548 LONGEVITY AND RELATIVE VIABILITY IX MAMMALS AND BIRDS, 



variations arise, that is to say changes in diathesis, structure, 

 function or habit, which enable the possessors of these to make a 

 better fight against enemies, or diseases, or any part of their hostile 

 environment, then we expect such variations to be preserved, and, 

 consequently, the constitution of the organism to be modified in 

 the direction of adaptation to its environment. It is no part of 

 my present argument to discuss whether such variations appear in 

 response to the environment, or by large or small stages ; these 

 a,re wide questions not germane to the present issue. But they 

 do arise, and in such a fashion there have come about increases or 

 decreases in size, advances in structui-e or degenerations, protective 

 colours and wea.pons — the whole armouiy offered by variation to 

 the choice of natural selection. Such improved powers of resistance 

 to the environment may increase the average specific longevity, 

 and indirectly may afiect the potential longevity. But the attain- 

 ment of the latter is too rare an event to come under the direct 

 control of natural selection. 



There is no more important part of the structure and function 

 of an organism than that associated with repi-oduction. The 

 impulses and instincts connected with fertilization, the time, 

 duration and frequency of rutting, the numbers of ova that 

 are fertilized and allowed to develope, the duration of embryonic 

 development, the amount and character of post-natal brood-care, 

 the whole apparatus and machinery of the reproductive process, 

 present an extraordinary series of modifications from species to 

 species, and from individual to individual. They are subject 

 to much variation amongst individuals at the present time, 

 and it may well be presumed that they have oftered a rich 

 material to natui^al selection in the past, Weismann ha,s 

 established a correlation between longevity and reproduction, 

 but I think his interpretation of the correlation must be reversed. 

 It is not longevity that has become adapted to reproduction, but 

 the rate of reproduction has been adapted not to potential 

 longevity, but to average specific longevity. Average specific 

 longevity is simply a measure of the pressure of conditions on a 

 species, and its constitution, including its mode of reproduction, 

 has been shaped in conformity with the pressure of these 

 conditions. Potential longevity is a bye-product of the process ; 

 it ciepends directly on the constitution, and is simply the result of 

 the fact that if the constitution be adapted to meet the average 

 mortality, it ixiust be able to outlive, a,nd will outlive, the average 

 duration of life in a number of fortunate cases. 



