318 STATISTICAL STUDY OF INHERITANCE 



of the stability of type, and may be thus expressed. This word 

 " type " has for its central idea the existence of a limited 

 number of recurrent forms — forms which have attained a con- 

 siderable degree of organic stability. A deviation from the 

 type may mean the attainment of a new position of organic 

 equilibrium, and many " sports " are said to be very stable ; 

 but it may also mean a position of instability from which a 

 regression to the old equilibrium is what might be expected. 

 Just as certain kinds of cells have very definite dimensions, 

 doubtless dependent in part on the optimum adjustment between 

 the volume and the surface, so many animals have a very definite 

 limit of growth, which doubtless represents a condition of con- 

 stitutional equilibrium. Where this is the case, it is easy to 

 understand that marked deviations in the direction of giants 

 or in the direction of dwarfs would tend to be unstable. Their 

 offspring may tend to regress to the position of stability simply 

 because it is the physiological optimum in given conditions. 

 The regulative phenomena in development would tend to secure 

 the regression, in the same mysterious way as they secure the 

 development of a perfect larva from a mutilated embryo. In 

 the particular case of human stature, a deviation of a few inches 

 may be quite immaterial, but it is easy to think of organisms 

 in which the proportions of the various bodily dimensions are 

 very important. 



The other reason which Galton gives for the occurrence of 

 regression is found in what may be called the fact of mosaic 

 inheritance. The child inherits partly from its parents, partly 

 from its ancestry. " In every population that intermarries 

 freely, when the genealogy of any man is traced far backwards, 

 his ancestry will be found to consist of such varied elements 

 that they are indistinguishable from a sample taken at haphazard 

 from the general population. The mid-stature M of the remote 

 ancestry of such a man will become identical with P [the mean 

 of the present population] ; in other words, it will be mediocre." 



