Now if the rite as it occurs in Australia is pure magic, and if religion is not a variety of magic but fundamentally different from it, then the rite which, as it occurs everywhere else, is religious, cannot be derived from, or a variety of, the Australian piece of magic; and the spring and harvest customs which are found in Australia cannot be "based on the {202} same ancient modes of thought or form part of the same primitive heathendom" as the sacramental rites which are found everywhere else in the world. The solemn annual eating of the totem plant or animal in Australia must have a totally different basis from that on which the sacrament and communion stands in every other part of the globe: in Australia it is based on magic, elsewhere on that which is, according to Dr. Frazer, fundamentally different and opposed to magic, viz. religion. Before, however, we commit ourselves to this conclusion, we may be allowed to ask, What is it that compels us thus to sever the Australian from the other forms of the rite? The reply would seem to be that, whereas the other forms are admittedly religious, the Australian is "a magical rite intended to assure the revival of nature in spring." Now, if that were really the nature of the Australian rite, we might have to accept the conclusion to which we hesitate to commit ourselves. But, as a matter of fact, the Australian rite is not intended to assure the revival of nature in spring, and has nothing magical about it. It is perfectly true that in spring in Australia certain proceedings are performed which are based upon the principle that like produces like; and {203} that these proceedings are, by students of the science of religion, termed--perhaps incorrectly--magical. But these spring customs are quite different from the harvest customs; and it is the harvest customs which constitute the link between the rite in Australia and the rite in the rest of the world. The crucial question, therefore, is whether the Australian harvest rite is magical, or is even based on the principle that like produces like. And the answer is that it is plainly not. The harvest rite in Australia consists, as we know it now, simply in the fact that at the appointed time a little of the totem plant or animal is solemnly and sparingly eaten by the headman of the totem. The solemnity with which the rite is performed is unmistakable, and may well be termed religious. And no attempt even, so far as I am aware, has been made to show that this solemn eating is regarded as magic by the performers of the rite, or how it can be so regarded by students of the science of religion. Until the attempt is made and made successfully, we are more than justified in refusing to regard the rite as magical; we are bound to refuse to regard it as such. But if the rite is not magical--and _à fortiori_ if it is, as Dr. Frazer terms it, sacramental--then it is {204} religious; and the ancient mode of