HUMAN LONGEVITY 



59 



and centenarians as of persons generally were produced by par- 

 ents who themselves had no longevous parentage whatever. It 

 seems clear beyond question of doubt that breeding counted 



S35 



c/y/hy unc/er* S 

 years o/'offe. 



A7ear?'(2^en'///ei//?7e o^ aye 



Fig. 7. Expectation of life in years (mean-after-lifetime) at ages 20, 40, 60, and 



80, of fathers of children dying (a) under 5 years of age (solid bars) and 



(b) 80 and over years of age (cross-hatched bars) 



mightily in the production of these nonagenarians and centen- 

 arians. 



Let us now turn to another method of approach, and a wholly 

 different material, to get still another view of the importance 

 of inheritance in the matter of longevity. Suppose one were to 



