MOVEMENTS OF BIRDS IN RELATION TO THE WEATHER. 385 



boro records show just twice as many instances of arrival during a 

 rising as during a falling temperature ; and the average temperature 

 of the two days before a bird arrives, when calculated for a series of 

 years, is always less than the average temperature of the day of 

 arrival. 



Every student of bird migration has noticed that in an unusually 

 early spring the first migrants arrive ahead of their average dates. 

 Thus in Washington during March and April, 1910, almost every 

 species anticipated its usual date of arrival. But it is seldom, if ever, 

 that such untimeliness continues throughout an entire season, nor is 

 any entire season likely to be later than usual. Indeed, seasons are 

 such combinations of warm and cold waves that the average date of 

 arrival for the whole migration period is remarkably uniform. At 

 Lanesboro the average date of arrival for all species is April 25; 

 the extremes vary less than four days, and the average variation from 

 year to year is only a single day. 



It has already been stated that each species prefers to arrive at its 

 breeding grounds when the average temperature is within certain 

 definite limits. Thus the Baltimore oriole arrives in southern Minne- 

 sota when the thermometer ranges around 55 F., but it does not 

 follow that the oriole will appear in spring as soon as the temperature 

 rises to that degree, nor that the bird never arrives before the temper- 

 ature reaches that point. One spring at Lanesboro the 10 days from 

 April 13 to 23 averaged 10 warmer than the oriole's preferred tem- 

 perature, but no orioles appeared until May. Another year the Bal- 

 timore oriole appeared at Lanesboro when for two weeks the ther- 

 mometer had not risen above 48 F. The point to be emphasized is 

 that a knowledge of weather conditions in any given season is not a 

 basis for deducing the time of arrival that season of any particular 

 species. 



BIRDS CAN NOT FORETELL WEATHER CHANGES. 



One morning in October the base of the Statue of Liberty in New 

 York Harbor was covered with the dead bodies of birds that had 

 struck against the light during the previous night. More than 175 

 were picked up, and a larger number had fallen into the sea, all 

 victims of this one light during a single night. Similar destruction 

 occurs during each storm of the migration season. Whirled by the 

 tempest until they lose all sense of direction, with the landmarks 

 hidden by enveloping clouds, the birds are lured to death by a beacon 

 light penetrating the mist. A continuous red light or a flashing 

 intermittent light of any color does not attract them, but a steady 

 white light is irresistible. 



Nor are man's beacons the only agents destructive to migrating 

 hosts. The night of October 10, 1906, flocks of migrants over Lake 

 Huron were caught by a snowstorm and forced into the waves, and 



