Ethnological Data in Folklore. 137
The conclusions to be drawn from this are, first, that the overlapping of the several examples (No. i overlapping No. 2 at a, b, c, d, No. 2 overlapping No. 3 at a^ b, g, h, No. 3 overlapping No, 4 at g, h, i) is the essential factor in the comparison ; secondly, that example No. 4, though possessing none of the elements of example No. i, is the same custom as example No. i.
These conclusions are not affected by the order in which the examples are arranged ; whether we begin with No. 4 or with No. i,the relationship of each example to the others, thus proved to be in intimate association, is the same.
But the distinction of the elements into two classes, which may be called radicals and divergences respectively, is of course an important point. As a rule, it will be found that the radical elements are the most constant parts of the whole group of examples, appearing more frequently, possessing greater adherence to a common form, changing ( when they do change ) with slighter variations ; while the divergent elements, on the other hand, assume many different varieties of form, are by no means of constant occurrence, and do not, even amongst themselves, tend to a common form. To these considerations, derived entirely from a study of the analysis, is to be added the fact that the radical elements are alone capable of being equated with customs or beliefs obtaining among savage or barbaric peoples. This enables us to take a very import- ant step, namely, to suggest that the divergences {g to m) mark the line of decay which the particular custom has un- dergone since it ceased to belong to the dominant culture of the people, and dropped back into the position of a survival from a former culture preserved only by a fragment of the people.
When any given custom or belief, having undergone this double process of analysis of component elements and classification of the individual examples, reveals a distinct parallel between its radical elements and the elements of a
