REASON AND LANGUAGE. 159 



the same gesture to other dogs, it is a singular fact, 

 which we are inclined to be sceptical about. We doubt 

 whether Mr. Romanes can be sure that the parrot did 

 so entirely drop the use of this sign. But whether it did 

 or not does not matter in the slightest degree for the 

 argument. The dropping of it could be no indication 

 of intellect. The recognition by a really intellectual 

 nature, of other dogs as being " dogs," would not make 

 the first known dog a bit less a dog, or cause it to be 

 considered less a dog. That the parrot could practically 

 distinguish between the familiar terrier and strange 

 dogs no person can doubt. Every dog who lives with 

 a cat in the house knows his friend " Tom " from all 

 other cats, and generally shows a disposition to treat the 

 latter very differently from the way in which " Tom " is 

 treated by him. In this anecdote, if we accept without 

 question all the facts stated, there is not a scintilja 

 of evidence of the possession by the parrot of an in- 

 tellectual nature ; there is nothing but what may be 

 entirely accounted for by that power of association and 

 consentient apprehension which we all allow that 

 animals possess. 



Mr. Romanes distinguishes * " four divisions of the 

 faculty of articulate sign-making — namely, meaningless 

 imitation, instinctive imitation, understanding words as 

 irrespective of tones, and intentional use of words as 

 signs." We do not quite understand how " understand- 

 ing words " can be a division of " sign-making," and 

 we object to his remark that the understanding of words 

 "implies, per se, a higher development of the sign- 



* p. U7- 



