116 EVOLUTION OF LIVING ORGANISMS. 



Instincts, or mental adaptations, are " hereditary " 

 in the popular sense (see p. 42) ; that is to say, will 

 reappear in succeeding generations under normal con- 

 ditions, since they correspond to metabolic processes 

 which depend on the interaction of certain constant 

 factors of inheritance with certain factors present in 

 the environment. They vary, and in all probability are 

 built up and preserved by natural selection according to 

 their usefulness in the struggle for existence. But simple 

 reflexes and tropisms need not be useful; on the con- 

 trary, as in the case of variation generally (p. 77), they 

 may be useless or even injurious. Moreover, a tropism 

 which is useful in the normal environment may be fatal 

 under other circumstances e.g. the moth and the candle. 

 In so far as they have been built up by selection, in- 

 stincts will be found to be advantageous. Apparent 

 exceptions may be due to their being relics of past 

 adaptations fallen into disuse, or to some change in the 

 environment. What we have said in previous chapters 

 about the development of the structural characters of 

 organisms applies equally to the evolution and specialisa- 

 tion of the metabolic processes to which the mental 

 characters correspond. 



Sequences of interlocking and co-ordinated reflexes, 

 each one of which sets off the next, give rise to the most 

 complicated instinctive behaviour, as we see in the 

 actions of our own internal digestive organs ; but in 

 the more elaborately differentiated instincts of animals 

 yet another element can be found, known as " associative 

 memory." For example, the homing instinct of insects, 

 enabling them to find their way back to the nest from 

 a distance, has been shown to be due to the retention of 

 effects produced by previous responses to visual stimuli. 

 This lasting and cumulative effect of responses to stim- 

 ulus enters very largely into the behaviour of the higher 

 animals and of man, but we know very little indeed of 

 the corresponding physico-chemical processes in such cases. 



