Eccentric First-Year Molt Patterns in Certain Tyrannid Flycatchers

ECCENTRIC FIRST-YEAR MOLT PATrERNS IN CERTAIN TYRANNID FLYCATCHE!.S

PETER PYLE, Point ReyesBird Observatory,4990 ShorelineHwy., StinsonBeach, California 94970

Most passerinesfollow a similarsequence of remex molt, replacementof the primariescommencing with the innermostand proceedingdistally, while that of the secondaries(except for the tertials, S7-S9) beginswith the outermostand proceeds proximally (Ginn and Melville 1983). Duringthe firstyear molts(here defined as all periodsof moltingbefore the secondprebasic molt),many passerines replace no remigeswhereas others replace some or all remiges(Pyle et al. 1987, Jenni and Winkler 1994). "Incomplete"first-year remexmolts occasionally result when the typicalmolting sequence is arrested, the distaljuvenal primaries and proximalsecondaries (excluding the tertials) being retained.This pattern of incompletemolt is found in severalNorth Americanpasserines, notably, the Phainopepla(Phainopepla nitens), North- ern Cardinal (Cardinalis cardinalis), and Lark Sparrow (Chondestes grammacus)(Pyle et al. 1987, Thompsonand Leu 1994). In a few passerinespecies, incomplete first-year primary molts (and in some casessecondary molts) that do not follow the typical sequencebut involvereplacement of centralor distalprimaries in varyingpatterns have been documented(Jenni and Winkler 1994, Thompson and Leu 1994). These replacementpatterns, termed "eccentric"by Jenni and Winkler (1994), are poorly understood.A common eccentricreplacement pattern involvesrenewal of the outer four to six primariesand inner three to five secondaries,feather replacementproceeding distally in the primariesand proximallyin the secondaries,as in the typicalsequence, but commencingat differentpoints along the wing (Jenni and Winkler 1994; Figure 1). This pattern has been documentedin severalNorth American passerines,notably, the LoggerheadShrike (Lanius ludovicianus),Yellow-breasted Chat (Icteria virens),and Passerina buntings (Miller 1928, Phillips1974, Thomp- son and Leu 1994). While examiningspecimens of North Americantyrannid flycatchers for molt-relatedage criteria, I found that first-yearbirds of severalspecies of differentgenera display eccentric remex-replacement patterns. Depending on the species,replacement of remigesoccurs during fall molting periods, spring molting periods, or both. Here I summarizethese patterns, and suggestseveral hypotheses for their occurrence.

MATERIALS AND METHODS

This studywas limited to birdscollected during their firstyear. Results were basedon datafrom 589 specimensof 22 north-temperatespecies examined at the CaliforniaAcademy of Sciences(CAS), Moore Laboratory of Zoology (MLZ), Museumof VertebrateZoology (MVZ), and WesternFoundation of VertebrateZoology (WFVZ). On each specimenI carefullyexamined the primaries,secondaries, and primary covertsfor feather-retentionpatterns

WesternBirds 29:29-35, 1998 29 ECCENTRIC FIRST-YEAR MOLT PATTERNS IN CERTAIN TYRANNID FLYCATCHERS

•..... ""'• 4 7 7 6 5 432 12 primaries secondaries

Figure1. Variationin the eccentricfirst prebasicmolt of the VermilionFlycatcher, fromnine remiges and no primarycoverts replaced (A) to allremiges and four primary covertsreplaced (C). Shadingdenotes replaced feathers. Note the differencein shape and wear betweenretained juvenal primary coverts(inner six feathersin C) and replacedadult primary coverts(outer four feathersin C). In many other tyrannid flycatchers,adults have primary covertsuniformly broad and fresh and are thus separatedfrom first-yearbirds, with entirelyor mostlynarrow, pointed, and worn coverts.See alsoPyle (1997). reflectingthe extent of previousmolts. Birdsthat had been undergoing activemolt when collected(as evidenced by growingor sheathedremiges) were excludedfrom analysesof extent,as extentcould not be determined from thesespecimens; however, I usedspecimens in molt for informationon the timingand sequence of featherreplacement. I examinedboth wings to assessmolt patterns,while all specificdata were recordedfrom the right wings. During this studyI discovered(as confirmed by replacementpatterns on specimenscollected during active primary molt) that most North American

3O ECCENTRIC FIRST-YEAR MOLT PA•FrERNS IN CERTAIN TYRANNID FLYCATCHERS tyrannidflycatchers retain mostor all of their juvenalprimary coverts until their secondprebasi½ molt and that juvenaland adult (definitive)primary covertscould be identifiedby differencesin shape,color pattern, and relative wear (Figure 1C; see also Pyle 1997). These differences,along with differencesin outerprimary shape, rectrix shape, feather wear, color of the wing bars, and color of the plumage (Pyle et al. 1987), allowedreliable separationof most first-yearfrom older birds through the second or definitiveprebasic molt. Even first-yearbirds with completelyreplaced remigesretain five or more inner primary coverts,allowing their accurate separationfrom adults. First-yearflycatchers show a varietyof moltingstrategies (e.g., Johnson 1963, 1974; Pyleet al. 1987). Remigescan be replacedduring the fall(July to November),the spring(March to May), or both (Dickeyand van Rossem 1938, Johnson1963, this study);thus, fall and winterbirds can showup to two generationsof remigesand springand summerbirds can showup to three generationsof remiges.On first-yearspecimens, I recordedeach remex and primary covertas either juvenal,"first fall" (feathersreplaced duringthe fall),or "firstspring" (feathers replaced during the spring).After the springmolting period, I distinguishedjuvenal, first-fall, and first-spring feathersby their relativewear, first-fallfeathers being fresher than juvenal feathersbut more worn than first-springfeathers.

RESULTS

Among first-yearflycatchers, eccentric replacement patterns were found in eight species(Table 1): the Olive-sidedFlycatcher (Contopus cooperi), Yellow-belliedFlycatcher (Empidonax fiaviventris), western Willow Fly- catcher(E. traillii brewsteri,adastus, and extirnus),Vermilion Flycatcher

'!'aisle! Numbers(Mean and Range) of ConsecutiveOuter Primaries, Inner Secondaries,and Outer PrimaryCoverts Replaced on the RightWing During EccentricFirst-Year Molts in TyrannidFlycatchers

Inner Outer Species n Primariesa Secondariesb PrimaryCoverts

Olive-sidedFlycatcher 28 7.8 (4-10) 6.9 (3-9) 0.9 (0-4) Yellow-belliedFlycatcher 12 7.3 (6-10) 7.3 (6-9) 0.0 (--) WesternWillow Flycatcher 37 8.1 (5-10) 7.2 (3-9) 0.0 (--) VermilionFlycatcher 82 9.0 (5-10) 8.2 (3-9) 1.3 (0-5) WesternTropical Kingbird 18 5.3 (5-6) 3.9 (3-6) 0.0 Cassin'sKingbird 14 1.7 (0-5) 3.7 (3-5) 0.0 (--) WesternKingbird 55 6.0 (5-7) 3.8 (3-6) 0.0 (--) Scissor-tailedFlycatcher 16 6.5 (4-7) 3.5 (3-5) 0.0 (--) aStartingpoints (primary number) for primarymolts can be calculatedas ten minusthe numberof feathersreplaced. bNotethat one or moretertials can be replacedtwice during the firstyear in kingbirds(see Table 2), but the totalsfor these speciesinclude each tertial once.

31 ECCENTRIC FIRST-YEAR MOLT PAl-fERNS IN CERTAIN TYRANNID FLYCATCHERS

(Pyrocephalusrubinus), western Tropical Kingbird (Tyrannus melancholi- cus occidentalis;other subspecieswere not examinedfor this study), Cassin's Kingbird (T. vociferans), Western Kingbird (T. verticalis), and Scissor-tailedFlycatcher (T. forficatus). Limitedspecimen evidence (n = 4) suggeststhat an eccentricreplacement pattern also occurs in first-yearGray Kingbirds(T. dorninicensis).Among other taxa of thesegenera, the West- ern (Contopussordidulus; n = 26) and Eastern(C. virens; n = 12) wood- pewees,eastern Willow (E. t. carnpestris/traillii;n = 6) and Alder (E. alno- rum; n = 22) flycatchers,and Eastern Kingbird (T. tyrannus; n = 12) replacedall remiges,whereas the GreaterPewee (C. pertinax; n = 15), the remainingErnpidonax flycatchers(virescens, n = 9; rninirnus, n = 18; harnrnondii, n = 28; oberholseri, n = 27; wrightii, n = 24; difficilis, n = 30; occidentalis, n = 16; and fulvifrons, n = 15), and the Thick-billed Kingbird(T. crassirostris;n = 16) replacedno remigesor primarycoverts, other than the tertials on some birds. The eccentricreplacement pattern is well illustratedin the Vermilion Flycatcher(Figure 1, Table1). The numberof remigesreplaced ranged from all secondariesand primaries(in 53.7% of 82 first-yearspecimens examined, identifiedby the retentionof juvenalprimary coverts)to nine of 19 feathers(five outer primaries and four inner secondaries;CAS 39678). The numberof outer primarycoverts replaced varied from none (23.1%) to five (2.4%). In all cases,replacement proceeded consecutively from the starting pointsto the outermostprimary and primary covertand to the innermost secondaryother than the tertials.The sequencewas confirmedwith eight specimenscollected during active molt. This pattern was also typical of firstyear remex molts in the other flycatcherspecies showing the eccentric pattern,except for mostCassin's Kingbirds (see below). All replacementof remigesby VermilionFlycatchers occurs in the late summerand fall; no springreplacement of flightfeathers was found. First-yearYellow-bellied and westernWillow flycatchersdiffer from the VermilionFlycatcher in that remexmolt occurs during the springrather than the fall (seealso Dickey and van Rossem1938, Mengel 1952, Johnson 1963); adultsof these forms replace flight feathersin the fall (Johnson 1963, P. Unitt pers. comm.). First-yearOlive-sided Flycatchers appear to have one protractedover-winter molt of flight feathers, commencingin September or October and concludingin March or April, althoughthis needsconfirmation as only three specimensI examinedwere collectedon the winter grounds.Among first-springOlive-sided Flycatcher specimens, 21.4% of 28 had replacedall remiges. In the kingbirds,remex molt beginsin the fall, is suspendedover winter, and resumesin the spring(Table 2). The outerprimary molt begins with one of P4, PS, or P6 (P7 in one Scissor-tailedFlycatcher), and the inner secondarymolt (excludingthe tertials,which typicallymolt before other secondaries)with one of S4, S5, or S6 (seeFigure 1 for remexnumbering). No specimenscollected in winter(November through February; see Table 2) were in active molt, confirmingthat remex replacementoccurs in stages rather than continuouslythrough the winter. Cassin'sKingbird differs from the other three speciesin that fewer remigesare replacedduring the first year (Table1), includingno primariesor secondaries(except tertials) in 36%

32 ECCENTRIC FIRST-YEAR MOLT PATTERNS IN CERTAIN TYRANNID FLYCATCHERS

'talkie 2 Numbers(Mean and Range)of RemigesReplaced During the First Fall and Spring Molts in Four Speciesof Kingbirds

Firstfall malt • Firstspring malt b

Outer Inner Outer Inner Species n primaries secondaries n primaries secondaries

TropicalKingbird 12 2.3 (1-5) 4.0 (3-6) 13 3.2 (2•,) 5.4 (4-7) Cassin'sKingbird 16 0.3 (0-1) 1.2 (0-3) 13 1.5 (0-5) 3.5 (3-5) WesternKingbird 27 1.6 (0-3) 2.9 (1•,) 45 4.9 (3-6) 3.5 (3-6) Scissor-tailed Flycatcher 21 1.9 (0-5) 3.0 (1-5) 13 4.7 (2-6) 3.1 (2-6) aBasedon specimenscollected from Novemberthrough February.Number of secondaries replacedduring the fall molt includesthe tertials,which are often replacedagain during the springmolt (seetext). Compare with the data presentedin Table 1. bBasedon specimenscollected from April through July. of 14 specimens,and that the molt is usually(in 77% of the nine specimens showingreplacement of primaries)arrested before the outermostprimaries are replaced.Thus, many first-spring Cassin's Kingbirds had replaced,e.g., only P6-P7 and S6 (CAS 46202) or P5-P7 and S5-S6 (MVZ 4299). The evidenceindicates that the replacementof primariesin spring continueswhere the fall molt is suspended.Numerous spring and summer kingbirdsshow three generationsof primaries,with the number and sequenceof first-fallfeathers (e.g. P5-P7 or P6-P7) being typical of fall replacementpatterns shown by mid-winterbirds. Among 27 Western Kingbirdscollected in their firstwinter, for example,the mean numberof primariesreplaced during fall moltswas 1.603 (_+0.907 [standarddevia- tion]),whereas the meannumber of first-fallprimaries detected on 45 spring andsummer birds was 1.589 (_+0.920). Thesesimilar figures (ANOVA, P = 0.824) supportthe premisethat no primariesreplaced during the fall molt are replacedagain during the springmolt. Among the secondaries,however, the tertialsappeared to be replaced duringthe fall molt and againduring the springmolt (seeTable 2). Among WesternKingbirds, only six of 45 springand summerbirds (mean 0.155 -+ 0.424) had retainedone or two first-falltertials, whereas 27 winter birdshad replaceda meanof 2.885 (-+0.824)juvenal tertials. This highlysignificant difference(ANOVA, P < 0.0001) in the numberof first-falltertials present in winterversus spring indicates that the tertialstypically replaced in the fall are replacedagain in the spring.The other three speciesof kingbirdsshow similarpatterns of fall and springreplacement of remiges(Table 2).

DISCUSSION

In severalrespects eccentric replacement patterns in tyrannidflycatchers differ from those documentedin other North Americanpasserines. The highly variable commencementpoints of this molt in the Vermilion and otherflycatchers, the suspendedeccentric pattern of the kingbirds,and the

33 ECCENTRIC FIRST-YEAR MOLT PATTERNS IN CERTAIN TYRANNID FLYCATCHERS suspendedand arrestedeccentric pattern of most Cassin'sKingbirds are strategiesthat have rarely been documented,even in Europe (Jenniand Winkler 1994). It hasbeen suggested that eccentricmolts occur in speciesthat are more exposedto brightsunlight or that sufferhigher feather wear due to harsh vegetation,the renewal of outer primariesand inner secondariesbeing neededfor protectionof underlyingfeathers and improvedflight (Dwight 1900, Mester and Prgnte 1982, Jenni and Winkler 1994). Willoughby (1991) suggestedthat vegetationmight be moreimportant than exposure to sunlightin the extentof first-yearmolts of $pizella sparrows.Many North American speciesthat have eccentricprimary molts, includingthe Verdin (Auriparus fiaviceps), White-eyed Vireo (Vireo griseus),Yellow-breasted Chat, Field Sparrow ($pizella pusilla), Passerinabuntings, and several species of wrens, thrashers, sparrows, and orioles (Willoughby1991, Thompson and Leu 1994, Pyle 1997), reside in harsh vegetation.The occurrenceof eccentricprimary molting patterns in other speciessuch as the tyrannidflycatchers, Loggerhead Shrike, Lark Bunting(Calamospiza melanocorys), and House Finch (Carpodacus mexicanus), extensively exposedto sunlightbut not to harshvegetation, suggests that exposureto sunlightmay alsoaffect the occurrenceof thesemolts. In addition,variation withintyrannid genera suggests a correlationbetween length of migration and the extent of remex molts, those taxa that winter in South America generallyreplacing their remigescompletely, those winteringin Central Americalargely showing eccentric patterns, and thosewintering in Mexico or the UnitedStates molting fewer remiges on average.This corresponds, in general, with patterns found in European migrants (Jenni and Winkler 1994), althoughit differsfrom what is foundamong Passerina buntings, whereboth migratory and residentspecies molt to a similarextent (Thomp- son 1991, Young1991, Thompsonand Leu 1995). Migrationdistance, as well as vegetation,exposure to sunlight,and otherfactors, should thus be considered when causes and extents of eccentric molts are addressed. Clearly, more studyis neededon eccentricmolts in North American passerines,their causes,and to what extentthese molts in differentgenera representhomologous patterns.

ACKNOWLEDGMENTS

I thank Luis Baptista and Karen Cebra (CAS), Jim Northern (MLZ), Ned K. Johnson,Carla Cicero, and BarbaraStein (MVZ), and Jon Fisherand Walter Wehtje (WFVZ)for permissionto examineflycatcher specimens under their care.Steve N. G. Howell helpedme in the collectionsand producedthe illustrationfor Figure1. Tom Schusterhelped me understandGerman references.The manuscripthas benefited from the comments of Ned K. Johnson, Jon R. King, Philip Unitt, and two anonymousreviewers. This is contribution700 of the PointReyes Bird Observatory.

LITERATURE CITED

Dickey,D. R., andvan Rossem,A. J. 1938. The birdsof E1Salvador. Field Mus. Nat. Hist. Zool. Ser. 23:1-609. Dwight,J. Jr. 1900. The sequenceof plumagesand molts of the passerinebirds of New York. Ann. New York Acad. Sci. 13:73-360.

34 ECCENTRIC FIRST-YEAR MOLT PATTERNS IN CERTAIN TYRANNID FLYCATCHERS

Ginn, H. B., and Melville, D. S. 1983. Moult in Birds. Guide 19. Br. Trust Ornithol., Tring, England. Jenni, L., and Winkler, R. 1994. Moult and Ageing of European Passerines. Academic Press, London. Johnson,N. K. 1963. Comparativemolt cyclesin the tyrannidgenus Empidonax. Proc. XIIIth Int. Ornithol. Cong., pp. 870-883. Johnson, N. K. 1974. Molt and age determinationin Western and Yellowish flycatchers.Auk 91:111-131. Mengel, R. M. 1952. Certain moltsand plumagesof Acadianand Yellow-bellied Flycatchers.Auk 69:273-283. Mester,H., andW. Pr•inte.1982. Die "sektorale"postjuvenile Handschwingenmauser der Carduelinenin S•ideuropa.J. f•ir Ornithol. 123:381-399. Miller, A. H. 1928. The molts of the LoggerheadShrike Lanius ludovicianus Linnaeus. Univ. Calif. Publ. Zool. 30:393-417. Phillips,A. R. 1974. The firstprebasic molt of the Yellow-breastedChat. WilsonBull. 86:12-15. Pyle,P. 1997. Moltlimits in NorthAmerican passerines. N. Am. BirdBander. 22:49-90. Pyle, P., Howell, S. N. G., Yunick, R. P., and DeSante, D. F. 1987. Identification Guide to North American Passerines.Slate Creek Press,Bolinas, CA. Thompson,C. W. 1991. The sequenceof moltsand plumages in PaintedBuntings and implicationsfor theoriesof delayedplumage maturation. Condor 93:209-235. Thompson,C. W., and Leu, M. 1994. Determininghomology of moltsand plumages to addressevolutionary questions: A rejoinder regardingemberizid finches. Condor 96:769-782. Thompson, C. W., and Leu, M. 1995. Molts and plumagesof Orange-breasted Buntings(Passerina leclancherii): Implications for delayedplumage maturation. Auk 112:1-19. Willoughby,E. J. 1991. Molt of the genus$pizella (Passeriformes,Emberizidae) in relationto ecologicalfactors affecting plumage wear. Proc. W. Found.Vert. Zool. 4:247-286. Young,B. E. 1991. Annualmolt and interruptionof the fall migrationfor moltingin LazuliBuntings. Condor 93:236-250.

Accepted8 September 1997