But there is another possible way of explaining the resemblance (granting that it is proved) of the Greek and Australian Pleiad myth.  The object of both myths is to account for the grouping and other phenomena of the constellations.  May not similar explanatory stories have occurred to the ancestors of the Australians, and to the ancestors of the Greeks, however remote their home, while they were still in the savage condition?  The best way to investigate this point is to collect all known savage and civilised stellar myths, and see what points they have in common.  If they all agree in character, though the Greek tales are full of grace, while those of the Australians or Brazilians are rude enough, we may plausibly account for the similarity of myths, as we accounted for the similarity of flint arrow-heads.  The myths, like the arrow-heads, resemble each other because they were originally framed to meet the same needs out of the same material.  In the case of the arrow-heads, the need was for something hard, heavy, and sharp—­the material was flint.  In the case of the myths, the need was to explain certain phenomena—­the material (so to speak) was an early state of the human mind, to which all objects seemed equally endowed with human personality, and to which no metamorphosis appeared impossible.

In the following essays, then, the myths and customs of various peoples will be compared, even when these peoples talk languages of alien families, and have never (as far as history shows us) been in actual contact.  Our method throughout will be to place the usage, or myth, which is unintelligible when found among a civilised race, beside the similar myth which is intelligible enough when it is found among savages.  A mean term will be found in the folklore preserved by the non-progressive classes in a progressive people.  This folklore represents, in the midst of a civilised race, the savage ideas out of which civilisation has been evolved.  The conclusion will usually be that the fact which puzzles us by its presence in civilisation is a relic surviving from the time when the ancestors of a civilised race were in the state of savagery.  By this method it is not necessary that ’some sort of genealogy should be established’ between the Australian and the Greek narrators of a similar myth, nor between the Greek and Australian possessors of a similar usage.  The hypothesis will be that the myth, or usage, is common to both races, not because of original community of stock, not because of contact and borrowing, but because the ancestors of the Greeks passed through the savage intellectual condition in which we find the Australians.