ALTITUDE OF THE MIGRATION FLIGHT 57 



heard faintly, but still distinctly, from heights to which, despite 

 every effort, it is impossible for the human eye to penetrate. 



At night, also, immense hosts of these birds, as well as of the 

 different species of Sandpipers, in scattered flocks as well as in end- 

 less streams, travel across Heligoland. In this case, however, the 

 altitude of flight of the birds docs not often exceed two hundred feet 

 above the cliff, as may be partially ascertained by observation within 

 the luminous circle of the lighthouse, but on a still greater scale 

 from the sound of their voices. It must not, however, be assumed 

 that birds migrate at lower elevations at night than during the day- 

 light. Cases in which this occurs must be regarded as distiu-bances 

 of the normal height of flight j^roduced by meteorological influ- 

 ences. This will be discussed in further detail in the chapter on 

 the meteoroloCTical conditions which influence misfration. 



Finally, the number of species whose migration jjroceeds usually 

 only a few hundred feet above the surface of the sea, or, in many 

 cases, in its immediate proximity, is extremely small. According 

 to my own experience of many years, it does not go beyond the fol- 

 lowing three, viz. the Hooded Crow, the Starling, and the Lark. The 

 last of these birds will, on clear days, rise to a height of from 600 

 to 1000 feet. Hooded Crows travel at such a height in exceptional 

 cases only, and Starlings extremely rarely. All three species migrate 

 at a greater elevation in spring than in autumn ; but during both 

 migration periods, especially in dull windy weather. Hooded Crows, 

 and still more markedly Larks, frequently fly immediately above the 

 surface of the water. I have never noticed this to happen in the 

 case of Starlings ; these, unless they intend to rest on the island, 

 hasten across it at a height of from two to three hundred feet, in 

 numerous densely-crowded swarms, and with a kind of impetus as 

 though each bird was endeavouring to outstrip the other. In 

 exceptional instances. Larks will on bright spring days travel at 

 such heights that, even under the most favourable conditions of 

 the atmosphere, one can only recognise them by their call-notes, 

 the birds themselves being completely out of the range of vision. 

 I have noticed the same phenomenon in the case of Jackdaws 

 and Rooks. 



A striking proof of the direct and important influence which 

 meteorological conditions exercise on the height of the migration 

 flight is furnished by the large numbers of migrants which are 

 captured at Heligoland during dark nights, being in part attracted 

 thither by the lantern of the lighthouse. 



A necessary condition for this capture is a dark, uniformly 

 overcast sky, especially if there be at the same time a very fine 

 precipitation of moisture. Under such conditions, Larks and 



