SOME REMARKABLE CUSTOMS, 283 



second degrees of relationship alone a bar to the marriage of 

 the Indians^ yet the Abipones, instructed by nature and the 

 example of their ancestors, abhor the very thought of marrying 

 any one related to them by the most distant tie of relationship. 

 Long experience has convinced me, that the respect to con- 

 sanguinity, by which they are deterred from marrying into 

 their own families, is implanted by nature in the minds of most 

 of the people of Paraguay," etc' 



It is likely that experience of the evils of marrying near re- 

 latives may be the main ground of this series of restrictions in 

 different parts of the world. Professor Lazarus, whose opinion 

 I asked about the matter, expressed this view, with the just 

 remarks that the observation and reasoning of savages are 

 often very accurate in practical matters requiring no instrument 

 for their observation, and that old people with a personal ex- 

 perience ranging through five or six generations would have a 

 fair ground for judging on such a question. If this physiolo- 

 gical objection, often exaggerated beyond reasonable limits, be 

 the principal basis of the series of restrictions, their various, 

 anomalous, and inconsistent forms may be connected with in- 

 terfering causes, and this one in particular, that the especial 

 means of tracing kindred is by a system of surnames, clan- 

 names, totems, etc. But this system is necessarily one-sided, 

 and though it will keep up the record of descent either on the 

 male or female side perfectly and for ever, it cannot record 

 both at once. In practice, the races of the world who keep 

 such a record at all have had to elect which of the two lines, 

 male or female, they will keep up by the family name or sign, 

 while the other line, having no such easy means of record, is 

 more or less neglected, and soon falls out of sight. Under 

 these circumstances, it would be quite natural that the sign 

 should come to be considered rather than the reality, the name 

 rather than the relationship it records, and that a series of one- 

 sided restrictions should come into force, now bearing upon the 

 male side rather than the female, and now upon the female side 

 rather than the male, roughly matching the one-sided way in 



' DobrizliofFer, vol. i. p. 63 ; vol. ii. p. 212. See Gumilla, Hist. Nat., etc., de 

 rOrenoque; Avignon, 1758, vol. iii. p. 269. 



