294 MENTAL EVOLUTION IN ANIMALS. 



eludes that the sense of direction cannot depend upon any- 

 process of dead-reckoning. At the suggestion of Mr. Darwin 

 he also tried the effect of attaching a magnetized needle to 

 the thorax of a bee ; but the bee having succeeded in getting 

 rid of the encumbrance, he did not repeat the experiment. 



Now, although the observations with the rotating box are 

 no doubt very interesting, they do not appear to me to sustain 

 the definite conclusion that the sense of direction is not due 

 to a process of dead-reckoning. It is of course impossible 

 to suppose that the bees could retain a register of all the 

 turns to which they were submitted in the sling, and, there- 

 fore, if it were certain that they found their way home by 

 means of their sense of direction, I should agree with 

 M. Fabre in concluding, once for all, against the theory of 

 dead-reckoning. But there is no evidence to show that the 

 bees which found their way home did so by means of their 

 sense of direction. It is quite possible that they found their 

 way home simply from their knowledge of land-marks ; for 

 the distance to which they were taken was only three kilo- 

 metres, and it is known that the hive-bee will go three times 

 that distance in its ordinary foraging excursions.* Moreover, 

 the fact that only a comparatively small number of the bees 

 succeeded in returning (about 22 per cent.), is suggestive of 

 the explanation that these were the only ones which, during 

 the random flight of the whole number in sundry directions, 

 happened to encounter land-marks with which they were 

 familiar. I am therefore inclined to feel that any sense of 

 direction which existed in these insects may very well have 

 been rendered useless by these experiments, and yet that the 

 results of the experiments might have been exactly those 

 which M. Fabre describes. 



Keturning, however, to the case of migration, I think it is 

 not very improbable that the sense of direction may be greatly 

 assisted by observing the direction of the sun with reference 

 to the appropriate line of flight. It is true that many migra- 

 tory birds fly at night ; but in this case, even if the moon is 

 not available to steer by instead of the sun, during much of 

 the night the directions of sun-set and sun-rise are clearly 

 iudicated by the light of the sky ; and it appears that on 

 very dark and cloudy nights migratory birds are apt to become 



* Sec Animal Intelligence, p. 150. 



