LIGHT AND MOVEMENT 



79 



in which the tree was located as if up to the last moment she used both the 

 tree and the block as landmarks. 



The prettj^ experiment of Tinbergen and Kruyt (1938) shows the astonishing 

 rapidity and precision with which the wasp learns to relate its nest to neighbouring 

 landmarks and to appreciate a situation as a whole. A ring of 20 pine-cones 

 was placed around the nest while she was inside ; on leaving she made a study 

 of the locality for 6 seconds only (Fig. 51) ; the pine-cones were then arranged 

 similarly one foot away from the nest ; returning after 90 minutes with a 

 captured bee, she alighted in the middle of the ring of cones, a choice repeated 



T 



6 



--a- 



Fig. 50. — The Path of the Digger Wasp, PHiLAyrHcs TRiAyocLcn. 



On the left, the training situation. The path of tlie wasp ^starting from 

 the circle) is directly to its nest at the angle of the block past the tree. 



On the right, the test situation. The block is rotated througli 4.5^ and 

 tlie tree moved tirst to A and then to B. The wasp alights alternately at 

 a and b (simplified from van Beusekom). 



13 times, and only found the nest after the original situation had been restored 

 (Fig. 52). A still more extraordinary ability is seen in the wasp, Ammophila, 

 which hunts caterpillars too heavy to be brought back on the wing ; as it 

 laboriously drags its prey to the nest it is apparently able, apart from occasional 

 exploratory flights, to utilize the memories of aerial observation, probably 

 aided by light-compass orientation (Thorpe, 1943-56). 



Behaviour of this type is obviously determined by the iitiHzation 

 of a number of stimuH and experiment has sho^\^l that in making its 

 decision the insect does not condition itself to every available landmark 

 but exercises some degree of choice on principles which would differ 

 from that of a human being. Moreover, the stimuli need not be 

 simultaneous for visual memories may be retained for some considerable 

 time ; thus the bumble-bee, Bombus, will fly regularly round a number 

 of fixed landmarks in the same sequence for weeks on end (Frank, 

 19-11). Such studies are of unusual interest but our knowledge of the 

 problems they raise is yet very imperfect ; indeed, experiment has just 



Bombus 



