490 TRANSMISSION 



slope is our coefficient of regression, and its value is expressed 

 by the ratio of /Wto PM. 



Galton l finds this ratio, when dealing with mid-parents and 

 mid-filial statures, to be approximately as 2 to 3 ; " that is to 

 say, the filial deviation from P (the common mean) is on the 

 average only two thirds as wide as the mid-parental deviation. 

 I call this ratio of 2 to 3 [he says] the ratio of ' filial regression.' 

 It is the proportion in which the son is, on the average, less 

 exceptional than his mid-parent." That is, the deviation of the 

 stature of children from the mean of the race is only about two 

 thirds as wide as that of their mid-parents. 



[ SECTION VII THE MEAN OF THE OFFSPRING NOT 

 , NECESSARILY THE SAME AS THE MEAN OF 



THE PARENTAGE 



At first thought it would seem axiomatic that, on the average, 

 the offspring as a whole would be the same as the parents, 

 unless the race is undergoing change. In the table of statures, 

 however, by comparing columns 16 and 17, row 0, we see that 

 the mean of all the parents was 68.6 inches (not counting ex- 

 tremes, or 68.7 inches including extremes), while the mean of 

 the adult children was but 68. o inches (68.1 inches including 

 extremes). This indicates a loss in stature of over a half inch 

 in a single generation, unless some other influence is at work to 

 counteract this discrepancy. That counteracting influences are 

 at work we shall shortly discover, but the fact remains that the 

 mean of the offspring is seldom identical with the mean of the 

 parentage. This fact is to be construed as meaning one (or both) 

 of two things, either that the race is undergoing transforma- 

 tion, or else that all grades are not equally productive. 



In this connection it is to be noted, first, that 68.6 is not to 

 be taken as the mean of the generation to which these mid-parents 

 belong ; that mean might have been either less or more, because 

 not all members of a generation become parents, nor is the 

 parent population a random draft from the generations of the 

 mid-parents. 



1 Gallon, Natural Inheritance, p. 97. 



