The same applies to the cases where we are told that the families own exclusively a certain area (Roth, G. S. Lang, Grey). But these cases were found to be not quite beyond question. In some instances when the local group is a larger unit, and there is no subdivision of land amongst families, several statements mention that the natives lived scattered in small groups, varying from two to four families perhaps. (Murray tribes according to Eyre; the Central and North-Central tribes according to Spencer and Gillen; the Moreton Bay tribes according to J. D. Lang; New South Wales tribes according to McDougall, Henderson and Hodgkinson; the Kabi and Wakka, West Australians according to Withnell, Browne, Scott-Nind.) In some cases there are reasons for supposing that the local group was larger (Bangerang, Western Victoria, at Encounter Bay, on the lakes; perhaps on the sea-shores in West Australia according to Grey). The remainder of our information (fifteen statements) does not give any clear answer to this question. From these approximately exact data we come to the conclusion that the majority of tribes lived in small groups of two or three families of six to nine individuals each, and only in a few tribes were there larger bodies living in actual daily contact. To get a more reliable answer on this point it is better to drop the less clear evidence and to take into consideration only such as is better and more reliable. If only the fully reliable and unambiguous statements be used, there are twelve affirming that aborigines live in small parties, which in some cases shrink to one family only (Howitt on the Kurnai; Eyre; R. Dawson; G. S. Lang; McDougall; Spencer and Gillen in the Central and North-Central tribes; Henderson; Hodgkinson; Rev. Matthew on the Kabi and Wakka; Withnell; Salvado). It should be noted that (1) some of these authorities are our best informants (Howitt Spencer and Gillen, Salvado); (2) that the area covered by these peoples is very extensive, and that the tribes in question are scattered over the whole continent. The statements which assert the mode of living in larger bodies are much less reliable. But it appears undoubted that the statements of Curr and Dawson, perhaps also those of Meyer, Schurman and Taplin (confirmed by Angas), are of quite unquestionable reliability. It is therefore clear that there were local differences in that respect. And such a geographical difference in the mode of living appears quite plausible, from general considerations. The reasons which must have determined the degree of aggregation in the Australian tribes were peculiarly economic ones: the scarcity of food supply was conditioned partly by the aridity of the soil, partly by the primitiveness of the means of procuring subsistence. Where the means of subsistence were plentiful and not easily exhausted, there larger groups could permanently aggregate. This was, in the first place, the case where fishing was at all possible.