HEREDITY IN MAN 367 



of the environment in the second sense, such as habit and what 

 may be regarded as tools, namely, modes of thought and so on. 

 In order to get at the characters we must disentangle or allow for 

 these tools. This difficulty does not arise in the case of physical 

 characters. When measuring the physical strength of two men 

 we are not likely to allow one man to throw an object with a 

 throwing-stick and another to use his own unaided strength. 

 But we are apt when comparing ability to forget that one man may 

 have learnt to distinguish between the categories when another 

 may not have done so. To this difficulty we shall refer again later. 

 These predispositions have their basis in the germinal consti- 

 tution and are therefore inherited. It is probable that we should 

 regard the germinal constitution as consisting of unit-factors. 

 However that may be, it is clear that either a differential birth- 

 rate (reproductive selection) or a differential death-rate (lethal 

 selection), will change the average germinal constitution of the 

 race among which they are operative. Bearing this in mind, we 

 may inquire when and where in human history we can detect 

 reproductive and lethal selection at work. And we must re- 

 member that in the first place not merely obvious factors such as 

 war have been and are at work, but also that subtle changes in 

 social life and social organization may have profound effects, and 

 that in the second place not merely obvious characters may be 

 favoured or eliminated, but the least striking and most minute 

 (though not necessarily the least important) characters both 

 mental and physical may be similarly affected. 



