SOME RESULTS OF EXPERIMENT 137 
But under domestication we seek to bring about a new working 
adjustment to conditions imposed byman. The skilful trainer 
utilizes the natural instinctive tendencies as a basis ; and, by 
a system of rewards and punishments, leads the intelligent 
modifications of behaviour along lines directed by his deliberate 
purpose. The conditions are largely those of experiment, 
and they bring out the play and range of intelligence in a way 
which would otherwise elude our observation. ‘The training 
of falcons for the chase affords a good illustration, since they 
cannot be bred in confinement, and the effects of training 
cannot therefore be hereditary. The falconer’s object is to 
modify the congenital instinctive behaviour of a bird of prey 
for the purposes of sport. She is trained to the lure at first 
at short distances, and step by step through longer flights ; she 
is taught by snatching away the lure to stoop at it repeatedly 
as often as it is jerked aside ; and then she is trained on living 
quarry, at first under easy conditions, till eventually she can be 
flown ata wild bird. And as a result a well-trained falcon will 
follow her master from field to field, regulating her flight by 
his movements, always ready for a stoop when the quarry is 
sprung. The fact that she can be thus educated for her work 
shows that her behaviour is plastic, and can be moulded by 
intelligence. Experimental conditions reveal the fact ; but’ 
under nature the moulding influence of intelligence is pre- 
sumably not less important, though it is more directly in line 
with the congenital instinctive tendencies. 
That much of the behaviour of the higher animals is guided 
by experience similar to that which plays so large a part in 
their training under the experimental conditions of domestica- 
tion is generally admitted. But what are the range and 
limits of animal intelligence, and whether it attains the level 
of rational conduct, in the restricted sense of the term 
*‘ rational,” are questions open to discussion, to which answers 
are more likely to be obtained through experiment than by 
chance observations. 
Before giving some of the results of such experiment it 
will be well to revert to the distinction, which was drawn in the 
