Chap. IV.] MANNER OF DEVELOPMENT. 127 



plenty of room. If such means were suddenly doubled 

 in Great Britain, our number would be quickly doubled. 

 "With civilized nations the above primary check acts 

 chiefly by restraining marriages. The greater death-rate 

 of infants in the poorest classes is also very important ; 

 as well as the greater mortality at all ages, and from 

 various diseases, of the inhabitants of crowded and mis- 

 erable houses. The effects of severe epidemics and wars 

 are soon counterbalanced, and more than counterbalanced, 

 in nations placed under favorable conditions. Emigration 

 also comes in aid as a temporary check, but not to any 

 great extent with the extremely poor classes. 



There is reason to suspect, as Malthus has remarked, 

 that the reproductive power is actually less in barbarous 

 than in civilized races. We know nothing positively on 

 this head, for with savages no census has been taken; 

 but from the concurrent testimony of missionaries, and 

 of others who have long resided with such people, it ap- 

 pears that their families are usually small, and large ones 

 rare. This may be partly accounted for, as it is believed, 

 by the women suckling their infants for a* prolonged 

 period ; but it is highly probable that savages, who often 

 suffer much hardship, and who do not obtain so much nu- 

 tritious food as civilized men, would be actually less pro- 

 lific. I have shown in a former work, E2 that all our do- 

 mesticated quadrupeds and birds, and all our cultivated 

 plants, are more fertile than the corresponding species in 

 a state of nature. It is no valid objection to this con- 

 clusion that animals suddenly supplied with an excess of 

 food, or when rendered very fat, and that most plants 

 when suddenly removed from very poor to very rich soil, 

 are rendered more or less sterile. We might, therefore, 

 expect that civilized men, who in one sense are highly do- 



59 ' Variation of Animals and Plants under Domestication,' vol. ii. pp, 

 111-113, 163. 



