[228] _Ibid._, 65, 163, 164. Similar results are reached by Spencer and Gillen, who have given a remarkably clear and minute account of matrimonial, tribal, and totemistic institutions in central Australia. So far as the Australians are concerned, the theory that the classificatory nomenclatures are merely terms of address is positively rejected.[229] "When, in various tribes, we find series of terms of relationship all dependent upon classificatory systems such as those now to be described, and referring entirely to a mutual relationship such as would be brought about by their existence, we cannot do otherwise than come to the conclusion that the terms do actually indicate various degrees of relationship based primarily upon the existence of inter-marrying groups.... Whatever else they may be, the relationship terms are certainly not terms of address, the object of which is to prevent the native having to employ a personal name. In the Arunta tribe, for example, every man and woman has a personal name by which he or she is freely addressed by others--that is, by any, except a member of the opposite sex who stands in the relationship of '_Mura_' to them, for such may only on very rare occasions speak to one another."[230] The fundamental idea of the Australian classes is "that the women of certain groups may marry the men of others." It is a device for limiting and defining the inter-marriage of persons supposed to be of common blood.[231] [229] SPENCER AND GILLEN, _Native Tribes of Central Australia_, 56. [230] _Ibid._, 56, 57. "A man can only marry women 'who stand in the relationship of _nupa_, that is, are children of his mother's elder brother's blood or tribal, or, what is the same thing, of his father's elder sister.'" The mother of a man's _nupa_ is "_mura_ to him and he to her, and they must not speak to one another." This applies to a possible mother, _i. e._, the sister of the father: _ibid._, 61, 62. [231] _Op. cit._, 58. Crawley likewise holds that the terms of the classificatory systems "are terms of kinship and not terms of address;" although being "in origin terms of _relation_," so far they are "terms of address also." For "all of the terms can be used as terms of address, just as our terms of relationship can be so used." The classificatory system, in some cases, appears clearly as a device to assist nature in confining marriage within the same generation.[232] [232] _Mystic Rose_, 473, 474. The results of the most recent research, therefore, seem to have advanced our knowledge of the early social condition of mankind; but not to have definitively settled the problem of the former existence of communistic marriage. One rises from an examination of the literature relating to totemism and to the classificatory systems of relationship with a feeling that much more material must be gathered and exploited before we shall escape entirely from the domain of speculation as to their full meaning.[233]