
"local" Mallards are available from this degree block for the years 
1952 am 1953, respectively. The distribution of recoveries of 
"localttMallards from degree block 51° * 109° is presented in table 6. 
Section A of table 6 gives the percentages taken in Canada and 
each of the flyway areas based on the total direct recoveries for 
each year. In sections B and C the percentages are figured on the 
basis of only the recoveries taken within the United States. 
Discussion 
The data presented in table || were used for the first time when 
making the 1956 fall-flight forecast. Although it is difficult to 
determine the additional accuracy attained through use of these data, 
there are considerable changes between these data and the rough 
approximations of distribution to the flyways that were available for 
use during prior years. There is every reason to believe, therefore, 
that considerable gain has been made, 
- As information accunulates, it is not unusual for complications 
to arise; the problem of determining distribution from breeding , 
areas to the four flyways is an example. If breeding ground survey 
data are to be used to predict changes in the size of fall flight 
in each of four flyways, the distribution to the four flyways from 
specific portions of the breeding grounds must be predictable in 
advance. Inorder for distribution to be predictable, it must be 
reasonably similar from one year to the next. Although individual 
birds are generally considered to follow rather fixed patterrs of 
migration from one year to the next, the waterfowl~banding date seem 
to show that variations occur. Recoveries of "locals" banded in 
degree block 51° - 109° can be used to illustrate this variations 
In table 6 it is evident that there are fluctuations from year 
to year in the percentages of birds taken in Canada, and in the per-= 
centages taken in each flyway, particularly the Central and the 
Mississippi. As a means of examining these fluctuations, grids have 
been corstructed showirg the distribution of direct recoveries by 
latitude and longitude for the years 195) and 1955 (grids l and 5). 
fhe overall pattern of distribution is similar for both, but there 
are differences in the distribution of recoveries along the flight 
lanes between the two years. 
Considering the east-west components only, it appears that the 
proportion going each way was approximately the same: 
Weatward Same area Eastward 
195) 9.0% 28.7% 62.3%. 
1955 8.6% 31.8% 59.6% 
