24 STUDIES IN BIRD-MIGRATION 



to identification was denied me. Nor could I walk 

 around the lantern and inspect them, as they fluttered 

 against the glass. A novel method for the capture of 

 specimens for determination was, however, resorted to 

 with considerable success : a sailor was stationed on 

 the sloping roof of the lantern, where, armed with an 

 angler's landing-net, he captured the birds, like so many 

 moths, as they streamed up the beams of light towards 

 him. In this way many birds, ranging from a Goldcrest 

 to a Rook, were secured for the purpose of identification. 



The weather conditions, under which the rays from 

 the lantern became conspicuous and attractive, were 

 identical with those I had noted at the Eddystone (see 

 Vol. I., p. 284) — namely, the existence in the atmosphere 

 of moisture not necessarily in the form of rain or haze, 

 but actually present, thoughnot visible, on dark starless 

 nights. In order to put my views on this subject to a 

 scientific test, I took with me to the lightship a hygro- 

 meter, with the object of ascertaining the actual 

 percentage of humidity In the air on such occasions. 

 I had not many chances of using the instrument, as 

 either rain or haze was, in most instances, present, but 

 on several occasions when it was not so manifested, I 

 found the hygrometer indicated a very large percentage 

 of moisture, in two instances reaching as high a figure 

 as 86 — a more pronounced result than I had anticipated. 



There was hardly a single occasion during my visit 

 on which the rays were conspicuous and the birds 

 absent ; on the other hand, there was not a single 

 instance of migrants visiting the light when the night 

 was bright and starlit or the moon was visible. 



The birds which appear at the lantern are, by some 

 authorities, considered to be birds that have lost their 



