Correlated Variation 199 



wool than did their ancestors, or do the newcomers.' 

 Now let us assume that what is ' alleged ' is true ; still 

 we find that it is not said that there is any evidence that 

 the young sheep inherit the harshness of wool ; it is only 

 alleged that the ' harshness increases with succeeding 

 generations,' and that ' the generations which have in- 

 habited such regions for several generations produce natu- 

 rally a harsher wool.' There is absolutely nothing here 

 to lead us to believe that the harshness has become con- 

 genital at all. Wool is not cut till the sheep has been 

 alive long enough to grow it. So it is natural to think 

 that each sheep acquires the harshness for himself. But 

 how account for the increasing harshness in succeeding 

 generations ? That is what has impressed Mr. Bather. 

 This might appear true from the fact that each shearing 

 of the wool of the same sheep was harsher than the last ; 

 for in order to compare the harshness of two generations, 

 sheep of the same age, measured by the number of times 

 they had been sheared, would have to be compared. But 

 again, waiving this, let us assume that there is a congenital 

 difference, the quality of wool being more harsh for 

 later generations, then how can we account for this 

 increased congenital tendency to harshness ? 



We might say that the increase in the harshness of the 

 wool in subsequent generations was due to the natural 

 selection of sheep with congenital variations ; this would 

 be open to the objections so frequently urged against nat- 

 ural selection, that it is not likely that such variations 

 would come in such great numbers; and also to the objec- 

 tion that the difference in harshness of the wool might not 

 be of utility. But the principle of organic selection act- 

 ing on correlated variations would meet both these objec- 



