the banding stations were in the United States. Only 11.3 percent 
of the recoveries (each station weighted equally) were in Canada. 
Unpublished information on the distribution of the mallard kill 
between Canada and the United States indicates that approximately 
35 percent of the kill is in Canada. ping this estimate, it appears 
that 32 percent of the Canadian kill ( x 100) was represented by 
pre-season bandings in 1959 and 1960. Hunting kill rates shown in 
table 12 reflect a 31 percent hunting kill (see next section), of 
which 27.5 percent was in the United States, since 11.3 percent of 
the total recoveries were in Canada. Since the kill in the United 
States should reflect 65 percent of the total kill rather than 
89 percent as the pre-season banding indicated, it was necessary to 
add the additional Canadian kill indicated in table 12. It is not 
known whether or not it is appropriate to add these birds to the 
pre-season total. The key to the problem is whether or not pre-season 
bandings largely in the United States truly reflect the total harvest | 
rate or whether the harvest rate is slightly low because of an un- 
represented Canadian kill. The solution to this problem is more 
adequate banding in Canada. 
In spite of the shortcomings of the banding data, which must be 
regarded as the weakest link in the chain of calculations outlined in 
table 12, the estimated pre-season population levels are of a reason-~ 
able magnitude. This estimate relates to the pre-season population 
which is represented by the banded samples listed in table ll, and 
these banding stations are associated with Central, Mississippi, and 
Atlantic Flyway harvest areas. | 
Importance of Hunting as a Mortality Factor 
If kill rates presented in table 12 truly reflect the kill in 
the Mississippi Flyway, then a substantial hunting mortality occurred 
in 1959 and 1960 despite comparatively restrictive regulations. 
About 64 percent of the band recoveries from banding stations listed 
in table 11 occurred in the Mississippi Flyway. The average harvest 
rate in the Mississippi Flyway was 16 percent (average value from 
table 12)3 therefore, the total kill rate for the population repre- 
sented by these stations is 16 percent divided by 64 percent or 25 
percent. If a crippling loss of 25 percent of the retrieved kill is 
added, this indicates that 31 percent of the pre-season population 
was harvested, Since all but one banding station was in the United 
States, the Canadian kill was not properly reflected in the recoveries, 
Therefore, the total harvest rate including the Canadian kill would 
be about 40 percent. 
