162 R. H. MATHEWS. 



It is evident that half of a man's possible wives and half 

 his possible families belong to the Willitji moiety, and half 

 his possible wives and families to the Liaritji moiety — the 

 section and the moiety of the children depending altogether 

 upon the mothers in every case. Whatever may be the 

 meaning of the terms Willitji and Liaritji it is quite clear 

 that they cannot be the names of two independent moieties, 

 because the sons of the Willitji men are liable to be scat- 

 tered up and down through all the sections of the entire 

 community. Exactly the same confusion will arise if we 

 take our example from the four "Husbands" comprising 

 the Liaritji moiety, see Table VI, p. 82. The partition of 

 the men into the two parts Willitji and Liaritji utterly 

 fails either to prove descent through the fathers, or to 

 establish exogamy of such moieties. These facts also con- 

 trovert the conclusion arrived at by Mr. N. W. Thomas 

 when he says : " The existence of phratriac [moiety] names 

 enables us to say definitely that the descent in this tribe 

 is in the male line." 1 



Corrections. 

 In Vol. xxxiv of this Journal, p. 129, and in Vol. xxxv, 

 p. 218, I reported a variety of totems appertaining to some 

 tribes about Cresswell Downs, Sturt's Creek and adjacent 

 country. The information was gathered for me by Mr. 

 Innes, Mr. Stretch, Mr. Wilson and other residents of those 

 regions. Upon continuing my inquiries through these men 

 and gathering further details, I find that the totems are 

 not arbitrarily attached to the particular pairs or quartettes 

 of sections mentioned in my former papers. All the totems 

 therein enumerated are found among the different sections, 

 but instead of being inherited from either parent, are 

 determined by the locality where the mother first became 

 aware that she was enceinte, in accordance with the beliefs 



1 Kindship and Marriage, p. 150. 



