Such a statement makes it clear how very crude and vague the deductions must have been from the observations of navigators, however accurately made, prior to the time when such tables as those of the _Nautical Almanac_ had been prepared. Fully to appreciate this, it is necessary to understand that the observations supplied in such profusion for the use of the mathematical astronomer are in themselves subject to errors that might seriously vitiate the results of the final computation. They must, therefore, be made with the utmost accuracy, and with instruments specially prepared for the purpose. The chief of these instruments is not the gigantic telescope but the small and relatively simple apparatus known as a transit instrument. This constitutes essentially a small telescope poised on very carefully adjusted trunnions, in such a way that it revolves in a vertical axis, bringing into view any celestial body that is exactly on the meridian, and bodies in this position only. To make observation of the transit--that is to say the passage across the meridian line--of any given body more accurate, the transit instrument has stretched vertically across the center of its field of vision a spider web, or a series of parallel spider webs; in order, in the latter case, that the mean time of several observations may be taken. So exceedingly difficult is it to manufacture and mount an instrument of requisite nicety of adjustment, that the effort has almost baffled the ingenuity of the mechanic. Sir George Airy, in making a transit instrument for use at the Royal Observatory at Greenwich, required the trunnions on which it was to be mounted to be ground truly cylindrical in form within a variation of one thirty-two-thousandth of an inch as determined by a delicate spirit level. Even when all but absolute decision has been obtained, however, it is quite impossible to maintain it, as the slightest variation of temperature--due perhaps to the application of the hand to one of the pillars on which the trunnions rest--may disturb the precise direction of the spider webs and so militate against absolute accuracy of observation. The instrument must, therefore, be constantly tested and its exact range of errors noted and allowed for. To devote so much labor to details, merely in the effort to determine the precise moment at which a star or planet crosses the meridian, would seem to be an absurd magnification of trifles. But when we reflect that the prime object of such observations is to supply practical data which will be of service in enabling navigators on all the seas of the globe to bring their ships safely to port, the matter takes on quite another aspect. We have here, obviously, another and a very striking illustration of the close relationship that obtains between the work of the theoretical devotee of science and that of the practical man of affairs. SOUNDINGS AND CHARTS