to refer these two groups of
folklore respectively to the earliest ages of man's existence as man
and to the neolithic stage of culture, they must be withdrawn from all
other classification. We cannot use the same items of folklore in two
totally different ways. The results of withdrawal are as important as
the results of allocation, and the necessity for the correct docketing
of all groups of folklore is thus at once illustrated.

The first point in the argument for ethnological data being
discoverable in folklore is that a survey of the survivals of custom,
belief, and rites in any given country shows one marked feature, which
results in a dividing line being drawn as between two distinct
classes. This feature is the antagonism which is discoverable in these
classes. On one side of the dividing line is a set of customs,
beliefs, and rites which may be grouped together because they are
consistent with each other, and on the other side is another set of
customs, beliefs, and rites which may be grouped together on the same
ground. But between these two sets of survivals there is no agreement.
They are the negations of each other. They show absolutely different
conceptions of all the phases of life and thought which they
represent, and it is impossible to consider that they have both come
from the same culture source. I have applied the test of ethnology to
such cases in Britain, and this appears to answer the difficulty which
their antagonism presents. It appears too to be the only answer.

The subjects which show this antagonism are all of vital importance.
They include friendly and inimical relations with the dead; marriage
as a sacred tribal rite and marriage as a rule of polyandrous society;
birth ceremonies which tell of admittance into a sacred circle of
kinsmen, and birth ceremonies which breathe of revenge and hostility;
the reverential treatment of the aged folk and the killing of them
off; the preservation of human life as part of the tribal blood, and
human sacrifice as a certain cure for all personal evils; the worship
of waters as a strongly localised cult, preserved because it is local
by whatsoever race or people are in occupation and in successive
occupation of the locality; totemic beliefs connected with animals and
plants contrasted with ideas entirely unconnected with totemism--all
this, and much more which has yet to be collected and classified,
reveals two distinct streams of thought which cannot by