By parallax, as here employed, is meant the error due to the difference in the apparent position of the sun as viewed by an observer at any point of the earth's surface from what the apparent position would be if viewed from the line of the center of the earth, from which theoretical point the observations are supposed to be made. In the case of bodies so distant as the sun, this angle is an exceedingly minute one, and in the case of the fixed stars it disappears altogether. The sun's parallax is very material indeed from the standpoint of delicate astronomical observations, but it may be ignored altogether by the practical navigator in all ordinary observations. There is one other correction that he must make, however, in case of sun observations; he must add, namely, the amount of semi-diameter of the sun to his observed measurement, as all calculations recorded in the _Nautical Almanac_ refer to the center of the sun's disk. PERFECTING THE CHRONOMETER The observation of the sun's height, with the various corrections just suggested, suffices by itself to define the latitude of the observer. Something more is required, however, before he can know his longitude. How to determine this, was a problem that long taxed the ingenuity of the astronomer. The solution came finally through the invention of the chronometer, which is in effect an exceedingly accurate watch. Time measurers of various types have, of course, been employed from the earliest times. The ancient Oriental and Classical nations employed the so-called clepsydra, which consisted essentially of receptacles from or into which water dripped through a small aperture, the lapse of time being measured by the quantity of water. At an undetermined later date sand was substituted for the water, and the hour glass with which, in some of its forms, nearly everyone is familiar, came into use. For a long time this remained a most accurate of time measurers, though efforts were early made to find substitutes of greater convenience. Then clocks operated by weights and pulleys were introduced; and, finally, after the time of the Dutchman Huygens, the pendulum clock furnished a timepiece of great reliability. But the mechanism operated by weight or pendulum is obviously ill-adapted to use on shipboard. Portable watches, in which coiled springs took the place of the pendulum, had indeed been introduced, but the mechanical ingenuity of the watchmaker could not suffice to produce very dependable time-keepers. The very idea of a watch that would keep time accurately enough to be depended upon for astronomical observations intended to determine longitude was considered chimerical.