bably the relic of the old sacrifices
by fire to the gods. In so far the superstition has a mythological
significance, but then, are we not driven back to the consideration
whether these gods were not actual personages in the minds of the old
Celtic worshippers, and not symbols of natural phenomena?

So much, however, for popular superstitions; and, as regards folk-tales,
we must, in speculating as to their origin,[1] 'look not into the
clouds, but upon the earth; not in the various aspects of nature, but in
the daily occurrences and surroundings.' The process of diffusion must
always remain uncertain. 'Much may be due to the identity everywhere of
early fancy, something to transmission,' but 'household tales occupy a
middle place between the stories of savages and the myths of early
civilization.'[2] And as nursery-rhymes are but the simplified form of
household or folk-tales, let us consider with Mr. Lang the relation
between savage customs and ancient myths.

The foundation of the method of comparative mythology is the belief that
'myths are the result of a disease of language, as the pearl is the
result of a disease of the oyster.' The method of inquiry is to examine
the names which occur in the stories, and having found or invented a
meaning for these names, to argue back from them to a meaning in the
myths. But then almost each scholar has his peculiar fancy in etymology,
and while one finds a Sanskrit root, another finds a Greek, a third a
Semitic, and so on. Even when they agree upon the derivation of the
proper names, the scholars seldom agree upon the interpretation of them,
and thus the whole system is full of perplexity and confusion to all who
approach its study with unbiassed minds. There is a further division
among the mythologists, for there are some who have a partiality for
sun-myths, others for cloud-myths, sky-myths and fire-myths, and each
seeks to work out an interpretation of an old-world story to suit his
own taste in myths. How can they be all right? And in whom can we have
confidence when we find so much disagreement, first, on the derivation
of names, and second, on their meaning after the derivation is
discovered? And then, how do we know that words had the same meaning to
the ancients as they have to us? Was the sky, for instance, to the
original story-makers 'an airy, infinite, radiant vault,' as it is to
us, or was it a material roof, or even a person? And, further, how is it
that we find