THE BANK-SWALLOW. 19 



pers to a point like a sailor's marlinspike, or, rather, 

 like the points of a pair of fine compasses when 

 shut. If we compare this little sharp borer, as we 

 may well call it, with the caliper-like mandibles of 

 the sand-wasps (Sphecida, LEACH), and of the bur- 

 rowing bees, which, like this swallow, excavate gal- 

 leries proportionable to their size in hard sand,* we 

 are compelled to confess that the bird is furnished 

 with the more efficient instrument. Its operation 

 also is very different. The insects alluded to gnaw 

 into the sand, or, rather, bite off a portion of it, and 

 carry it out of the hole in their mouths ; but the 

 bank-swallow, as we have had an opportunity of 

 observing, works with its bill shut. This fact our 

 readers may verify by observing their operations 

 early in the morning, through an opera-glass, when 

 they begin in the spring to form their excavations. 

 In this way we have seen one of these swallows 

 cling with its sharp claws to the face of a sandbank, 

 and peg in its bill as a miner would do his pickaxe, 

 till it had loosened a considerable portion of the hard 

 sand and tumbled it down among the rubbish below. 

 In these preliminary operations it never makes use 

 of its claws for digging ; indeed, it is impossible it 

 could, for they are indispensable in maintaining its 

 position, at least when it is beginning its hole.f 



We have farther remarked that some of this swal- 

 low's holes are nearly as circular as if they had been 

 planned out with a pair of compasses, while others 

 are more irregular in form ; but this seems to depend 

 more on the sand crumbling away than upon any 

 deficiency in the original workmanship. The bird, 

 in fact, always uses its own body to determine the 

 proportions of the gallery, the part from the thigh 

 to the head forming the radius of the circle. It 

 does not trace this out as we would do, by fixing a 

 point for the centre around which to draw the cir- 

 cumference. On the contrary, it perches on the 



* See Insect Architecture, chap, iii., &c. t J. Rennie. 



