ANIMAL INDUSTRIES 139 



tion of instinct, which must necessarily cover this 

 wholly unexplained power of local divination. So 

 far as a recognised theory of the origin of instinct 

 may be said to exist, there is very little to be 

 added to the form in which it is presented in M. 

 Houssay's treatise. He considers that by a care- 

 ful comparison and classification of observed facts, 

 it is possible to find in animals all the intermediate 

 stages between a deliberate, reflective action, and 

 an act that has become instinctive, and so inveterate 

 to the species that it has reacted on its body, and 

 produced new and special organs. ' If an individual 

 is led to reproduce the same series of actions, it 

 contracts a habit; the repetition may be so frequent 

 that the animal comes to accomplish it without 

 knowing it ; the brain no longer intervenes ; the 

 spinal cord or chain of nerves alone govern this 

 order of acts, to which has been given the name 

 of reflex actions. This tendency to reflex action 

 may be transmitted by heredity to the descendants, 

 and then becomes an instinct' On the case of the 

 bee, assuming this hypothesis to be the true one, 

 the determination of the hexagonal form for its 

 cells, which is just that which calculation shows to 

 demand the least quantity of wax for the storage 

 of the greatest quanity of honey, must originally 

 have been due to trial and reflection.^ No one 



