Summary 
For 19,365 male mallards and 15,73 female mallards, banded 
from 1927 to 191445 in British Columbia, Montana, and South Dakota, 
recovery rates were 23.5 and 19.2 per cent respectively. The differ- 
ence is believed to reflect the tendency of hunters to shoot males 
in preference to females, and closely follows that reported for 
Illinois-banded birds in the 19h0's. A slightly higher man mortality 
rate for males was found in two samples of British Columbia and Mon- 
tana-banded birds. Mortality rates for Missouri- ad Oklahoma-banded 
mallards tended to show females as having the higher rates, but the 
difference was not conclusive, 
Analyses of banded samples of mallards disclosed no striking 
differences in mortality rates between adults banded at different 
localities in a given geographic region; in a number of cases the 
similarity was striking. An equally striking similarity was evident 
in five years for birds banded at widely separated places. 
Mortality ratas derived for specific years from life tables 
based upon a single source of data (like hunting) are quite unreali- 
able if this mortality factor is violently fluctuating from one year 
to another. 
During the latter part of the 1920's, male mallards banded 
on the Pacific seaboard suffered annual mortality rates around 65 per 
cent. Males banded in western Montana had rates of the same magnitude. 
Samples of both sexes banded in the Mississippi Valley sustained mach 
lower rates, their mean being 46.7: per cent. The evidence thus far 
assembled points to different mortality rates for different flyways 
during this period. 
After 1930, adult male mallard mortality rates dropped 
about one-third on the Pacific seaboard, particularly in response to 
limitations on the length of hunting season in the United States. 
A split season of 30 days in 193k seems to have been fully as 
lethal to these birds as the 60-day season that preceded it and con- 
tributed to an overall annual mortality rate that was about one- 
fourth or one-fifth higher than in the year that followed. The 
same phenomenon was evident for birds banded in the Mississippi 
Valley. From 1935 to 190, mortality rates for Illinois—banded 
birds ran close to ll; per cent. Larger samples from numerous 
bandings in the Mississippi Valley yielded crude estimates of 
overall rates of 7 to 53 per cent during the 1940s when hunting 
regulations were being relaxed. Recent fluctuations in adult 
survival rates were noticed to be associated in two or three years 
with hunting regulations and in one year with a marked increase in 
the production of young. 
153 
