But, discarding this old philological hypothesis, how did the pristine groups get their totem names? We ought first to return to our conjecture as to what these pristine groups were like. They must have varied in various environments. Where the sea, or a large lake, yields an abundant food-supply, men are likely to have assembled in considerable numbers, as ‘kitchen middens’ show, at favourable stations. In great woods and jungles the conditions of food-supply are not the same as in wide steppes and prairies, especially in the uniform and arid plateaus of Central Australia. Rivers, like seas and lakes, are favourable to settlement; steppes make nomadism inevitable, before the rise of agriculture. But, if the earliest groups were mutually hostile, strongly resenting any encroachment on their region of food-supply, the groups would necessarily be small, as in Mr. Darwin’s theory of small pristine groups, the male, with his females, daughters, and male sons not adult.[40] A bay, or inlet, or a good set of pools and streams, would be appropriated and watchfully guarded by a group, just as every area of Central Australia has its recognised native owners, who wander about it, feeding on grubs, lizards, snakes, rats, frogs, grass-seeds, roots, emus, kangaroos, and opossums. The pristine groups, we may be allowed to conjecture, were small. If they were not, the hypotheses which I venture to present are of no value, while that of Mr. Atkinson shares their doom. Mr. McLennan, as far as one can conjecture from the fragments of his speculations, regarded the earliest groups as at least so large, and so bereft of women, that polyandry was the general rule. Mr. Darwin, on the other hand, began with Polygyny and Monogamy, ‘jealousy determining the first stage.’[41] This meant that there was a jealous old sire, who kept the women to himself, as in Mr. Atkinson’s theory. As we can scarcely expect to reach certainty on this essential point, anthropology becomes (like history in the opinion of a character in _Silas Marner_) ‘a process of ingenious guessing.’ But, embarking on conjecture, I venture to suggest that the problem of the commissariat must have kept the pristine groups very small.[42]