598 HIRUNDO RIPARIA. 



sion, and state it to be my opinion that they take up their 

 abode in situations favourable to mining, whether there be 

 water near them or not. 



In the face of these banks the little creature commences its 

 operations by clinging with its sharp claws to a selected spot, 

 and digging with its very short and pointed bill, as with a 

 pike. One may convince himself that it is a very efficient in- 

 strument for the purpose, by employing one of similar form 

 and hardness, for a few minutes, against a section of sand or 

 loam. The circular cavity is gradually deepened, the bird 

 always forming its extremity of a conical or hemispherical shape, 

 and after some progress is made, the debris are ejected with the 

 feet. The hole, which is horizontal, generally extends to a 

 depth of about two feet, and is seldom quite straight, although 

 it makes no abrupt bend, and seems to become sinuous more 

 from accident than design. Frequently deserted holes are 

 found, some just commenced, others in which considerable pro- 

 gress has been made ; and it is probable that the cessation has 

 in them been caused by the occurrence of some insuperable 

 obstacle, a stone or bit of hard ground. 



The form of the aperture varies according to the nature of the 

 materials of which the bank is composed. When the ground 

 is clayey and tenacious, it is circular ; wiien very loose, and 

 the sand is disposed in thin layers, it is often rectangular. It 

 also, and for the same reason, varies in size, sometimes being 

 not much larger than is necessary for the passage of the bird, 

 and sometimes having a diameter of three or four inches. 



The extremity of the hole is wider than the rest, rounded, 

 and scooped out beneath, its diameter about five or six inches. 

 The nest is shallow, forming a small segment of a hollow sphere, 

 and is composed of dry grass, rudely put together, with a lining 

 of a few large feathers. In a specimen found in a sand-pit near 

 Edinburgh I can distinguish Poa trivialis, Festuca duriuscula, 

 Poa pratensis, in long stalks, with their roots, leaves, and 

 panicles, besides a spike of unripened wheat, and a leaf of Po- 

 tentilla anserina ; the feathers those of the domestic duck. 

 The eggs are of an elongated oval form, about eight-twelfths in 

 length, five and a half in breadth, white, without spots, and 

 four or live in number. 



