v.] IN THE THEORY OF NATURAL SELECTION. 311 



but the growth is not such as to crush the embryo, which 

 remains free in the cavity, the so-called larval chamber, which 

 is formed around it. It would be out of place to discuss here 

 the question as to how we can conceive that the plant is thus 

 compelled to produce a growth which is at any rate indifferent 

 and may be injurious to it ; and which, moreover, is exactly 

 adapted to the needs of its insect-enemy. But it is at all events 

 obvious that this cannot be an example of a self-protecting 

 reaction under a stimulus, and that therefore an organism does 

 not always respond to external stimuli in a manner useful to 

 itself. 



But even if we could accept the suggestion that the purpose- 

 ful reaction of an organism under stimulation is a primary and 

 not a secondarily produced character, such a principle would 

 by no means suffice for the explanation of existing adaptations. 

 Nageli attempts to explain certain selected cases of adaptation 

 as the direct results of external stimuli. He looks upon the 

 thick hairy coat of mammals in arctic regions, and the winter 

 covering of animals in temperate regions, as a direct reaction of 

 the skin under the influence of cold. He considers that the 

 horns, claws, and tusks of animals have arisen directly as 

 reactions under stimuli applied to certain parts of the surface 

 of the body in attack and defence \ This interpretation is 

 similar to that offered by Lamarck at the beginning of this 

 century. At first sight such a suggestion appears to be 

 plausible, for the acquisition of a thick hairy covering by the 

 mammals of temperate regions is actually contemporaneous 

 with the cold season of the year. But the question arises as to 

 whether the production of a larger number of hairs at the begin- 

 ning of winter is not merely another instance of a secondary 

 character, like the assumption of a green colour by the tree-frog 

 under the stimulus exerted by strong light. 



In the case of the hairy coat it is only necessary to produce 

 a larger number of structures such as had existed previously . 

 but how can it have been possible for the petals of flowers, 

 with their peculiar and complex forms, to have been developed 

 from stamens as a direct result of the insects which visit them 

 in order to obtain pollen and nectar ? How could the creeping 

 of these insects and the small punctures made by them 



1 I. c, p. 144. 



