214 CERTAIN LAWS OF VARIATION. 



of these eggs is distinctly asymmetrical, due, one may 

 imagine, to the weeding out of a larger proportion of 

 the birds producing the smaller eggs. The curve of 

 distribution of the English eggs is practically sym- 

 metrical, so that selection in them probably removed 

 extreme individuals to an equal extent in either direc- 

 tion. It should be borne in mind, however, that the 

 skewness of the curve of distribution of the American 

 eggs may, after all, be due only to Reproductive Selec- 

 tion, i. e. y to a greater fertility of the birds producing 

 the larger eggs. 



To what, then, are the diminished size of the eggs and 

 their increased variability due? Probably they are 

 nothing more than a direct response to the change in the 

 environmental conditions. These conditions, though, 

 by virtue of the absence of enemies and other checks, 

 more suitable to increase in the numbers of the spar- 

 rows, are probably more variable, and, on the whole, not 

 so suitable to individual growth. Thus in England the 

 sparrows have been subjected to a very keen competi- 

 tion for a very long time, and so have become closely 

 adapted to their surroundings. In other words, their 

 environment is as favourable as it can be for their 

 maximum growth, and for a minimum of variability. 



Another series of observations made by Bumpus * to 

 prove his views may from their intrinsic interest be 

 cited also. They concern the variations of the peri- 

 winkle, LiHorina littorea. This mollusc appears to 

 have been first carried to American shores about 1850, 

 and since then it has spread with great rapidity along 

 the New England coast. For statistical purposes, 1000 

 *Zo51. Bulletin, vol. i. p. 247. 



