J. Field Ornithol., 54(2):152-159

ANALYSIS OF PLUMAGE VARIATION IN THE CANADA WARBLER

BY JOHN H. RAPPOLE The plumagesof the CanadaWarbler (Wilsoniacanadensis) have been describedin detailby Oberholser(1974), Dwight(1900), Roberts(1936), and others.Although Oberholser(1974:793) presentsinformation on 10 different plumagesin the , the Canada Warbler is generally consideredto be dimorphicwith "adult male" and "female"type plum- ages(e.g., Verner and Willson 1969, Rohwer et al. 1980). The male plumageis gray aboveand yellowbelow with a seriesof black streaks acrossthe breast.The femaleplumage is similarexcept the breaststreak- ing is gray (see Peterson1980:234). The speciesis consideredto be monotypicthroughout its geographicrange (A.O.U. 1957:518). As part of a study of transient migrants, I examined a number of Canada Warblers and discovered considerable variation within the ma- jor plumagetypes. The purposeof this paper is to describe,categorize, and evaluatethe significanceof plumagevariation in the CanadaWar- bler.

METHODS Field work for this studywas done at the Welder Wildlife Foundation, 48 km N of CorpusChristi, Texas, from 15 Aug.-23 Oct. 1973;4 Mar.- 24 May 1974; 20 Aug.-13 Nov. 1974; 29 Mar.-27 May 1975 (Rappole 1978) and in the Tuxtla Mountains of southern Veracruz, Mexico from Aug. 1973-May 1975 (Rappoleet al. 1979). were capturedin mistnets (12 m x 2.6 m x 30 mm mesh)and each was examinedfor plumageaberrancies. Age in fall birds wasde- termined by degreeof skullpneumatization following a method similar to that of Schneider (1981) except that ethanol was used for wetting skull feathers. Wing chord measurementswere taken as describedby Oberholser (1974) to the nearest millimeter. I captured, banded, and released 634 birds and collected 87 birds. I examinedan additional 54 specimensof known age and sex from the Bell Museum of Natural History, Universityof Minnesotaand the University of Museum of Natural History. The total specimen samplewas divided into 6 categorieson the followingbasis: (1) testes reportedpresent, skull pneumatized, captured Aug.-Oct. = adult male, Basicplumage; (2) testesreported present,skull not completelypneu- matized,captured Aug.-Oct. = immaturemale, Basic plumage; (3) testes reportedpresent, skull pneumatized, captured Mar.-Jun. = adult male, Alternate plumage;(4) ovary reported present,skull completelypneu- matized,captured Aug.-Oct. = adult female, Basicplumage; (5) ovary reported present, skull not completelypneumatized, captured Aug.- Oct. = immature female, Basic plumage; (6) ovary reported present, skull completelypneumatized, captured Mar.-Jun. = adult female, Al- ternate plumage.

152 Vol.54, No. 2 CanadaWarbler Plumages [ 153

T^BLE1. Descriptionand assignedscores for colorcharacters used in analysisof plumage variationin individualCanada Warblers of different age, sex,and plumagestatus.

Character and Score Description

Crown 0 Yellowish-greenor yellowishbrown. 1 Gray tinged with yellowish-brownor green. Gray tinged with yellowish-brownor green, someblack flecks. Gray fleckedwith black. Black edged with gray. Forehead 0 Yellowishor yellowish-green. 1 Yellowish-greentinge on gray. Gray or yellowish-greenflecked with black. Gray fleckedwith black. Black.

Side of Neck 0 No defined spotsalong border betweengray nape and yellowthroat. 1 Defined grayishspots along border betweengray nape and yellowthroat. 2 Definedblack spots along border betweengray nape and yellowthroat. 3 Blackborder betweengray nape and yellowthroat. Cheek 0 Mostly yellowish. 1 Gray tinged with yellow. 2 Gray. 3 Black and gray. 4 Black.

Breast 0 No gray or black on breast. 1 Someill-defined grayish streaks. 2 Distinctgrayish or grayish-blackstreaks. 3 Grayishor blackishstreaks with more or lesswell-defined black spots. 4 Black streaks.

Wing chord measurementsof the specimensin the different age, sex, and plumagecategories were comparedstatistically. Plumage characters were analyzedusing an index type scoringmethod similar to that used by Short (1965) (Table 1).

RESULTS

Plumagescores of malesin both Basicand Alternateplumage individ- uals varied from a stronglymale-like score of 19 to a scoreof 6 well withinthe femalerange (Table 2). Femalesin Alternateplumage varied from a score of 3 to a male-like score of 13. The main differencesamong the four Basicplumages (Fig. 1) are 154] J.H. Rappole J.Field Ornithol. Spring 1983

T^i•L•. 2. Analysisof color and pattern variation in individualsof different sex,age, and plumagecategories in Canada Warblers.

Averagescore and range for individuals Basicplumage Alternateplumage Adult Adult ImmatureImmature Plumage Male Female male female male female character (65)' (18) (7) (10) (19) (22)

Crown 3.82 (2-4) a 1.8 (0--3) 2.9 (0-4) 1.4 (1-2) 0.7 (0--2) 0 (0) Forehead 3.9 (2-4) 1.4 (1-3) 2.7 (0-4) 1.0 (0--2) 0.6 (0--2) 0.1 (0--2) Side of neck 2.7 (2-3) 0.2 (0-2) 2.6 (1-3) 0.5 (0--1) 1.4 (0--2) 0.3 (0--3) Cheek 3.0 (2-4) 1.3 (0-3) 3.1 (2-4) 1.4 (1-2) 1.1 (1-2) 0.8 (0--2) Breast 3.8 (3-4) 1.8 (1-3) 3.9 (3-4) 2.2 (1-3) 3.0 (3) 1.3 (0--3) Total score 17.2 (11-19) 4 6.5 (3-13) 15.2 (6-19) 6.5 (4-9) 6.8 (4-10) 2.5 (0--9)

Samplesize. Averagescore. Rangeof scores. Averageand rangeof total scoresfor individualswithin eachplumage category.

• 4 'f

A B

Fmta•E 1. Plumagesof the CanadaWarbler. A--immature female, Basic;B•adult fe- male,Basic; C---immature male, Basic; D•adult male,Basic (note that a fifth plumage type, adult male, Alternate, is very similarto D but distinguishable---seeTable 3). Vol. 54, No. 2 Canada WarblerPlumage• [155

TABLE 3. Key to Canada Warblersin Basicplumage.

1. (a) Necklace(breast markings) of thick (1-2 mm) blackstreaks. Forehead black or blue-grayflecked with black.Skull completely pneumatized ...... adult male (Fig. 1D).• (b) Necklace not as above ...... 2. 2. (a) Necklaceof distinctblack or grayishspots. Forehead usually (84%) yellowish- green.Skull not completelypneumatized (Jul.-Nov.) .... immaturemale (Fig. 1C). (b) Necklaceof mostlygrayish streaks ...... 3. 3. (a) Skull completelypneumatized (Jul.-Nov.). Generally (90%) with someblack in necklace.Forehead grayish, sometimes tinged with yellow.2 ...... adult female (Fig. lB). (b) Skullnot completely pneumatized (Jul.-Nov.). Generally (77%) without black in necklace.Forehead yellowish-green (95%) ...... immaturefemale (Fig. 1A).

• Malesin Alternateplumage in contrastto thosein Basic,generally lack yellow edgings on grayand blackfeathers of crown,back, neck, and breastpresumably because of feather wear. 2 Femalesin Alternateplumage are indistinguishableto my eye from adult femalesin Basicplumage.

summarizedin the key in Table 3. Using this key 93% (54 of 58 birds) of the totalspecimen sample of birdsin Basicplumage would have been correctlyidentified to age and sex on the basisof plumagealone. Use of skullpneumatization information increased the accuracyto 56 of 58 birds (97%). Three individuals(.4%) of the total sampleof bandedbirds and spec- imens examined (775 individuals)had aberrant plumages.One imma- ture male had throat, breast, and forehead flecked with black and yel- low, and a gray rather than white crissum.One adult female in Basic plumagehad creamcolor in all areaswhich would normally be yellow. Areasof gray color and the pattern of necklaceand face are normal in this bird. One adult male in Alternate plumagehad pale yellowfrom necklaceto crissuminstead of normal bright yellow.Other areasof this bird's plumagewere normal in coloration. The wingsof malesaveraged 5-6% longerthan thoseof females,but there wasconsiderable overlap (Fig. 2). The wingsof 89% of fall males were 65 mm or longer, while all fall femalesin the specimensample had wings62 mm or lessin length.Wing lengthof 98% of springmales was 64 mm or more while a bird with a chord of 61 mm or less was a female for 92% (11 of 12) of the spring sample.Fall birds in Basic plumagetended to havelonger wings than springbirds of the samesex in Alternate plumage(1%). Remigeswere moltedonly during the pre- Basicmolt, so the springbirds had 7-8 monthsmore wear on the wings than fall birds. However, I wasunable to distinguishbirds hatchedthe previousyear from older birdsin Alternateplumage based on degree of wing wear (cf. Rohwer et al. 1980:420-421). 156] J.H. Rappole J.Field Ornithol. Sp•ing 1983

AM n=65 ALTERNATE PLUMAGE AF n:18

AM n=7

AF n:10 BASIC PLUMAGE

IAA n=19

IF J J n=22

58 60 62 64 66 68 70 WING CHOR D(mm) FIGURE2. Comparisonsof wing measurementsfor Canada Warblersin different age, sex, and plumage categories.(Horizontal line = mean, verticalline = range, rectan- gle = one standarddeviation.)

DISCUSSION The CanadaWarbler is polymorphicwith respectto plumagecolor- ation. This conclusionis suggestednot only by my data but by the de- scriptionof the different plumagesof this speciesin Oberholser(1974) and other meticulousworks where both the norm and the range of variation within the main plumagecategories are describedin detail. The Canada Warbler has five separablepost-Juvenal plumage types: Adult male--Basic, Adult male--Alternate, Adult female--Basic and Alternate, Male---first Basic, Femalesfirst Basic. However, there are severalvariants within eachmajor type whichresemble norms of other plumage types•males that resemblefemales (2.7% of specimensam- ple), immaturesthat resembleadults (2.4%), femalesthat resemblemales (3.5%), etc. There are no data with which I am familiar on other speciesfor comparisonof plumageanalysis results. However, cursoryexamination of specimensof other "dimorphic"species of (Verner and Willson 1969, Rohwer et al. 1980) for which I have good samplesizes from Texasand Mexicoindicates a comparabledegree of variationwith- Vol.54, No. 2 CanadaW(trbler Plu•nages [ 157 in the supposedlydistinct "male" and "female"plumage categories (Wil- son'sWarbler Wilsoniapusilla, Northern Parula Parula americana,Bay- breasted Warbler Dendroica castanea, Chestnut-sided Warbler Dendroica pensylvanica,Northern Oriole Icterusgalbula, Yellow-breasted Chat Icteria virens,Kentucky Warbler Oporornisformosus). Theseobservations indicate that plumagecategories, e.g., "adultmale, Basicplumage," are not discreteentities. There is variation within each categoryand overlapsbetween categories. The evolutionarysignificance of sexualdimorphism has been studied in detail and someworkers have examined the tendencyof youngmales in so•nespecies to look like females(Rohwer et al. 1980, Rohwer 1978). However,other plumagevariations, e.g., the tendencyfor somefemales to closelyresemble •nales, have not been thoroughly exa•nined even though knowledgeof thesephenomena is not new. Oberholser(1974: 791) describes females of the ( citrina) that havethe blackthroat andjugulum of the male-typeplumage. "Normal" female plumage in this speciesis a drab greenishabove and yellowish below (Peterson 1980:242). I proposethat the plumagepolymorphisms found in the CanadaWar- bler and other speciesmentioned above can be attributed to the bal- ancing effectsof selectionfactors that favor opposingdirections in the evolutionof plumagecoloration. These factorscan be divided into those favoring a "brighter" plumage (one in which there is a large degree of contrastbetween colors, making the bird more visible)rs. thosefavoring a "duller" plumage(less contrast). Factorsfavoring a "bright" plumageinclude: (1) intra-malecompeti- tion for mates and/or territories (Wallace 1889, Rohwer et al. 1980:407), (2) epigamicselection (Darwin 1859:44, 1871:529,Hamilton and Barth 1962,Selander 1965, 1972,Lowther 1975, Fisher 1930), (3) intraspecific competitionfor limiting resources(Fretwell 1972:109-113, Ketterson and Nolan 1976, Schwartz1964, Rappoleand Warner 1976, 1980). "Dull" plumage may help to reduce aggressivenessof adult males toward the dull-plumagedbird. This would be advantageousfor adult femalesby reducingpairing time (Hamilton 1961), for unmated females entering breeding territories of mated birds, and for unmated young males(Rohwer et al. 1980). "Dull" plumagesalso reduce the likelihood of predation on the individual itself or on its eggsor nestlings(Darwin 1871:504). Someselection factors, e.g., speciesrecognition (Sibley 1957) may fa- vor neither a "bright" nor a "dull" plumage,merely one that is different from other species. Good data on molts and plumagesof birds like the Canada Warbler have been availablefor a number of years.Researchers like Ridgway, Brewster,Oberholser, and Dwight were well aware of the variety and sequenceof plumagesdemonstrated in different arian species.Yet, the possibleevolutionary significance of these different plumageshas not been analyzedfor the mostpart. Although male/femaledimorphism has 158] J.H. Rappole J.Field Ornithol. Spring 1983 stimulatedconsiderable interest, as have subadultmale plumages,most other plumagestrategies have not been closelyexamined (though see Morton 1976,Hamilton and Barth 1962).Plumage variants deserve con- siderablymore attentionas they reflect important aspectsof species' biologythat have been overlooked.

SUMMARY Museumspecimens of 141 CanadaWarblers of known sex,age, and plumage status,and 634 wild caught individualswere examined. Anal- ysisof the specimensample revealed 5 visuallydifferent plumagecat- egories:adult male--Basic, adult male--Alternate, adult female--Basic and Alternate, immature malesBasic, immature female--Basic. A vi- sual scoringmethod was used to characterizevariation within each of the plumage categories.The norm of each plumage type was easily separablefrom the norm of all other types.However there were variants within eachage--sex--plumage group whichresembled individuals from other groups: male-like females, female-like males, adult-like irama- tures. The major plumage patternsas well as the variantswithin each group can be explainedon the basisof balancingselection forces (e.g., intrasexualcompetition during the breedingseason, predation, epigam- ic selection,and intraspecificcompetition during the non-breedingsea- son).

ACKNOWLEDGMENTS I thank ElizabethC. Rappole,Christopher P. Barkan, Bruce A. Fall, Mario A. Ramos,Richard 3. Oehlenschlager,and Robert Zink for help with field work. I thank the Directors, Staff and Trustees of the Welder Wildlife Foundation, in particular Clarence Cottam, W. C. Glazener, Eric Bolen,and Gene Blacklockfor their sunportthroughout this study. Eugene S. Morton revieweda copy of the manuscriptand I am most grateful for his suggestions.Dr. Dwain W. Warner made specimens from the Bell Museum of Natural History, Univ. of Minnesota,available for examination.

LITERATURE CITED

AMERICAN ORNITHOLOGISTS' UNION. 1957. A.O.U. checklist of North American birds. 5th ed. Port City Press,Inc., Baltimore, Maryland. DARWIN,C. 1859. On the origin of speciesby meansof natural selection,or the pres- ervation of favoured races in the struggle for life. EncyclopediaBritanica, Great Booksof the Western World, Vol. 49, reprint, 1952, Chicago. ß 1871. The descentof man and selectionin relation to sex, First Edition. Ency- clopediaBritannica, Great Booksof the Western World, Vol. 49, reprint, 1952, Chi- cago. DWIGHT,J., JR. 1900. The sequenceof plumagesand moultsof the passerinebirds of . Ann. N.Y. Acad. Sci. 13:73-360. FISHER,R.A. 1930. The geneticaltheory of natural selection.Clarendon Press, Oxford. FRETWELL,S.A. 1972. Populationsin a seasonalenvironment. Princeton Univ. Press, Princeton, N.J. Vol.54, No. 2 CanadaWarbler Plumages [ 159

HAMILTON,T.H. 1961. On the functionsand causesof sexualdimorphism in breeding plumageof North Americanspecies of warblersand orioles.Am. Nat. 95:121-123. , ANDR. H. BIRTH,JR. 1962. The biologicalsignificance of seasonchange in male plumageappearance in someNew World migratorybird species.Am. Nat. 96:129- 144. KETTERSON,E. D., ANDV. NOLAN,JR. 1976. Geographicvariation and its climaticcor- relatesin the sexratio of easternwintering Dark-eyedJuncos (]unco hyemalis hyemalis). Ecology57:679-693. LOWTHER,P.E. 1975. Geographicand ecologicalvariation in the family Icteridae.Wilson Bull. 87:481-495. MORTON,E. 1976. The adaptive significanceof dull colorationin Yellow Warblers. Condor 78:423. OBERnOLSER,H.C. 1974. Bird life of Texas. Univ. Texas Press, Austin. PETERSON,R.T. 1980. A field guideto the birds.Houghton Mifflin Co., Boston,Mass. RIPPOLE,J. H. 1978. Seasonaldistribution notes on birdsfrom the Welder Refuge,San PatricioCounty, Texas. Bull. Texas Ornithol. Soc. 11:30-34. , M. A. RAMOS,R. J. OEHLENSCHLAGER,D. W. WARNER,AND C. P. BARKAN.1979. Timing of migrationand route selectionin North Americansongbirds. Pp. 199-214, in Proceedingsof the First Welder Wildlife FoundationSymposium. L. Drawe,ed., Welder Wildlife Foundation, Sinton, Texas. , ANDD. W. WARNER. 1976. Relationshipsbetween behavior, physiology, and weather in avian transientsat a migration stopoversite. Oecologia26:193-212. , AND--. 1980. Ecologicalaspects of arian migrant behaviorin Veracruz, Mexico.Pp. 353-393 in Migrant birds in the Neotropics:ecology, behavior, distri- bution, and conservation, A. Keast and E. S. Morton, eds., Smithson. Inst. Press, Washington,D.C. ROBERTS,T.S. 1936. The birds of Minnesota.Univ. MinnesotaPress, Minneapolis. ROHWER,S. 1978. Passerinesubadult plumages and the deceptiveacquisition of re- sources:test of a criticalassumption. Condor 80:173-179. , D. M. NILES,AND S. D. FRETWELL.1980. Delayedmaturation and the deceptive acquisitionof resources.Am. Nat. 115:400-437. SCHNEIDER,K.J. 1981. Age determinationby skull pneumatizationin the Field Sparrow. J. Field Ornithol. 52:57-59. SCHWIRTZ,P. 1964. The Northern Waterthrush in Venezuela.Living Bird 3:169-184. SELINr)ER,R. K. 1965. On mating systemsand sexualselection. Am. Nat. 99:129-141. ß 1972. Sexualselection and dimorphismin birds.Pp. 180-230 in Sexualselection and the descentof man. B. Campbell,ed., Aldine, Chicago. SHOR•r,L. L., JR. 1965. Hybridizationin the flickers(Colaptes) of North America.Bull. Am. Mus. Nat. Hist. 129, No. 4. SIBLEY,C.G. 1957. The evolutionaryand taxonomicsignificance of sexualdimorphism and hybridizationin birds. Condor 59:166-191. VERNER,J., INO M. F. WILLSON. 1969. Mating systems,sexual dimorphism, and the role of male North American passerinebirds in the nestingcycle. Ornithol. Monogr. 9. WILLICE,A.R. 1889. Darwinism:an expositionof the theoryof naturalselection with someof its applications.Macmillan, London. CaesarKleberg Wildlife Institute, Texas A&I University,Kingsville, Texas 78363. Received11 Feb. 1982; accepted22 Mar. 1983.