BANK SWALLOW OR SAND MARTIN. 191 



their summer residence, great numbers die in their holes, and many have 

 been found there in a state bordering on torpidity. 



Their food, which consists of small insects, principally of the hymenopterous 

 kind, even during winter in the Floridas, is procured on wing. They very 

 seldom approach walls or the trunks of trees to seize them, but frequently 

 snatch them from the tops of grasses or other plants on which they have 

 alighted. They also seize small aquatic insects; but, although I suspect that 

 they disgorge in pellets the harder parts of these, I have no proof, obtained 

 from actual observation, that they do so. 



The holes perforated by this species for the purpose of breeding require 

 considerable exertion and labour. They are usually bored at the distance of 

 two or three feet from the summit of the bank or surface of the ground, to 

 the depth of about three feet, but sometimes to that of four or even five. 

 They are near each other or remote, according to the number of pairs of 

 Swallows that resort to that place, and the extent of the bank. In one situa- 

 tion you may find not more than a dozen pairs at work, while in another 

 several hundreds of holes may be seen scattered over some hundreds of 

 yards. On the bluffs of the Ohio and the Mississippi there are many very 

 extensive breeding-places. While engaged in digging a sand-bank on the 

 shore of the Ohio, at Henderson, for the purpose of erecting a steam-mill, I 

 was both amused and vexed by the pertinacity with which the little winged 

 labourers continued to bore holes day after day, whilst the pickaxes and 

 shovels demolished them in succession. The birds seemed to have formed 

 a strong attachment to the place, perhaps on account of the fine texture of 

 the soil, as I observed many who had begun holes a few hundred yards off 

 abandon them, and join those engaged in the newly opened excavation. 

 Whether the holes are frequently bored horizontally or not I cannot say, but 

 many which I examined differed in this respect from those described by 

 authors, for on introducing a gun-rod or other straight stick, I found them to 

 have an inclination of about ten degrees upwards. The end of the hole is 

 enlarged in the form of an oven, for the reception of the nest, and the accom- 

 modation of the parents and their brood. 



W r hen the birds have for awhile examined the nature of the bank, they 

 begin their work by alighting against it, securing themselves by the claws, 

 and spreading their tails considerably, so as, by being pressed against the 

 surface, to support the body. The bill is now employed in picking the soil, 

 until a space large enough to admit the body of the bird is formed, when the 

 feet and claws are also used in scratching out the sand. I have thought that 

 the slight ascent of the burrow contributed considerably to enable the bird 

 to perform the severe task of disposing of the loose materials, which are seen 

 dropping out at irregular intervals. Both sexes work alternately, in the 



