Phenological comparability in each of the four grouping charac¬ 
teristics would provide a maximum agreement index of 8. Results 
of this treatment are presented in table 2. 
To compare counts obtained in 1955 and 1956 with counts in 
1963 and 1964, data in each zone may be selected from one of the 
two earlier years that are most comparable to data from one of the 
two later years. In the southern zone, the highest agreement index 
(8) was obtained from comparing 1956 with 1963. Those years showed 
an 8-percent drop in black ducks per segment between the mid-1950's 
and the early 1960's. In the central zone, the procedure indicated 
that the years 1956 and 1963 were most comparable (agreement index 
= 7). Comparison of ducks per segment indicated a decrease of 21 
percent between 1956 and 1963. In the northern zone, 1955 and 1964 
appeared most similar (agreement index = 6) and showed a decrease 
of 10 percent between 1955 and 1964. 
To compare consecutive early years (1955-56) or consecutive 
late years (1963-64), we selected the zone which showed the highest 
agreement index and assumed that this zone was a .valid indicator 
of differences. Between 1955 and 1956, the southern zone showed 
an index of 7, while the northern zone had an index of 6 and the 
central zone an index of 5. The comparison with the highest index 
(the southern zone) indicated a drop of 8 percent between 1955 and 
1956. When we examine 1963 and 1964, the southern zone again had 
the highest agreement index (7), and the data indicated an increase 
in ducks per segment of 38 percent. 
In summary, the above procedure showed a decrease in black 
duck populations from the mid-1950's to the early 1960's, a decrease 
from 1955 to 1956, and an increase from 1963 to 1964, if the assump¬ 
tion is correct that the zones and years selected for comparison are 
representative. 
Adequacy of Current Survey Procedures 
Indexes available for examining black duck population trends 
include the winter survey, densities of black ducks observed during 
surveys on the breeding grounds, and the results of the group- 
characteristic procedure applied to breeding population survey data. 
Table 3 shows the indexes from the above sources for the years 
1955, 1956, 1963, and 1964. The trend of each index is presented 
graphically in figure 5. 
8 
