Even though the "index" figures are not a measure of total populations, 
it is believed that they are representative of relative population levels 
to the extent that data from one location can be accumulated with those 
from another, and that year to year changes can be detected. Although 
a measure of total population would have certain advantages, a deter- 
mination of relative change seems adequate for the purpose of practical 
management. 
Needless to say, the breeding ground surveys are cooper- 
ative in nature. The Fish and Wildlife Service, the Canadian Wildlife 
Service, the Provincial game branches, and Ducks Unlimited, combine 
their manpower and equipment to cover all of the important waterfowl 
breeding areas in Canada. Service biologists cover the important areas 
in Alaska, while the State conservation agencies, with some help from 
the Service, carry on surveys in about 25 States. 
Waterfowl Kill 
During the 1952-53 waterfowl shooting season the Fish and 
Wildlife Service inaugurated a new: method of measuring the waterfowl 
kill. The method functions through the cooperation of the Post Office 
Department and provides for a sampling of the hunters in each flyway in 
proportion to their occurrence in the various States. The objectives of 
the kill survey are to determine for each flyway, (1) the number of birds 
taken by hunters with an error not to exceed five percent; (2) the size of 
the average daily bag; and (3) the evoreee number cf times a hunter went 
afield during the season. 
he mailing addresses for the questionnaire survey are 
obtained at the time duck stamps are purchased at Post Offices. The 
questionnaires are mailed out on the closing date of the shooting season 
in each State. Three weeks later, a follow-up questionnaire is mailed 
to those who have not answered the first questionnaire. The number of 
questionnaires mailed out and the number returned in each flyway is 
shown in the following table. 
