Mental Processes in Animals, 353 



tion." * And, speaking of the sole, the same observer 

 saySjt " In searching for food the sole creeps about on 

 the bottom by means of the fringe of fin-rays with which 

 its body is edged, and, thus slowly moving, it raises its 

 head upwards and sideways, and gently pats the ground 

 at intervals, feeling the objects in its path with the peculiar 

 viliform papillae which cover the lower (left) side of its head 

 and face. In this way it will examine the whole surface 

 of the floor of the tank, stopping and going back to in- 

 vestigate pieces of stick, string, or other objects which it 

 feels below its cheek." 



If we admit the fact that carp come to be fed at the 

 sound of a bell, we have evidence that some fishes can 

 associate an arbitrary sound with the advent of things 

 good to eat. But it is, perhaps, better at present to regard 

 the fact as one requiring verification. 



That some birds can associate arbitrary signs with their 

 percepts will be admitted by all who have watched their 

 habits. And from its peculiar and almost unique power 

 of articulation, the parrot shows us that not only may the 

 words suggest a construct, but that the sight of the con- 

 struct may suggest the word that it has heard associated 

 with the object by man. Mr. Romanes gives evidence 

 which satisfies him that a parrot which had associated the 

 word " bow-wow " with a particular dog, uttered this sound 

 when another dog entered the room. The word was here 

 suggested at sight, not of the same object, but of an object 

 which the bird recognized as similar. A somewhat similar 

 case is furnished by one of my own correspondents (Miss 

 Mabel Westlake). " We left London," she says, *' in 

 December, 1888, and brought our grey parrot with us ; but 

 left behind with a friend our favourite cat, a dark tortoise- 

 shell with a white breast, the forehead clearly marked with 

 a division down the middle to the tip of the nose. This 

 led to our calling her * Demi.' For a week or two after 



* Journal of Marine Biological Association, New Series, toI. i. No. 2, 

 J). 214. I should not myself have used the word " explanation." 

 t Ibid. vol. i. No. 3, p. 240. 



2a 



