The original idea was simple enough. At first, it was merely that of a vessel of water, having a small hole in the bottom, so that the liquid dripped out drop by drop. As the level within the jar was lowered, it showed the time upon a scale. Thus, if the hole were so small and the vessel were so large that it would require twenty-four hours for the water to drip away at an absolutely steady rate, it may be seen that the side of the vessel might easily have been marked with twenty-four divisions to indicate the hours. It may also be seen that the water would drip as rapidly at night or in shadow as in sunlight. And the clepsydra could be used indoors, which the sun-dial could not, although it required attention in that it must be regularly refilled and the orifice must always be kept completely open, because the slightest stoppage would retard the rate of dripping and the "clock" would run slow. The sun, which, with the other heavenly bodies, had therefore been the sole reliance of the human race in its time-reckoning could now be ignored and the would-be timekeeper called to his aid another mighty servant from the forces of nature—that of gravitation. The most interesting human fact, however, about the clepsydra is that it involved an entirely different conception of the marking of time. Now it was not so much a question of _when_ as of _how long_. A good sun-dial set in a proper position would always indicate three o'clock when it _was_ three o'clock, but the clepsydra might do no such thing. It would merely show how many hours had elapsed since last it was filled, and the steady drip, drip, drip of the escaping water could—and did—lower the surface quite as evenly at one time of day as at another. We have already seen that the first purpose in marking time was merely for making appointments, but the clepsydra shows that, with its invention, mankind had already made some progress toward a new point of view. One important factor in this change was the very practical need of telling time at night, in stormy weather, or indoors, where the sun-dial could not be used. The clepsydra, on the other hand, worked equally well at any hour or place, and in all sorts of weather. Nevertheless, it, too, proved to have certain faults. After a time, people noticed the interesting fact that water ran faster from a full vessel than from one which was nearly empty; this was, of course, because of the greater pressure. Since such a variation interfered with calculations, they hit upon the idea of a double vessel; the larger one below containing a float which rose as the vessel filled, thus marking the hours upon the scale, and the smaller one above, the one from which the water dripped, being kept constantly filled to the point of overflow.