520 Wright, Morning Awakening and Even-song. [bet. 



when Mr. Richard M. Marble recorded the awakening with me, 

 shows that a Song Sparrow sang once at 2.45; that a Chipping 

 Sparrow trilled once at 2.46, followed directly by a second bird; 

 that Song Sparrow sane again at 2..">2, and a second bird twice at 

 2.f>4; that the first Robin gave a few notes at 2.54 and began to 

 sing freely at 3.04, a second Robin beginning to sing one minute 

 later; that Yeery first called at 3.05 and sang at 3.11; that King- 

 birds gave a few calls at 3.07; that Alder Flycatcher sang at 3.09; 

 that Tree Swallow was in song flight at 3.10; that Vesper Sparrow 

 sang at 3.11, and Bluebird at 3.12 o'clock. The record then con- 

 tinues in quite the usual way. On June 26, 29, July 7, 10, 13, 17, 

 and 28 similar records were taken. The last two have not been 

 incorporated in the averages, because they were too late in the 

 singing season for satisfactory results. 



And an occasional individual record in the illustrative records 

 which have just been presented has not been made use of, because 

 not in harmony with the series of records of the species. Some 

 censorship of this kind is quite necessary when a series of records 

 is examined, as it sometimes happens that a late record can be 

 accounted for by the fact that the bird was not near enough to be 

 heard in its first songs. Experience must be availed of in deciding 

 such questions. 



The period within which all the records were obtained, which 

 have been combined, extends from the time of earliest sunrise, 

 which is 4.02 o'clock at Jefferson, to the time of sunrise at 4.16 

 o'clock; the variation in sunrise, therefore, is fourteen minutes. 

 Had it been possible to procure the desired records between June 

 7 and 22, there would have been no variation in the time of sunrise 

 to reckon with. As it is, an exact basis has been established by 

 adopting the number of minutes before or after sunrise on each 

 day of record for the time-record of each first song and by averaging 

 the times on this basis. The clock-time when a species awakes to 

 sing can thus be determined for any date. I have also averaged 

 the clock-times, as taken by the watch, of first songs of each spe- 

 cies throughout the series of records of each to obtain the average 

 clock-time within the prescribed period. But this is not an un- 

 varying basis, since it varies according to the dates of records; 

 for it makes a difference whether the records of a species are ob- 



