BIOPSYCHOLOGICAL 609 



learn in cold blood, but with an interested activity, whetted by 

 appetite. 



The "homing" of horses and some other mammals is probably 

 kinesthetic ; and as for cats that are transported to a distance in a 

 basket, more precise information is required. One requires to know, 

 for instance, the percentage of failures, and whether a cat that found 

 its way in a few days from Ayrshire to Fife could have repeated 

 the journey the following week. The "homing" power of ants and 

 bees is in the main the outcome of learning the region and taking 

 advantage of landmarks; the case of homing pigeons is by itself, 

 since man secures a graduated education for the birds; the homing 

 of migrant birds remains a puzzle. As indicated in another section, 

 the experiments made with brooding terns, removed in closed 

 baskets from the Tortugas and taken on board steamer for hundreds 

 of miles into unknown waters, whence a variable percentage returned 

 in safety, seem to prove conclusively that there is a "sense of 

 direction" whose nature and location are quite unknown. All this, 

 however, is but to emphasise the point that habituation is a very 

 important factor, and that the "learning" process need not be 

 intelligent. Very clear evidence of this is afforded by the behaviour 

 of starfishes, which learn to free themselves more and more rapidly 

 from staples which bind their arms to the experimenting table, of 

 course without pain or actual contact. There is a gradual eHmination 

 of useless movements; but we dare not speak of anything like 

 inteUigence when we are dealing with animals that have not a single 

 nerve-ganglion in their bodies. The nerve-cells are not integrated 

 into centres. Nor, on the other hand, can we put our learning to 

 ride a push-bicycle or to play a game like tennis at the simple 

 level illustrated by the starfish, for although there is the important 

 common feature of eliminating useless movements, man is con- 

 tinually demanding and receiving intelligent reasons for learning 

 as he does. One of the clearest instances of non-inteUigent "learn- 

 ing" is afforded by some experiments on earthworms which were 

 confined in T-shaped tubes filled with soil, one path leading to a 

 mild electric shock and the other to satisfaction. After 120-180 

 lessons, about six per day, the earthworms learned to avoid the 

 electric path, making only one mistake or two out of twenty trials. 

 But the lesson was learned not less thoroughly by headless earth- 

 worms! We need have no hesitation in this case in speaking of pre- 

 intelligent learning. 



Prof. Agar of Melbourne made some interesting experiments with 

 water-fleas {Daphnia carinata), water-mites (Eylais and others), 

 and the Australian freshwater crayfish {ParachcBraps bicarinatus). 

 Each animal experimented on was placed in conditions unfavourable 

 to its normal activities, such as confined space, water too shallow 

 to allow of unrestricted swimming movements, or with an excess 



VOL. I RR 



