The poem of Aratus enjoyed an immense popularity in classical times and throughout the Middle Ages, and no doubt helped to stereotype the forms whose origin was already forgotten when he wrote. He was not an astronomer, however, and the poem is only a popular paraphrase of a lost work by Eudoxus. This Greek astronomer had lived a hundred years earlier, and it is thought that he himself copied from an older source. The attempt to discover this source has been the object of many ingenious conjectures, and much research among ancient monuments and writings. Some of the old constellations are met with in Isaiah and Job, in Homer, on tablets found at Nineveh, and an immense antiquity is sometimes claimed for them. Dupuis, writing at the end of the eighteenth century, thought he had conclusively proved that the figures of the zodiac were designed in Egypt 15,000 years ago![12] Miss Plunkett, in her “Ancient Calendars and Constellations” assigns them to the seventh millenium before Christ. ------------------------------------------------------------------------ [12] _Origine de tous les cultes, ou Religion universelle_, by C. F. Dupuis. ------------------------------------------------------------------------ [Illustration: THE OLD CONSTELLATION FIGURES. SOUTHERN HEMISPHERE.] [Illustration: NORTHERN HEMISPHERE. THE OLD CONSTELLATION FIGURES ACCORDING TO ARATUS. _In Ptolemy’s Catalogue Equuleus and Corona Australis are added to these._ _(From Peck’s “Constellations and How to Find Them.”)_] An ingenious theory, suggested independently by Schwartz and Proctor, and developed by Mr. E. W. Maunder,[13] is founded on an examination of the space round the South Pole which was left blank by the ancient constellation designers. From its extent, Proctor concluded that they cannot have seen further south than about 40° from the South Pole, and therefore that they must have lived in a latitude of about 40° north of the equator (say, central Asia or Asia Minor); from the position of its centre, which must have been the Pole, he concluded that the date was about B.C. 2200. For the centre of the circular patch seems to lie near the star _Delta Hydri_, which was the South Pole star at that time. (This movement of the Pole among the stars, due to “precession,” will be explained later). This date does not differ much from that found by Robert Brown, from the position of the celestial equator among the stars, as described by Aratus; he says B.C. 2084.[14] Schwartz gave B.C. 1400; Mr. Maunder, from additional considerations of the positions of various constellation figures, says that they must all have been originally designed about B.C. 2800. ------------------------------------------------------------------------ [13] _Journal of the British Astronomical Association_, “The Oldest Astronomy,” July 1898, June 1899, April 1904, May 1909; _The Observatory_, December 1898; _Knowledge_, October 1904; and elsewhere.