Birds by Land and Sea 



On the loth September sudden gales and rains 

 caused the river to rise about fifteen feet in one night, 

 and when I went out early the following morning I 

 found the water in places pouring over the high 

 embankment into the meadows. Strangest sight of 

 all was the companies of sand-martins flying on their 

 usual beats, and with their usual cheerful note, at 

 the edge of the swollen river, picking from the 

 water the insects which had been washed up from 

 the submerged grassy banks. Although their nest- 

 ing-holes were now some five or six feet beneath the 

 surface of the water, the birds still recognized the 

 old landmarks, and continued to beat up and down 

 on the former lines, apparently unconscious of the 

 catastrophe which had befallen them. However, 

 they probably suffered no great loss by the inun- 

 dation, for I had noticed that for some time previously 

 they no longer entered their nesting-holes. Cold 

 days and nights succeeding, the birds drew off 

 rapidly, the greater portion of them leaving on 

 the 1 4th September, and only a few remaining 

 until the morning of the I5th. By noon of that 

 day these, too, had disappeared, so that punctually 

 at the middle of September the sand-martin left us. 



Although what is known of the migration of 

 birds is meagre enough as compared with what 

 remains to be learned, there are few chapters in 

 the history of bird science which are more engaging 

 in the retrospect than the gradual formation and con- 

 firmation of the theory that certain species of birds 



