Division of the Illinois State Natural History Survey under the direc- 
tion of A. S. Hawkins, and the work of John Jedlicka for the Forest 
Preserve District of Cook County, Ill. A report on the former is in 
process of preparation (Bellrose in litt.); a mimeographed report on 
the latter has been written by Mann, Thompson, and Jedlicka (19)7). 
The 40,000 mallard recoveries in the files of the U. S. Fish 
and Wildlife Service represent an enormous research opportunity that 
may not be fully utilized for years to come. While I found the use- 
fulness of the recoveries severely restricted by the unimown age of 
most of the birds and the statistically awkward dates of banding (i.e., 
during the hunting season), I was certainly impressed with the cursory | 
nature of my own analyses of mallard recoveries and the great amount 
of unpublished information still remaining in these files. 
The mallard banding work, as of 1946, was particularly un- 
representative of young birds banded in the best breeding range 
before the start of the hunting season. The two mst notable geo- 
graphic gaps, in terms of banding stations, were Alaska and Texas. 
Banded literature on eurvival 
At the time this study was undertaken, no detailed survival 
studies of wild-trapped mallards in North America had been published. 
An important report by Bellrose and Chase (1950) on the survival of 
Illinois mallards appeared after this chapter was written; it will be 
discussed in Part III of this report. Munro (1943) has given the age 
distribution of 3387 British Columbian birds banded from 1928 to 190 
and later recovered up to January 31, 193; the oldest individual had 
at least reached age interval 10-11. 
H8hn (1948) analyzed the mortality reported for the European 
race of this species (Anas p. plat hos). Banding reports on 305 
adult-banded birds gave a mean annual mortality rate of 65 per cent, 
and a mean after lifetime of 1 year and 2 months. For 828 birds ringed 
as young (67 per cent of which were known to be hand-reared), "89% 
died during the first year, 9.6% during the second year, 0.6% during 
the third year and 0.8% during succeeding years. The average period 
of survival was 1.5 months after ringing.” Héhn points out that these 
results were based almost entirely on shooting, and that-—while it 
cannot be assumed that the banded always recorded the hand-reared 
status of young birds—-there is no a priori reason to assume a dif- 
ference in survival between hand-reared and wild-reared mallards. 
Considering the ease with which imprint is reported to 
occur in waterfowl (Lorenz 1937), it seems E me Tet it would be 
more conservative to assume that-—-until facts are available--some 
hand-reared waterfowl will adopt human beings as "social companions" 
and become unduly vulnerable to hunting as a result. A re-oxamination 
of H8hn's data is therefore pertinent. From table 27, in which. the 
first age interval starts as of the date of banding, it is evident 
that no difference exists in first-year mortality rates, but the 
66 
