1908.] NATURAL SCIENCES OF PHILADELPHIA. 135 



Germantowi) . Washington. Difference. 



Wood Thrush May 1 April 26 5clays. 



Black-throated Blue Warbler May 6 May 2 4 " 



Ovenbird May V April 23 8 " 



Maryland Yellow Throat April 29 April 21 8 '' 



These dates being the averages of a number of years, would seem to be 

 sufficiently accurate for the purpose of estimating the time of flight of 

 the species mentioned between Washington and Philadelphia, and by 

 comparing them we find that it is respectively five days, four days, 

 eight days and eight days. The Germantown records quoted from 

 Prof. Cooke's papers are based upon schedules which I filled out for 

 the Department of Agriculture from 1883 to 1890. I now find that 

 my dates vary from those obtained by other observers in the neighbor- 

 hood of Philadelphia from 1901 to 1907, just as the latter have been 

 shown to vary from each other. 



Had any of the other records from the vicinity of Philadelphia been 

 used in place of the Germantown series, as would have been perfectly 

 justifiable, a very different result would have been obtained; and 

 there is no doubt but that the dates of several individual observers in 

 the vicinity of Washington would show just as much diversity as is 

 shown in our Philadelphia series, which would still further vary the 

 results. 



In a number of instances moreover the difference between the average 

 date of arrival at Washington and Philadelphia, as given in Prof. 

 Cooke's papers, is no greater than that between two stations well within 

 the Philadelphia ten-mile circle. 



In comparing the dates of arrival of species for several consecutive 

 years we also find a considerable variation in the records of nearby 

 stations which we should expect to show uniformit5^ 



For instance, taking the eleven species given in the table on page 134, 

 and computing the average dates of arrival for the six years 1901 to 

 1906 at each of the three stations, and then comparing these with the 

 dates of arrival at each cf the stations in 1907, we find that at station 

 No. I the 1907 dates averaged three days late, while at station No. II 

 they averaged one day late and at station No. Ill they averaged 

 exactly normal, and yet each one of these stations was represented by 

 several accurate observers, and there is nothing in their relative geo- 

 graphic position to warrant any difference. 



Combination of Individual Records. 

 After discrediting the value of individual records, one must natur- 

 2 Omitted in Prof. Cooke's paper, and supplied from my own memoranda. 



