~ 
1900.] MATHEWS—THE AUSTRALIAN ABORIGINES. 569 
prudent and politic to amalgamate for the purposes of mutual defense. 
This alliance could have been easily accomplished by the inter- 
change of wives between the members of the opposite confederacies. 
Koopungie could have taken the wife of Kellungie and could have 
given his own wife to a Kellungie man. Cheekungie and Karpun- 
gie could have exchanged wives in a similar manner, but there was 
no alteration made in the names of the offspring in any case. For 
example, the children of a Koopungie man were distinguished as 
Karpungie the same as before the coalition and so on, as in the 
following table, a careful perusal of which will make my meaning 
more clear: 
TaBLE No. 4. 
LPhratry. Husband, Wife. Child. 
A Koopungie Cheekungie Karpungie 
Kellungie Karpungie Cheekungie 
B J Cheekungie Koopungie Kellungie 
L Karpungie Kellungie Koopungie 
In examining the above table, together with the two preceding 
ones, it will be observed that a man, Koopungie, of the A phratry 
in Table No. 2 marries a woman, Cheekungie, of the B phratry in 
Table No. 3, and wce versa; and that a Karpungie man of the B 
phratry in Table No. 2 marries a woman, Kellungie, of the A phra- 
try in Table No. 3, and wce versa. It is also seen that the people 
belonging to the two A phratries go together, and the two B phra- 
tries together in Table No. 4. That is to say, the men of the 
sections Koopungie and Kellungie of the A phratry in Table No. 4 
marry the women of the two sections Cheekungie and Karpungie of 
phratry B in this new organization, and conversely. The men and 
women of the respective phratries are therefore mutually related as 
brothers-in-law and sisters-in-law. It is therefore apparent that 
whether the organization consist of two divisions, as among the 
Yowerawarrika, or of four sections like the Warkeemon, the com- 
munity still retains the two primary intermarrying phratries, A and 
B.. Aggregates-of fofems, chosen from the fauna and the flora, are 
attached to each division. 
It now remains to explain the probable origin of the present 
system of dividing a tribe into ezg#fsections. If our theory is of 
any value it should hold good in this system as well as in that of 
four divisions already explained. ‘The eight divisions of the 
