Body composition of male and female Ring-necked Ducks (Aythya collaris) changed substantially over the course of reproduction, with the largest nutrient reserves occurring before the onset of laying. Ring-necked Ducks used lipid reserves acquired before occupancy of nesting areas to meet subsequent reproductive requirements. Paired males reduced feeding and expended lipid reserves while attending prelaying and laying mates. Females depleted lipid reserves during ovarian follicle growth; after ovulation they were highly dependent on ambient food resources. Female Ring-necked Ducks obtain protein required for egg production from the diet during ovarian follicle growth. This conclusion was supported by (1) the limited extent of body protein fluctuation; (2) the absence of a relationship between body protein and protein contained in recrudesced reproductive tissues and expended in eggs; and (3) similarities in the rate of change in body protein by date between males and females, and among females before and after laying. Female energy expenditures during ovarian follicle growth are influenced by costs associated with obtaining nutrients needed for egg production. The level of lipid reserve needed for reproduction, therefore, changes with food availability. Deficiencies in female lipid reserve at the initiation of ovarian follicle growth further increase exogenous requirements for reproduction and may result in adjusted breeding patterns or deferred breeding. The contribution of body reserves to reproduction was partially a function of nutrient levels attained during the arrival period. Variations in Ring-necked Duck body-weight change, by year and by female age, suggest that nutrient acquisition and allocation for waterfowl nesting in temperate regions may be more indeterminant than has been implied previously. Received 18 March 1985, accepted 20 September 1985.

Department of Fisheries and Wildlife, University of Minnesota, St. Paul, Minnesota 55110 USA ENERGETIC costs of reproduction for ducks and geese are large relative to those of other major bird groups (King 1973, Ricklefs 1974). In gen- eral, birds have a substantial capability for stor- age and utilization of energy reserves, primar- ily in the form of lipids (Blem 1976). Such reserves play a prominent role in waterfowl reproduction (Barry 1962; Ryder 1967, 1970; Harvey 1971; Korschgen 1977; Ankney and Macinnes 1978; Raveling 1979; Krapu 1981; Drobney 1982; Reinecke et al. 1982; Ankney 1984; Tome 1984). Ring-necked Ducks (Aythya collaris) are small- bodied (500-900 g), inland diving ducks (Tribe Aythyini), which have been studied principal- ly in the northeast (Mendall 1958). They com- monly nest in northern bog marshes, perma- nent wetlands characterized by low primary production (Reader 1978). I examined changes Present address: Delta Waterfowl Research Sta- tion, Rural Route 1, Portage la Prairie, Manitoba R1N 3A1, Canada. in body weight and body composition of breeding Ring-necked Ducks by sex and age. Attention was focused on body condition of returning birds, with the prediction that en- dogenous reserves accumulated before occu- pancy of wetlands used for nesting were criti- cal to successful reproduction by this species. METHODS Breeding Ring-necked Ducks were studied from April to August 1978-1980 on Roseau River Wildlife Management Area (WMA) in northwestern Minne- sota. Birds were collected by shooting, decoy trap- ping (Anderson et al. 1979), nest trapping (Weller 1957), or night-lighting (Lindmeier and Jessen I961). Body weights (+5 g) were taken immediately after collection. Feathers (5th primary, 5th secondary, 12th and 13th greater upper secondary coverts) were re- moved in 1979 and 1980, and age (yearling = first- year bird, before prebasic II molt; adult = after-first- year bird, after prebasic II molt) was assigned on the basis of discriminant function analysis of feather measurements (Hohman I984). Live-trapped birds were marked individually with a plastic nasal-saddle (Doty and Greenwood 1974). Birds collected by shot- gun were frozen in sealed plastic bags for later dis- section. Females and their mates were assigned to the fol- lowing reproductive categories, based in part on Kra- pu (1974): expressed as a percentage of maximum mean body weight (Reinecke 1977). Mean body weights by re- productive status and 10-day interval were compared between 1979 and 1980 using sign tests (Conover 1980). Regression procedures followed Weisberg (1981). Significance level was set at P < 0.05. Arrival.--Females returning to the breeding grounds with ovary weight <3 g. Prelaying.--Preovulating females with ovary weight >-3g. Laying.--Ovulating females. Early incubation.--Postovulating hens captured on nest, estimated days of incubation -<9 days as determined by field candling (Weller 1956) or embryonic inspection (Caldwell and Snart 1974). Middle incubation.--Eggs incubated 10-18 days. Late incubation.--Eggs incubated > 18 days. Early brood.--Hens attending age class I ducklings (Gollop and Marshall 1954). Late brood.--Hens attending age class II ducklings. Postbreeding.--Birds collected in July and August on molting areas before remigial feather molt. Females initiated ovarian follicle growth after 6 May, as verified by back-dating on hens of known repro- ductive status. Consequently, all birds handled be- fore 6 May were placed in the arrival category. Pair- ing status was assigned to marked males on the basis of behavioral observations made following capture. Paired individuals were birds showing active asso- ciation, i.e. copulation, mutual display, female tol- erance of male, or nonrandom spacing. Unmarked males were considered unpaired if observed on col- lection before 6 May without a mate. Predictive indices were used to estimate ash-free, lean dry weight ("protein"), and lipid composition of whole carcasses less reproductive tissues (Hohman and Taylor 1986). Body lipid was regressed on wet weights of abdominal fat and skin tissues. Protein was estimated from eviscerated carcass, right breast muscle and right leg muscle weights, and bill length measurement. Lipid and protein content of ovary and oviduct tissues were derived using wet weights of these organs and estimates of composition of the ovary (35.2% lipid, 17.7% protein; Romanoff and Romanoff 1949) and of the oviduct (22.4% protein; Drobney 1980). To monitor change in mineral composition, ashed (550øC, 48 h) weights were taken on female femurs and tibiotarsi. Statistical comparisons of body weight, body lipid and protein composition, and weight of ashed leg bones by reproductive status and 10-day Julian date interval were made using one-way analysis of vari- ance. Multivariate (factors = age, year, date, and re- productive status) analysis of variance was per- formed on body-weight data using BMDP-4V statistical software (Dixon and Brown 1981). The maximum extent of seasonal body-weight change was RESULTS BREEDING PHENOLOGY Paired Ring-necked Ducks began to arrive in the vicinity of Roseau River WMA in early April, soon after snowmelt. Sheet water on ag- ricultural fields adjacent to the management area was used by birds initially. Birds remained on seasonal wetlands until about late April, when the study area became ice-free. Laying was initiated from mid-May to early June, and broods usually were evident by the fourth week of June. Males left Roseau River WMA by early July; females departed 4-6 weeks later, after their broods had attained class II stage of plum- age development. Breeding phenology for Ring-necked Ducks in northwestern Minnesota was similar to that observed by Sarvis (1972) in Michigan, but somewhat later than that recorded in Maine (Mendall 1958). Breeding phenology in north- western Minnesota did not vary appreciably among years. BODY WEIGHTS Female.--Body weights were obtained from 263 breeding female Ring-necked Ducks. Female body weight changed substantially through reproduction (Table 1). Weight in- creased significantly from arrival to maximum seasonal levels at the initiation of ovarian fol- licle growth. Body weight declined after the onset of laying. Body weight recovered signif- icantly in hens with class II ducklings after late July from seasonal minima observed in early brood rearing. The maximum body-weight change amounted to 29% (229 g) for breeding females. The pattern of body-weight change differed for yearling and adult females, and varied among years. Adult females collected early in the arrival period were significantly heavier than yearlings in both 1979 and 1980 (Table 2). Mean body weights for adult and yearling birds, however, did not differ by reproductive status T^ILE 1. Body weight and body composition of breeding female Ring-necked Ducks collected in north- western Minnesota, 1978-1980. Stage of reproduction Body weight Lipid Protein Arrival 703 + 8 a 88.0 + 7.0 119.7 + 1.1 n 69 35 35 pb * * * NS NS Prelaying 803 + 35 118.2 + 24.1 123.8 + 3.2 n 8 4 4 P NS ** *** Laying 782 + 12 51.0 + 4.0 116.1 + 1.2 n 16 14 14 P *** NS NS Early incubation 666 + 14 35.5 ñ 4.1 111.9 + 0.9 n 5 4 4 P NS NS NS Midincubation 653 ñ 15 37.8 ñ 5.7 111.9 + 2.5 n 9 2 2 P * * NS NS Late incubation 594 + 8 27.9 + 3.2 106.9 + 7.9 n 23 2 2 P NS NS NS Early brood rearing 574 + 7 28.1 + 3.5 103.3 + 2.9 n 54 3 3 P NS NS NS Late brood rearing 588 + 7 40.1 + 4.2 101.3 +1.4 n 42 5 5 a Mean + $E (g). b p = probability (from one-way ANOVA) that adjacent means within a column are significantly different. ** = P < 0.01; *** = P < 0.001; NS = P > 0.05, not significant. within years, suggesting that adult females re- turned heavier or gained weight more rapidly, or both, compared with yearlings. Mean body weight by 10-day interval fluc- tuated more widely in 1979 than in 1978 and 1980 (Fig. 1). Differences among years were greatest early in the reproductive cycle. Before 6 May, adults and yearlings were significantly lighter in 1980 than early-arrival birds collect- ed in the previous years (Table 2). Females by reproductive status (sign test, T = 5, df = 5, P < 0.05) and by 10-day interval (sign test, T = 9, df = 10, P < 0.01) were generally heavier in 1979 than in 1980. Outside of the arrival peri- od, however, mean female body weight by 10- day interval or by reproductive status did not differ significantly between years. Male.--Body weights were taken on 334 T^ILE 2. Body weights of breeding Ring-necked Ducks collected in northwestern Minnesota early in the arrival period (before 6 May). Female Male Year Yearling P' Adult All Yearling P Adult All 1978 b 700 ñ 13 c 731 ñ 8 n 15 44 1979 715 ñ 20 * 777 ñ 18 742 ñ 16 772 ñ 21 NS 793 ñ 15 787 ñ 12 n 9 7 16 6 16 22 P ** ** NT NT NT *** 1980 632 ñ 12 *** 705 ñ 9 670 ñ 11 700 ñ 12 NS 726 ñ 7 718 ñ 6 n 12 13 25 20 50 70 a p = probability that adjacent means within a column or row are significantly different. * = P < 0.05; ß * = P < 0.01; *** = P < 0.001; NS = P > 0.05, not significant; NT = no test. b Age was not determined. c Mean ñ SE (g). 8oo I /\ FEMALE ß / X --- 1978 (n = 38) / . -- 1979 (n= 123) ! / . ..... ,, 6 550 , , 3/4 , . 850 750 650 550 MALE --- 1978 1979 (n-lo3) .......... 1980 (n= 98) / APRIL MAY JUNE JULY Fiõ. 1. Mean body weiõhts of breedinõ female and male Ring-necked Ducks by 10-day interval and year, 1978-1980. breeding males. Male body weight changed significantly with their mates' reproductive status and by 10-day interval (Table 3). Birds were heaviest during their mates' arrival and prelaying reproductive periods. A significant decrease in body weight corresponding to the period of ovarian follicular growth occurred in mid- and late May. Seasonal minima in male body weight were recorded after their females began incubation. The maximum extent of change in male body weight amounted to 10% (73 g). No significant age-related differences in male body-weight change were detected, but male body weight by 10-day interval changed dif- ferently in 1979 and 1980 (Fig. 1). Seasonal fluctuation in mean body weight by 10-day in- terval was greater in 1979 than in either 1978 or 1980. Males collected early in the arrival pe- riod were significantly heavier in 1979 than in 1978 and 1980 (Table 2). Weight was gained between mid-April and early May 1979, and similar gains were shown by recaptured indi- viduals. Similar increases were not observed before the onset of ovarian follicle growth in other years. Outside of the arrival period the patterns of seasonal weight change for all years were alike. BODY COMPOSITION Female.--Lipid and protein composition was estimated for 72 breeding females. Lipid stores were largest in arrival and prelaying birds be- fore mid-May (Table 1). Body lipid declined significantly when females began laying. Sea- sonal minima occurred during late incubation and early brood rearing, mid-June to late July. No significant differences between adult and yearling females, or among years, were appar- ent in body lipid by reproductive status or by 10-day interval, although small sample sizes in some categories may have prevented the detec- tion of differences. Body protein declined 19% (23 g) from pre- laying to late brood rearing (Table 1). Seasonal decline in body protein was approximately constant (Fig. 2) and unrelated to changes in reproductive protein, i.e. protein contained in recrudesced reproductive tissues and expended in eggs (r = -0.3467, df = 15, P > 0.05). Minerals required for eggshell formation ap- parently were deposited in the skeleton during the period of ovarian follicle growth. Fernoral and tibiotarsal mineral content increased sig- nificantly from arrival to prelaying and laying. Ashed leg-bone weights before and after the period of ovarian follicle growth did not differ, and were equivalent in adult and yearling fe- males. Male.--Lipid and protein composition was estimated for 78 breeding males. Body lipid in males decreased significantly between their mates' arrival and laying reproductive stages (Table 3). Lipid level in males was greatest be- fore mid-May, when some females initiated laying. Seasonal minima were recorded in mid- June at the termination of the laying period. Body lipid remained at low levels while birds were on the study area. Changes in male lipid and protein composition were not significantly influenced by age, year, or pairing status. Body protein declined in males from May to July (11%, 14 g) at the same rate observed in females (Fig. 2). T^BLE 3. Body weight and body composition of breeding male Ring-necked Ducks collected in northwestern Minnesota, 1978-1980. Stage of mate's reproduction Body weight Lipid Protein Arrival 735 + 5 a 102.5 + 6.8 125.6 + 1.0 n 144 38 38 pb NS NS NS Prelaying 752 + 22 93.0 + 44.3 126.6 + 2.1 n 6 3 3 P NS NS ** Laying 707 + 11 54.6 + 4.5 117.8 + 1.8 n 13 12 12 P NS NS NS Incubation 679 + 19 49.6 + 4.6 117.2 + 2.2 n 12 5 5 P * * NS NS Postbreeding 735 + 22 65.4 + 11.1 121.2 + 2.0 n 9 9 9 ß Mean + $E (g). b p = probability (from one-way ANOVA) that adjacent means within a column are significantly different. ß * = P < 0.01; NS = P > 0.05, not significant. DISCUSSION Body composition of male and female Ring- necked Ducks changed substantially over the course of reproduction, with the largest nu- trient reserves occurring before the onset of laying. Ring-necked Ducks use lipid reserves acquired before occupancy of nesting areas to meet subsequent reproductive requirements. Females incur substantial costs from egg pro- duction. Lipid reserves are depleted during ovarian follicle growth; after laying, females are highly dependent on ambient food re- sources. Paired males reduce feeding (Hohman unpubl. data) and expend lipid reserves while attending prelaying and laying mates. Close at- tendance of the female minimizes the chances of a male being displaced by another male, or having his mate inseminated by other males. Moreover, male defense of the female against intrusions by other birds and vigilance against potential predators probably facilitate female feeding (Ashcroft 1976). This may be especially important during ovarian follicle growth, when female nutrient demands are greatest. Protein stored in muscle sarcoplasm (Kendall et al. 1973) is used by Common Eiders (Somateria mollissima; Korschgen 1977) and some arctic- nesting geese (Ankney and Macinnes 1978, Raveling 1979) to produce eggs. Over 80 g of protein are required for production of an av- erage clutch and biosynthesis of oviduct tissue by Ring-necked Ducks (Hohman 1984). This is equivalent to about 67% of the protein content of female carcasses before ovarian follicle growth. Female Ring-necked Ducks obtain pro- tein required for production of eggs almost ex- clusively from the diet during ovarian follicle growth. This conclusion is supported by (1) the limited extent of body protein fluctuation; (2) the absence of a relationship between body protein and protein contained in recrudesced reproductive tissues and expended in eggs; and (3) similarities in the rate of change in body protein by date between males and females, and among females before and after laying. Other small-bodied waterfowl, such as Wood Ducks (Aix sponsa; Drobney 1982), American Black Ducks (Anas rubripes; Reinecke et al. 1982), Mal- lards (Anas platyrhynchos; Krapu 1981), and Ruddy Ducks (Oxyura jamaicensis; Tome 1984), also obtain protein required for egg production from the diet during prelaying and laying. Female energy expenditures during ovarian follicle growth are influenced by costs associ- ated with obtaining nutrients for egg produc- tion. Depletion of body lipid by prelaying/lay- ing Mallard and Wood Duck hens may exceed the amount contained in a completed clutch of eggs (Krapu 1981, Drobney 1982). This differ- ence was attributed to additional foraging costs incurred by females specializing on inverte- brate foods. Hohman (1985) suggested that se- lecting food on a noncaloric basis (e.g. protein, balanced amino acid composition, minerals) may reduce net foraging gains and result in oMae 135- ø ß Female o o 125- ', P, ~,.,...~ o ;' ß % ß 105- . ß 95 APRIL MAY JUNE JULY Fig. 2. dy protein composition by date for breeding male and female Ring-necked Ducks. Regressions significant at P < 0.0001. expenditure of stored lipids, as noted in Mal- lard (Krapu 1981) and Wood Duck (Drobney 1982) hens. Birds may minimize the costs of obtaining reproductive protein requirements by timing laying to coincide with peak inverte- brate abundance. This was suggested by a sig- nificant positive correlation between nonmol- lusc invertebrate biomass and the weight of female reproductive tissues (Hohman 1984). Body weights are a useful index for exam- ining nutrient changes in waterfowl (Bailey 1979, Raveling 1979, Wishart 1979, Ankney 1982, Drobney 1982). Body-weight data suggest that nutrient levels in Ring-necked Ducks en- tering reproduction may vary by year and by female age. Reduced reserves in females at the initiation of ovarian follicle growth increases reliance on exogenous sources for reproductive nutritional requirements. Deficiencies in fe- male lipid reserve (i.e. lipid needed for egg production and energy costs associated with obtaining protein requirements) increase fur- ther the exogenous requirements for reproduc- tion and may result in breeding adjustments such as altered time-activity patterns, delayed laying, or clutch-size reduction. Inadequate re- serves coupled with reduced food availability could cause deferred breeding. Yearling fe- males are lighter than adults early in the arriv- al period, and probably are more dependent than adults on food resources in wetlands used for nesting. Biomass and density of foods in these wetlands change seasonally and vary among years (Hohman 1984). Adult and year- ling females entered reproduction in 1980 at reduced body weights, but available nutrients (reserve + ambient food resources) apparently were inadequate for yearlings, most of which did not nest in 1980 (Hohman 1984). The contribution of body reserves to repro- duction was partially a function of nutrient levels attained during the arrival period. Whereas body weight during the arrival period was constant in 1980, weight gained before ear- ly May constituted over 50% (50% female, 62% male) of the total body-weight fluctuation ob- served in 1979. Before occupancy of the wet- lands used for nesting, male and female diets consisted mostly of plant material, including seeds of native grasses and sedges, annual plants, and waste grain that birds gathered on temporally flooded wetlands (Hohman 1985). The rate of caloric intake by feeding Ring- necked Ducks was greater in temporary wet- lands than in the semipermanent and perma- nent wetlands used for nesting. Deposition of lipids undoubtedly was facilitated by the use of high-carbohydrate plant foods, which were readily available in temporary wetlands. The lower body weights observed in 1980 were probably a consequence of drought, which re- duced the area of temporary wetlands and, thereby, prevented access to seeds of wet- meadow vegetation. Events occurring over winter and during mi- gration also may influence the reserve status of returning birds. Ring-necked Ducks in south- ern Florida put on body fat and showed a con- stant weight gain through the wintering peri- od (Hohman et al. 1986). The rate of increase was significantly greater in yearlings than adults, but varied annually. Yearlings in fall were initially lighter than adults, but body weights were equivalent by late winter. Ring - necked Ducks defer pairing until spring mi- gration (Weller 1965) and, thus, simultaneously incur energetic costs of migration and court- ship in March/April. Reductions in the body weight of returning birds from winter maxima presumably reflect these costs. Similarities in the arrival body weights of males by age, and differences in adult and year- ling female body weight (adult > yearling) early in the arrival period, suggest sex- and age- related differences in prearrival and arrival costs. Familiarity with food resources on stag- ing and breeding areas may enable adult fe- males to feed more efficiently than inexperi- enced yearling hens. Adult and yearling males following mates to their mates' natal areas were presumably also naive regarding food resource availability in nesting wetlands and showed similar body weight changes. Krapu (1981) suggested that age-related differences in lipid reserve carried by Mallard females during the nesting season may result because inexperi- enced yearlings are less efficient at capturing invertebrates than adult females (i.e. incur greater energy costs). Differences in body weights of yearling and adult female Ring- necked Ducks, however, appeared before birds shifted to invertebrate foods. Variation in the chronology of pairing also might account for initial differences in adult and yearling female body weight. Pair-bond formation entails costs and benefits that differ between the sexes. Males incur considerable energy costs during courtship and while at- tending mates. The advantages of early pair formation to the male, therefore, must be weighed against the risks of subsequent dis- placement by a bird in superior condition. Fe- males benefit from mate attendance and may be ready to pair before males (Afton and Sayler 1982). Male investment probably is influenced by female quality. Thus, experienced females may obtain mates, and begin to accrue pairing benefits, before yearlings. Indeed, intensive courtship activity on nesting areas appears to involve primarily yearling individuals (Hoh- man unpubl. data). Extended courtship costs, or intrusions by unpaired males testing weak pair bonds and disrupting feeding, probably impair yearling female ability to accumulate nutrient reserves after their return to breeding areas. Patterns of nutrient acquisition and alloca- tion for ducks nesting in temperate regions, as evident in this study, may be more indetermi- nant than previously implied (Owen and Re- inecke 1979). Waterfowl are long-lived occu- pants of variable environments and exhibit temporally dynamic reproductive strategies (Nichols et al. 1976). While ultimate control of factors such as clutch size and time of laying probably are genetically based (Bait and Prince 1979), expressed breeding patterns are the re- sult of interplay between local environmental conditions (e.g. food availability) and the nu- trient-reserve status of the parent. Changes in local environmental conditions or reserve sta- tus of the parent may result in altered, subop- timal breeding patterns. Studies that link the reproductive performance of individuals with their environment are a necessary first step to- ward the interpretation of the broad pattern exhibited by the species (Drent and Daan 1980). Future research examining the bioenergetics of temperate-breeding waterfowl must consider the individual variation, as well as behavioral plasticity, in birds that enables individuals to adjust breeding patterns to changing environ- mental conditions. ACKNOWLEDGMENTS This paper represents a portion of my Ph.D. dis- sertation, work supervised by M. W. Weller and sub- mitted to the University of Minnesota at St. Paul. I was also advised by E. F. Cook, J. A. Cooper, F. McKinney, and J. R. Tester. I am grateful to R. T. Alisauskas, C. D. Ankney, G. L. Krapu, M. W. Tome, and M. W. Weller for their comments on earlier drafts of the manuscript. Financial support was provided by the North American Wildlife Foundation through the Delta Waterfowl Research Station, Minnesota De- partment of Natural Resources, University of Min- nesota (Department of Fisheries and Wildlife, Com- puting Center, Agricultural Experiment Station, and Graduate School), and the Minnesota Waterfowl As- sociation. LITERATURE CITED AFTON, A.D., & R. D. $AYLER. 1982. Social courtship and pair bonding of Common Goldeneyes (Bu- cephala clangula) wintering in Minnesota. Can. Field-Natur. 96: 295-300. ANDERSON, M. G., R. D. $AYLER, & A.D. AFTON. 1979. A decoy trap for diving ducks. J. Wildl. Mgmt. 44: 217-219. ANKNEY, C. D. 1982. Annual cycle of body weight change in Lesser Snow Geese. Wildl. Soc. Bull. 10: 60-64. 1984. Nutrient reserve dynamics of breed- ing and molting Brant. Auk 101: 361-370. , & C. D. MACINNES. 1978. Nutrient reserves and reproductive performance of female Lesser Snow Geese. Auk 95: 459-471. ASHCROFT, R. E. 1976. A function of the pair bond in the Common Eider. Wildfowl 27: 101-105. BAILEY, R.O. 1979. Methods of estimating total lipid content in the Redhead Duck (Aythya americana) and an evaluation of condition indices. Can. J. Zool. 57: 1830-1833. BARRY, T.W. 1962. Effect of late seasons on Atlantic Brant reproduction. J. Wildl. Mgmt. 26: 19-26. BAIT, B. D. J., & H. H. PRINCE. 1979. aying dates, clutch size, and egg weight of captive Mallards. Condor 81: 35-41. BLEM, C.R. 1976. Patterns of lipid storage and uti- lization in birds. Amer. Zool. 16: 671-684. CALDWELL, P. J., & A. E. SNART. 1974. A photograph- ic index for aging Mallard embryos. J. Wildl. Mgmt. 38: 298-301. CONOVER, W.J. 1980. Practical nonparametric sta- tistics. New York, John Wiley & Sons. DIXON, W. J., & M. B. BROWN (Eds.). 1981. BMDP statistical software, P-series. Berkeleyß Univ. Cal- ifornia Press. DOTY, H. G., & R. J. GREENWOOD. 1974. Improved nasal-saddle marker for Mallards. J. Wildl. Mgmt. 38: 938. DRENT, R. H., & S. DAAN. 1980. The prudent parent: energetic adjustments in avian breeding. Ardea 68: 223-252. DROBNEY, g.D. 1980. Reproductive bioenergetics of Wood Ducks. Auk 97: 480-490. ß 1982. Body weight and composition changes and adaptations for breeding Wood Ducks. Con- dor 84: 300-305. GOLLOP, J. B., & W. H. MARSHALL. 1954. A guide for aging duck broods in the field. St. Paulß Missis- sippi Flyway Council Tech. Sect. Rept. HARVEY, J. M. 1971. Factors affecting Blue Goose nesting success. Can. J. Zool. 49: 223-234. HOHMAN, W. L. 1984. Aspects of the breeding bi- ology of Ring-necked Ducks (Aythya collaris). Unpublished Ph.D. dissertationß St. Paul, Univ. Minnesota. ß 1985. Feeding ecology of Ring-necked Ducks in northwestern Minnesota. J. Wildl. Mgmt. 49: 546-557. --, & T. $. TAYLORß 1986. Indices of fat and protein for Ring-necked Ducks (Aythya collaris). J. Wildl. Mgmt. in press. ß --, & M. W. WELLgR. 1986. Annual body weight change in Ring-necked Ducks (Aythya collaris). In Proc. Waterfowl in Winter Symp. (M. W. Weller, Ed.). Galveston, Texas. KENDALL, M.D., P. WARDß $. BACCHUS. 1973. A protein reserve in the pectoralis major flight muscle of Quelea quelea. Ibis 115: 600-601. KING, J.R. 1973. Energetics of reproduction in birds. Pp. 78-107 in Breeding biology of birds (D. S. Farner, Ed.). Washington, D.C.ß Natl. Acad. Sci. KORSCHGEN, C. E. 1977. Breeding stress of female Eiders in Maine. J. Wildl. Mgmt. 41: 360-373. KRAPU, G.L. 1974. Feeding ecology of pintail hens during reproduction. Auk 91: 278-290. 1981. The role of nutrient reserves in Mal- lard reproduction. Auk 98: 29-38. LINDMEIER, J.P., & R. L. JESSEN. 1961. Results of capturing waterfowl in Minnesota by spotlight- ing. J. Wildl. Mgmt. 25: 430-431. MENDALL, H.L. 1958. The Ring-necked Duck in the northeast. Orono, Univ. Maine Stud., 2nd Series No. 73. NICHOLS, J. S., W. CONLE3/4, B. BATT, & A. R. TIPTON. 1976. Temporally dynamic reproductive strate- gies and the concept of r- and k-selection. Amer. Natur. 110: 995-1005. OWENß R. B., JR., & K. J. REINECKE. 1979. Bioener- getics of breeding dabbling ducks. Pp. 71-93 in Waterfowl and wetlands--an integrated review (T. A. Bookhutß Ed.). Madison, Wisconsin, Proc. Symp. North Central Section, The Wildlife Soc. RAVELING, D. G. 1979. The annual cycle of body composition of Canada Geese with special ref- erence to control of reproduction. Auk 96: 234- 252. READER, R.J. 1978ß Primary production in northern bog marshes. Pp. 53-62 in Freshwater wetlands: ecological processes and management potential (R. E. Goodß D. F. Whigham, and R. L. Simpson, Ed$.). New York, Academic Press. REINECCE, K.J. 1977. The importance of freshwater invertebrates and female energy reserves for Black Ducks breeding in Maine. Unpublished Ph.D. dissertation, Orono, Univ. Maine. ß T. L. STONE, & R. B. OWEN. 1982. Seasonal carcass composition and energy balance of fe- male Black Ducks in Maine. Condor 84: 420-426. RICKLEFS, R. F. 1974. Energetics of reproduction in birds. Pp. 152-292 in Avian energetics (R. A. Paynter, Jr., Ed.). Publ. Nuttall Ornithol. Club No. 15. ROMANOFF, A. L., & A. J. ROMANOFFß 1949. The avian egg. New Yorkß John Wiley & Sons. RYDER, J.P. 1967. The breeding biology of Ross' Goose in the Perry River region, Northwest Ter- ritories. Can. Wildl. Serv. Rept. Ser. No. 3. ß 1970. A possible factor in the evolution of clutch size in Ross' Goose. Wilson Bull. 82: 5-13. $ARVIS, J. E. 1972. The breeding biology and ecol- ogy of Ring-necked Ducks in northern Michi- gan. Unpublished M.S. thesis, Logan, Utah State Univ. TOME, M. W. 1984. Changes in nutrient reserves and organ size in female Ruddy Ducks breeding in Manitoba. Auk 101: 830-837. WEISBERG, S. 1981. Applied linear regression. New Yorkß John Wiley & Sons. WELLER, M.W. 1956. A simple field candler for wa- terfowl eggs. J. Wildl. Mgmt. 20: 111-113. --. 1957. An automatic nest-trap for waterfowl. J. Wildl. Mgmt. 21: 456-458. --. 1965. Chronology of pair formation in some nearctic Aythya (Anatidae). Auk 82: 227-235. WISHART, g.A. 1979. Indices of structural size and condition of American Widgeon (Anas ameri- cana). Can. J. Zool. 57: 2369-2374.