A STUDY OF A CRUSTACEAN. 59 



the estuary, with one or two negligible exceptions, made no attempt 

 to reach the crustaceans ; the faithful surface of the sand 

 recorded every disturbance of the burrows ; and although it was 

 natural to think that a restless party of birds were actually 

 feeding, a subsequent examination of the spot, or careful obser- 

 vation through our glasses, proved that the Corophium burrows 

 were not disturbed. 



The Black-headed Gulls, although feeding exclusively on 

 Corophium for many days, were powerless to reach the crusta- 

 ceans inhabiting the uncovered sands. They could catch only 

 those living in sand covered by water, and if the water was more 

 than three inches deep they were again powerless. Each pool, 

 however, shrank visibly from day to day, until the water vanished 

 completely ; and thus each day brought a smaller ring of new 

 ground within reach of the birds. The method employed in 

 feeding, although often observed, and described in several 

 northern journals, seems to have escaped wider notice in the 

 regular literature of ornithology. 



The Gull stands in the water, and, holding its body horizon- 

 tally, dances vigorously with alternate steps for a minute or 

 more, but with no change of position. This action on the sand, 

 possibly by filling up the burrows, alarms the crustaceans, which 

 rise to the water and scatter in flight. As soon as they appear 

 the bird stops its dance for a second or so, and, still remaining 

 precisely in the same spot, snaps in the water at the swimming 

 animals. On imitating the action with the tips of my fingers, and 

 of course with the same result, I found that the crustaceans were 

 readily detected by the sense of touch as they struggled to the 

 water; and this suggested the possibility of the webs of the 

 Gull's toes being used as tactile organs. The point is well worth 

 consideration, for if established it would explain the presence of 

 highly developed webs in birds which are addicted to wading but 

 rarely use their feet in swimming. 



Sometimes a Gull would remain in one spot for so long a 

 period as half an hour, gravely and patiently dancing the whole 

 of the time. The result would be a crater-like depression six 

 inches in diameter and an inch deep ; but, if the birds were not 

 disturbed, they would move gradually backwards, and in the 

 course of a few hours make shallow furrows varying in length 



F 2 



