It is, of course, not possible to fix either time or latitude very closely, since the limits of the unmapped space are a a little vague. But it is significant that if we take a celestial globe, arranged so as to represent the heavens for the time 2,700 B.C., and for north latitude 40°, we find several striking relations. First of all, the Great Dragon then linked together the north pole of the celestial equator, and the north pole of the ecliptic; it was as nearly as possible symmetrical with regard to the two; it occupied the very crown of the heavens. With the single exception of the Little Bear, which it nearly surrounds, the Dragon was the only constellation that never set. Next, the Water-snake (see diagram, p. 200) lay at this time right along the equator, extending over 105° of Right Ascension; or, to put it less technically, it took seven hours out of the twenty-four to cross the meridian. It covered nearly one-third of the equatorial belt. Thirdly, the intersection of the equator with one of the principal meridians of the sky was marked by the Serpent, which is carried by the Serpent-holder in a very peculiar manner. The meridian at midnight at the time of the spring equinox is called a "colure,"--the "autumnal colure," because the sun crosses it in autumn. Now the Serpent was so arranged as to be shown writhing itself for some distance along the equator, and then struggling upwards, along the autumnal colure, marking the zenith with its head. The lower part of the autumnal colure was marked by the Scorpion, and the foot of the Serpent-holder pressed down the creature's head, just where the colure, the equator, and the ecliptic intersected (_see_ diagram, p. 164). It is scarcely conceivable that this fourfold arrangement, not suggested by any natural grouping of the stars, should have come about by accident; it must have been intentional. For some reason, the equator, the colure, the zenith and the poles were all marked out by these serpentine or draconic forms. The unmapped space gives us a clue only to the date and latitude of the designing of the most southerly constellations. We now see that a number of the northern hold positions which were specially significant under the same conditions, indicating that they were designed at about the same date. There is therefore little room for doubt that some time in the earlier half of the third millennium before our era, and somewhere between the 36th and 40th parallels of north latitude, the constellations were designed, substantially as we have them now, the serpent forms being intentionally placed in these positions of great astronomical importance.