names of the phratries mean Eagle Hawk and Crow. Each
born Crow must marry an Eagle Hawk; each born Eagle Hawk must marry a
Crow. The names are derived through the mothers. One obvious result is
that no two persons, brother and sister maternal, can intermarry; but
the rule also excludes from intermarriage great numbers of persons in
no way akin to each other by blood, who merely share the common phratry
name, Crow or Eagle Hawk.

In each phratry are smaller sets of persons, each set distinguished by
the name of some animal or other natural object, their 'totem.' The
same totem is never found in both phratries. Thus a person marrying out
of his or her phratry, as all must do, necessarily marries out of his
or her totem.

The same arrangements exist among tribes which derive phratry and totem
names through the father.

This derivation of names and descent through the father is regarded by
almost all students, and by Mr. J. G. Frazer, in one passage of his
latest study of the subject, as a great step in progress. ['The Beginnings
of Religion and Totemism among the Australian Aborigines,' FORTNIGHTLY
REVIEW, September 1905, p. 452.] The obvious result of paternal descent is
to make totem communities or kins local. In any district most of the
people will be of the same paternal totem name--say, Grub, Iguana, Emu,
or what not. Just so, in Glencoe of old, most of the people were MacIans;
in Appin most were Stewarts; in South Argyll Campbells, and so on.

The totem kins are thus, with paternal descent, united both by supposed
blood ties in the totem kin, and by associations of locality. This is
certainly a step in social progress.

But while Mr. Frazer, with almost all inquirers, acknowledges this, ten
pages later in his essay he no longer considers the descent of the
totem in the paternal line as necessarily 'a step in progress' from
descent in the maternal line. 'The common assumption that inheritance
of the totem through the mother always preceded inheritance of it through
the father need not hold good,'[IBID. p. 462.] he remarks.

Thus it appears that a tribe has not necessarily made 'a great step in
progress,' because it reckons descent of the totem on the male side. If
this be so, we cannot so easily decide as to which tribe is socially
advanced and which is not.

In any case, however, there is a test of social advance. There is an
acknowledged advance when a tribe is divided into, not two, but four or
eight divis