days after the accident) to have been recently put in new. It is perfectly free from flaw or patch, and would certainly have run at least 100,000 miles. The same may also be said with respect to the outer shell, which is nearly of the original thickness. The engine had been in the repairing shop the three months previous to the accident; and the iron fire-box stays, about which so much has been said, were tested by the hammer in the usual way, and were considered, both by the workman and the foreman, Wheatley, to be all sound. When originally made, they were 11ths in diameter, and were equal to a strain of at least ten times the force they had to sustain. With the exception of one stay, which was on the top row, the one most reduced from oxidation was half-inch diameter; and supposing the hold on the copper box to have been good, it was capable of resisting a strain of rather more than 63 times the working pressure, equal, say, to 390 lbs. per square inch. The only point therefore which could admit of doubt as to the safety of the boiler, was with respect to the hold which the stays might have in the copper box; but it appears, from experiments which I have since made, and which are about to be repeated by Mr. Fairbairn, that from the force required to pull some of the old stays out of a copper plate similar to the fire-box, into which they had been screwed by the old threads only, and not riveted, the boiler could not have burst under a pressure of less than 300 lbs. per square inch. One of the old stays, which had had the thread partially damaged from being ripped out of the copper box by the explosion, was screwed by hand into a copper plate, by the old thread, to a depth equal to the thickness of the fire-box plate, but not riveted, and it required a dead weight of 8204 lbs. to pull it out; and as each stay has to support a surface of 5 inches × 53 inches, say 27 square inches only, it follows that a pressure of 8204÷27=303.85 lbs. per square inch would have been required to strip it.

Another stay, which had not been stripped by the explosion, but which was screwed out of the old box, was similarly treated, and required a force of 9184 lbs. to strip it, equal to 340 lbs. per square inch.'

Since the experiments here referred to were made, I have repeated them with great care; and taking into account the

·

tensile strength of the stays, in their corroded state, and of the side of the fire-box, which to appearance was the first to give way, I find that a force of 380 lbs. upon the square inch would be required to effect rupture; and the results of the experiments on the resistance of stays screwed into the copper fire-box fully confirm those already made by Mr. Ramsbottom. Assuming therefore that the ends of the screws were riveted, and sound in other respects, we may reasonably conclude that a strain of not less than 450 to 500 lbs. upon the square inch would be

FIG. 2.

A

B

required to strip the screws, or tear the stays themselves asunder. I have founded these facts upon the experiment of the resisting powers of the iron stay screwed into a portion of the copper cut out of the ruptured fire-box, and another experiment of a similar stay first screwed and then riveted, as shown in the annexed sketch.

The stay marked A, 3ths of an inch in diameter, in the first experiment required a force of 18,260 lbs.=8·1 tons to strip the screw, and draw it out of the copper; and the stay B, of exactly the same dimensions, but riveted over the end, required a force

=

of 24,140 lbs. 107 tons before it was dislodged.* Taking therefore the mean of those experiments, including those of Mr. Ramsbottom, and we arrive at the results given above, namely, a resisting power of 785 lbs. on the square inch, to burst or produce fracture in the stays and side of the fire-box. In locomotive engines of more recent construction, where the stays are thicker and formed into squares of 4 to 4 inches, the resisting powers will probably be increased to 850 or 900 lbs. on the square inch, that is, 7 or 8 times the working pressure.

On a careful examination of the fire-box and every other part of the boiler, it was found that the stays and copper were perfect, and that they were able to sustain a pressure much exceeding 207 lbs. upon the square inch, as given in the following table.

In these experiments, the top of the fire-box sank a little, owing to the breakage of a bolt of one of the cross-bars; but the fire-box stays were quite perfect, and to every appearance would have sustained nearly double that pressure. If the firebox stays had been new and the top well stayed, it is more than probable that a force from 800 to 900 lbs. on the square inch would have been required to cause rupture.

As much stress has been laid upon the weakness of the stays which unite the flat surface of the boiler to the sides of the fire-box, the following experiments clearly indicate that the fire-box stays are not the weakest parts of a locomotive boiler, and that we have more to fear from the top of the furnace, which under severe pressure is almost invariably the first to give way. Great care should therefore be observed in the construction of this part, as the cross-beams should not only be strong, but the bolts by which the crown of the fire-box is suspended should also be of equal strength, in order that no discrepancy should exist, and that all the parts should be proportioned to a resisting force of at least 500 lbs. on the square inch.

Finding our knowledge with regard to the power of resistance of locomotive boilers to strain exceedingly imperfect, I availed myself of the present opportunity to determine by actual expe

* Vide Experiments, Note, p. 338.

riment the laws on which these powers are founded; and for this purpose the Directors of the London and North-Western Railway Company placed in my hands an engine of the same age, constructed by the same makers, and in every respect a fac simile of that which exploded. This engine was subjected to hydraulic pressure as follows::

Experiment made May 4, 1853, to determine the Resisting Powers of the Fire-box and Exterior Shell of No. 2 Engine on the Eastern Division of the London and North-Western Railway.

In this experiment, the boiler was furnished with a valve, A, of exactly 1-inch area, and a lever of the annexed dimensions, as per sketch, fig. 3. This lever, 15: 1, gave as the weight upon

From the above, it is evident that the boiler which led to these experiments could not have burst under a pressure of less than 300 to 350 lbs. upon the square inch, as the failure of a single bolt in one of the cross-bearers above the fire-box, under a pressure of 207 lbs. on the square inch, was not the measure of its strength, but one of those accidental circumstances which is calculated to weaken, but not absolutely destroy, its ultimate powers of resistance. I have been led to this conclusion from the fact of finding the upper part of the fire-box in every re