Fall and winter age ratios.--The accumlation of data on 
age ratios among waterfowl bagged by hunters almost took on the 
aspects of a major industry during the latter part of the 190's. 
The data are extremely difficult to analyze, and I want to acknowledge 
at this time the vast amoung of unpublished information that A. S. 
Hawkins, H. A. Hochbaum, and L. K. Sowls supplied to me on this 
subject. 
“There are at least two important reference points that 
life-table analyses can provide in the analysis of these phenomena. 
The first of these is some sort of average figure on the ratio of 
young to adults alive in the population as of September 1. The 
1.6 young per adult implied in table 66 (293 in c, divided by 180 
in f) must be regarded as a preliminary estimate only. This statistic 
owes its origin to table 28, where 163 wild-reared young birds were 
shown to. be alive on September 1 as against 103 adults. A better 
estimate awaits the creation of a larger sample and a clearer pic- 
ture of juvenile vulnerability to the gun. | 
A second reference point is the age ratio probably needed 
to keep mallard populations balanced; this can now be approximated 
as holding for birds alive at the end of the hunting season. The 
reasoning is quite simple. Over a period of 21 vears, four large- 
scale banding operations are known to have involved a mean annual 
me a rate of 8.7 per cent per year for adult mallards (table 
7). 
Table 67.--Adult Mallard Mortality Rates in the Aggregate 

erio umber um- ean Annua 
Where of Alive ber Mort. Rate 
Banded Study atstart Dying (% Per Year) Reference 
Pacific Coast 1927-16 3622 1760 48.6 Table 57 
‘TT1l.. and Mo. 192-37 2547 8911185 6.5 Table 63 
Western Montana 1928-37 951 597 62.8 This Study 
Illinois 1933=H6 1869 835 bh 7 Table 6) 
Totals and Mean a :):):} a aed ; 
Now if we can assume that this 20-year period contains 
about an even number of fluctuations in population level both up 
and down, we can say that this average mortality rate is our best 
presently available estimate of the average percentage of adults 
annually lost in this species. Under these conditions, stable 
populations will be maintained if by February 1 they average 51.3 
per cent adult and 8.7 per cent young. Thus a mid-winter age 
ratio of 0.95 young per adult is our best estimate of what is. 
required to balance mallard populations under "average" conditions. 
These ratios refer to birds alive in the population. 
Their relation to age ratios among birds that are dead can be 
seen in table 68. Under whatever conditions we select, the ratio 
159 
