114 THE WILSON BULLETIN— September, 1921 



altliougli under favorable conditions one eonld be praeti- 

 callv sure in the case of some birds. 



Onr method of taking- a Eoadside Censns is as follows : 

 Ave have a small notebook with a pencil attached by a 

 string, in this at the top of a i)age we write the date, hour, 

 number of miles, county, jdace of starting and destination, 

 and also a note as to weather. The temperature we have 

 obtained later by writing to the local stations of the 

 Weather liureau. We always began a new census with a 

 new county although sometimes it was difficult to knovr 

 the county boun<laries; often it was necessary to calculate 

 from the nmp the number of miles from a certain town to 

 the border of the county and turn over a page when the 

 speedometer showed that we had reached that spot. As 

 to the birds, we jot them down as they appear, using ap- 

 proximately the order of the A. O. U. Checklist, so as to 

 know where to expect the name of e^ch bird, an important 

 matter when one sees several ditferent species at a time. 

 The numbers seen are put down after each bird's name and 

 of course changed when occasion arises. There are usually 

 a number of birds that have to go down under the caption 

 ^' unknow^n " ; with us the difficulty generally lay in being 

 sure whether a few individuals were English Sparrows, 

 Dickcissels or (Grasshopper Sparrows. (In making up the 

 totals for this paper we divided these '' unknowns *' be- 

 tween the three species in the same proportion as the 

 ''known" birds; other ''unknowns'' were lelt as such.) 

 Naturally the person who <lrives the car cannot take a 

 Koadside Census although he can often see birds on his 

 side that the census taker misses. AVe find it important 

 to coi)y our censuses in a permanent notebook each evenijig 

 when the exi)erience of taking them is fresh in our minds, 

 for they are naturally difficult to read. That is one reason 

 why it is better not to take too long a census at one time, 

 but to start again after twenty or thirty miles because it 

 is often hard to be sure of figures when so many liave been 

 written one after another when jolting along. 



On our censuses we probably seldom saw more than a 

 third of the breeding birds of a region ; occasionally the 



