206 Cooke, The Relation of Bint Migration to the Weather. [^JJjj 



constantly in the field so as to note the birds immediately on their 

 arrival; it is also necessary that these observations be continued 

 long enough to make possible the computation of reliable averages. 

 A great advantage would be to have these records taken in a dis- 

 triet free from mountains, valleys, or any other physical features 

 that would tend to interfere with the free and uninterrupted north- 

 ward movement of migration. The fulfilment of all these condi- 

 tions was found in the work of Dr. J. C. Hvoslef at Lanesboro, 

 Minnesota. An ardent student of bird life, a close observer with 

 a good knowledge of birds, his profession as a physician with a 

 large country practice, kept him daily in the field and made it 

 probable, that few birds would escape his acute observation. Dr. 

 Hvoslef contributed migration records for ten consecutive years, 

 1884 1893. At the same time notes were received from several 

 towns in Iowa — notably Grinnell, Iowa City and Coralville — 

 whose records are especially valuable as supplementary and coirobo- 

 rative evidence. 



As is well known the weather comes usually in alternate cold and 

 warm waves. If therefore the weather is the controlling factor 

 in bird migration, then the progress of migration should be in waves 

 corresponding to those of the weather, birds arriving freely when 

 the temperature rises above normal and checking their advance 

 when it falls below. While a sort of general correlation can be 

 noted between the waves of weather and migration, the exceptions 

 are many and striking. The accompanying chart gives the course 

 of the weather and migration for three years at Lanesboro, Minne- 

 sota. The first year, L885, shows two pronounced waxes of bird 

 arrival coinciding with two waves of decided warm weather; it 

 also shows the biggest migration wave of the whole season coming 

 at the coldest part of a sharp cold snap that stmt the temperature 

 far below normal. The second year, L889, shows a close agreement 

 between the larger waves of migration and the wanner waxes of 

 temperature. The third year, L892, shows all the large bird move- 

 ments as occurring not on account of the weather but in spite of it. 



The bird wave of May 7. 1885, is particularly to be noted. On 

 this day a storm of snow with a north wind forced the temperature 

 below the freezing point, yet on the morning of May 7 "the woods 

 and rixer bottoms seemed to be almost alive with small bird-." 

 Among these were the following seen for the first time: 



