1886.] 



Family Likeness in Stature. 



45 



" squadron " numbering 1000 statures each way, the whole standing 

 upon a square base. Our squadron may be considered as made up of 

 ranks (parallel to the plane of zx) as in Z, or of files (parallel to the 

 plane of zy) as in A. The ranks, as we have seen, are all similar 

 stature-schemes, the files are all rectangles which have the same 

 breadths but are of dissimilar heights. 



It is now easy to give a general idea, to be developed as we proceed, 

 of the way in which any large sample of a population gives rise to a 

 group of distant kinsmen in any given degree, who are statistically 

 (in all respects except numbers) undistinguishable as regards their 

 statures from themselves. I must suppose for convenience of ex- 

 planation, that tall, short, and mediocre men are equally fertile 

 (which is not, however, strictly the case, the tall being somewhat 

 less fertile than the short*), and then on referring to fig. 2, the 



Fig. 2. 



A 



fortunes of the distant descendants of two of the rectangular files of 

 squadron A will be seen traced. 



As the number of kinsmen, in any remote degree we please to 

 specify, of the men in each of the two files is about the same ; I take 

 1000 of them in each case. Again, as the stature-schemes of those 

 kinsmen are identical with those of equal numbers of men taken at 

 random, as samples of the general population, it follows that they 



# Oddly enough, the shortest couple on my list hare the largest family, namely 

 sixteen children, of whom fourteen were measured. 



