1 62 The Study of Animal Life PART n 



all the mice of a house or birds of a district. The wild rabbit 

 is extremely timid, but the domesticated variety is as tame 

 as possible. In explanation of such cases we might easily 

 invoke the aid of " the principle of cessafion of natural selec- 

 tion " when the safety of the species ceased to depend upon 

 wildness, but we prefer to suppose that the direct action 

 of intelligence is to a great extent operative. 



As already noted, what are perhaps the most striking 

 examples of instinct of the second class occur among in- 

 sects. The comb-building of bees, the wars, the slave-using, 

 the agricultural pursuits of ants, have been so often de- 

 scribed that they need not detain us here. The brain of an 

 ant was to Darwin the most wonderful piece of matter in 

 the world. Wonderful, indeed, it would be if we supposed 

 that all the acts of an ant were truly instinctive, that is, 

 that the nervous machinery of co-ordination was ready, 

 waiting only the appropriate stimulus to evoke any one of 

 that series of nicely-adjusted actions. But if we suppose 

 that individual intelligence has a considerable share in that 

 co-ordination, then the brain of an ant, though still very 

 . wonderful, is not to us quite so astounding an arrangement 

 of particles as it was to Darwin. 



The third class of instincts, those connected with 

 reproduction, comprise such actions as the building of 

 homes and nests, the storage of food for the use of 

 young that may never be seen, and the care of young after 

 birth. 



The nest-building of birds would form a very good sub- 

 ject upon which to experiment, in order to determine how 

 far such a complex act may be truly instinctive, and how 

 far it is perfected by training, by imitation, and by intelli- 

 gent practice and observation. The method would be to 

 isolate young birds from their parents and from all other 

 creatures of their kind, so as to preclude training and 

 imitation, and then see how far the nests that they 

 built resembled the typical nest of their species. Then one 

 might remove other birds from their parents but allow them 

 the society of the members of an allied species, but one 

 whose nests differed to some observable extent from those 



