Direct recovery rates of bandings in each reference area are 
presented in table 46. Recovery rates appeared to vary as a result 
of both the location of banding and route or terminus of migration. 
Thus, bandings in Alaska have produced recovery rates that are much 
lower than those from British Columbia, although the location of re¬ 
coveries was quite similar for both (fig. 5). A similar example is 
provided by the low recovery rates resulting from bandings in Mackenzie 
Territory as compared with those from Alberta. The differences in both 
of these examples may be attributed to the location of the banding 
station. The low recovery rates in the northern areas probably resulted 
from the low recovery rate in the vicinity of the banding station. 
Recovery rates from bandings in the prairie regions of Alberta 
and Saskatchewan were relatively uniform but much lower than the 
recovery rates resulting from bandings in the adjoining Dakotas and 
Minnesota. Again, the geographical distribution of recoveries can 
be seen to be quite similar (figs. 6 to 13), and the major part of 
the difference must result from the location of banding. 
If we consider recovery rates by flyway, it is evident that 
reference areas contributing mainly to the Pacific and Mississippi 
Flyways tended to have higher recovery rates than those contributing 
to the Central and Atlantic Flyways. The differences may be caused 
by one or several factors, such as number of hunters, band reporting 
rates, or differences in hunting regulations. 
Recovery rates are summarized by year of banding for the important 
mallard production areas of the Canadian Prairie Provinces and the 
North Central States in table 47. Recovery rates were relatively 
consistent for each area during the period 1954 to 1958, but dropped 
sharply in most areas as a result of the restrictive regulations 
initiated in 1959. Recovery rates declined significantly for band¬ 
ings in southwestern Saskatchewan and in South Dakota in 1957 and 
slightly in southeastern Alberta, southeastern Saskatchewan, and 
southwestern Manitoba. Because regulations were as liberal in 1957 
as in 1956, the cause of the decreased recovery rates must lie 
elsewhere. The most evident factor occurring in 1957 was the failure 
of many birds to migrate to normal southern wintering areas. The 
changed proportions of recoveries in the Central and Mississippi 
Flyways as a result of this altered distribution were considered in 
a previous section. The present data indicate that not only the 
relative proportions of the kill in the two flyways were altered, 
but the decreased kill in the Mississippi Flyway was not offset by 
an increase in the Central Flyway sufficient to prevent a general 
decrease in recovery rates. 
22 
