[Illustration.] Figure 2 “The moment primitive man began to observe anything, he must have taken note of the stars, and as soon as he began to talk about them he must have started by defining, in some way or other, the particular star he meant.... Observers would first consider the brightest stars and separate them from the dimmer ones; they would then discuss the stars which never set (the circumpolar constellations) and separate them from those which did rise and set. Then they would naturally, in a northern clime, choose out the constellation of the Great Bear or Orion, and for small groups, the Pleiades (_op. cit._ p. 132).... A few years’ observation would have appeared to demonstrate the absolute changelessness of the places of the rising and setting of the same stars. It is true that this result would have been found to be erroneous when a long period of time had elapsed and when observation became more accurate, but for hundreds of years the stars would certainly appear to represent fixity, while the movements of the sun, moon and planets would seem to be bound by no law ... would appear erratic, so long as the order of their movements was not known.” The reflection that Ursa Major was probably the first constellation which made any deep impression upon the mind of prehistoric man in America, as elsewhere, lent an additional interest to the star-group, as I concentrated my mind upon its form and endeavored to imagine it in four equidistant positions, corresponding to the numerals in the symbol _Ollin_ of the calendar-stone of Mexico (fig. 2, no. 2).