shifting about of the populations. In 1950, nesting studies in south- 
central and northwestern Wisconsin indicated a phenological difference 
of about three weeks. In view of this south-north phenological gradient, 
the state was arbitrarily cut into four east-west bands, and the mail 
carrier counts were run on four consecutive weeks in the four bands. 
Thus, the southernmost band was run in the first week, and each band 
northward was run progressively a week later. In this manner it was 
hoped that the counts were being made during the same period in the 
nesting season throughout the state. 
Carriers were asked to make counts on three days during the week, 
preferably during good weather. They were asked to record weather con- 
ditions, time of day route was run, length of route, exact location of 
route (township(s) and distances and directions from nearest towns), 
and willingness to make more counts in the future. Advantages of this 
method are the slight expense and small time involved in getting a large 
volume of data. 
The results of this survey were extremely gratifying. Out of an 
estimated 1,000 carriers in the state, 871 returned forms, 767 of which 
were usable. The great majority were very cooperative and indicated 
willingness to make more counts in the future. 
It was not possible to plot all 767 counts on a map. Therefore 
only the 119 high counts (arbitrarily, those of 10 or more doves per 
hundred miles) were plotted. Figure 2 is the resultant map. 
From this map the tendency for high populations to occur along 
rivers is quite striking. Since there is surely no shortage of water 
in Wisconsin for doves, I suspect that the birds are drawn to streams, 
not by the water itself, but by the sandy outwash plains of the rivers. 
In Wisconsin doves appear to prefer sandy regions, and several of the 
larger rivers in the state such as the Mississippi, Wisconsin, Chippewa, 
and others have sandy outwash plains, in places several miles wide. 
f second generalization to be drawn from the map is the tendency 
for high populations to occur south and west of a line drawn between 
the northwest and southeast corners of the state. This line roughly 
coincides with the tension line between the original prairies and 
deeiduous forest (Curtis, 1951). Doves in Wisconsin, then, tend to 
occur in the prairie half of the state. This coincides with other 
reports which suggest that high dove populations occur in the prairie 
states (McClure, 193; recent call count findings; others). 
49 
