394 The Origin of Totem Names mid Beliefs.
that, so far as we know, savages never deliberately destroy the totems of their enemies is a strong point against the "soul-box" theory. In Torres Straits, as in Australia, a group (or, as one is more in the habit of saying, a "clan") may have one chief totem and a number of sub- sidiary totems ; though the exact relation of these to each other is at present obscure. Mr. Lang's hypothesis, or, as he says, guess," is that the people of any small group call themselves " men," and call one another by animal nick- names. The " guess " which I should like to make is as follows : Primitive man moved about in small groups, and each small group was necessarily composed of related individuals. No savage peoples have an unlimited amount of space to wander in ; their area is limited by their coming in contact with similar groups, and almost every part of the savage world is marked out into hunting-grounds beyond which the proprietary groups do not go. Thus there would be a certain number of families living along a coast, others up the river, others in the jungle, and so there would come about groups of people each living mainly on one kind of food. In Australia and New Guinea we find there is a distinct idea that food of any kind can hardly grow naturally by itself, but that it is necessary to assist nature by means of magic ; hence a number of magical ceremonies are practised for the purpose of increasing the food-supply. Of course the most important fact in life for man, whether savage or civilised, is his food ; so the first aim of man would always be to get enough food for him- self. Therefore each group of people would naturally direct its magical ceremonies to the increase of the par- ticular kind of food on which it chiefly depended. When these groups, by force of circumstances, came to mingle together, there would be family groups each having, in addition to its own speciality in food-magic, a peculiar sympathic relation with a particular group, or groups, of a;nimals and plants ; because the livelihood of each de- pended on the knowledge of the habits of particular animals, or of the flowering and fruiting seasons of particular trees. I have myself seen shrines for making fruitful coco-nut trees, bananas, and other plants, and I know that this is the business of definite groups of people. Thus we get departmental group-magic, for, when these primitive totem- groups came to be synthesised into tribes, each would still
