The obvious measure of the sun's movements was the moving shadow cast by the sun itself. And the earliest device for recording time was naturally the sun-dial. Its origin fades into the twilight of antiquity. Long before we know anything about him, primitive man measured the moving shadow of some tree. And it occurred to him to set up a post or pillar in some convenient place, and mark out the positions into which the shadow swung. The earliest sun-dials were of this pattern, with a vertical pointer of _gnomon_, and the hours marked upon the ground. And it is related of the early Greeks that they told the time individually by marking and measuring the length of their own shadows. But the measure of time by the length of a shadow is very irregular at best, because of the yearly motion of the sun. The shortest shadow of the day will indeed fall at noon. But that noon shadow will vary in length according as the sun's noon is high in Summer or low in Winter; and so the whole scale of lengths will be different for every day in the year. If a three foot shadow means mid-afternoon today, it will mean quite another time tomorrow. And for measuring by the _direction of the_ shadow, the vertical gnomon is more irregular still. For the swing of the shadow would depend not only upon the sun's motion across the sky from East to West, but also upon his slant North and South along the sky. And this would change from day to day. The difficulty was to make a dial of which the shadow would move as regularly as the sun moves. [Illustration: ANCIENT GREEK HEMICYCLE] This the ancients accomplished in a very simple and ingenious way. The sun moves in the sky as it were upon the inner surface of a hollow globe or sphere. So they made the dial a little hemisphere, place with its hollow side up toward the sky as a bowl stands on a table. The pointer was placed above and to the South of this, on the side toward the sun; and the Time was marked by the shadow of the tip end of the pointer which was a little ball or bead. The path of this shadow across the bowl reproduced exactly on a small scale the path of the sun across the great bowl of the heavens. And it was then an easy matter to mark off the bowl into equal divisions which the shadow would cross at equal intervals of the day. Of course, the track of the shadow changed with the season of the year. But it moved always as the sun moved, and just as regularly, giving a true measure of the solar day.