The breeding season over, the birds in each locality unite in a small flock composed of twenty or thirty to fifty or more pairs and start their wandering life. Those in the north migrate or drift south, and vast numbers, as we see, spend the winter in the southern counties. And here they have their favourite roosting-places and are accustomed to assemble in tens and hundreds of thousands. But the original small flock composed of a few pairs, is never broken up--never absorbed by the multitude. Each morning when it is light enough, the birds quit the roosting-wood, but not all together; they quit it in flocks, flock following flock so closely as to appear like a continuous stream of birds, and the streams flow out in different directions over the surrounding country. Each stream of birds is composed of scores and hundreds of units, and each unit drops out of the stream and slopes away to this or that side, to drop down on its own chosen feeding-ground, to which it returns morning after morning through the winter. When all the units have dropped out and settled on their feeding areas for the day, it may be seen that the whole country within a circuit of ten or twelve or more miles from the roosting-
BIRDS IN CORNISH VILLAGE 285
place has been occupied, that each flock has its own territory, where it splits up into some groups and spends its short hours flying about and exploring every green field, and one might almost say "every grass." One can only explain this perfect distribution by assuming that each unit instinctively looks for unoccupied ground in its winter habitat, and that consequently there is very little overlapping. It must also be assumed that at the place of assembly in the evening each flock has its own roosting-place--its own trees and bushes where the members of the flock can still keep together and to which after each aerial performance they can return. The flock comes back to sleep on its own tree, and no doubt every couple roosts side by side on its own twig.
On the return of Spring the birds do not migrate in a body, but slip away, flock by flock, to reappear about the end of April in their old breeding-place in the North Country, with, perhaps, the loss of a few members--the one that was old and died in the season of scarcity; and one that was taken at the roost by a brown owl, and one that had its feet frozen to the perch; and was killed by a jackdaw when struggling to free
286 BIRDS IN TOWN AND VILLAGE
itself; and one that was struck down by a sparrow-hawk on his homeward journey.
What I have so far been unable to trace is the career of the young after August. We see that once they are able to fend for themselves they club together in small flocks and continue together during their "brown thrush" stage, but by and by they get the adult plumage and language and are no longer distinguishable as young. Do they, then, join the old birds before the wandering and migrating south begins? And do they pair or not before the winter?
