
the mallard and pintail. The longest nesting season was 1951, which began 
early and, with the exception of the blue-winged teal, continued late; it 
is felt that the large number of water areas present through Jume stimulated 
continued nesting effort. On the other hand, the 1952 season began early 
but was not prolonged, possibly because of the rapid drop in water levels 
starting in early June. In 1953, a wet and cold spring, the early nesting 
mallards and pintails were retarded ~- pintails to an even greater extent 
than in 1950. After the period of bad weather in April and early May, con- 
ditions became more favorable for nesting and the normally late-nesting blue- 
winged teal and gadwall were not noticeably retarded. Nesting for all species 
in 1953 was compressed into a shorter period than in any other of the four 
years. However, the productivity of the nesting population was at its highest 
in 1953, and there were fewer unsuccessful hens available for late renesting, 
even though conditions were favorable for it. 
The total length of the brood season and its chronology is a highly 
variable item, as is shown by figures 14-18. These figures indicate by species 
the percent of the total brood population visible at any date during the 
season. They are an accumulation of all broods produced, each spaced along 
the abscissa with regard to its dates of hatching and flying. The percent of 
the year's production present on any date may be read directly from the curves. 
The number of broods per square mile of each species produced on the study 
area is indicated below the year dates. 
The longest brood seasons were those of the mallard and pintail, which 
nest early, require considerable time to reach flying age, and are persistent 
renesters. In 1951, there were mallard broods present on the area for 148 
days, and there were either mallard or pintail broods present for 162 days, 
or more than 5 months, from May 20 to October 29. This is in contrast to 
1953 when broods were present for only 125 days, or a little over 4 months. 
The total span of the nesting season is also reflected in the percentage 
of total brood production present in a flightless state at the period of peak 
density. For all species this varied between 82 percent in 1953 and 72 per- 
cent in 1951. When the species are considered individually, this variation 
is even more pronounced. For instance, 95 percent of all mallard production 
could be measured at the peak in 1953, but.only 65 percent in 1951. If a 
specific date is considered, another variable is added, in that brood seasons 
of equal duration may peak at different dates. For instance, considering all 
broods combined, a count on July 20 of 1950 would have measured only 36 per- 
cent of the season's production, while in 1953 a count on the same date would 
have measured 67 percent. This difference can be decreased by making a count 
more nearly at the average peak date. A count on August 10 each year would 
have measured from 67 percent of all broods produced in 1951 to 75 percent in 
1952 and 1953. 
There is, therefore, considerable variation in the timing of brood use 
of an area, the early nesting species being more variable than the late 
nesters which are not so greatly affected by early spring weather. 
In no case is it possible to pick out more than one definite peak of 
hatching. The only possibility is suggested by the relatively few pintail 
broods in 1951 and 1952. In all other cases, staggered dates of nest be- 
27 
