AMERICAN INSTITUTE. 501 



fingered families of men, and it is easy to conceive, that if any 

 instinctive family aversion existed against such race in early times, 

 compelling them to associate and intermarry only with each other, 

 a perpetuity of the peculiarity would have ensued. In perform- 

 ing amputations of the lower extremity, surgeons do not always 

 save all they can of the limb ; experience has shown that a stump 

 is most useful when the division is made at certain points; so it is 

 probable that these accidental variations are disposed to be per- 

 manent in certain ways by the same law of adaptation, now not 

 well undersood. 



From what has been said we may conclude that, 



I. The construction of the different species of animals and plants 

 is such that no one individual can be taken as the type of the race, 

 there being to the original type a margin to allow of variation, and 

 that margin being so wide as to be covered by no one individual 

 form. 



II. This variation is produced to meet necessities by the law of 

 development, the exercise of any organ increasing its growth. 



III. When the variation occurs it is attended with a change in 

 the chemical composition of the animal or plant, based on a change 

 in the chemical composition of its food. 



IV. If the food be defective, or cannot be assimilated, the modi- 

 fication does not occur but the animal dies. 



V. These changes are always made in a direction to adapt the 

 subject of them more perfectly to such new conditions as require 

 them. 



VI. There is a tendency to reproduce these variations in the 

 progeny. • 



VII. The variations go fui-ther as they are reproduced in the 

 race. 



VIII. They stop at the line of species, and never pass that line. 



IX. While the pressure of circumstances urges them against 

 that line they are permanent. 



X. By crossing they may be carried over the line, but the re- 

 sulting hybrid is unstable, and always returns, after a few gene- 

 rations to one or other of the parent species. 



[Am. Inst.] 33 



