1 64 ANIMALS' ILLUSIONS 



to go through. This is an obvious instance of the 

 disadvantage at which most animals stand in regard 

 to means of physical experiments. The horse, for 

 instance, need only feel the lintel to find out that it is 

 fixed and does not move, and is not alive and waiting to 

 hit him. But, except his lips, which are sensitive, he 

 has no member with which he can make this experiment. 

 Except the elephant and the monkey, most of the 

 ' higher ' animals suffer from this lack of the means of 

 experiment. The wonder is not that they suffer from 

 illusions, but that they make so few mistakes. 



The routine of chemical experiment gives some idea 

 of the common means by which we guard against mis- 

 taking one thing for another. The inquirer notes the 

 taste, scent and colour, and judges of the weight, 

 solubility, and, in the case of crystals, of the shape of 

 the object he wishes to identify ; he tries if it is brittle or 

 tough, he heats it or cools it. In common everyday 

 experience the number of c tests ' unconsciously applied 

 by men to prevent illusion and identify objects 

 approaches much more nearly to the number prescribed 

 for scientific inquiry than to the simple experiments 

 used by animals. There is even a test for a ghost, 

 which, since quoting Latin to it fell into disuse, usually 

 takes the form of seeing if it is ' sensitive to percussion. 1 

 Now, even this simple experiment is denied to a horse 

 when uncertain as to the reality of a figure seen by 



