We have generally made a practice, upon the completion of the train for a fine clock, to put in the place of the escape-wheel a very light, well-balanced fly, to prevent “backlash,” and a very fine soft cord on the barrel; then hang on a very light weight; so slight that--all of the wheels being balanced, and no oil upon the pivots--the fly will move so slowly that its revolutions may be counted. By taking care that the weight be not too much in excess of the resistance, the least inaccuracy in the wheels and pinions may be discovered by the difference in the velocity of the fly, or by its suddenly stopping, which will be occasioned by any inequality in the train teeth, which would not have been discovered by the closest scrutiny. It was by means of this test that we discovered an inaccuracy in a pinion, caused by hardening, which could not have been discovered by a less delicate test. The wheels in the train should be as light as possible, for as the whole train is stopped every time a tooth drops on the pallets, it is plain that the driving weight must overcome the inertia as well as the friction of the train at every beat. To this end it has been customary to “arm out” the wheels, leaving a very light rim supported by light arms, the wheels being generally of cast brass, turned up, and cut, then lightened. We followed this plan for some time, but abandoned it, as we found great difficulty in making a perfectly round wheel. The arms serve as posts to support the rim in cutting or turning, but the space between is very apt to spring down. We prefer making the wheels of fine hard-rolled sheet brass; it is superior to cast brass, much finer, harder, and more durable, and is freer from flaws. After the wheels are cut, they are turned out on each side, leaving a thin web in the centre; they can be made lighter, finished easier, and are round. As to the shape of the teeth in clock-wheels, the subject has been so ably treated by Reid, Dennison, and Prof. Willis (who has invented an instrument to assist in laying out the curves for the teeth of wheels), that we shall not attempt it in this paper; besides, there is so little of the entire theory that can be applied to a clock-wheel of two and a half inches in diameter, with 120 to 140 teeth, farther than to leave the wheel and pinion of the proper diameter, that we consider it unnecessary; for if makers of regulators and other fine clocks will use pinions of 16 or 20 teeth, the friction or driving is all after the line of centres, and the whole subject of cycloids, epicycloids, and hypocycloids is reduced to a very small point, and might be said to “vanish into thin air.” Having given only a few practical hints, and not yet crossed the threshold of the subject, we propose to continue from month to month--if the readers of the JOURNAL do not weary--the discussion of the various parts that go to make the sum total of a fine clock, with notices of the various clocks made in this country.