Not one of the theories here summarised, except the Dieri and Woeworung myth, explains why members of the various totem kins are exogamous, may not marry other members of the name. Suppose you do get your totem name from that of a distinguished male ancestor, why may you not marry another descendant? If because the common name, say Emu, is taken to indicate some sort of blood-relationship, why may you not marry a blood relation, even if there be no traceable kinship between you and her? A Douglas may marry a Douglas, a Smith may marry a Smith; but an Emu is often capitally punished if he marries an Emu. Suppose you get your totem name from the beast for which you do magic. Why may you not marry a person who bears the name of the same beast, and whose male kindred do magic for it? Because it is sacrosanct to you and her? But you are actually breeding it for the food-market. The answer must be that you may not marry a person who bears your own totem name, and is in the same branch of the Co-operative Magical Stores, because her beast and yours are in the same phratry, and phratry mates may not intermarry. But why may they not marry? The reply will probably be, because the legislator divided the previously undivided commune into two intermarrying exogamous phratries. But that theory we have shown to be untenable. Thus not one of the extant hypotheses of the origin of Totemism explains why totem kins are exogamous, unless Mr. Haddon supposes that the totem names, once given from without, came to be explained by myths asserting the sacred character and tribal kindred of the totem. Mr. Haddon has not said anything about a previous exogamous tendency in each of the groups which, by his scheme, received totem names from without. By my hypothesis, these groups had already a strongly exogamous tendency, which later was hall-marked and sanctioned by the totem, with its myths and tabus. This advantage of explaining the exogamous attribute of the totem, my scheme possesses, and its rivals lack. STATEMENT OF THE PROBLEM