128 PROCEEDINGS OF THE ACADEMY OF [April, 



METHODS OF RECORDING AND UTILIZING BIRD-MIGRATION DATA. 

 BY WITMER STONE. 



The custom of recording the dates of arrival of migrant birds has 

 been practised for a great many years in various countries, and more 

 recently attempts have been made to encourage the keeping of such 

 records on a uniform plan and to gather them together for the purpose 

 of study and comparison. 



In America this work was begun in 1884 under the direction of the 

 American Ornithologists' Union, and since 1885 has been conducted 

 by the Division of Biological Survey (formerly Ornithology and Mam- 

 malogy) of the U. S. Department of Agriculture. 



All the published records with which I am familiar represent the 

 work of one individual at each station, and until very recently there 

 has been no attempt made to compare the records of several observers 

 at practically the same locality. 



The meagerness of the data that it is possible for one individual to 

 gather on bird migration, compared with the magnitude of the phenom- 

 enon, must be apparent to all, and yet we are constantly attempting 

 all sorts of estimates — as to the rapidity of flight, the relation of fluctu- 

 ation of migration to temperature variation, etc. — based for the most 

 part upon the records of individual observers. 



In 1901 the Delaware Valley Ornithological Club of Philadelphia 

 organized a corps of observers for the study of bird migration in this 

 vicinity. This corps now numbers sixty-three, of which thirty-five are 

 located within ten miles of the center of Philadelphia. 



The study and comparison of the yearly records of these observers 

 throws some interesting light upon the accuracy of individual records 

 and suggests some methods by which a more correct index of the pro- 

 gress of migration may be obtained. 



Many of the records are presented in detail each year in Cassinia 

 the annual publication of the DelaAvare Valley Ornithological Club, 

 and to these, as well as to the original schedules returned by the ob- 

 servers, I am indebted for the data discussed in the present paper. 



In a paper read before the American Ornithologists' Union in New 

 York City in November, 1905, and later published in The Condor, I 



