of the governor in flour-mills, but has not succeeded. The earliest publication he has as yet found is the specification of Thomas Mead's patent of 1787 (No. 1628), ‘Regulator for Wind and other Mills.’ A reader of this specification must certainly come to the conclusion that Mead was (or that he believed himself to be) the inventor of the implement, and not merely the suggester of its application to mills.
The writer has not been able to ascertain when Watt first applied the governor to his steam engines. Farey in his book on the steam engine, published in 1827, says, at p. 437: ‘In the years 1784 and 1785 Messrs. Boulton and Watt made several rotative engines … One of the first of these was set up at Mr. Whitbread's brewery in Chiswell Street … Mr. Whitbread's engine was set to work in 1785. In their general appearance these engines were very much like that represented in plate xi, having the same kind of parallel motion, sun and planet wheels, and governor.’ If this statement about the governor be correct, then Watt was using governors three years before the date of Mead's patent. It must, however, be remembered that Farey was writing between forty and fifty years after the period under consideration. At p. 435 Farey, describing the governor, says: ‘It was on the principle which had been previously used in wind and water mills.’
Having regard to Watt's silence on the question of the governor, to the fact that he did not patent it, nor even its application to the steam engine; having regard also to the statements (unsupported, it is true) of many writers that the implement was used as applied to flour-mills before the date of its application by Watt to the steam engine, it appears the probabilities are largely against Watt being the inventor of the governor. Watt applied it to the steam engine, and devised a particular kind of valve, the ‘throttle valve,’ which, being balanced on each side of a central spindle, was capable of being moved by a comparatively weak agent, such as the centrifugal governor.
There is another very useful adjunct to the steam engine—the indicator—the whole invention of which is also commonly but erroneously attributed to Watt. The indicator is an implement by which a pencil, controlled by a spring, is made to move forwards or backwards in accordance with the pressure prevailing in the engine cylinder at any moment, while a card, or nowadays a paper, is caused to traverse transversely to the movement of the pencil, and thus there is drawn on the card by the pencil, a diagram, which shows and records the varying pressures in the cylinder at all parts of the stroke of the piston, and thus enables the work done on the piston and the quantity of steam used to be determined. No doubt this implement has been of the greatest value in the developing of the various improvements which have been made, and are still going on, in the steam engines. Watt's share in the invention of the indicator was confined to the simple and comparatively useless vertical motion of the pencil in accordance with the pressure in the cylinder, and was a mere substitution for a glass tube containing mercury; the transverse motion, by which alone the diagram could be obtained, was due, it is believed, to the genius of John Southern, one of Boulton & Watt's assistants. So long as steam engines were used only for raising water, it was extremely easy to state the amount of work they were doing and to compare one engine with another. Thus, if engine A were raising a hundred gallons per minute from a depth of a hundred fathoms, and engine B were raising two hundred gallons from the same depth, B was obviously doing double the work of A; but when engines were employed to drive mill-work, there was no such record of ‘work done’ obtainable; it became necessary, therefore, to devise some standard. Prior to the use of the steam engine rotary motion on the large scale was derived from water-wheels, and on a small scale from windmills or from horse-wheels. Watt therefore, following Savery, determined that the horse-power should be the standard. Savery had come to the conclusion that it would need a stock of three horses to provide one always at work. He does not appear to have determined the ‘work’ of a horse; but if there were required four horses at work to drive, say, a pump, and Savery made an engine competent to do the same duty, he called that a 12-horse engine, as it was equivalent to the twelve horses that needed to be kept to provide four horses always at work. Watt, however, did not follow Savery in his rule-of-thumb determination, nor did he credit his engine with the idle horses. He satisfied himself that an average horse could continue to work for several hours when exerting himself to such an extent as would raise 1 cwt. to a height of 196 feet in a minute, equal to 22,000 lb. one foot high. In order that a purchaser of one of his engines should have no ground of complaint, he proportioned these machines so that for each of his horse-powers they should raise half as much again, or 33,000 lb. one foot high per minute. As regards the confusion into which the ques-
