SELECTION: ORGANIC AND SOCIAL 203 



death-rate in man. A certain number of people 

 are killed every year in Britain by lightning; 

 their death is purely fortuitous. But this cannot 

 be said of phthisis, where the ehmination is in part 

 quite definitely discriminate. Even from the fact 

 that longevity is truly heritable it is evident that 

 there must be a selective death-rate in mankind. 



Lessening the Burden op the Theory. — 

 another change which seems now coming about as 

 the result of discussion and investigation is ex- 

 pressed in a growing tendency to lessen the burden 

 that has been hitherto laid — faute de mieux — on 

 the shoulders of Natural Selection. We cannot 

 do more than illustrate how different bodies of 

 workers are arriving at the same general conclusion. 



(a) If we find increasing warrant for postulating 

 the occurrence of mutations of considerable magni- 

 tude and for believing that they are not readily 

 lost when they once emerge, then it is not necessary 

 to suppose that every character has arisen by the 

 accumulation of minute steps. It goes without 

 saying that mutations must pass through the 

 selection sieve. 



(&) Whether we postulate mutations or fluctua- 

 tions, we cannot but sympathise with the heresy 

 which is often whispered, that it is very difficult 

 to give a concrete selectionist interpretation of 

 what may be called the " big lifts " * in evolution. 



'■ The difficulty in regard to " big lifts " is bound up with the 

 question whether there are ajay quahtative steps in evolution, or 

 whether change apparently qualitative may not be due to the 

 accumulation of minute quantitative changes. It is said that 

 there are no transitions between a sledge and a wheeled cart, and 

 that a new unity cannot arise piecemeal. It is difficult, however, 

 to feel confidence in these arguments from analogy. Conscience, or 

 the habit of judging our actions by a standard, is a very distinctive 



