Northern Prairie Wildlife Research Center
We identified the sex, in addition to the species, of 1,376 adult dabbling ducks and of 30 adult diving ducks found at dens. Seventy-six percent of the dabbling ducks and 40% of the diving ducks were hens. Most breeding duck populations in the midcontinent area contain more drakes than hens; the disparity is greatest among diving ducks (Bellrose et al. 1961, Bellrose 1976:24-25). Hence, fox predation on diving ducks tends to reflect the sex composition of diving duck populations much more than fox predation on dabbling ducks, which is strongly hen selective. We attribute this difference to differences in nest site selection. Most dabbling duck hens nest in uplands where the risk of fox predation is high, whereas most diving duck hens nest over water where they are relatively secure from foxes (Sergeant 1972, Bellrose 1976). Although lesser scaups nest in uplands (Bellrose 1976:349) and were the most vulnerable diving duck to foxes, only 6 of 15 with sex determined were hens. Our sample of lesser scaups is too small to reach meaningful conclusions about differential sex vulnerability, but the preponderance of males probably reflects the much higher proportion of males in diving duck populations than in dabbling duck populations and identification bias (lesser scaup females were difficult to identify). In addition, because the lesser scaup is one of the latest nesting ducks (Bellrose 1976:352) and the fox denning season ended before the end of the duck nesting season, hens taken from nests would be under represented in our samples.
Only the Nebraska data were inconsistent with these findings for dabbling ducks. Twelve of 16 sex-identified dabbling ducks (all mallards) found at dens in Nebraska were drakes. The Nebraska sample was unique in that it represented fox use of wintering and migrant ducks rather than predation on nesting ducks. The preponderance of drakes in that sample reflected the high proportion of drakes among wintering mallards. For example, 67% of 7,900 mallards surveyed in Nebraska during winter 1972-73 were drakes (Central Flyway Technical Committee, unpubl. minutes, March 1973).
The sex composition of dabbling ducks found at dens in North Dakota and South Dakota revealed that although fox predation was hen selective in all species, there were important differences among areas and species (Table 5). Sixty-five percent of west river mallards and northern pintails found at dens (species combined to increase sample size) were hens compared with 7656 hens (P < 0.05) for the east river area. Causes for this difference are unknown but may relate to wetland differences. Man-made stockponds with steep shorelines and little emergent vegetation predominate in the west river area in contrast to natural marshes with gradually sloping shorelines and often with much emergent vegetation in the east river area. Wetland characteristics affect duck loafing, nesting, and feeding habits and thereby affect the vulnerability of both hens and drakes to foxes.
Notable differences in the relative vulnerability of hens of different species to foxes were especially evident in the east river data. Sex compositions of ducks found at dens in that area ranged from 64% hens (1.8 hens/drake) for gadwalls to 90% hens (9.0 hens/drake) for northern shovelers (Table 5). The differences among species relate primarily to differences in nesting chronologies. For example, the low percentage of hens among gadwalls reflected the late season nesting habits of that species. By avoiding nesting early in the season, gadwall hens spend less time on nests during the fox denning season than do hens of early nesting species. Although time and duration of nesting are the most important factors affecting sex vulnerability, other factors such as behavior and nest site selection are also important. These other factors may be the cause of the higher percentage of hens among northern shovelers (a midseason nesting species) as compared with either mallards (P < 0.05) or northern pintails (P < 0.001) (both early season nesting species) and of the higher percentage of hens among mallards as compared with northern pintails P < 0.05).
The effect of nesting chronology on sex compositions of ducks taken by foxes is also evident in comparisons of percentage of hens in early and late season samples of the 5 principal dabbling duck species found at dens in our 3-county intensive study area (Table 6). Seventy-five percent of all dabbling ducks found at dens in that area were hens. The proportion of hens for each species increased from the early to late sample period. The seasonal difference was greatest for the gadwall, the latest nesting species, which increased from 50 to 65% hens. Also, the relative proportion of gadwalls in the sample more than doubled from the early to the late period. Gadwalls begin arriving on the nesting grounds in mid-April but do not nest until late May (Oring 1969, Dwyer 1974).
Few data were obtained on reproductive and physical conditions of ducks taken by foxes because nearly all ducks found at dens were represented by only a few feathers or by limited remains in various stages of decomposition. Reproductive state was positively determined for 4 hens. Two, a northern pintail and a blue-winged teal, were laying; the other 2, a mallard and a gadwall, were each 2-4 days into incubation. Other observations contributed information on 2 more hens, an incubating gadwall was found in a fox food cache, and fox predation on a mallard who had just laid her seventh egg was witnessed. These hens all appeared to be in good physical condition. Though scant, these data are consistent with our other findings that indicate most dabbling duck hens taken by foxes were nesting birds.