The Roadside Census 117 



Two aspects of the weather affected the iiiimhers of 

 birds to be seen : first, the noon heat, and second, rain. 

 All the censnses bnt two were taken on days whose maxi- 

 nninj tenii>eratnre ranjjed from 90° F to '.)r)° F. The ex- 

 ceiitions were one cooler day — May 3Uth, when the maxi- 

 mum was S>i° F. and one hotter day — July 3rd, when the 

 thermometer reached 99° P\ All bnt two of the censuses 

 were taken on snnny days. Thus there were remarkably 

 uniform weather conditions on thirteen of the sixteen 

 days on which censuses were taken. The thirteen counts 

 at or near noon occurred Avlien the weather was unde- 

 niably Ii(jt and the inactivity of the birds is nmrked, for 

 only an avera<>e of .').9 birds ]>er mile was seen in con- 

 trast to the 0.2 birds per mile noted in the cooler parts of 

 the day. The effect of an all day rain, even though the 

 temperature rose to 91° F., is even more striking in driving 

 the birds to cover than the noon heat on sunny days, for 

 only 1.4 birds were seen per mile instead of the 5.2 ob- 

 served in pleasant weather. 



V^e have very little data on the difference in the num- 

 ber of birds to be seen at different times of the year, as 

 there aie only 47 days between our first and last census. 

 Our results stand as follows : .")() miles of censuses on May 

 oOtli gave us an average of 10.2 birds a mile; 228 miles 

 from June 9th to 16th gave 3.5 birds a mile and 196 miles 

 from July 12t]i to 15th gave 6.7 birds a mile. The distance 

 travelled in May is so short that it is hardly comparable 

 to the other two sets of figures. As for the June and 

 July results, tliey represent the same numl)er of censuses — 

 eleven — , much the same number of miles, equally pleas- 

 ant weather with practically the same distance travelled 

 near noon ■ — 73 and 76 miles, respectively, — and both were 

 taken entirely in ]»iairie country. Therefore they would 

 seem as comparable as one could wish with one exception; 

 they were taken in different parts of the State, the June 

 censuses in the southwest, in Cleveland, Gradj^ and Com- 

 anche counties; while the July censuses taken in the east, 

 in Leiiore, Sequoyah, Muskogee, AVagoner, Creek, Okniul • 



