142 BRITISH BIRDS. 



denoting heavy showers all around — this is the sort of 

 weather that takes a heavy toll of the feathered passengers 

 across the waterway. Lured from the land by the 

 deceitful appearance of the sun, by the light and pleasant 

 breeze, they start their journey. Ere long, with the 

 suddenness so characteristic of this class of weather, 

 they find themselves in a deluge of rain, great heavy 

 drops, one after another beating down their wing strokes 

 and soaking their plumage. Their view obliterated — 

 perhaps the greater consideration with species whose 

 plumage by reason of its gloss or oily nature is less 

 susceptible to wet than that of the majority of Passer es — 

 little wonder the birds are forced to descend to lower 

 planes. Then, what a piece of luck to find a ship to 

 rest on to ehake out the plumage and greet the re-appear- 

 ance of the sun, for it has only been a ten-minutes' 

 shower, but woebefcide their less fortunate companions. 



Hardly a single instance of a really abundant arrival 

 of birds on board the " Venus " at sea in the Mediter- 

 ranean (and I might also include the Red Sea and off 

 the coast of Spain between Gibraltar and. St. Vincent), 

 but we were able to attribute it definitely to weather 

 such as I have described. Indeed, so strongly do I feel 

 impressed with a sense of certainty of the result effected 

 by this sort of weather upon migrating birds that, 

 accepting it as an axiom, I would infer from it that the 

 fact of birds never appearing on board ship during 

 thunder-shower weather in these two large sea-areas 

 (II. and IV., Figure VI.) is satisfactory evidence that birds 

 do not normally travel over those areas, and with regard 

 to the observed fact that it is only when fairly close to 

 land that birds come on board in large numbers, I would 

 suggest that it is because they are normally over the 

 central part of the sea during the night time, when a ship 

 is invisible to them. 



The following account, the facts of which were recorded 

 directly after the events had taken place, will serve to 

 illustrate my point. 



