THE INTELLIGENCE OF MAMMALS 251 
is the syllogism in the dog’s mind. The dog probably did 
not reason the matter out in the explicit fashion indicated 
by the words employed by Mr. Stone, and perhaps the writer 
would not insist that he did. The bark may have made the 
small dog aware that the large dog was out of the room and 
that it was safe for him to seize the bone. He may not have 
thought that therefore he could safely seize the bone, but 
the large dog being away the small one just went after the 
bone because he was no longer restrained. It is possible, 
though less likely, that the bark simply served to direct the 
attention of the small dog from whatever object it was 
bestowed upon, and that it was then directed to the bone 
which was seized because the large dog was out of the way. 
Or it may be that when the small dog betook himself to the 
corner he fell half asleep and was brought to himself by the 
bark of the other dog. How varied are the interpretations 
that can be made of the contents of the dog’s mind! We 
may feel convinced from our general knowledge of dog 
behavior and the special circumstances of the case that 
there was something in the small dog’s mind corresponding 
to “large dog outside, I can now get the bone;”’ but our con- 
viction does not constitute proof. And so it goes with story 
after story. 
While the proof of the existence of explicit inference may 
be difficult though by no means impossible, there is a singular 
lack of conclusiveness in the arguments sometimes employed 
to prove its absence. It is argued by Thorndike that the 
gradual descent of the time curves of learning in his experi- 
ments showed the absence of reasoning. But when we 
examine these curves it becomes apparent that their shape, 
in a considerable proportion of cases, is far from gradual, 
as in most of the figures on pp. 18, 20 and 24. Small finds 
in the curves of learning of the rat that there is, as a rule, a 
