Ultraviolet Reflectance of Colored Plastic Leg Bands

J. Field Ornithol., 70(2):236-243

ULTRAVIOLET REFLECTANCE OF COLORED PLASTIC LEG BANDS

K•wrqJ. McGP,•W•, GEOFFREYE. HILL, ANDAMBER J. K•sE}•2 Departmentof Zoologyand Wildlife Scienceand Alabama Agricultural ExperimentStation 331 Funchess Hall Auburn University,Alabama 36849 USA

Abstract.--The bright colorationof plasticleg bandsused in avianresearch has been shown to affect the socialinteractions of birds by augmentingthe ornamentation that traditionally mediatesintersexual mate choice and/or intrasexualcompetition for mates.With ultraviolet (UV) vision and plumage reflectanceplaying a role in the mating decisionsof somespecies, it would seemimportant to examine the UV reflectanceproperties of colored bands,as these bands may be providing strong and meaningful UV color signalsthat are invisibleto the human eye and are potentiallyunaccounted for in most avian behavioralstudies. In this study,we used a fiber optic spectrometerand a UV/visible light sourceto collect spectral reflectance data for the commonly used types and colors of bands and showedthat none exhibit peak reflectance values throughout the range of UV wavelengthsthat is visible to birds. Instead, UV reflectanceof colored bands at thesewavelengths is comparableto levels of backgroundreflectance found elsewherein the avianvisible spectrum. Consequently, colored bandslack discreteUV color signalsand can safelybe used in studiesof wild birdswith a minimal risk of influencing intraspecificsignaling that is basedon ultraviolet ornamentation. However,caution must be exercisedwhen usingthese bands under experimentalcon- ditions, as certain treatments (e.g., indoor lighting, one-wayglass, sunscreens) may maskthe backgroundlevels of UV reflectancethat still contributeto the overallband color,effectively inducing a color change in the bands that could provide unnatural visualcues to the birds and unintentionallyinfluence their socialinteractions.

REFLECTANCIA ULTRAVIOLETA DE LAS BANDAS PI_•STICAS DE COLORES PARA LAS PATAS.

Sinopsis•Se ha demostradoque la coloraci6nbrillante de las bandaspl•sticas para las patas usadasen investigaci6nesde avesafectan las interaccionessociales de estasal aumentar la ornamentaci6n que tradicionalmentemedia la selecci6nintersexual de parejasy/o la com- petenciaintrasexual por parejas.Como la visi6n de luz ultravioleta (UV) y la reflectanciadel plumajejuegan un rol en las decisionesde apareamientode ciertasespecies, pareceria im- portante examinar la reflectanciaen UV de las bandasde colores,ya que podrian producir serialesfuertes y significativasen color UV que son invisiblesal oho humano y que no son consideradasen la mayorCade los estudioseto16gicos. En este estudiousamos un espectr6- metro de fibra 6ptica y una fuente de luz visible/UV para obtener datos de reflectancia espectralpara los coloresy tipos de bandascomunmente usadas y mostramosque ninguna exhibe valoresde reflectanciamayores a trav6sde los largos de onda UV que son visible a las aves.Por el contrario, la reflectancia de las bandasde coloresen estoslargos de onda son similaresa los nivelesde reflectanciadel fondo halladosa trav6sdel espectrovisible de las aves. Consecuentemente, las bandas de colores carecen de seriales de color UV discreto y se pueden usar de forma seguraen estudiosde avessilvestres con un minimo de riesgode influenciar las serialesintraspecificas basandose en ornamentaci6nultravioleta. Sin embargo, se debe tener cuidadocuando se usen estasbandas bajo condicionesexperimentales ya que ciertostratamientos (e.g., lucesinteriores, espejos de visibilidadde un lado, cremasprotec- toras de pied pueden enmascararlos niveles de reflectanciade UV del fondo que todav•a

1 Currentaddress: Department of Neurobiologyand Behavoir,Seeley G. •VluddHall, CornellUni- versity,Ithaca, New York 14853 USA. 2 Currentaddress: Department of Genetics,Life SciencesBuilding, Universityof Georgia,Athens, Georgia30602 USA.

236 Vol.70, No. 2 UVReflectance of Colored Bands [237 contribuyeal color general de la banda, induciendoefectivamente un cambio en color en las bandasque podian producirsefias visuales nonaturales alas avese influenciarno inten- cionalmente sus interacciones sociales.

The classicmeans by which individual birds are identified in both field and captivestudies is through the applicationof coloredplastic leg bands (Spencer 1978). This techniqueassumes that colored bandsdo not sig- nificantly alter any aspectof avian socialbehavior. In some cases,it has been demonstrated that colored bands do not affect the interactions of birds (Watt 1982, Beletskyand Orians 1989, Belthoff and Gauthreaux 1991, Weatherhead et al. 1991, Cristol et al. 1992, Hannon and Eason 1995). In other studies,however, colored leg bandshave been shownto influence a varietyof bird behaviors,including mate choice (Burleyet al. 1982, Burley 1986a,b;Johnson et al. 1993, Swaddle and Cuthill 1994, Bennettet al. 1996,Fiske and Amundsen1997), mate guarding(Johnsen et al. 1997), and intrasexualcompetition (Metz and Weatherhead1991). Thus, indiscriminantuse of these bands could undermine many of the behavioral studiesin this field. In fact, some researchersavoid using leg colors that are similar to the ornamental coloration of the bird under study.These concernswarrant further, careful considerationof the sig- naling propertiesof colored leg bands. To addressthis issue fully, we must consider not only colors in the visible spectrum (400-700 nm), but also those in the ultraviolet (UV) (290-400 nm). Given our inability to see UV wavelengthsof light (Wald 1952), reflectancein the ultraviolethas rarely been considered.However, severalavian species are known to possessspecialized UV-sensitive retinal cones (Chen et al. 1984, Chen and Goldsmith 1986) and associated oil droplets (Goldsmith et al. 1984) that extend their vision into the short- wavelengthportion of the spectrum (Parrish et al. 1984). Peak visual sensitivityin this range appearsto be between 350 and 380 nm (Emmer- ton and Delius 1980, Chen et al. 1984, Maier and Bowmaker 1993), with a lower limit of around 325 nm (Govardovskii and Zueva 1977). Addi- tionally, the plumage of a number of specieshas been shownto reflect UV light (Burkhardt 1989, Burkhardt and Finger 1991, Maier 1993, Blei- weiss 1994, Andersson 1996, Bennett et al. 1996, 1997, Andersson and Amundsen1997, Johnsen et al. 1997,Andersson et al. 1998, Keyserand Hill 1999), and ultravioletvision and plumagereflectance can play a role in mate choice (Maier 1993, Bennett et al. 1996, 1997, Andersson and Amundsen 1997, Andersson et al. 1998, Hunt et al. 1998). Hence, it appearsboth valuable and necessaryto characterizethe de- gree of ultravioletlight reflectancefor the colorsand typesof plasticleg bands used in avian research,as the bands may provide meaningful UV color signalsto the birds that are invisible to humans and have subse- quentlygone unaccountedfor in both field and captivestudies. This issue hasbeen partiallyaddressed byJohnsen et al. (1997), who usedspectro- radiometryto quantifythe UV reflectanceof coloredbands used in their studyof wild Bluethroats (Luscinia svecica)and found low to moderate 238] K.J. McGrawet al. j. FieldOrnithol. Spring 1999 reflectancelevels in four colorsof hobby beads.However, as of yet no one has investigatedthe UV signalingcapacity of all of the bandsused by field ornithologists.Here we used a fiber optic spectrometerand a UV/visible light sourceto providespectral reflectance data for the colors and brandsof plasticleg bands that are commonlyused in avianstudies.

METHODS In searchof color signals,we work under Andersson's(1996) assump- tion that "colourfulness... is related to deviations from uniform reflectance."The absenceof color is typicalof the two shades,black (no light reflectance)and white (near-perfectreflectance), that both showuniform reflectance.In contrast,colors are characterizedby the presenceof peaks at certain wavelengths;for example, an object is perceivedto be blue when it exhibitspeak reflectanceat a wavelengthbetween 425 and 475 nm. In this study,we looked for discretereflectance peaks within the range of UV wavelengthsknown to be visible to birds (325-400 nm) and abovetypical background readings throughout the rest of the spectrum. The three most commonlyused typesof colored leg bands are hobby beads, split-ringbands, and wrap-aroundbands. Although hobby beads are manufacturedby a number of companies,Perler Beads© (Novacon Corp., Santa Rosa, California) are often used in avian research.These beadscome in solid,translucent, fluorescent and glitterystyles (Hill 1992) and fit mostpasserines and other specieswith similarsized tarsi. Wrap- aroundbands from the NationalBand and Tag Company(Newport, Ken- tucky) and split-ringbands from A. C. Hughes(Hampton Hill, Middlesex, UK) have alsobeen traditionallyused. In our spectrometricanalyses of the three band types,we have only included the most commonlyused solid colors. We collectedreflectance data with an Ocean Optics, Inc. S2000 fiber optic spectrometer(Dunedin, Florida). The unit is fitted with a grating that actsas a prism and divideslight reflectedfrom the surfaceinto its spectralcomponents from 250-880 nm. However,we only presentdata from 300-700 nm, which spansthe range of UV and visiblelight wavelengths.The coloredbands were illuminatedsimultaneously by a visible (tungsten-halogen)light source and an ultraviolet (deuterium) light source (AnalyticalInstrument Systems,Inc., Flemington, New Jersey). Readingswere madewith a reflectanceprobe that washeld at a 90ø angle to and 5 mm from the reading surfacefor all samples.A metal sheath mountedon the probeexcluded all externallight. Coloredband readings were taken from an area 1.75 mm in diameter.Data integrationtime was set at 55 ms and opticalpixel resolutionwas obtained using data smooth- ing level 10. Reflectancewas measured as the percent of light reflected

FIGURE1. Reflectancespectra for the commonly-usedcolors of: a) hobbybeads, b) wrap- around bands,and c) split-ringbands. Each curverepresents the averageof 20 spectra. Vol.70, No. 2 UV Reflectanceof ColoredBands [239

80t- 70÷

0 400 500 600 700 Wavelength(nm)

lle. _

lea.

90. 80. B yellowørange'-'• 70-

60.

-- blue

300 400 500 600 700 Wavelength(nm)

lie

lea

90.

80- 70.

50.

30.

20-

e- 400 500 600 700 Wavelength(nm) 240] K.j. McGrawet al. J.Field Ornithol. Spring 1999 from the surfaceof the band relativeto a 99% Spectralonwhite reflectance standard. For each band, 20 spectrawere measured and then av- eraged without lifting the probe tip from the band surface.

RESULTS AND DISCUSSION Reflectancespectra from the most commonly used colorsand typesof leg bandsshow that none of the bandsexhibit discretereflectance peaks in the ultravioletportion of the spectrum (Fig. 1); thus, thesebands lack hidden UV color signals.Instead, reflectancepeaks are confined to the portion of the spectrumthat is visibleto humans.Interestingly enough, all colorsand typesof bands reflect some amount of UV light. This re- flectanceis often lessthan or equal to the baselineintensity at which light is reflected throughout the rest of the spectrum, and this passivereflec- tance at shorterwavelengths contributes to the overall,full-spectrum color that is exhibited by each band. These resultsare in agreementwith those of Johnsenet al. (1997), who found a wider range of UV reflectance intensitiesamong four colors of hobby beads, but also failed to detect substantialUV color peaks in the beadsused in their study. Comparing the relative reflectanceintensities of the band types,wrap- around bandsreflect more intenselythan do split-ringbands, which in turn are more reflectivethan hobbybeads (Fig. 1). Acrossall band types, light-colored bands (e.g., white, yellow) seem to exhibit more intense reflectivepeaks than darker colors (e.g., blue, green) (Fig. 1). The glossy plasticsurfaces of certain bands are largely responsiblefor the generally- high reflectancevalues across the entire spectrum,and the differencesin shininessamong the three typesare noticeable to the human observer (pers. obs.). Abrading the surfacesof the bands with sandpapersignifi- cantly reducesreflectance intensitiesacross the entire spectrum,but does not change the overall shapesof the curves (pers. obs.). Colored bands differ both in the wavelengthsof peak reflectance(the true color) and in the heights of the peaks (the reflectanceintensity), and avian biologistscan use the data presentedhere to weigh the benefits and risksassociated with usingbands of different color and manufacturer and selectthose bands that leastinfluence the socialsystem being studied, while maximizing band detectibility.Others may be interestedin actually testingthe influence of leg bandsin a controlled situation,where match- ing plumage color with band color might be important. In this case,spectral data from feather patches can be compared to the colored band spectra we present here. Given the absenceof UV reflectancepeaks among the typesand colors of plastic leg bands analyzed here, we now know that the use of these bands in avian studiesof behavior in the wild should not disrupt intraspecificsignaling systems based on ultraviolet ornamentation. However, it is imperative that we not ignore the UV reflectancethat is in fact present in all of the colored bands examined. Although merely at background levels,the ultravioletwavelengths of light reflected by the bands are an integral part of the overall color that is perceived by birds. Elimi- Vol.70, No. 2 UV Reflectanceof Colored Bands [241 nating such reflectance would create a deviation in uniform reflectance, effectivelyinducing a color change that could influence the visualcom- munication systemunder study. Conditionslike these that maskbackground levels of UV reflectance are common in the experimentalsettings under which UV signalingin birds is currently investigated.In their studiesof zebra finches,Bennett et al. (1996) usedUV light filters to preventthe UV reflectanceof plumage and colored bandsfrom being transmittedto the signalreceivers. Anderssonand Amundsen(1997) haveused human sunblockto prevent UV light from being reflected from body regions of male bluethroats, consequentlypreventing females from seeingareas that may reflect in the UV and from being able to usesuch plumage cues when selectinga mate. While thesestudies have employed such manipulations to explorethe function of UV vision and plumage reflectance,any haphazarduse of thesetechniques or others,in conjunctionwith coloredbands, could un- knowinglyprovide unnatural cuesto the birds and influence the observed social interactions in an unintended fashion. In fact, Hunt et al. (1997) haveemphasized the differencesin the coloredband preferences of zebra fincheswith respectto changingUV light conditions.Implicated in this scenariohave been the use of one-wayglass and mirrors that often absorb UV light (Swaddleand Cuthill 1994), aswell as the weak emissionsof UV light wavelengthsby standardindoor lighting (Wyseckiand Stiles1967). This examplehighlights the need to carefullyconsider the experimen- tal conditionsto which studyindividuals are subjectedwhen exploring the role of UV reflectance in social situations. As researchers become increasinglyinterested in the function of UV plumage ornamentation, the apparentchallenge is to designcontrolled experiments that adequate- ly manipulate UV signalsthat we cannot see. While our study demon- strates that biologistscan continue to use colored bands, albeit more carefully,in avianresearch, we hope that it concurrentlyemphasizes the importance of UV vision in the behavioral interactions of birds.

ACKNOWLEDGMENTS

Support for this work was provided by the National Science Foundation (Grant #IBN9722171to G.E.H.), and by the Collegeof Scienceand Mathematics,the Department of Zoologyand Wildlife Scienceand the AlabamaAgricultural Experiment Station at Auburn University.We extend a specialthanks to the Hill lab and I. Cuthill for providing comments that improved the quality of the manuscript.

LITERATURE CITED

ANDERSSON,S. 1996. Bright ultraviolet colouration in the Asianwhistling-thrushes (Myiophon- us spp.). Proc. R. Soc. Lond. B 263:843-848. , AND T. AMUNDSEN. 1997. Ultraviolet colour vision and ornamentation in bluethroats. Proc. R. Soc. Lond. B 264:1587-1591. , I. ORNBORG,AND M. ANDERSSON.1998. Ultravioletsexual dimorphism and assortative mating in blue tits. Proc. R. Soc. Lond. B 265:445-450. BELETSKY,L. D., ANDa. H. ORIANS.1989. Red bands and red-wingedblackbirds. Condor 91: 993-995. 242] •c.j. McGrawet al. J.Field Ornithol. Spring 1999

BELTHOFF,J. R., AND S. A. GAUTHREAUX,JR. 1991. Aggressionand dominance in House Finchesß Condor 93:1010-1013. BENNETT,A. T. D., I. C. CUTHILL,J. c. PARTRIDGE,AND E.J. MAIER.1996. Ultraviolet vision and mate choice in zebra finchesß Nature 380:433-435. --, --, --, ANDK. LUNAU.1997. Ultraviolet plumagecolors predict mate pref- erencesin starlings.Proc. Natl. Acad. Sci. USA 94:8618-8621. BLEIV,•ISS,R. 1994. Behaviouraland evolutionaryimplications of ultraviolet reflectanceby gorgetsof sunangelhummingbirds. Anita. Behar. 48:978-981. BURKHARDT,D. 1989. UV vision:a bird's eye view of feathers.J. Comp. Physiol.A 164:787- 796. --, AND E. FINGER. 1991. Black, white and UV: how birds see birds. Naturwiss. 78:279- 280. BURLEY,N. 1986a. Sexual selectionfor aesthetictraits in specieswith biparental care. Am. Nat. 127:415-445. ß 1986b. Comparison of band colour preferencesof two speciesof estrilid finches. Anita. Behav. 34:1732-1741. --, G. KRANTZBERG,AND P. RADMAN.1982. Influence of colour-bandingon the conspe- cific preferencesof zebra finches.Anim. Behar. 30:444-455. CHEN,D., J. s. COLLINS,AND T. H. GOLDSMITH.1984. The ultraviolet receptor in bird retinas. Science 225:337-340. --, ANDT. H. GOLDSMITH.1986. Four spectralclasses of cone in the retinas of birds.J. Comp. Physiol.A 159:473-479. CRISTOL,D. A., C. S. CHIU, S. M. PECKHAM,AND j. F. STOLL.1992. Color bandsdo not affect dominance statusin captiveflocks of wintering dark-eyedjuncos. Condor 94:537-539. EMMERTON,J., ANDJ. D. DELlUS.1980. Wavelengthdiscrimination in the 'visible' and ultraviolet spectrumby pigeons.J. Comp. Physiol141:47-52. FISKE,P., ANDT. AMUNDSEN.1997. Female bluethroatsprefer males with symmetriccolour bands. Anim. Behav. 54:81-87. GOLDSMITH,T. H., J. s. COLLINS,AND S. LICHT. 1984. The cone oil dropletsof avianretinas. Vision Res. 24:1661-1671. GOVARDOVSIm,V. I., ANDL. V. ZUEV^.1977. Visual pigments of chickenand pigeon.Vision Res. 17:537-543. HANNON,S. J., mXDP. EASON.1995. Colour bands, combs and coverablebadges in willow ptarmigans.Anim. Behav.49:53-62. HILL, G. E. 1992. An inexpensivesource of colored leg bands.J. Field Ornithol. 63:408-410. HUNT, S., I. C. CUTHILL,J.P. SWADDLE,AND A. T D. BENNETT.1997. Ultraviolet visionand band-colourpreferences in female zebra finches, Taeniopygiagutrata. Anim. Behav.54: 1383-1392. , A. T. D. BENNETT, I. C. CUTHILL, AND R. GRIFFITHS. 1998. Blue tits are ultraviolet tits. Proc. R. Soc. Lond. B 265:451-455. JOHNSEN,A., J. T. LIFJELD,AND P. A. ROHDE.1997. Coloured leg bands affect male mate- guarding behaviourin the bluethroat. Anim. Behav.54:121-130. JOHNSON,K., R. DALTON,AND N. BURLEY.1993. Preferencesof female American Goldfinches (Carduelis tristis) for natural and artificial male traits. Behav. Ecol. 4:138-143. KES•3ER,A. J., AND G. E. HILL. 1999. Condition-dependentvariation in the blue-ultraviolet colorationof a structurallybased plumage ornament. Proc. R. Soc.Lond. B 266:771- 778. MAIER,E.J. 1993. To deal with the "invisible": on the biological significanceof ultraviolet sensitivityin birds. Naturwiss.80:476-478. , ANDJ. K. BOWMAKER.1993. Colour vision in the passeriformbird, Leiothrixlutea: correlationof visualpigment absorbanceand oil droplet transmissionwith spectralsen- sitivity.J. Comp. Physiol.A 172:295-301. METZ, K. J., ANDP. J. WEATHERHEAD.1991. Color bands function as secondarysexual traits in male red-wingedblackbirds. Behav. Ecol. Sociobiol.28:23-27. PARRISH,J. w., J. A. PTACEK,AND K. L. WILL. 1984. The detectionof near-ultravioletlight by non-migratoryand migratory birds. Auk 101:53-58. SPENCER,R. 1978. Ringing and related durable methodsof marking birds. Pp. 41-53, in B. Vol.70, No. 2 UVReflectance of Colored Bands [243

Stonehouse,ed. Animal marking:recognition marking of animalsin research.McMillan Press,London, United Kingdom. SW^DDI-E,J.P., ^,','DI. C. CUTHII_I_.1994. Preference for symmetric males by female zebra finches. Nature 367:165-166. WALD,G. 1952. Alleged effectsof the near ultra-violeton human vision.J. Opt. Soc. Amer. 42:171-177. WATT,D.J. 1982. Do birds use color bands in recognition of individuals?J. Field Ornithol. 53:177-179. WEATHERHEAD,P.J., D.J. HOYSAK,K.J. METZ,AND C. G. ECKERT.1991. A retrospectiveanalysis of red-bandeffects on red-wingedblackbirds. Condor 93:1013-1016. WYSECI•I,G., )•r•) W. S. S•rILES.1967. Color Science.New York:John Wiley. Received19 Dec. 1997; accepted17 Jun. 1998.