es as kinsmen, or brothers and sisters; they have the same
totem mark and are bound to protect each other. The totem bond is
stronger than any blood tie, while the sex totems are even more sacred
than the clan totems.

Much confusion has arisen out of the attempts to explain the
Australian system; and for long the close totem kinship was supposed
to afford evidence of group marriage, by which a man of one clan was
held to have sexual rights over all the women in another clan. But
further insight into their customs has proved the error of such a
view, which arose from a misunderstanding of the terms of relationship
used among the tribes. Nowhere is marriage bound by more severe laws;
death is the penalty for sexual intercourse with a person of a
forbidden clan. And it is certain that there is no evidence at all of
communism in wives.[105]

[105] _See_ Westermarck, _op. cit._, pp. 54-56.

A system of taboos is very strongly established, and as we should
expect the women appear to be most active in maintaining these sexual
separations. If a man, even by mistake, kills the sex-totem of the
women, they are as much enraged as if it were one of their own
children, and they will turn and attack him with their long poles.

In Australia it is easy to recognise a very early stage in human
society. The organisation of the family group into the clan is still
taking place. Moreover, the most primitive patriarchal conditions have
not greatly changed, for the males are great individualists and cannot
readily suffer the rights of others than themselves. Mother-right can
hardly be said to exist, and the position of women is low. It is not
the custom among any tribes for the husband to reside in the home of
the wife; this in itself is sufficient to explain the power of the
husbands. Wives are frequently obtained by capture, and fights for
women are of common occurrence. Here it would seem that progress has
been very slow. Indeed, it is the chief interest of the Australian
tribes that we can trace the transformation from the early patriarchal
conditions to the communal clan.

There is still another fact of very special interest. In the large
majority of tribes known to us descent is traced through the mother;
the proportion of these tribes to those with father-descent being four
to one. Now, the question arises as to which of these two systems is
the earlier custom? As a rule it is assumed that in all cases descent
was originally