4 BULLETIN 1165, U. S. DEPARTMENT OF AGRICULTURE. 



The Biological Survey lias advocated counting tlie singing males 

 as the most convenient way of taking a bird census, and this method 

 has given excellent results. The observer starts at daylight some 

 morning at the height of the breeding season and zigzags back and 

 forth across the selected tract, counting the singing birds. At that 

 season, when the migration is over and the birds are settled on their 

 breeding grounds, each male bird may be safely considered to rep- 

 resent a nesting pair, and early in the morning, before the insects 

 are flying and the birds begin feeding, the male is usually to be found 

 in the vicinity of the nest. The count should be repeated once or 

 twice at intervals of a few days to be sure that no birds have been 

 missed and that all the birds counted are actually nesting on the area. 



If the enumerator lives close by, one day's count may be checked 

 by subsequent observations throughout the breeding season. When- 

 ever possible, it is well to go over the land again late in the season to 

 catch any late nesting species, such as goldfinches and waxwings. 

 To locate every nest is not necessary; unless the enumerator lives on 

 or very near the tract it is practically impossible to do so, and the 

 time required in any case is enormous. Experience has shown that 

 a count of 50 acres can be made in three hours by the method outlined 

 and that subsequent observations throughout the summer make 

 almost no change. However, in the case of unusual species and 

 especially those outside their known breeding range, it is insisted 

 upon that the nest be located or other satisfactory evidence found as 

 proof that the birds were actually breeding in the locality and were 

 not merely wanderers or delayed migrants. 



ESSENTIALS OF A SATISFACTORY BIRD CENSUS. 



The observer must be thoroughly familiar with the birds breeding 

 in his locality, both by sight and song, in order that all the birds 

 found nesting on the selected area may be positively identified, or a 

 recognizable description given of such as can not be readily named. 

 Otherwise some species may be omitted and the report be not usable 

 since It does not tell the whole truth. 



For a number of years the Biological Survey has been receiving 

 each year from several hundred volunteer observers throughout the 

 country reports on the arrival and departure of birds during the 

 spring and fall migrations. Some of these observers are well ac- 

 quainted with the birds, while others know only a few; in the case of 

 the latter, however, if their knowledge is accurate regarding the few 

 they do know, and if they are well situated to watch the birds and 

 note their first arrival, the fact that they do not know all the species 

 in no way detracts from the value of their records for the arrivals 

 and departures of the species noted. In making a bird count, how- 

 ever, such partial knowledge is worse than useless, and some reports, 

 doubtless entirely correct for the species listed, have had to be rejected 

 because the absence of the smaller and less conspicuous species 

 showed the reports to be incomplete. A census of breeding hirds to le 

 of any value must tell the ivhole trutli; it is not sufficient that it shall tell 

 the truth as far as it goes. 



Care must be exercised that the same individual be not counted 

 more than once, as there is danger of doing in the case of species in 

 which the two sexes are nearly or quite alike. This is one reason 

 in favor of counting only singing birds, in so far as is compatible 



