My fourth engine, of the same size, had been spoiled for practical use by having the upper half of the cylinder and steam-chest planed off, to show the cylinder and valves in section. It was belted from the large engine to run very slowly, and thus exhibited the valves and gear in motion to the end of the exposition. Mr. Whitworth wanted his friend Mr. Owen to purchase this model for the South Kensington Museum, but it appeared to Mr. Owen that Mr. Whitworth ought to present it to the museum. This I learned from Mr. Hoyle. What was finally done with it I have forgotten, if, indeed, I ever knew. My fifth engine, of the same size, 6×12 inches, I got up to show what the capabilities of high speed really were, so far as smooth and safe running were concerned. The reciprocating parts, which weighed altogether only 40 pounds, were exactly balanced. I did this by rolling the crank-disk on a boring-table, with 40 pounds hung on the crank-pin, and cutting out the lead from the hollow disk opposite the pin, where I had purposely put it in somewhat in excess, until the pin came down to the horizontal position. This brought the inertia of the reciprocating parts of the engine, at every point in the revolution, into equilibrium with the horizontal component of the centrifugal force of the revolving counterweight. The vertical component of this force, or rather its upward stress, for downward it would be resisted by the whole mass of the earth, remained to be dealt with. To prevent the whole engine from being lifted at the crank end by this stress at every revolution might have been accomplished by putting on a heavy fly-wheel; but for my use I wanted a very small one. The fly-wheel I put on the shaft was a solid disk, 18 inches in diameter and ¹⁄₂ inch thick, with a rim 1 inch square. The bed of the engine I filled with lead, and set it on a block of Caen stone 3 feet thick and wide and 5 feet long. To this stone it was firmly bolted, and I was ready for business. The governor was speeded to hold the engine at 500 turns per minute. As it might be difficult for some persons to count this speed, I put a little pinion on the end of the shaft, engaging with a larger wheel, one to ten. Fifty revolutions per minute could be accurately counted, and the speed was put beyond dispute. I was guilty of one oversight: I did not protect this gear. A French gentleman had the skirt of his frock-coat caught in it, and I thought it never would be got out. The engine had been running only two or three days, but the speed being then well established, I took off the gear. I ought to have protected it instead, and have had it to substantiate the big story I am going to tell, but it never occurred to me.