are males. While the hens, incubating their eggs and caring for their young, are subject to high © 
mortality, the drakes gather in bands in large marshes or on extensive lakes where such deci- 
mating factors as predation and drought apparently cause fewer losses. Differences in the sur- 
vival rates of mallard drakes and hens are presented graphically in fig. 3. 
100 
90 
80 
70 
60 
—— DRAKES 
50 === HENS 
40 
30 
PER CENT ALIVE 
0 | 2 3 4 5 6 7 8 
YEARS FOLLOWING BANDING 
Fig. 3.--Survival rates of mallard drakes and mallard hens banded at 
Lake Chautauqua, 1939-1944. 
Banding traps are selective in that in most situations they take a disproportionate 
number of individuals in one or more of the various age and sex groups (usually the adult male 
group). Therefore, life tables based on banding data in which figures for all age and sex groups 
are lumped together usually are not representative of the population involved. 
However, at McGinnis Slough, trap selectivity was at a minimum because a relatively 
low population density allowed the ducks there to feed into the traps in such a way that seldom 
did an age or sex group dominate the situation, and the trapped population was fairly represen- 
tative of the population using the area. Because of this relative lack of selectivity in trapping 
and because the number of ducks involved was small, the data on all age and sex groups of each 
of the three species treated in this paper have been combined in determining mortality losses, 
tables 5, 6, and 7. 
Among McGinnis Slough mallards, the mortality losses for all age and sex groups, 
table 5, approximated losses suffered by mallard hens banded at Lake Chautauqua, table 4, and 
were slightly higher than mortality losses suffered by the entire mallard group banded at Lake 
Chautauqua. 
16 
