at a disadvantage, to a certain extent, but the amount of this disadvantage is small. In a 26-feet course, which is a good medium, giving

the over-head draught the full advantage of the right-angle-90°-the other will draw at an angle of about 72° with the radius or beam; and it is easy to show that the amount of disadvantage arising from this is as 21 to 20. If the draught of a horse in a wheel amounts to 170 lbs. under the favourable position, it will require an exertion of 178 lbs. from the same horse, when yoked unfavourably, that is, by a swing-tree. With this disadvantage, which is but small, if we compare the freedom of action and uniformity of the resistance in the case of the swingtree draught, with the constrained action and jolting effects which the horse undergoes in the over-head yoke; and to these, if we add the chances of disadvantage to horses of low stature, being constrained to draw at an unfavourable vertical angle, we shall soon find an amount of disadvantage greater than in the former case. The question is not now of that importance that it once possessed, in consequence of the extensive application of steam; but it appears still to be deserving

of consideration.

1716. In the construction of the horsewheel, also, a question arises as to the diameter of the actual wheel, whether it should be equal in diameter to the entire horse-walk, and work as a spur-wheel, or have a diameter considerably under the former, and be applied as a face or bevelled wheel. It appears to me that the large spur-wheel, of 25 or 30 feet in diameter, has been conceived under a false impression, and that, on principle, its application is erroneous. It is also probable, that a consideration of the overshot water-wheel, which, from its construction, and the nature of the element employed, requires that its power should be given off at or near the extremity of its arms, may have given rise to this formation; but the causes that combine to render this not only advisable, but imperative, in the water-wheel, if every thing is duly considered, do not apply to horse power. When the horsewheel has a diameter larger than the mean diameter of the horse-path, it gives the first motion a higher velocity than that of the moving power, by its more extended radius; and if any inequality occurs in the moving power, it will sensibly affect the succeeding motions. Horses do not

exert a perfectly uniform force when yoked in a wheel-the very act of stepping forth, by removing the exertion from one shoulder to the other, produces small increments and decrements alternately to the power, and these must be communicated to the wheel which extends beyond that point of the lever by which the horse draws. Besides this effect on the machine, it must have an equally bad effect upon the horses; for, in consequence of the construction of the large wheel, and from the yoke being applied to a point where all elasticity is removed, the draught becomes what is termed dead, that is to say, there are no elastic or yielding parts betwixt the power and the first impulse, that might tend to soften the sudden strains that come upon the horses, unless other means are resorted to to produce that result. Wheels of this construction will, therefore, be found more fatiguing to the horses than those of smaller diameter.

1717. Of horse-wheels with a small circle of teeth, the diameter best suited for all purposes, and which might produce a maximum effect, has not yet been defined; but from analogy, and taking into view the properties of the centre of percussion, we may infer that the radius of the segments forming the toothed wheel should be two-thirds of the radius of the beam, measuring to the centre of draught, which may be taken at 11 feet when the course is 26 feet diameter, giving to the toothed segments a diameter of 14 feet 3 inches. The diameter thus found is subject to modification, arising from considerations of strength, and the too great obliquity of the diagonal braces of the wheel, that would follow upon a large diameter. Such considerations will induce a reduction of diameter to 12 or 13 feet, as a good medium size of wheel. The projection of the horse-beams beyond the point of action of the toothed segments, produce that degree of elasticity pointed out in 1716 the absence of which forms a defect in wheels of large diameter.

1718. The horse-wheel represented in fig. 138, is constructed on data derived from the foregoing considerations, and is an elevation of the wheel. It is constructed for four horses,-the course is 26 feet diameter within the pillars, and the wheel is

side towards the stack-yard, but generally the former. In the figure, the barn-wall is marked a, and of the two main pillars, which support the main collar-beam, one is marked b; of the two minor pillars, erected solely for completing the bearings of the roof, one only is seen at c, and d is the floor or horse-walk. The footstep of the horse-wheel is supported on the stone block e, the step being adjustable by four screws, to bring the wheel to the true level; and ƒ is the collar-beam, which is laid upon and bolted to the main pillars, and carries the plummer-block for the head of the central shaft. The sheers g are framed into the collar-beam, and resting on the wall a; and these have two diagonal braces, not seen in the figure, to resist the shake from the action of the wheel upon the pinion of the lying shaft; and the cast-iron bridge i is bolted down upon the sheers carrying the end of the lying shaft. The flanges of the central shaft k, and the one at the top not seen, form the foundation of the wheel; to the latter are bolted the horizontal arms, as well as the horse-beams o, and these are supported by the diagonals n n, seated in and drawbolted to the flange k. The horizontal braces pp, of the horse-beams are framed into the ends of the horizontal arms, and secured with cast-iron knee-plates at their junction with the wheel and with the horse-beams. The yoke-bars qq, are made of hard wood, tapering towards the lower end, strongly bolted to the horse-beams,

and are each mounted with an iron pulley near the lower end, over which the draught-chain passes, the height of which from the horse-path should be 3 feet 6 inches, liable to slight variation, arising from the stature of the horse that is to be yoked into it. The wheel r is now always made of cast-iron, in segments, and, when the wheel is very carefully made, are fitted and bolted to a bed-plate of the same material, previously bolted to the arms and horse-beams, and is 13 feet diameter. The horse-wheel pinion s is mounted on the lying shaft st, whose inward bearing is upon the barn wall, in an opening formed for the purpose, and this shaft carries the spur-wheel u inside the barn. The calculations of this machine would stand thus:The horses will walk the course three times in a minute, being at the rate nearly of 2 miles per hour, the lying shaft st will make 11 revolutions for one of the wheel, or 33 per minute, and if the drum pinion, which is driven by the wheel u, is made 8.6 inches diameter, the wheel being 7 feet, would give the drum 320 revolutions per minute; a fair average velocity for a 4horse machine, which can be increased by a

quicker step of the horses, say to 24 miles per hour, which would give 340 revolutions per minute to the drum.

1719. Some horses, when yoked in a wheel, are observed, after a short practice, to take advantage of lagging back, and allowing those who are more willing to

take the heavy end of the work. To counteract this, methods have been adopted to make the horses draw by chains, so arranged as to make them work against each other in pairs; or make any number of them draw from a ring-chain common to the whole. Another method was to make each horse draw against a certain weight suspended over pulleys; but all these have their imperfections in one way or another. A new and more perfect arrangement of the ring-chain was introduced by Mr Christie, Rhynd, Fifeshire, which received the approbation of the Highland and Agricultural Society of Scotland. This arrangement is exhibited in the figure by the doted lines under the horse-beams; but I do not enter into the details of the arrangement. Suffice it for me to say that the principle of the arrangement is, that the ring-chain forms a figure of so many equal sides or angles as there are horses in the wheel, and that the angles always remain equal. The defects of any other method of this kind which has been tried being, that the angles vary according to the sluggish or active temper of the horses.

1720. A method of equalising the resistance to the shoulders of each individual horse has been long practised, and which, from its simplicity as well as its beneficial effects upon the horses, is deserving of general adoption. The apparatus consists of an iron lever with equal arms, which is suspended upon a bolt by a perforation through the centre of the lever forming the fulcrum, and the ends are formed into hooks to which the draught-chains are attached. Fig. 139 represents the application of this to the horse-beam, wherein a is a part of the beam, and bb the yoketrees; c is the lever above described, suspended upon the back of the horse-beam; dd the draught-chains, hooked to the lever, and passed under the pulleys of the yoketrees, beyond which the horse is yoked to the extremity of the chains. The advantages of this mode of yoking will at once be obvious; for suppose that, from inadvertence, the horse may have been unequally yoked, whenever he exerts his force, the chain that had been yoked short -suppose it to be the left shoulder-will

1721. The Water-wheel.-Water, when it can be commanded, is the cheapest and most uniform of all powers; and on many farms it might be commanded by carefully collecting and storing in a dam. Waterwheels have been commonly treated as of two kinds; but, with great deference, I conceive they may be classed under two heads. The under-shot, or open float-board wheel, which can only be advantageously employed where the supply of water is considerable and the fall low; it can therefore rarely answer for farm purposes, and need not be discussed. The second is the bucket-wheel, which may be over-shot or breast, according to the height of the fall. It is this wheel that is adopted in all cases where water is scarce or valuable, and the fall amounting to 6 or 7 feet or more, though it is sometimes employed with even less fall than 6 feet. It is the most effective mode of employing water, except where the fall is excessively high, or ex