150 
THE NATURAL HISTORY 
[LETT. 
east and south-west) are too shallow, the nests are washed down 
every hard rain ; and yet these birds drudge on to no purpose from 
summer to summer, without changing their aspect or house. It 
is a piteous sight to see them labouring when half their nest is 
washed away, and bringing dirt " to patch the ruins of a fallen 
race" — " generis lapsi sarcire ruiDas." Thus is instinct a most 
wonderful but unequal faculty ; in some instances so much 
above reason, in other respects so far below it ! Martins love 
to frequent towns, especially if there are great lakes and rivers 
at hand ; nay, they even affect the close air of London. And I 
have not only seen them nesting in the Borough, but even in 
tlie Strand and Fleet Street ; but then it was obvious from the 
dinginess of their aspect that their feathers partook of the filth 
of that sooty atmoshhere. Martins are by far the least agile 
of the four species ; their wings and tails are short, and there- 
fore they are not capable of such surprising turns and quick 
and glancing evolutions as the swallow. Accordingly, they 
make use of a placid easy motion in a middle region of the 
air, seldom mounting to any great height, and never sweeping 
long together over the surface of the ground or water. They 
do not wander far for food, but affect sheltered districts, over 
some lake, or under some hanging wood, or in some hollow vale, 
especially in windy weather. They breed the latest of all the 
swallow kind : in 1772 they had nestlings on to October the 
21st, and are never without unfledged young as late as 
Michaelmas. 
As the summer declines the congregating flocks increase in 
numbers daily by the constant accession of the second broods, 
till at last they swarm in myriads upon myriads round the 
villages on the Thames, darkening the face of the sky as they 
frequent the aits of that river, where they roost. They retire, 
the bulk of them I mean, in vast flocks together about the 
beginning of October ; but have appeared of late years in a 
considerable flight in this neighbourhood, for one day or two, as 
late as November the 3rd and 6th, after they were supposed 
to have been gone for more than a fortnight. They therefore 
withdraw with us the latest of any species. Unless these birds 
are v«ry short-lived indeed, or unless they do not return to the 
4 
