pirrauru circle and making it equivalent to the noa group, sometimes narrowing the noa to make it coincide with the pirrauru group (see Fig. 1). In Folk-Lore, xvii., p. 104, Dr. Howitt says that the tribes whose kinship terms he has just quoted had pirrauru marriage; among the tribes in question is the Arunta, and the term which Dr. Howitt mentions as in use among them is the word unawa (=noa). Now, as my diagram shows and as Dr. Howitt cannot but admit when he is challenged, the noa group is not as a rule co-extensive with any one pirrauru group, though it includes it; pirrauru is therefore used in this passage in the sense of noa-group-marriage. When, therefore, Dr. Howitt speaks of pirrauru, we are uncertain whether he means the extant Dieri custom or the conjectured institution which he asserts on p. 181 to have been restricted by the kandri ceremony.
Conversely, Dr. Howitt speaks of pirrauru as group-marriage (xvii. 185, xviii. 171, 185, etc.), and at the same time asserts the former existence of another kind of group-marriage among the Kurnai, whose terms maiaubra, as I shall show below, correspond not to pirrauru, but to noa, in all essentials, Dr. Howitt's affirmation notwithstanding. There are therefore not only two kinds of pirrauru, but also two kinds of group-marriage, and Dr. Howitt leaves his readers to guess which he means in any particular passage. If he is misunderstood, his blood is on his own head.
(3) I now pass on to the third point of those mentioned above—the origin of the marital terms. Dr. Howitt asserts (p. 170) that "the (group) terms, husband and wife, father and mother, son and brother, all arise out of the pirrauru family." If by this Dr. Howitt means pirrauru in its only proper sense, that in which it is used by the Dieri, this statement is unfortunately absolutely misleading. Dr. Howitt has shown nothing of the sort and can show nothing of the sort, for the
