Fig. 2 represents the wheels and pinions in this clock. A is the first or great wheel; it contains 120 teeth, and turns round in 12 hours. On its axis is the plate on which the 12 hours above-mentioned are engraved.

This plate is not fixed on the axis, but only put tight upon a round part thereof, so that any hour, or part of an hour, may be set to the top of the fixed index A, without affecting the motion. of the wheel. For this purpose, twelve small holes are drilled through the plate, one at each hour, among the quarter divisions; and, by putting a pin into any hole in view, the plate may be set, without affecting any part of the wheel-work. This great wheel A, of 120 teeth, turns a pinion B of 10 leaves round in an hour; and the minute-hand B (Fig. 1) is on the axis of this pinion, the end of the axis not being square, but round; that the minute-hand may be turned occasionally upon it, without affecting any part of the movement. On the axis of the pinion B is a wheel C of 120 teeth, turning round in an hour, and turning a pinion D of 6 leaves in 3 minutes; for 3 minutes is a 20th part of an hour, and 6 is the 20th part of 120. On the axis of this pinion is a wheel E of 90 teeth, going round in 3 minutes, and keeping a pendulum in motion that vibrates seconds, by pallets, as in a common clock, where the pendulum wheel has only 30 teeth, and goes round in a minute. But, as this wheel goes round only in 3 minutes, if we want it to show the seconds, a thin plate must be divided into 3 times 60, or 180 equal parts, and numbered 10, 20, 30, 40, 50, 60; 10, 20, 30, 40, 50, 60; 10, 20, 30, 40, 50, 60; and fixed upon the same axis with the wheel of 90 teeth, so near the back of the dial-plate, as only to turn round without touching it; and these divisions will show the seconds, through the opening e f g h in the dialplate, as they slide gradually round below the point of the fixed fleur-de-lis C.

As the great wheel A, and pulley on its axis over which the cord goes (as in a common 30 hour clock) turns round only once in 12 hours. this clock will go a week with a cord of common length, and always have the true hour, or part of that hour, in sight at the upper end of the fixed index A on the dial-plate. These are two advantages it has beyond Dr. Franklin's clock; but it has two disadvantages of which his clock is free. For, in this, although the 12-hour wheel turns the minute-index B, yet,

if that index be turned by hand, to set it to the proper minute for any time, it will not move the 12 hour plate to set the corresponding part of the hour even with the top of the index A; and therefore, after having set the minute-index B right by hand, the hour-plate must be set right by means of a pin put into the small hole in the plate just below the hour. 'Tis true, there is no great matter in this; but I have some suspicion that the pendulum wheel E having 90 teeth instead of the common number 30, may be some disadvantage to the 'scapement, on account of the smallness of the teeth; and 'tis certain, that it will cause the pendulum-ball to describe but small arcs in its vibrations. Indeed some men of science think small arcs are best; but if they really are, I must confess myself ignorant of the reason. For, whether the ball describes a large or a small arc, if the arc be nearly cycloidal, the vibrations will be performed in equal times; the time then depending entirely on the length of the pendulum-rod, not on the length of the arc the ball describes. The larger the arc is, the greater is the momentum of the ball; and the greater the momentum is, the less will the times of the vibrations be affected by any unequal impulse of the pendulum wheel upon the pallets.

But the worst thing about this clock (and what every one will allow to be a disadvantage) is, that the weight of the flat ring on which the seconds are engraved, will load the pivots of the axis of the pendulum-wheel with a great deal of friction, which ought by all possible means to be avoided; and yet I have seen one of these clocks (lately made) that goes very well, notwithstanding the weight of this ring. For my own part, I think it might be quite left out; for they are of very little use in common clocks, not made for Astronomical observations, and table-clocks never have them." (Vide Ferguson's "Select Mechanical Exercises." Lond., 1773, 1st Edit., pp. 4—11).

In his recently discovered "Common Place Book," Ferguson says, "I showed this Clock to Dr. Franklin, who approved very much of the alteration." "Common Place Book," p. 98 (Col. Lib. Edin.)

MANUSCRIPT Copy of FERGUSON'S ASTRONOMY.-Under date 1757, page 224, allusion is made to three manuscript books, in Ferguson's autograph, which were discovered near the end of the

236

year 1865. The second one of the three, in the order of date,—is
a quarto copy of his Astronomy, carefully and beautifully written
-an exact copy of his quarto Astronomy then in print. It is illus-
trated by engravings same as in the printed copies, with four ad-
ditional ones, done by him in pen and ink,-viz. A view of the
Dial-face of an Astronomical Clock ;-Plan of the wheel-work of
his large Orrery;—and two projections of the Solar Eclipses in
1761 and 1764. The title-page of this volume is the same as
in the printed copy (see page 214), and has the date MDCCLVIII
lettered with the pen at the foot of the page. It has been sup-
posed that this MS. is the original from which the first edition
of his Astronomy was printed (in 1756)—if we admit this, then
Ferguson must have added the title-page to it in 1758. We
are of opinion that the MS. was written in 1758, from the
second and corrected edition of 1757. The MS. is very
carefully and neatly written, free from erasures and inter-
lineations, such as are usually to be found in copy for the press
-too clean to have passed through the printing office.
We can form no opinion as to Ferguson's motive for writing
this MS.

The manuscript volume is a thick quarto of 328 pages, followed by 24 pages of printed matter, and Tables by Tobias Mayer. It is full bound in calf, with gilded edges-in excellent preservation-size, 103 by 8 inches, and 13 inch thicknow in College Library, Edinburgh, with the other two volumes. For an account of the third MS. Vol., see date 1776.

The year 1758 opens with a long and interesting letter from Ferguson to his much esteemed friend, the Rev. Mr. Irvine of Elgin. The year closed on Ferguson with the sad intelligence of his friend's death. The Rev. Alexander Irvine died at Elgin on the 22d December, 1758, in the 59th year of his age, and 34th of his ministry. (See also note 69).

1759.

BIRTH JOHN FERGUSON born. Our first memorandum of Ferguson, for the year 1759, refers to an "interesting domestic occurrence," viz. the birth of a son, baptized John. The following extract of the birth is in the autograph of Ferguson, taken from a small pocket Bible once belonging to the Ferguson family, and to which we have been indebted for a record of all

the births in his family. As previously noted, this Bible is now the property of Dr. George, Surgeon, Keith.

"JOHN; (born), Tuesday, 27th Feby. 1759, N.S."

It is very probable that Ferguson's son, John, was named after either his grandfather, whose name was John, or in compliment to his uncle John, who, at the time, was residing in the parish of Keith, Banffshire. Ferguson had now a family of one daughter and three sons, viz. Agnes, James, Murdoch, and John; there was no after addition to his family.222

PYROMETER. In 1759, Ferguson invented and made a Pyrometer, for showing the expansion of metals by heat. The following description, and annexed figure of this instrument are from Ferguson's " Lectures on Select Subjects."

"AA is a flat piece of mahogany, in which are fixed four brass studs B, C, D, L; and two pins, one at F, and the other at H. On the pin F turns the crooked index E I, and upon the pin H, the straight index G K, against which, a piece of watch-spring R bears gently, and so presses it toward the beginning of the scale M N, over which the point of that index This scale is divided into inches and tenth parts of an inch; the first inch is marked 1000, the second 2000, and so on. A bar of metal O, is laid into notches in the top of the studs C and D; one end of the bar bearing against the adjusting screw P, and the other end against the crooked index E I, at a 20th 222 For notices of John Ferguson, see date 1773; and the Appendix.

moves.

part of its length from its centre of motion F. Now it is plain, that however much the bar O lengthens, it will move that part of the index E I, against which it bears, just as far; but the crooked end of the same index, near H, being 20 times as far from the centre of motion F, as the point is against which the bar bears, it will move 20 times as far as the bar lengthens. And as this crooked end bears against the index G K at only a 20th part of the whole length G S from its centre of motion H, the point S will move through 20 times the space that the point of bearing near H does. Hence, as 20 multiplied by 20 produces 400, it is evident that if the bar lengthens but a 400th part of an inch, the point S will move a whole inch on the scale ; and as every inch is divided into 10 equal parts, if the bar lengthen but the 10th part of the 400th part of inch, which is only the 4000th part of an inch, the point S will move the tenth part of an inch, which is very perceptible.

To find how much the bar lengthens by heat, first lay it cold into the notches of the studs, and turn the adjusting screw P until the spring R brings the point S of the index G K to the beginning of the divisions of the scale M; then, without altering the screw any farther, take off the bar and rub it with a dry woollen cloth till it feel warm; and then, laying it on where it was, observe how far it pushes the point S upon the scale by means of the crooked index EI; and the point S will show exactly how much the bar has lengthened by the heat of rubbing. As the bar cools, the spring R bearing against the index K G, will cause its point S to move gradually back towards M, in the scale; and when the bar is quite cold, the index will rest at M, where it was before the bar was made warm by rubbing. The indexes have small rollers under them at I and K; which, by turning round on the smooth wood, as the indexes move, make their motion the easier, by taking off a great part of the friction, which would otherwise be on the pins F and H, and of the points of the indexes themselves on the wood." 223 (Vide Ferguson's "Lectures on Select Subjects." Lond. 1760, p. 246).

223 This Pyrometer is "defective both in principle and execution;" it was found so shortly after its introduction. Mr. Charles F. Parkington, who edited an edition of Ferguson's Mechanics, in 1825, has a foot-note regarding this Pyrometer, in which he says, The Pyrometer here described has been justly considered as defective both in principle and execution. If it be merely intended to illustrate the expansion of metals by heat, the fact may be shown by fitting a piece of large (thick) wire into a hole, and it will be found that after it has been