INTRODUCTION 
A continent wide pre-hunting season banding program was begun in 
1959 in an effort to measure the shooting pressure to which each age 
and sex component in the pre-season population is subjected. To do 
this, adequate samples of birds must be banded at each of a number 
of banding stations distributed across northern United States and 
southern Canada. Because of the similarity of arrangement between 
this network of banding stations and the network of radar stations 
spaced across Canada, this banding project has been called the 
"D. Es We Line” banding program (Distant Early Warning). Some pre- 
season banding has been done in the past, but not as an organized 
effort across the continent; these earlier data were not available 
for this report because they were involved in the reconstruction 
of the damage caused by the fire at Patuxent in June 1959. This 
report outlines the objectives of the program and summarizes the 
results from the banding of mallards and black ducks in 1959 and 
1960. It also includes a discussion of the problems and difficulties 
associated with the program. Examples are given to show how the 
data can be used in measuring the size, age composition and harvest 
rates of the pre-season duck population. 
OBJECTIVES OF PRE~SEASON BANDING 
The banding of flying adults and immatures shortly before the 
hunting season opens provides the best source of information on the 
following subjects: 
le Annual rates of hunting kill. 
2. Mortality rates. 
3. The importance of hunting as a mortality factor. 
4. The relationship between hunting regulations and the 
proportion of the waterfowl population harvested, 
5. The relative degree to which the various ages, sexes, and 
species are taken by hunting. 
Mortality due to hunting is measured most reliably from pre~season 
banding because ducks banded at other times of the year are affected to 
a greater extent by non-hunting season mortality. Pre-season banding 
may have one serious limitation, however. At present no population 
surveys are made at this time, and if the data prove to be highly 
variable, interpretation of the results will require that the size 
