DISTRIBUTION OF MEMORY 221 



attacked and killed. Hence chemical stimuli of 

 certain volatile substances will excite the ants. In 

 this case we do not need to assume intelligence any 

 more than we do in the case of the tentacles of Ac- 

 tinians which, as we have seen, will immediately carry 

 a piece of filter paper soaked in meat-juice to the mouth 

 while they ignore a piece of paper soaked in sea-water. 

 The assumption of machine-like irritable structures is 

 quite sufficient here to explain the reaction. Mem- 

 ory is quite unnecessary. Possibly the behaviour of 

 the ant may be explained in the same way. Bethe 

 was able to prove by special experiments that these 

 reactions of ants are not learned by experience, 

 but are inherited. The "knowing "of "friend and 

 foe " among ants is thus reduced to different reactions, 

 depending upon the nature of the chemical stimulus 

 and in no way depending upon memory. 



Memory and intellect are supposed to be responsible 

 for the fact that an ant is able to find its way back to 

 the nest and that when " foragers " have discovered 

 honey or sugar the other ants of the nest soon go to 

 it in great numbers. The ability to communicate in- 

 formation was assumed in this case. Bethe, however, 

 was able to determine by means of ingenious experi- 

 ments that an ant, when taking a new direction from 

 the nest for the first time, always returns by the same 

 path. This shows that some trace must be left be- 

 hind which serves as a guide back to the nest. If 

 the ant returning by this path bear no spoils, Bethe 

 found that no other ants try this direction. But if it 



