
of the population. In 1951, on the other hand, blue-winged teal populations 
were decreasing by late September, as shown in table 6. 
Although the study was terminated before the hunting season in 1952 and 
in 1953, data for the other two years indicate that hunting rapidly drives 
the birds off the study area. On only one occasion was the hunting-season 
population known to exceed 25 percent of the preseason population, and on 
one occasion during the peak of migration it dropped to only 5 percent, later 
to rise to 36 percent of the preseason level as new migrants came in. It is 
not known where these birds (over 300 per square mile) go when the shooting 
starts, but they do not show up on the Waubay Refuge or the nearby lakes. 
The effect of hunting is further demonstrated by the fact that on six Fridays 
the population averaged 55 percent greater than on 10 weekends when hunting 
was heaviest. These differences were in spite of the fact that hunting 
pressure was extremely light, about a dozen parties of visitors on opening 
weekend and three or four on subsequent weekends, plus about the same number 
of local residents. With the exception of the opening day, shooting during 
the week was a rarity. This low pressure was undoubtedly due to the ban on 
the shooting of waterfowl by nonresidents of the State. Such hunting as was 
done was relatively successful, and 37 parties interviewed in 1950 and 1951 
averaged 0.7 duck per man-hour of hunting effort. 
Chronology of the season 
Brood data provide much more accurate and useful information for study- 
ing successful nesting than does a study of either spring populations or 
nests. Since nearly all broods were seen and the ages of a high percentage 
were estimated, there was constant intensity of sampling throughout the 
season, and it was possible to determine quite accurately the chronology of 
successful nesting effort. Each brood tallied as produced on the area 
(Appendix I; Analysis of Production) was dated back to the start of its nest. 
This was done by calculating the hatching date from the estimated age of the 
brood, back-dating a number of days equal to the incubation period for the 
species, then back-dating a number of days equal to the average clutch size 
for the species. For instance, a brood estimated as 4 weeks old on July 15, 
of a species with an incubation period of 26 days and an average clutch size 
of 9 eggs, would have hatched June 17, begun incubation May 22, and come 
from a nest started May 13. Figures 12 and 13 show the chronology of suc- 
cessful nesting effort obtained by this method. These data refer only to 
successful nests, and each curve is cumulative. For example, 60 percent in- 
dicates the intensity of successful nesting effort for that period. 
The year 1950 was conspicuous for retarded nesting in the blue-winged 
teal and gadwall, though not markedly so in the mallard and pintail. Except 
for the blue-winged teal, nesting in 1950 was not prolonged to a later-than- 
normal date, even though both the blue-winged teal and the gadwall showed an 
increase in the rate of nesting in late Jume. An unusual number of birds 
could be seen in mid-June of that year loafing in groups on mud flats, of 
drying potholes, apparently having abandoned further attempts at nesting. 
This was dssociated with a rapid drop in water levels and a decrease in number 
of potholes at that time. A partial recovery of water levels late in the 
month did not appear to result:’in a resumption of normal nesting effort for 
26 
