232 Lamarckia7i Heredity and Teleology 



case it be real, the outcome of the evolution movement 

 might still, on the whole, be the same as if it were due to 

 the natural selection of favourable variations from a great 

 many cases distributed fortuitously or by the law of prob- 

 ability. This has been shown, in fact, to be the case in 

 recent investigations in moral statistics ; e.g.^ suicides are 

 distributed in accordance with the law of probability, and 

 vary with cHmate, food-supply, etc., in a way which can 

 be plotted in a curve, despite the fact that each suicide 

 chooses to kill himself. That is, the result is as regular 

 and as liable to exact prediction, if we take a large popu- 

 lation, as are deaths from disease or accident, or other 

 ' natural ' events in which purpose and choice have no 

 part. In such cases, indeed, we have results which are 

 subject to laws as definite as those of mechanics, although 

 the individual data are teleological in the sense of following 

 individual purpose. This case and the reverse, indicated 

 above, show the fallacy of claiming that the exercise of 

 individual purpose is necessarily bound up with a teleologi- 

 cal movement in evolution. 



§ 5. Natural Selection not Unteleo logical 



But there is another supposition open to objection in 

 the view which requires Lamarckian heredity, in order to 

 secure teleology in evolution ; the position that natural 

 selection, working on so-called * fortuitous ' or ' chance ' 

 variations, is 'blind' and unteleological. It has been 

 found that biological phenomena — variations in particular 

 — follow the definite law of probability; in short, that 

 there is no such thing in nature as the really fortuitous or 

 unpredictable. Natural selection, therefore, works upon 

 variations which are themselves subject to law. If this 



