200 
NATURAL HISTORY 
irritans^), swarming at the mouths of tliese holes, like bees 
on the stools of their hives. 
The following circumstance should by no means be 
omitted — that these birds do not make use of their caverns 
by way of hybernacula, as might be expected ; since banks 
so perforated have been dug out with care in the winter, 
when nothing was found but empty nests. 
The sand martin arrives much about the same time with 
the swallow^ and lays, as she does, from four to six white 
eggs. But as this species is cryptogam e, carrying on the 
business of nidification, incubation, and the support of its 
young in the dark, it would not be so easy to ascertain the 
time of breeding, were it not for the coming forth of the 
broods, which appear much about the time, or rather some- 
what earlier than those of the swallow The nestlings are 
supported, in common like those of their congeners, with 
gnats and other small insects ; and sometimes they are fed 
with Libellulce (dragon-flies) almost as long as themselves. 
In the last week in June we have seen a row of these sitting 
on a rail near a great pool as perchers, and so young and 
helpless as easily to be taken by hand; but whether the 
dams ever feed them on the wing, as swallows and house 
martins do, we have never yet been able to determine ; nor 
do we know whether they pursue and attack birds of prey. 
When they happen to breed near hedges and enclosures, 
they are dispossessed of their breeding holes by the house 
sparrow, which is on the same account a fell adversary to 
house martins. 
These Hirundines are no songsters, but rather mute, 
making only a little harsh noise when a person approaches 
their nests. They seem not to be of a sociable turn, never 
with us congregating with their congeners in the autumn. 
Undoubtedly they breed a second time, like the house mar- 
tin and swallow ; and withdraw about Michaelmas. 
' The flea of the sand martin, although so similar to the bed flea as to 
be scarcely distinguishable from it, is really distinct. It appears even 
to be distinct from the flea of the swallow, Pulex hirundinis (Stephens), 
and has been described as P. bifjLSC.iatus (Curtis). — Ed. 
