Recovery Rates and Relative Recovery Rates 
Table 3 presents direct recovery rates and relative recovery 
rates for mallards banded at the 10 stations having the largest 
banded samples in 1959 and 1960. Comparable information for black 
duck bandings at Perch Lake in New York also is included in this 
table. 
Major hunting regulation changes between 1959 and 1960 consisted 
of a shift in the legal shooting hours from sunrise in 1959 to one- 
half hour before sunrise in 1960. This change affected all Flyways. 
In the Pacific Flyway, daily bag limits were reduced from five ducks 
to four in four States. Possession limits were reduced by two or 
more ducks in each of the seven States in the Flyway. In California, 
the daily bag limit was increased from five ducks in 1959 to six in 
1960. 
There were no pronounced changes in rate of kill that could be 
associated with changes in regulations (table 3). Including black 
duck recovery rates from Perch Lake bandings, the adult recovery rates 
increased in seven cases and decreased in four. Immature recovery rates 
also increased in seven cases and decreased in four. These changes 
suggest an increased rate of kill in 1960; they cannot be considered 
Significant, however. | 
Recovery rates and relative recovery rates in table 3 indicate 
considerable variability among locations. If it could be assumed 
that. the rates truly reflected the kill characteristics of pre-season 
populations at each banding station, and that these banding stations 
representatively sample the pre-season mallard population, then it 
would be possible to determine the number of stations required to 
reliably estimate recovery rates relating to the continental popula- 
tion. Some of the variability in table 3 can be attributed to sample 
size; some to band collecting activity. Lack of adequate pre-season 
banding in Canada leaves the representativeness of the locations open 
to question. It is hoped that in future years samples will be larger 
and their geographic distribution better so that 3 realistic appraisal 
of the required number of pre-season banding stations can be made. 
Tables 4 through 10 present information on regional differences 
in relative recovery rates for those stations having sufficient 
bandings to examine in this way. Only Agassiz and Rice Lake National 
Wildlife Refuges had sufficient bandings for comparison of annual 
differences. Other rates in these tables were obtained by combining 
recovery information from 1959 and 1960. When recoveries from a 
banding station are divided into two or more harvest areas, the 
problem of sampling error becomes acute; therefore, estimates of 
telative recovery rates for a specific harvest area must be regarded 
as very imprecise, 
