HEREDITY 491 



No fact is better known among statisticians than that wives 

 differ from daughters and mothers differ from wives, which 

 means that all women (daughters) do not marry, and that not 

 all wives are mothers (selection) ; that is to say, parents are 

 a selected draft from the entire population, and we should not 

 expect to find their mean the same as that of their offspring, 

 which closely approaches that of the general population. 



SECTION VIII EXTREMES OF A RACE RELATIVELY LESS 

 PRODUCTIVE THAN THE MEANS 



In the table of statures a strong tendency is evident toward 

 increased height and a still stronger one toward decreased 

 stature (see sect, ix, " Progression "). This being true, the racial 

 distribution would rapidly spread, if not entirely divide, into two 

 races, giants and dwarfs, unless prevented by some principle such 

 as natural selection. This principle in this particular instance is 

 evidently relative infertility, a principle easily deduced in at 

 least two ways : 



1 . One hundred and three children are recorded at or below 

 64.2, and only one parent below 64.5 (see table of statures). 

 Clearly most short children do not become parents, else the race 

 would rapidly degenerate as to size. This agrees with common 

 observation, which is that dwarfs do not marry. When, however, 

 the principle is applied to degeneracy and crime, the case is dif- 

 ferent, for criminals often produce more than their normal ratio, 

 and many of their offspring, by the principle of progression 

 (see sect, ix), are frightful degenerates. 



To what extent giants marry is not so clearly indicated by 

 this table, but that they are less fertile than the average when 

 they do marry is clearly shown by comparing columns 16 and 17. 



2. The average of fertility in this table as a whole is almost 

 exactly 4^ (928 -r- 205) per mid-parent. This average is well 

 sustained in the lower and middle statures, rising in many 

 cases to 5 per mid-parent ; but it rapidly lessens in the higher 

 statures, for from 70.5 up it is approximately 3. This, too, 

 agrees with common experience, namely, that extremes of a race 

 are generally less fertile than the means. Indeed, it is commonly 



