The Condor 96:909-9 I S 0 The Cooper OrnithologicalSociety 1994

VARIATION IN BODY MASS OF WILD CANVASBACK AND REDHEAD DUCKLINGS ’

JANEE. AUSTIN National Biological Survey, Northern Prairie ScienceCenter, Jamestown,ND 58401-7317 JEROMER. SERIE U.S.Fish and Wildlife Service,Ojice of Migratory Bird Management, Patuxent Environmental ScienceCenter, Laurel, MD 20708

Abstract. We assessedvariation in body massof ducklingsin single-and mixed-species broodsof wild Canvasbacks(Aythya valisineria) and Redheads(Aythya americana) 20-50 daysold. Bodymass of canvasbackducklings was not affectedby yearand season(early vs. late hatchdate) despite changes in waterconditions. Mean body massof male and female Canvasbacksdid not differ in ClassIIA but did differ in older ageclasses. Within-brood differencesin body masstended to be higherin ClassIIA ducklings(6-7% of mean body massfor Canvasbacks,9-l 1% in Redheads)and generallydeclined to 4-6% in ClassIIC and olderducklings. Some within-brood differences were as high as 20-30% of mean body mass.Tests to assesssources of within-broodvariation (age, sex, and season)in body mass for Canvasbackswere inconclusive.Variation within broodswas generallyless than that amongbroods for bothCanvasbacks and Redheads.The lackof differencesin ducklingbody massbetween single- and mixed-speciesbroods for any ageclass, sex, or speciessuggests that masswas not affectedby interspecificbrood parasitism. Key words: Aythya americana;Aythya vahsineria;duckling; brood: body mass; Can- vasback;Redhead.

INTRODUCTION data from captive birds (e.g., Oring 1968, Rhy- Body mass strongly influences duckling survival. mer 1988b, Lightbody and Ankney 1984). Vari- Heavier ducklings generally are better able to ation in body massamong captive ducklingstends withstand periods of food deprivation (Rhymer to be small (Rhymer 1988b), probably because 1988b) and cold weather (Koskimies and Lahti of uniform environmental conditions and access 1964, Kear 1965, Samuel and Goldberg in press) to ad libitum food. Information on variability in and are less susceptible to predators in the first body mass under natural conditions is sparse, 10 days (Swennen 1989) than lighter ducklings. however, and few researchershave assessedthis Factors affectingbody mass and growth of duck- variability within or among broods. Low vari- lings include genetics(Prince et al. 1970, Rhymer ability in body mass within broods may be ex- 1988a), eggmass (Rhymer 1988b, Holmberg and pected becausebrood mates are genetically sim- Klint 1991) food availability (Street 1978, Hun- ilar and have encountered similar environmental ter et al. 1984, Rattner et al. 1987), cold stress conditions and foraging opportunities. However, (Samuel and Goldberg in press),disease (Samuel two studies reporting within-brood variability and Goldberg in press), and contaminants found sizable differencesin body masses.Dzubin (Swennen 1991, Cain and Pafford 198 1, Hoff- (1959) reported within-brood differences of 5 g man et al. 1992). The effect of egg mass dimin- (11% of mean body mass) at hatch, 16 g (17%) ishes during the fist l-3 weeks (Holmberg and in nine-day-old, and 120 g (12%) in 50-day-old Klint 199 1, Rhymer 1988b). Sexual differences wild Canvasback (Aythya valisineria) ducklings. in body mass usually do not appear until near In wild Gadwall (Anus strepera), Mallard (Anus fledging (Greenwood 1974, Lightbody and An- platyrhynchos), and Blue-winged Teal (Anus dis- kney 1984, Lokemoen et al. 1990). cm-s) ducklings 19-4 1 days old, differences in Most studies reporting body mass, growth body mass within broods ranged as high as 140 curves, or measurements of ducklings have used g for females and up to 95 g for males (Lokemoen et al. 1990). Canvasback nests and broods in the prairie * Received28 January1994. Accepted 18 July 1994. pothole region are often parasitized by Redheads 19091 910 JANE E. AUSTIN AND JEROME R. SERIE

(Aythyu americana) (Stoudt 1982, Serie et al. using Pesola spring scales,the accuracy of which 1992). The effect of interspecific brood parasit- was regularly checked with known weights. Age ism on duckling growth and body mass is un- of each duckling was determined in hand using known, although some researchershave assessed plumage classesdescribed by Gollop and Mar- its effect on survival to fledging. Canvasbacks shall (1954). Most captured ducklings were 20- reared in broods parasitized by Redheads had 50 days old (Classes IC-IIC). Some ducklings lower survival rates than Canvasbacks reared in were classifiedonly as Class III; becausewe were unparasitized broods (Leonard 1990). In con- uncertain if they were Class IIIA or later, we trast, survival of Common Goldeneye (Buce- separately delineated them as Class IIID. Statis- phala clangula) ducklings was not affected by tical comparisons of body mass among and with- interspecific parasitism (Eadie and Lumsden in broods were made only with data from broods 1985). which would be readily identified from others on Data collected from Canvasback and Redhead the same pond. Within each identified brood, we ducklings during banding operations in the 1970s determined mean and range of mass differences provided an opportunity to assessbody mass of by comparing each possible pair of ducklings 20- to 50-day-old wild ducklings. BecauseCan- within the brood. vasback nests were commonly parasitized by We used analysis of variance (ANOVA) to as- Redheads(Serie et al. 1992) data included broods sessthe effectsof year, age class, sex, and season of both single and mixed (containing Canvasback on mean duckling mass for Canvasbacks. Season and Redhead ducklings) species.My objectives refers to early- or late-hatching ducklings, where were to (1) assessvariability of duckling mass early-hatched ducklings are those hatched from within and among broods of Canvasbacks and nests initiated during the first 15 days of known Redheads, (2) evaluate factors influencing duck- Canvasback nest initiation each year (Serie et al. ling mass, and (3) determine whether ducklings 1992). The design was a split-plot; broods within raised in a single-speciesbrood are heavier than year-by-season-by-age class combinations were ducklings in mixed-species broods. the whole plots, sex was the subplot level, and individual ducklings were the subsampling units. METHODS Mean separations following significant effects(P During 1974-1978, more than 1,100 Canvas- 5 0.05) were done using Fisher’s protected LSD back and Redhead ducklings were weighed in values at the 0.05 level of significance (Milliken conjunction with a study of the breeding biology and Johnson 1984). of Canvasbacks(Serie et al. 1992). The work was We assessedfactors affecting variances of mass conducted in the prairie pothole region near within Canvasback broods using a means model Minnedosa, in southwestern Manitoba, Canada approach (Milliken and Johnson 1984) in an (5O”l SN, 99”5O’W). Wetlands, topography, veg- ANOVA. We first analyzed all testable hypoth- etation, and densities of breeding waterfowl in esesfor yearly, seasonal, age class, and sex dif- the area have been describedby Kiel et al. (1972) ferences simultaneously in an unbalanced split- Adams and Gentle (1978) and Stoudt (1982). plot design. Broods served as the whole plot and Careful observations and notes made over sev- sex as the subplot. We then tested individual eral days before and immediately during the cap- hypotheses(contrasts) using single degreeof free- ture operation on each pond were used to locate dom F-tests. broods and determine brood age (Gollop and We used the following ANOVA model to es- Marshall 1954) size, and speciescomposition. timate variance components attributable to Usually only a single brood occurred on a pond; within versus among broods and to compare be- where several broods occurred on a pond, they tween pure and mixed-species broods (brood usually could be distinguished by their different type) for each species: ages. In a number of cases,the brood hen was marked and the hatch date of the brood was known. Ducklings were caught in drive traps where (Cowan and Hatter 1952, Trauger 197 1) in late July and August. Ducklings were weighed im- Y,, = weight of duckling k in brood j of mediately after capture, as soon as the down was type i dry. Mass was determined to the nearest gram Jo,= mean mass of ducklings of type i, VARIATION IN BODY MASS OF CANVASBACKS AND REDHEADS 911

TABLE 1. Mean differences in body mass (g) of Canvasback and Redhead ducklings within broods by age c1ass.a

Difference mass (g) Number of Number of in body sex Age class ducklings broods nb + SD Range

Canvasbacks: F IIA 54 24 56 41.2 26.9 O-120 IIB 111 47 114 49.9 42.2 O-230 IIC 86 41 78 50.7 53.9 O-290 IIIA 9 10 74.2 46.8 10-170 IIID 16 ; 12 45.8 40.2 O-120 M IIA 55 26 48 37.0 35.6 O-165 IIB 102 55 77 39.0 34.5 O-145 IIC 104 47 116 49.0 45.9 O-225 IIIA 14 8 7 49.3 34.0 lo-95 IIID 22 8 21 63.2 86.8 O-325 F vs. M IIA 98 25 103 51.5 43.4 O-250 IIB 218 52 243 62.1 46.8 O-230 IIC 153 37 192 77.1 54.2 O-290 IIIA 60 6 24 90.0 64.1 5-245 IIID 13 7 43 107.1 62.4 O-395 Redheads: F IIA 36 23 51.5 40.0 O-l 10 IIB 48 29 33.6 44.0 O-180 IIC 30 20 17 46.5 36.0 5-115 IIIA 6 4 2 37.5 31.8 15-60 IIID 6 4 2 52.5 46.0 20-85 M IIA 43 25 42.8 32.0 O-130 IIB 53 29 :: 50.0 36.6 O-145 IID 19 13 7 35.0 48.8 10-145 IIIA IIID 10 7 4 15.0 10.8 5-30 F vs. M IIA 61 16 67 46.9 37.5 O-135 IIB 17 53.0 58.0 O-315 IIC :; 11 :: 57.5 31.5 O-120 IIIA 9 3 4 37.5 18.5 25-65 IIID 5 2 3 48.3 50.8 lo-105

=Where broods contained ducklings of > I age class, the data were summarized for the dominant age class of that sex. b Number of all possible comparisons withii brood. r DuckliingsinClass IIID could only be aged to Class III.

b,, = random variable for brood j of type differences tended to be highest in Class IIA i, and ducklings (6-7% in Canvasbacks,9-l 1% in Red- d,cico, = random variable for duckling k in heads) and generally declined to 4-6% in Class brood j of type i IIC and older ducklings. Assessmentof factorsaffecting mean body mass The variance of Yi,k = vb2 + vd*, where vb2 is the in Canvasback ducklings was complicated by the variance among broods and vd2 is the variance lack of data in Class IIC and in 1976. By ex- within broods. Unbiased estimators of mean mass cluding age classIIC and 1976 in the first ANO- of ducklings in pure and mixed broods are given VA, we were able to balance the design to assess as least squares means (LSMEANS). All statis- effectsof age class, sex, year, and season.Of the tical testswere conducted using SAS (SAS Insti- 4 factors, age class (I; = 38.39; df = 1, 69; P = tute, Inc. 1989). O.OOl)andsex(F= 14.89;df= 1,46; P= 0.0004) were significant. All other main effects(year and RESULTS season)and interactions were not significant. Be- Most mean differences in body mass within cause season did not seem to account for vari- broods were around 35-50 g (Table 1). When ation in mass, we conducted a second ANOVA expressedas a percentage of mean body mass, including Class IIC and 1976 and ignoring sea- 912 JANE E. AUSTIN AND JEROME R. SERIE

TABLE 2. Mean body mass (g) of Canvasback duck- TABLE 3. Variance in body mass of Canvasback and lings, expressed as LSMEANS, by age class and sex. Redhead ducklings among and within broods, by age class. Percent contribution to total variance is in pa- Number rentheses. Age N??ber of class broods ducklings Sex P SE

IIA 33 61 5 549SA 14.81 31 62 562.2A 18.05 IIB 56 129 F 691.9A 10.86 62 113 M 739.1B 12.20 Canvasback IIC 48 96 F 863.5A 10.39 F IIA 54 21 3,670 (75) 1,212 (25) 52 114 M 930.OB 12.35 IIB 111 36 6,432 (63) 3,184 (37) IIC 86 28 2,285 (42) 3,166 (58) * Meanswithin ageclass followed by a commonletter are not signili- cantly different(P > 0.05) using Fisher’s protectedLSD value. IIIA 9 ; -90(O)’ 3,738 (100) IIID 16 6,111 (79) 1,588 (21) M IIA 6,137 (82) 1,391 (18) son. Effectsof age class(F = 168.75; df = 2, 16 1; 55 22 IIB 102 39 6,051 (63) 3,502 (37) P = O.OOOl), sex (F = 56.75; df = 1, 97; P = IIC 104 32 5,593 (59) 3,845 (41) O.OOOl),and sex-by-age class (F = 3.75; df = 2, IIIA :“2 108 3,693 (69) 1,655 (31) 97; P = 0.0271) were significant. Within age IIID 794 (12) 5,953 (88) classes,mean mass of male and female Canvas- Redhead back ducklings did not differ in Class IIA but did F IIA 36 18 3,912 (65) 2,115 (35) differ in older age classes(Table 2). IIB 48 24 3,496 (57) 2,590 (43) Analysis of factors affecting variance of body IIC 30 18 2,668 (59) 1,825 (41) 4 - mass within Canvasback broods was inconclu- IIIA IIID : 4 -66; (0) 1,906 (100) sive. Although the overall test of no difference M IIA 43 19 3,596 (47) 4,023 (53) was significant (F = 10.08; df = 34, 25; P = IIB 53 21 4,863 (60) 3,231 (40) 0.000l), when the main effects(year, season,age IIC 19 13 4,733 (71) 1,890 (29) class, and sex) were analyzed separately, no ef- IIIA 3 3 - fects or interactions were significant (P > 0.35). IIID 10 7 3,194 (94) 189 (6) Variance within broods was less than that * variance componentnot differentfrom 0. among broods for both Canvasback and Red- head ducklings, except for Class IIC Canvasback females and Class IIA Redhead males (Table 3). Comparison of variability among broods may Variance was highest among broods for Class be confounded by the use of plumage develop- IIIA female and IIID male Canvasbacksand IIID ment for ageing, especially for ducklings around female Redheads, but sample sizes were small. 2-3 weeks old. Lokemoen et al. (1990) suspected We found no significant differences in body the high variation in body massof Mallards, Blue- mass between single- or mixed-species broods winged Teal, and Gadwall within the same plum- for any age class, sex, or species(Tables 4 and age classwas due in part to inaccuracy of ageing 5). birds from plumage. Schneider (1965) found er- rors of more than two weeks within and among DISCUSSION broods if age at which teleoptiles first emerge (2- Differences in body mass within broods in this 3 weeks of age; Southwick 1953) was used for study were similar to those reported for wild age classification. Once feathering began, the se- Canvasbacks (Dzubin 1959) and wild Mallards quence and timing of emergence seemed to be and Gadwalls (Lokemoen et al. 1990). As ex- constant. Retarded growth or poor body condi- pected, within-brood variation in body masswas tion may be reflected in delayed plumage devel- generally lessthan that among broods. However, opment. We were able to assessaccuracy of age- in some groups, the difference between within- ing in this study by comparing 49 ducklings from and among-brood variation was very small. Some 12 broods of known hatching date with age class- of these exceptions may have been due to intra- es as outlined by Dzubin (1959). Of these, 14% specific parasitism, which can be extensive in had been misclassified:one was estimated as one both species (Sorensen 1990). Tests to assess week younger, five as one week older, and one sourcesof within-brood variation in body mass as two weeks older than actual age. for Canvasbacks, however, were inconclusive. Age classand sex were the only factors affect- VARIATION IN BODY MASS OF CANVASBACKS AND REDHEADS 913

TABLE 4. Mean body mass (expressed as least squares TABLE 5. Mean body mass (expressed at least squares means) of Canvasback ducklings by age, ’ sex, and brood means) of Redhead ducklings by age, sex, and brood type (mixed vs. pure species). type (mixed vs. pure species).

Mean body mass(g) Mean body mass@) Age Ase Sex class Purebrood (n)” Mixed brood (n) F Sex class Pure brood (nr Mixed brood (n)” P

F IIA 516 (1918) 584 (35/15) 0.056 F IIA 415 (13/6) 480 (23/15) 0.094 IIB 694 (46/17) 705 (65125) 0.646 IIB 682 (22/8) 630 (26/22) 0.081 IIC 890 (40/14) 852 (46120) 0.133 IIC 797 (6/2) 741 (24/18) 0.159 IIIA 957 (6/2) 928 (3/3) 0.591 IIIA 867 (6/4) IIIDb 976 (7/4) 995 (9/5) 0.776 IIIDd 822 (3/2) 8 14 (3/2) 0.782 M IIA 516 (19/9) 576 (36/15) 0.155 M IIA 461 (2318) 489 (20/17) 0.441 IIB 739 (54/19) 739 (48128) 0.996 IIB 708 (21/7) 660 (32/22) 0.229 IIC 922 (66/21) 919 (38/17) 0.939 IIC 836 (5/2) 784 (14/l 1) 0.309 IIIA 1,068 (4/3) 1,019 (1015) 0.413 IIIA 849 IIID 1,077 (10/4) 1,124 (12/6) 0.289 IIID 885 (2/2) 9 15 (815) 0.606 * Ageclasses as definedby Gallop and Marshall (I 954). * Ageclasses as definedby Gallop and Marshall (1954). b Samplesize = (no. canvasbackducklings/no. broods). b Samplesize = (no. redheadducklings/no. broods) cHo: + = ~md. ‘H,:~=,L,_+ d Classdiedonly as ClassIII. d Class~Eedonly as ClassIII.

ing body mass in Canvasbacks. Sexual differ- in this study (20-30% of mean body mass) could ences in mass became significant after Class IIA affectsurvival, particularly during periods of cold (about four weeks of age), somewhat earlier than stress. Unfortunately, researchers investigating reported for captive Canvasbacks and Redheads duckling thermoregulation have assessedcold (Lightbody and Ankney 1984, Lightbody 1985). stressonly during the first l-2 weeks after hatch Canvasbacks were similar to Blue-winged Teal (Untergasserand Hayward 1972, Rhymer 1988b); (Dane 1965) but differed from Mallards (Rhymer little information is available about the capabil- 1988b) in the lack of a seasonaleffect on duckling ities of older ducklings relative to body mass. body mass. The nesting period of Mallards is The lack of differencesin body mass between longer than that of Canvasbacks or Blue-winged ducklings in pure or mixed broods suggestedthat Teal and provides greater opportunities for re- masswas not affectedby interspecific brood par- nesting and any associated changes in egg or asitism. However, we do not know whether body duckling mass. mass differed relative to brood type in one- to The lack of a year effect on duckling mass sug- two-week-old ducklings or whether such differ- gestshabitat quality and food availability were ences affected early survival. Leonard (1990) not limiting, despite differencesin water condi- found that lower survival of Canvasback duck- tions among years(Serie et al. 1992). Canvasback lings in mixed-species broods occurred during broods moved extensively among ponds of var- the first week after hatch, survival did not differ ious sizesand types (Austin and Serie 199 1). The between pure- and mixed-species broods in 2-9 mobility of the broods allows broods to adapt to weeks after hatch, which coincides with the ages changing food availability and thus minimize the of ducklings in this study. effectof food availability on growth and fledging. Although ducklings within a brood encounter In studiesusing captive ducklings, the greatest the same environmental conditions and foraging variability in duckling growth rates and body opportunities, they can differ by as much as 34% mass occurred at about 2-6 weeks after hatch, of average body mass even within sexes. Such the period of most rapid growth (Schneider 1965, large differencescannot entirely be attributed to Prince et al. 1970, Brown and Fredrickson 1983). sex or size at hatch becausethese differencesde- Brown and Fredrickson (1983) suggestedthat the cline throughout development to nonsignificant high variability of body mass during this period levels after 3-4 weeks (Rhymer 1988b, Holm- may be related to high energy and nutrient de- berg and Klint 1991). Other factors that may mands and that stress would have its greatest contribute to body mass variation within broods effect on gains in body mass, and possibly sur- include behavior, injuries or disease, parasite vival, during this period. Certainly, one would loads, and different parentage (intraspecific par- expect that differencesof the magnitude reported asitism [Sorensen 19901). If ducklings differ, for 914 JANE E. AUSTIN AND JEROME R. SERIE example, in their ability to learn various feeding EADIE, J. M., AND H. G. LUMSDEN. 1985. Is nest skills, they may develop differencesin their for- parasitism always deleteriousto goldeneyes?Am. Nat. 126:859-866. aging efficiency. GOLUIP,J. B., ANDW. H. MARS=. 1954. A guide Few data are available to assesswithin-brood for aaina duck broods in the field. Miss. Flvwav< - variability in body massduring the first two weeks Cot&l-Tech. Sec. Rep. (Dzubin 1959), when duckling mortality is great- GREENWOOD,R. J. 1974. Reproductive aspects, est (Sargeantand Raveling 1992). The influences growth, and development of Greenland Mallards. Condor 76~223-225. of low body mass in this early period on body HOFFMAN,D. J., C. J. SANDERSON,L. J. LECAPTAIN, mass and survival later in development are un- E. CROMARTIE,AND G. W. PENDLETON.1992. In- certain. Swennen (1989) found that during the teractive effectsof arsenate,selenium, and dietary first 10 days after hatch lighter ducklings were protein on survival, growth, and physiology in Mallard ducklings.Arch. Environ. Contam. Tox- less alert and reacted more slowly to alarm calls icol. 22155-62. than heavier ducklings; these lighter ducklings HOLMBERG,K., AND T. KLINT. 1991. Growth in re- were most likely to be taken by gulls. Lighter lation to eggweight and sex in Mallard ducklings, ducklings in the first l-2 weeks after hatch are Chapter 5. In K. Holmberg [ed.], Mallard ducks: mate choiceand breedingsuccess. Ph.D.diss., Univ. more susceptible to cold stress than heavier of Stockholm. ducklings (Koskimies and Lahti 1964, Unter- HUNTER, M. L., JR., J. W. WITHAM, AND H. Dow. gasserand Hayward 1972, Rhymer 1988b). 1984. Effectsof a carbaryl-induceddepression on Insight into the factors influencing body mass the growth and behavior of American Black Ducks of ducklings and, in turn the influence of body and Mallard ducklings.Can. J. Zool. 62:452456. KEAR,J. 1965. The internal food reservesof hatching mass on survival, would be valuable in our ef- Mallard ducklings.J. Wildl. Manage. 29:523-528. forts to provide brood habitat and enhance wild KIEL, W. H., JR., A. S. HAWKINS,AND N. G. PERRET. duck production. A combination of experimen- 1972. Waterfowl habitat trends in the aspenpark- tal studies of ducklings in captivity and in the land of Manitoba. Can. Wildl. Serv. Reo. Ser. 18. Kos~rmrnsJ., AND L. LAHTL 1964. ColdIhardiness wild is needed to addressthese questions. of the newly-hatched young in relation to ecology ACKNOWLEDGMENTS and distribution in ten speciesof Europeanducks. Auk 8 1:28l-307. C. E. Korschgen,J. T. Lokemoen,P. J. Pietz, G. A. LEONARD,J. P. 1990. Survival and movements of Baldassarre, and an anonymous reviewer provided Canvasbackducklings-impact of brood density. valuable comments on earlier drafts. R. Khan-Mahlik M.Sc.Thesis. Michigan State Univ.. Lansine. MI. and W. Newton provided statistical analysesand re- LIGHTBODY,J. P: 1985: Growth rates’and de&lop- view. ment of Redhead ducklings.Wilson Bull. 97:554- 559. LITERATURE CITED LIGHTBODY,J. P., ANDC. D. ANKNEX. 1984. Seasonal influence on the strategiesof growth and devel- ADAMS,G. D., AND G. C. GENTLE. 1978. Spatial opment of Canvasback and Lesser Scaup duck- changesin waterfowl habitats. Can. Wildl. Serv. lings. Auk 101:121-133. Occas. Pap. 38. L~KEMOEN,J. T., D. H. JOHNSON,AND D. E. SHARP. AUSTIN,J. E., ANDJ. R. Santa. 1991. Habitat useand 1990. Weights of wild Mallard Anus pkztyrhyn- movements of Canvasbackbroods in southwest- chos,Gadwall Anas strepera.and Blue-wingedTeal em Manitoba. Prairie Nat. 23:223-228. Anas discorsduring the breeding season.Wildfowl BROWN,P. W., ANDL. H. FREDRICKSON.1983. Growth 41:122-130. and molt progression of White-winged Scotor MILLIKEN,G. A., ANDD. E. JOHNSON.1984. Analysis ducklings.Wildfowl 34: 115-l 19. of messy data. Vol. I: designedexperiments. Van CAIN, B. W., AND E. A. PAFFORD. 1981. Effect of Nostrand Reinhold. New York. dietary nickel on survival and growth of Mallard ORING,L. W. 1968. Growth, molts, and plumagesof ducklings. Arch. Environ. Contam. Toxicol. 10: the gadwall. Auk 85:355-380. 737-745. PRINCE,H. H., P. B. SIEGEL,AND G. W. CORNWELL. COWAN,I. M., AND J. HATTER. 1952. A trap and 1970. Inheritance of eggproduction and juvenile technioue for the capture of diving waterfowl. J. growth in Mallards. Auk 87:342-352. Wildl. Manage. 16:43844 1. - RATTNER,B. A., G. M. HARAMIS,D. S. CHU, C. M. DANE. C. W. 1965. The influence of see on devel- BUNCK,AND C. G. SCONES.1987. Growth and opment and reproductive capabilityYdf the Blue- physiological condition of ducklings reared on winged Teal (Anus discorsLinnaeus). Ph.D.diss., acidified wetlands. Can. J. Zool. 65:2953-2958. Purdue Univ., Lafayette, IN. RHYMER,J. M. 1988a. Geographic variation and ge- Dzuar~, A. 1959. Growth and plumagedevelopment notype-environment interactions in avian growth of wild-trapped juvenile Canvasback(Aythyu vul- and develooment. Ph.D.diss.. Florida State Univ.. isineria). J. Wildl. Manage. 23:279-290. Tallahasee:FL. VARIATION IN BODY MASS OF CANVASBACKS AND REDHEADS 915

RHYMER,J. M. 198813. The effect of egg size vari- head and Canvasback ducks. Ph.D. diss., Univ. ability on thermoregulation of Mallard (Anus of Minnesota. plutyrhynchos)offspring and its implications for SOUTHWICK,C. 1953. A system of age classification survival. Oecologia (Berlin) 75:20-24. for field studies of waterfowl broods. J. Wildl. SAMUEL,M. D., ANDD. R. GOLDBERG.In press. My- Manage. 17:l-8. coplasma anatis and cold stress on hatching of STOUDT,J-H. 1982. Habitat use and productivity of Mallard ducklinas.J. Wild. Dis. Canvasbacksin southwesternManitoba. U.S. Fish SARGEANT,A. B., .&& D. G. RAVELING. 1992. Mor- Wildl. Serv. Spec. Sci. Rep. No. 248. tality during the breeding season,p. 396-422. In STREET,M. 1978. The role of insects in the diet of B. D. J. Batt. A. D. Afton. M. G. Anderson. C. D. Mallard ducklings-an experimental approach. Ankney, D. H. Johnson, J. A. Kadlec, and G. L. Wildfowl 29:93-100. Krapu [eds.], Ecology and management of water- SWENNEN,C. 1989. Gull predation upon eider So- fowl. Univ. of Minnesota Press. Minneanolis. materia mollissimaducklings: destruction or elim- SAS INSTITUTE,INC. 1989. SASSTAT’s user’s guide, ination of the unfit? Ardea 77r21-45. version 6, 4th ed. Cary, NC. S~ENNEN,C. 1991. Fledgling production of eiders SCHNEIDER,K. B. 1965. Growth and plumage de- Sometaria mollissima in the Netherlands. J. Or- velopment of ducklings in interior Alaska. M.S. nithol. 62~427-437. thesis, Univ. of Alaska, Fairbanks. TRAUGER,D. L. 1971. Population ecology of Lesser SERIE.J. R.. D. L. TRAUGER.AND J. E. AUSTIN. 1992. Scaup(Aythya u&zis) in subarctictaiga. Ph.D. diss., Inkuence of age and selected environmental fac- Iowa State Univ., Ames, IA. tors on reproductiveperformance of Canvasbacks. UNTERGASSER,G., AND J. S. HAYWARD. 1972. De- J. Wildl. Manage. 56:546-556. velopment of thermoregulationin ducklings.Can. SORENSEN,M. D. 1990. Parasitic egg laying in Red- J. Zool. 50: 1243-1250.