Australian Marriage Customs. 311
simple reason that, as my diagram shows, the noa group is not co-extensive with any one pirraurii group, but more extensive, whereas use of the kinship terms men- tioned by Dr. Howitt is limited by the noa group and not by the pirvanrit. group. A boy, for example, in the Dieri tribe applies the term ngaperi to the primary husband of his mother (x Y in diagram ^), whether he is actually his progenitor or not, and ngaperi-waka to all his father's tribal brothers (NOAG), whether they are his mother's pirraiirii (pirn) or not. Dr. Howitt in fact admits as much on p. 174, line 21 ; yet he argues all through as if the pirratirit group were the limit of these kinship terms. Dr. Howitt's argument on p. 179 about the "group-mother" is vitiated by precisely the same error ; no one who reads the passage would gather that a boy applies the term which we translate by the word mother, not only to his actual mother and to all the pirraurii spouses of his father, but also to all the women of his father's noa group, even to babies in arms ; yet such is the case, though Dr. Howitt's argument is thereby reduced to an absurdity. Put in bald terms it comes to this : that the twenty-seven women of the noa group who are not pirraurii to a given boy's father are addressed by that boy as mothers, because eight or nine otiier women have relations with his father. Comment is needless.
In connection with marital terms, I must once more refer to my point as to ngaperi and mungan (xvii. 303). I charged Dr. Howitt with being guilty of a grave confusion in asserting that to the ngaperi-zvaka (Dieri) who is also pirraiiru, is analogous in position the breppa-mungan (Kurnai). But little of Dr. Howitt's reply has any bearing on my contention, and what little does bear on it leaves my position absolutely untouched.
' For the text here llie «<?a-group must be taken as composed of males, the wife belonging to class 4.
