on such matters; but he is a very exceptional man if he has any considerable knowledge of the principles underlying the construction of the most common chronometers, so far as they employ principles not found in common clocks. In illustration of this ignorance of the subject, even among workmen who are thoroughly competent to treat all the disorders of time-pieces properly, and that to the degree of constructing broken or missing parts, or of mending fractures so nicely as to leave no trace of a break, the following instances may be given:
The clock most commonly used by the watch-makers as a "regulator" is one with what is called a "gridiron" pendulum. This consists of nine brass and steel rods, side by side, with their couplings so arranged that, in the changes of temperature, the variation of the brass counteracts that of the steel. Now, it is the fact that a large per cent of these pendulums are simply false "gridirons," while a very small per cent of the watch-makers are able to tell the difference. One might suppose that the running of the clock would reveal this at once. But it should be remembered that the variations of a clock on account of temperature are very slight. An abrupt change of ten degrees maintained through twenty-four hours wall cause a seconds pendulum to vary but a little over two seconds. Bearing in mind, then, that a pendulum may be timed to a mean temperature, and that thus the variations would tend to equalize each other; that, if the clock should thus come within a second a day, it would satisfy most watch-makers; and that as a matter of fact they ordinarily do change their "regulators" several times a year—we have little difficulty in accounting for the wide-spread ignorance of theory among them. To this may be added the consideration that there is no money in knowing about these things, and that to know about them takes time that might become money.
Again, we shall find our jeweler almost equally ignorant of the principles of compensation in the balances of watches. There are but few of them, indeed, who can not tell the genuine balance from the spurious, but there are hundreds of them who could not state, if life depended upon it, why the brass must needs be on the outside and the steel inside in such a balance, or why the converse arrangement would not be equally good, or whether the screws in the rim have anything to do with the compensation.
If, then, those who might be expected to know know so little, where shall we look for information? In view of the general ignorance of this matter and interest in it, a plain and untechnical account of the difficulties in the way of measuring time accurately with clocks or watches, and the progress that has been made in obviating them, may be profitable.
The whole matter of the accurate measurement of time turns, of necessity, on the manner of controlling the rate of escape of the mechanism which indicates the time. Thus far there are only two
