BIRD MIGRATION. 7 



distance, then nearer, and finally joining the throng within the circle 

 of light. 



Sept. 4. A. clear cool day ; the evening perfectly clear up to ten 

 o'clock, when a heavy curtain of clouds rolled overhead from the north- 

 west, and it became very dark. An hour later dense fog set in, and at mid- 

 night it began to rain, heavy showers succeeding one another at frequent 

 intervals. Wind south ; puffy, at times strong. 



As soon as the sky became overcast small birds began to come about 

 the light. Their numbers increased steadily from ten to eleven o'clock, but 

 during this time the majority kept at a safe distance, and only two or three 

 struck. With the advent of the fog they multiplied tenfold in the course of 

 a few minutes. For the next hour from fifty to a hundred were constantly in 

 sight, and from one to eight or ten dashing at the lantern. About seventy-five 

 per cent struck the glass with slight force, fluttered down the pane, and 

 dropped to the platform beneath, exhausted, but uninjured. After a 

 moment's rest these would make off with uncertain flight, usually dis- 

 appearing in the darkness, sometimes returning and striking again, 

 always harder the second time. About twenty per cent struck so 

 forcibly as to injure themselves beyond hope of recovery, often, however, 

 fluttering off the platform and down to the ground beneath. Not over five 

 per cent were killed outright. None of the killed or wounded were defaced 

 externally, but all had one eye slightly bulged out and more or less blood 

 settled about it under the skin. With the beginning of the rain their num- 

 bers diminished rapidly,* but a larger percentage struck, and they also 

 struck harder. 



I remained on the lighthouse from ten o'clock until two the next 

 morning. During this time fully two hundred birds came against the lan- 

 tern. Of these at least fifty were killed or disabled, and I caught and 

 examined probably fifty more which were too wet or exhausted to fly after 

 dropping on the platform. About forty per cent of the specimens identified 

 were Geothlyfis trichas, forty per cent more Vireo olivaceus, with the re- 

 maining twenty per cent made up of Turdus alicicB, T. swainso^ii, Den- 

 droica fensylvanica, D. maculosa, D. castanea, D. blackbtirnia, Seto- 

 ■phaga ruticiUa, and Sylvania canadensis. 



At the height of the 7nelee the scene was interesting and impressive be- 

 yond almost anything that I ever witnessed. Above, the inky black sky; 

 on all sides, dense wreaths of fog scudding swiftly past and completely en- 

 veloping the sea which moaned dismally at the base of the cliffs below ; about 

 the top of the tower, a belt of light projected some thirty yards into the 

 mist by the powerful reflectors ; and in this belt swarms of birds, circling, 



♦They were probably obliged to seek the nearest shelter, for many that came to the lantern 

 had their plumage so water-soaked that they could hardly fly. 



