2 3S Evolution and Adaptation 



izing the flower with pollen of another variety having pods 

 of a different color. In these cases we do not know whether 

 the color of the fruit is influenced directly by the foreign 

 pollen, or whether the influence is through the embryo that 

 develops from the egg-cell. The action may appear to be the 

 same, however, in either case ; but because it seems probable 

 here that there is some sort of influence of one tissue on 

 another, let us not too readily conclude that this is brought 

 about through any such imaginary bodies as gemmules. It 

 may be directly caused, for instance, by some chemical sub- 

 stance produced in the young hybrid plant. If this is the 

 case, the result would not be different in kind from that of 

 certain flowers whose color may be influenced by certain 

 chemical substances in the soil. 



In the cases amongst animals, where the maternal tissues are 

 believed to be influenced by a previous union with the male, as 

 in the oft-cited case of Lord Morton's mare, a reexamination of 

 the evidence by Ewart has shown that the case is not demon- 

 strated, and not even probable. Several years ago I tried to 

 test this view in the case of mice. A white mouse was first 

 bred to a dark male house-mouse, and the next time to a 

 white mouse, but none of the offspring from the second union 

 showed any trace of black. If the spermatozoa of the dark 

 mouse are hypodermically injected into the body-cavity of the 

 female, the subsequent young from a white male show no evi- 

 dence that the male cells have had any influence on the ovary. 



The following facts, spoken of by Darwin himself, are 

 not in favor of his hypothesis of pangenesis : " But it 

 appears at first sight a fatal objection to our hypothesis 

 that a part of an organ may be removed during several 

 successive generations, and if the operation be not followed 

 by disease, the lost part reappears in the offspring. Dogs 

 and horses formerly had their tails clocked during many gen- 

 erations without any inherited effect; although, as we have 

 seen, there is some reason to believe that the tailless condi- 



