48 PROCEEDIKGS OF THE 



I. Bristol, Burlington, Beverly, Rancocas. 

 II. Moorestown. 



III. Haddonfield. 



IV. Swarthmore, Lansdowne, Secane, Media. 



V. Overbrook, West Fairmount, Ardmore, Haverford, 

 Bryn Mawr. 

 VI. Germantown, Wissahickon, Olney, Oak Lane, Frank- 

 ford. 

 Comparing the earliest dates of arrival in each of these sec- 

 tions for twenty-one common species, we find that eight were 

 first seen in the Ardmore section, seven in the Germantown sec- 

 tion, six in the Swarthmore, and one each in the Moorestown 

 and Bristol sections. Tabulating this data as in former reports 

 we have the following result: 



Bristol. Moorestown. Swarthmore. 



First . . 1 1 6 



Second .7 3 5 



Third . . 6 5 4 



Fourth .4 8 6 



Fifth . . 3 4 1 



Or by giving a rank of 21 to the section which might have 

 been first to observe all the species, we get relative ranks as fol- 

 lows: Germantown 46, Swarthmore 53, Ardmore 5(i, Bristol 

 64, Moorestown, 74. The Haddonfield record being too defi- 

 cient to warrant consideration. 



This comparison would seem to indicate a tendency on the 

 part of arriving birds to spread over the entire Philadelphia dis- 

 trict simultaneouslj', the apparent earlier arrival in the imme- 

 diate vicinity of the river, indicated in previous years, being 

 less marked. The variability in the time of arrival as recorded 

 at nearby stations is, however, remarkable, indicating how 

 stragglers may arrive here and there days before the siJecies is 

 recorded at the bulk of the stations. Of course, this may be 

 due in some degree to careless observation, but not usually, as 

 can be seen by the general unanimity of arrival at a number of 

 stations on the same day, dates of arrival later than this "bulk 

 movement" are, of course, due to lack of opportunitj' for con- 

 tinuous observation. 



