9. Small populations of mallards in the Central Flyway that 
are near the southern margins of the breeding range are 
largely nonmigratory and contribute primarily to the kill 
in the State of banding. 
10. Mallards produced in all States in the Mississippi Flyway 
and regions to the north contribute primarily to the kill 
within the flyway. 
11. Relatively small populations of mallards within the Atlantic 
Flyway and in regions to the north contribute primarily to 
the kill in the Atlantic Flyway. In areas from Chesapeake 
Bay south, birds produced in the Mississippi and Central 
Flyway areas may be more important to the total kill. Birds 
from the Canadian Prairies that reach the Atlantic Flyway 
appear to move primarily through the Tennessee Valley and 
nearly all returns are from south of 37° latitude. Mallards 
from the Great Lakes region seem to move south from Lake 
Ontario into Chesapeake Bay. 
12. Annual variations in the distribution of direct recoveries 
of mallards banded as locals were not great, but some 
annual fluctuations may occur. It appears that the direction 
of migration from a given reference area is stable, but 
that the speed or timing, and ultimate destination, of 
migratory movements may vary as a result of weather con 
ditions during the period of migration. Where migration 
routes involve crossing extensive portions of one flyway 
to reach another, the speed and timing of migration may 
cause changes in the relative recovery rates between the 
two flyways. Such changes have occurred in the Central and 
Mississippi Flyways. 
13. Direct recoveries of both sexes of mallards banded as locals 
are nearly identical in distribution, the only difference 
being a slightly greater proportion of females that are 
recovered in the degree block of banding. The geographical 
area from which indirect recoveries have been reported was 
nearly identical to that for direct recoveries, but the pro¬ 
portion of the recoveries occurring in the various portions 
of this area differed. The primary difference between re¬ 
coveries of adults and immatures was the smallet proportion 
of adults that were recovered in the breeding areas. In 
general, the distribution of recoveries of immatures was 
more similar to recoveries of adult females than to those 
of adult males. 
24 
