The same remark applies to some of the portable or agricultural boilers which have exploded, such as those shown in Fig. 30 and Fig. 31, No. 43, 1868, and No. 12, 1869.

Much mischief is often caused by bad imitation of well planned boilers. Thus in boilers of the Cornish form, the ends are made sometimes so rigid as to give no allowance for the expansion of the tube, and the result is such continued strain as to cause constant leaking and the consequent risk of fracture. In furnace boilers the tops of the crowns of the inside tubes are often made flat, as in Fig. 37, instead of being domed; or the inside tube is of undue size, as in Fig. 21, No. 23, 1870, see page 73. Furnace boilers have been made with the omission of the stays that are so peculiarly necessary in that form, whereby both ends have been left free to bulge outwards with the pressure, as in Fig. 32.

Cornish boilers are often altered to the plain cylindrical form, without compensation being made for the loss of strength caused by the removal of the tube; this has led to such explosions as shown in Fig. 33, No. 47, 1869, where two tubes where taken out, and Fig. 34, No. 42, 1867, where one tube was taken out. One of the most frequent and serious causes of loss of strength is the repairing of externally fired boilers. Not only are the patches sometimes only bolted on in a temporary manner, as in Fig. 35, No. 29, 1869, but even where they are rivetted on there is an entire want of bond or crossed joint, as in the case of the exploded boilers shown in Fig. 15, Fig. 16, and Fig. 17, see page 70 and 71, No. 10, 1869, and Fig. 36.

An attempt is made in Fig. 38 to show the effect of wear and tear of boiler plate in an ordinary upright furnace boiler, such as is shown in Fig. 37. The external surface is exposed to intense heat and consequently expands, while the internal surface is kept cool by contact with the water and expands to a much less degree. The continued repetition of this process produces the same effect of cracking the surface as that seen in the anvil blocks of steam hammers; and the strength of the plate is reduced in proportion to the destruction of the continuity of its surface. The deleterious effect of this process is much increased if the boiler is subject to alternate heating and streams of cold air on opening the fire-doors. To avoid it the flame should have room to spread over as large a surface as possible, without impinging on one particular point, and the firing should be as regular as possible; and hence the greater freedom from injury in boilers mechanically fired or heated by gas. The above action is quite distinct from the overheating of the plates that occurs when no water is in contact with them, which simply softens them and reduces their strength, as in Fig. 39. It is believed that many boilers suffer from overheating without being short of water: and an attempt is made to show this action in such an upright boiler as is represented in Fig. 37, page 78, by the enlarged section of the side shown in Fig. 40. The flame is shown impinging on a limited surface, as before, and the steam rises so rapidly from the inner surface as to maintain a continuous stratum of steam between the iron and the water, and the plate consequently becomes overheated at that part. When the intense flame subsides by an alteration of the working of the furnace, the stream of steam diminishes, and the water returns and suddenly cools and contracts the plate, but often not before it has commenced to get out of shape. This has perhaps led to the explosion shown in Fig. 59, No. 37, 1868, page 82. The same thing may happen to the crowns of tubes of internally fired boilers when over fired, as in Fig. 41. Success has attended the use of internal linings to boilers, arranged so as to ensure a rapid circulation over the most heated parts, and also to catch all the mud and loose scale.

In order to enable boiler minders to make proper periodical examinations, it is necessary that care should be taken to arrange both the boilers and the flues with that view; and this can be done without materially injuring the efficiency of the boiler. Ordinary plain cylindrical boilers can be entered easily, as in Fig. 42; and although the small spaces between the tubes and the shells of Cornish and Lancashire Boilers, as shown in Fig. 43, render the complete examination troublesome, there is no difficulty in seeing those parts most likely to need examination, such as the crowns of the tubes and the end plates and angle iron. It is in the external flues that greater accommodation is needed, as in many cases these are so narrow that the boiler is quite inaccessible without pulling down the brickwork, as in Fig. 44 and Fig. 45. The loss of heating effect caused by the use of wider flues is so little, that it is far outweighed by the greater security obtained from the more efficient examination that is thereby rendered practicable. The flues of the plain cylindrical boiler are easily made wide enough for a man to pass through them. The flues of Cornish and Lancashire boilers should be made as shown in Fig. 46 and Fig. 47, so that a man can enter them without such inconvenience as in Fig. 48. One point of danger being the use of wide mid-feather walls, on which corrosion is apt to take place, these should be narrowed and the weight of the boiler supported on side brackets; the top of the mid-feather and side walls can then be constructed with sight holes as at A A in Fig. 49 and Fig. 50, so as to give the means of examining the plates near each seam by simply removing loose bricks.

The explosions of fourteen Domestic or Heating-Apparatus Boilers are included in the list of explosions, Table III; and some notice is required to be taken of these, because they have led to the loss of the lives of those who could not be expected to know their construction or how to guard against accident; and as these boilers are seldom seen or examined after they are once set, they should be the more carefully constructed. In one or two cases these boilers were of a rectangular shape, as in Fig. 51, No. 41, 1868, ill adapted to bear internal pressure, and yet placed in connection with cisterns in the roofs of lofty houses, so as to expose them to a hydrostatic pressure almost up to their bursting strength without any addition of steam pressure. The most usual cause of explosion is the lighting of the fire during frosty weather in a house that has been left vacant, so that steam pressure accumulates in the boiler whilst the exit is frozen up, as was the case in Fig. 52, No. 6, 1870.

The cast-iron boilers commonly used, Fig. 53, end of 1869, are capable of bearing but little pressure; and the wrought iron boilers, as in Fig. 54, No. 7, 1870, are found often so badly welded as to be but little stronger; but even if they were as strong as they could be made, the stoppage of the pipes by ice would lead to explosion. Steam pressure may be guarded against by a safety valve; but as this may become set fast in a little time, it would be far better to avoid all chance of steam accumulation by such an arrangement as that shown in Fig. 55, where the circulating boiler is placed within an open-topped boiler behind the kitchen fire, and only receives its heat through the hot water surrounding it, and therefore cannot itself become sufficiently hot to generate steam.

A few remarks may be useful as to those faults arising in working which fall under the department of the boiler minders. Not a few of the explosions during the last four years have occurred from acts of simple carelessness, such as where a blow-off pipe was left open, so that the boiler was nearly emptied of water while at work; or in another case where two boilers were fed at the same time through a common pipe without a back valve, and the water from one "kicked" over into the other. Undue pressure has been allowed to accumulate by safety valves being tied down, as in the agricultural boiler, Fig. 56, No. 16, 1867; or by an extra weight being put upon the safety valve, as in an instance where three bricks were fastened to the lever and the fires were lighted earlier than usual, under the idea that an accumulation of steam could be raised during the night to make a good start in the morning. Another explosion was caused by working a boiler at more than three times its proper pressure to meet a temporary emergency. In not a few cases of explosion there was no pressure gauge on the boiler, or the gauge was out of repair in consequence of being placed on the steam pipe, so that it vibrated with every stroke of the engine; as in the examples shown in Figs. 24 and Fig. 16, No. 35, 1868, No. 32, 1870, pages 70 and 74.

Corrosion has been the direct cause of many of the explosions. In one or two cases the corrosion was known to exist, but the renewal of the boiler was too long delayed, as in Fig. 57, No. 8, 1869, in others it took both owners and minders by surprise, as in Fig. 1, No. 12, 1870, page 63. It is said that to produce rapid rusting of iron there must be present oxygen, water, and carbonic acid; and as all these are present in a boiler flue when there are leaks, it is not surprising that so many cases occur of explosions from corrosion.

Much mischief is often done by the injudicious use of compositions in the boiler that are designed to prevent incrustation, especially where there is no blow-off cock or where its use is neglected. A hard deposit on the boiler plates is, in the writer's opinion, not so injurious as the soft and muddy deposit produced by the use of such compositions. A hard scale is equivalent to thickening the plate; and although this is sufficiently mischievous, the injury to the plates is much more rapid when a thicker but spongy deposit entirely prevents contact of the water and impedes the transmission of the heat. An attempt to illustrate this is given in Fig. 58, which is an enlarged view of a portion of such a boiler as is shown in Fig. 37. The money spent in boiler compositions would be better applied in securing a supply of proper water, or in filtering and purifying the water before it enters the boiler.

The writer has had to mention only faults in boilers; but it is not to be inferred that all boilers are working in actual danger. A very small percentage perhaps are so; but without periodical examination no one can feel sure of the condition of any boiler. It is not likely that explosions in future will be from exactly the same causes as those now described, because the known faults will be avoided. For instance no new Balloon, Wagon, or Butterley boilers are now made; and the peculiar faults and the weakness of the tubes in Cornish and others of the better classes of boilers are now so well known as to be generally avoided; and as information spreads, many evils will become things of the past.

As periodical examination has been so strongly advocated, it might seem natural to desire that it should be enforced by government authority; but this is by no means recommended. A select parliamentary committee has been recently investigating the subject, with a view to ascertain whether that would be desirable, but has adjourned for the session without coming to any decision on this point. Even if a perfect system of government inspection could be contrived and perfectly administered, it would have the effect of taking the responsibility from the owners, who are the natural guardians of the safety of their boilers. Although the loss of 70 lives per annum by boiler explosions is sufficiently deplorable, the deaths by railway accidents are more than three times that number; yet very little inspection of railways is held to be necessary, and that inspection takes place chiefly before the commencement of working or after accidents. A coercive system may introduce more evils than it cures, especially as at present so much difference of opinion exists respecting the causes of boiler explosions. In the opinion of the writer, far more real good arises from the calm discussion of the facts and from the spread of correct information by such societies as this Institution, than from enforcing by law any action which is not perhaps believed by the majority of steam users to be at all necessary or useful. It has been at times suggested to increase the power and responsibility of coroners in holding inquests upon those killed by boiler explosions, by requiring them to obtain scientific evidence and to insist that the causes of the explosions shall be added to the verdicts of juries. But it is believed that this would only encumber an important institution, because a jury who might well decide whether a person had been killed by any criminal carelessness would not be a suitable tribunal to decide between possibly conflicting scientific evidence; and also, as an inquest may result in a verdict of manslaughter, the eliciting of information on such an occasion is checked by the natural fear of inadvertently involving some one in so serious a charge. The public at large, and steam users generally, would gain more information and guidance from the scientific evidence itself than from the verdict of a coroner's jury; and it is believed much good has resulted in preventing locomotive boiler explosions by publishing the reports of the government inspecting engineers, who have gained their knowledge of the facts in conversation with all those concerned, and have added recommendations which have been promptly acted upon.

The writer's object has been that the boilers found most convenient and best suited for the different purposes for which they are used should be made to work with safety, rather than that reliance should be placed upon the qualities of any particular kind of boiler or fittings. No form of boiler at present admits of absolute reliance upon its freedom from risk.

The following general conclusions appear to arise from the consideration of the records of boiler explosions.

1. That the force accumulated in an ordinary boiler is enough to account for the violence of an explosion.
2. That no form of boiler, however well constructed and fitted, is free from the liability to explosion, if allowed to get out of order; and that boilers which bear the hydraulic test may still be dangerous.
3. That the condition of a boiler can be satisfactorily ascertained only by periodical examinations, and that no boiler should work without being thoroughly examined at short intervals.
4. That the cost of periodical examination is so little as to be far outweighed by the greater security obtained; and that the settings of all boilers should be constructed with a view to facilitate examination.
5. That the surest way to make systematic examination general is to spread as widely as possible correct information as to the facts and ascertained causes of boiler explosions, and to inform boiler owners and minders what dangers to guard against; and that this is preferable, and more likely to lessen explosions than enforcing any system of inspection by legal enactment.

TABLE I.

TABLE II.

TABLE III.

TABLE IV.

BOILER EXPLOSIONS IN 1866.

No. 1. Nottingham.

January 1st.

none injured.

Locomotive, 110 lbs., standing with steam up near a platform. All but the fire box was blown away, the main portion being thrown a distance of 400 yards. The first rent took place at a longitudinal seam of the barrel where grooving had gone on very rapidly, which was not discovered when examined and tested a short time before.

No. 2. Walsall. (Fig. 1.)

January 2nd.

2 injured.