From all this we may conclude that marriage was not as a rule an ephemeric occurrence among the Australian natives. In the majority of cases it lasted for life; anyhow, for a long period. To supply here the _experimentum crucis_, let us quote some contradictory instances. Lumholtz says that the women usually change their husbands so often that the children do not generally know their fathers and never grow very attached to them (_loc. cit._, p. 193; comp. below, p. 245). Salvado, speaking of the unhappy lot of an aboriginal beauty, mentions that she has very often to pass from hand to hand, being continually coveted and captured by some new lover who is stronger than her actual possessor. The same is related by Grey. Lumholtz's cursory statement is not explicit enough to enable us to judge whether it were not formed from observations of "civilized blacks." He was only a short time in personal contact with the natives, and what he gathered from the settlers applied probably in the main to blacks corrupted by contact with civilization. Salvado's and Grey's information applies only to exceptional cases when the belle excited special passions by her personal charms. Besides, from all we know, elopement, and still more capture, were not every-day occurrences which would follow each other in the case of the same woman. On the contrary, if an exceptionally desirable woman were taken away by some strong and influential aggressor from her lawful husband, the former would have power enough, personal and social, to retain her, if he had enough to secure her. That elopements occurred and that they were more frequent in the case of a beautiful and useful woman is beyond doubt. Still the picture that we would form from these three statements does not seem to fit the framework of the other facts.