MIGRATION OF BIRDS. 283 
ingeniously used to account for the fairly constant appearance 
of odd specimens of North American species in Great Britain. 
It is pointed out that a certain number of these species breed 
in Greenland, not far away from species which migrate 
south-east to Europe, instead of south-west to America. An 
American bird straying slightly might fall in with voyagers 
taking the European route and follow in their wake to Great 
Britain instead of taking its proper course. 
There is no doubt that the perils of the journey are many, 
and take a heavy toll of the migrating hosts. A bird may 
start from Ushant with a fair wind and encounter a gale before 
Scilly or the Eddystone is reached. Fog is another source 
of danger, especially when there are lighthouses near. When 
the nights are clear, the light does not exercise so great an 
attraction, but in haze, drizzle, or fog it seems to act in a 
mesmeric manner and lures thousands of birds to death. They 
dash at the lantern, strike the glass, fall stunned or with broken 
wings, and so perish. 
Weather, in fact, has a most important bearing on migration. 
Migrants have in the main a fairly regular time table. Most 
of the English summer visitors, for instance, start south before 
the necessities of inclement weather and a failing food supply 
drive them away, but an unusual warm spell in autumn may 
make birds put off their departure or linger en route, while a 
sudden cold snap will hasten their going. The great thing to 
remember is that it is the weather at the point of departure 
which influences them, especially at the start of a long overseas 
journey. The direction of the wind does not seem to have a 
great influence, provided it is not too strong, but boisterous 
weather holds migrants up. Most of the wholesale destruction 
which at times overtakes migrating flocks is due to their 
suddenly meeting with unfavourable weather after they have 
once committed themselves to an extended flight. 
In the light of these main facts concerning migration in 
general, we may now consider the special conditions applying 
to migration in Ceylon. I am afraid that in this part of my 
paper many of my remarks have to rest on inference and 
analogy. Little detailed observation on the subject has been 
done in the Island, nor can I find much on record regarding the 
