of the present day. On p. 174, and again at the top of p. 177, without perceiving that he is contradicting his previous statement, he asserts that "the application of the term ngaperi to the other brothers who have not become pirrauru" (i.e. to all the men of the noa group) appears to be a vestigiary survival of what was once a fact. That is to say that between the "absolute promiscuity of the undivided commune" and the pirrauru marriage of the present day there was a stage, in which all the men of one noa-group were de jure husbands of all the women of another; and this view he emphasises on p. 181, where he asserts that the kandri ceremony is a restriction of the range of license within the noa group and creates the pirruru group, while the noa relationship is itself a restriction on a former wider range of licence (i.e. absolute promiscuity). In order to make Dr. Howitt's error quite clear, I now quote from p. 172 (cf. the passage at the bottom of p. 177, contradicting that at the top) a sentence in which he affirms the view stated in the first of these three passages: "I consider the noa relationship as having restricted the range of an earlier and wider license to the present limits of the pirrauru marriage."
On p. 174, on p. 176 and on p. 183, Dr. Howitt charges me with not understanding the facts of noa and pirrauru. If this were in fact so, I should have ample justification in the confusions just quoted; but in fact the three passages from my remarks on those pages are absolutely accurate, and would have been clear to Dr. Howitt, even if he did not agree with them, had he read them in the light of my definitions and not tried to take my terminology in a sense of his own.
On p. 171 Dr. Howitt remarks, "Throughout my paper I spoke of pirrauru as group-marriage." As a matter of fact Dr. Howitt uses both group-marriage and pirrauru in two different senses, sometimes enlarging the
