594 The Descent of Man. Fabt 111. 



which savages are exposed, and some of their habits, are favour- 

 able to natural selection ; and this comes into play at the ssme 

 time with sexual selection. Savages are known to suffer severely 

 from recurrent famines ; they do not increase their food by 

 artificial means ; they rarely refrain from marriage,"^ and generally 

 marry whilst young. Consequently they must be subjected tb 

 occasional hard struggles for existence, and the favoured indi- 

 viduals will alone survive. 



At a very early period, before man attained to his present 

 rank in the scale, many of his conditions would be different from 

 what now obtains amongst savages. Judging from the analogy 

 of the lower animals he would then either live with a single 

 female, or be a polygamist. The most powerful and able males 

 would succeed best in obtaining attractive females. They would 

 also, succeed best in the general struggle for life, and in defend- 

 ing their females, as well as their offspring, from enemies of all 

 kinds. At this early period the ancestors of man would not bo 

 sufBciently advanced in intellect to look forward to distant 

 contingencies ; they would not foresee that the rearing of all 

 their children, especially their female children, would make the 

 struggle for life severer for the tribe. They would be governed 

 more by their instincts and less by their reason, than are savages 

 at the present day. They would not at that period have 

 partially lost one of the strongest of all instincts, common to all 

 the lower animals, namely the love of their young offspring ; and 

 consequently they would not have practised female infanticide. 

 Women would not have been thus rendered scarce, and poly- 

 andry would not have been practised ; for hardly any other cause, 

 except the scarcity of women seems sufficient to break down the 

 natural and widely prevalent feeling of jealousy, and the desire of 

 each male to possess a female for himself Polyandry would be a 

 natural stepping-stone to communal marriages or almost pro- 

 miscuous intercourse ; though the best authorities believe that 

 this latter habit preceded polyandry. During primordial times 

 there would be no early betrothals, for this implies foresight. Nor 

 would women be valued .merely as useful slaves or beasts of 

 burthen. Both sexes, if the females as well as the males were per- 

 mitted to exert any choice, would choose their partners not for 

 mental charms, or property, or social position, but almost solely 

 from external appearance. All the adults would marry or pair, 



'« Burchell say.=i ('Travels in S. Azara ('Voyages dans I'Am^rique 



Africa,' vol. ii. 18'24, p. 58), that Mei'id.' torn. ii. 1809, p. 21) maltes 



among the wild nations of Southern precisely the same remark in regan! 



Africa, neither men nor women ever to the wild Indians of Sctith Ara» 



pass their live^ "ji a state of celihacy. rica. 



