Recovery Patterns 
The general characteristics of recovery patterns were described 
for reference areas in the preceding section. These general charac¬ 
teristics are summarized in table 25 which provides the percentage 
of direct recoveries in each flyway from each reference area, using 
as a base only those recoveries from the United States. Recoveries 
from Canada are given as a percentage of the total recoveries. Tables 
26 through 33 summarize the distribution of recoveries by State and 
Province from each reference area. Figures 5 through 16 illustrate 
the distribution of recoveries by degree block. These tables and 
figures are based on total direct recoveries of both sexes, and are 
assumed to represent the average distribution of the kill, although 
the number of recoveries may have varied considerably from one year 
to another as banding effort fluctuated. Differences in band 
reporting rates between regions, as discussed by Bellrose (1955) and 
Geis and Atwood (1961) have not been considered in this analysis, 
and may bias the proportions cited in the tables to an unknown degree. 
Annual variation 
The annual distribution of direct recoveries by flyway and 
latitude for the period of 1954 to 1961 is summarized for the most 
important production areas in tables 34 to 40. The proportion of 
young mallards recovered in Canada was fairly consistent for given 
conditions and areas, but was somewhat higher subsequent to the onset 
of drought in 1958 and 1959 and the decline in populations in the 
Prairie Provinces. It appeared that even with somewhat more restrictive 
regulations, hunters in Canada were able to take a relatively greater 
proportion of young mallards during the period of low populations 
than in years of normal or high waterfowl populations. The relatively 
low recovery rates in the United States during the period of low 
populations may have resulted from the more restrictive regulations 
there than in Canada. Overall recovery rates declined in both Canada 
and the United States. 
An important segment of the populations from Alberta and western 
Saskatchewan migrate to the Pacific Flyway (tables 34 to 36). The 
proportion of recoveries in this flyway did not appear to vary statis¬ 
tically from year to year. The relatively high variability of the 
samples probably resulted from the small sample size. 
The annual distributions of direct recoveries between the Central 
and Mississippi Flyways have a reciprocal relation to each other as 
noted by Low (1957). However, the only statistically significant shift 
in area of recovery that could be detected for any of the reference 
areas was from 1957 bandings in southeastern Saskatchewan. In this 
17 
