Beak coloration of starling (Sturnus unicolor) males depends on the length of their throat feathers

Beak colou a ion o s a ling (S u nus unicolo ) males depends on he leng h o hei 1 h oa ea he s 2 3 Manuel Azcá a e-Ga cía1, Magdalena Ruiz-Rod íguez1, C is ina Ruiz-Cas ellano1, 4 Sil ia Díaz-Lo a2, Gus a o Tomás1, Manuel Ma ín-Vi aldi2,3 & Juan José Sole 1,3 5 6 7 1 Depa amen o de Ecología Funcional y E olu i a, Es ación Expe imen al de Zonas 8 Á idas (CSIC), Alme ía. Spain. 9 2 Depa amen o de Zoología, Facul ad de Ciencias, Uni e sidad de G anada, G anada. 10 Spain. 11 3 Unidad asociada (CSIC): Coe olución: cucos, hospedado es y bac e ias simbion es. 12 Uni e sidad de G anada, 18071-G anada, Spain. 13 14 15 16 17 Co esponding Au ho : 18 Manuel Azcá a e-Ga cía 19 ADDRESS: Es ación Expe imen al de Zonas Á idas: C a. de Sac amen o s/n, La 20 Cañada de San U bano, 04120, Alme ía (Spain) 21 TLF: (+34) 660058398 22 E-MAIL: mazca a ega c[email p o ec ed] 23 24 25 C oss e Funding Da a Regis y 26 This wo k was suppo ed by he cu en ly named Minis e io de Ciencia, Inno ación y 27 Uni e sidades and Eu opean (FEDER) unds (CGL2013-48193-C3-1-P, CGL2017-28 83103-P, CGL2017-89063-P). MAG was inanced by a p edoc o al con ac (BES-2014-29 068661) om he Spanish Minis e io de Economía y Compe i i idad, and GT by he 30 Ramón y Cajal P og amme. 31 32 Acknowledgmen s 33 We hank Na alia Juá ez Ga cía-Pelayo and Ca men Sole Zamo a o he help in 34 cap u ing bi ds and collec ing da a om he ield. The esea ch g oup bene i s om 35 acili ies, including and apa men , p o ided by he ci y hall o Guadix whe e a small lab 36 o quickly p oceed he samples was ins alled. 37 38 E hical No e 39 We pe o med he s udy ollowing he ele an Spanish na ional (Dec e o 40 142/2013, 1 de oc ub e) and egional guidelines. The e hics commi ee o he Spanish 41 Na ional Resea ch Council (CSIC) app o ed he p o ocol, and he Conseje ía de Medio 42 Ambien e de la Jun a de Andalucía, Spain, p o ided all he necessa y pe mi s o nes and 43 bi ds manipula ions (Re : SGMN/GyB/JMIF). The p o ocols adhe ed o he ASAB/ABS 44 Guidelines o he Use o Animals in Resea ch. The au ho iza ion o ca ying ou he 45 expe imen al wo k on animals in he ield was g an ed by he Di ección Gene al de 46 Ges ión del Medio Na u al o he Conseje ía de Medio Ambien e o Jun a de Andalucía. 47 48 Au ho con ibu ions 49 Concei ed and designed he expe imen s: JJS, MRR, GT and MMV. Fieldwo k: 50 MAG, JJS, MRR, GT, CRC and SDL. Analysed he da a: MAG and JJS. Con ibu ed 51 eagen s/ma e ials/analysis ools: subs an ial con ibu ion om all au ho s. MAG w o e 52 he i s e sion wi h supe ision o JJS and MRR. All au ho s subs an ially con ibu ed 53 o inal e sion. 54 55 Da a accessibili y 56 Da a used in his pape can be ound in CSIC Ins i u ional Reposi o y, wi h he 57 accession numbe s <xxxxxxxx>. 58 59 1 Lay summa y 1 2 The use o signals o indica e he indi idual quali y is widesp ead in na u e. Howe e , 3 al hough mos species show mo e han one signal, he ela ionships be ween di e en 4 signals ha e almos ne e been expe imen ally s udied. He e, we demons a ed ha he 5 expe imen al educ ion o h oa ea he s leng h condi ioned he beak colou a ion o 6 spo less s a ling males a he ime o ep oduc ion. Ou esul s a e he i s expe imen al 7 e idence o wo sexually dimo phic ai s being ela ed in na u al condi ions. 8 2 Beak colou a ion o s a ling (S u nus unicolo ) males depends on he leng h o hei 9 h oa ea he s 10 Abs ac 11 12 Wi hin he con ex o complex sexual signalling, mos esea ch has ocussed on explo ing 13 he associa ions be ween se e al signals and/o hei ela ionships wi h di e en p oxies 14 o indi idual quali y. Howe e , e y ew s udies ha e ocused on checking whe he he 15 exp ession o one signal is condi ioned by he exp ession o he o he s. He e, by 16 expe imen ally sho ening he h oa ea he s o male spo less s a lings (S u nus 17 unicolo ), we e alua ed he in luence o his ai on he colou exp ession o he beak 18 base. In addi ion, we es ed he ela ionship be ween hese wo sexually dimo phic 19 cha ac e s wi h ai s indica ing indi idual quali y such as body condi ion and colou 20 e lec ance a he wa eleng h ela ed o ca o enes in he ip o he beak. Ou esul s show 21 ha he colou a ion o he beak base in males, bu no in emales, is posi i ely ela ed o 22 body condi ion and o he leng h o o namen al h oa ea he s. Mo eo e , he 23 expe imen al sho ening o h oa ea he s in males had a nega i e e ec on he blue 24 ch oma in ensi y o hei beak base one yea a e manipula ion. These esul s suppo o 25 he i s ime a causal link be ween he exp ession o wo sexually dimo phic cha ac e s, 26 which is essen ial o unde s and hei unc ionali y in a mul iple signalling amewo k. 27 28 Keywo ds: Beak colou , Body condi ion, In e ac ing signals, Mul iple signals, 29 O namen al ea he s leng h, Sexually dimo phic cha ac e s. 30 31 32 3 In oduc ion 33 Animals use a wide a ay o signals o in o m abou hei pheno ypic o gene ic 34 quali y o conspeci ics in social in e ac ions, in con ex s such as ma e choice o 35 compe i ion o esou ces (Kokko 2003, Ande sson and Simmons 2006, K aaije eld e 36 al. 2007, Lyon and Mon gome ie 2012, Edwa d 2015). In con ex s o sexual selec ion, 37 males ypically possess mul iple ai s ha may con ey independen in o ma ion o 38 ecei e s (Mølle and Pomiankowski 1993). Al hough mos esea ch on he e olu ion and 39 unc ion o signals has ocused on single ai s, he impo ance o s udying hese 40 cha ac e s wi hin he heo e ical and mo e ealis ic amewo k o mul iple signals has 41 been highligh ed (Candolin 2003, Hebe s and Papaj 2005). Di e en cha ac e s migh , o 42 ins ance, imply mul iple o edundan messages (i.e., in o ma ion), o migh be mo e 43 e icien in pa icula en i onmen s o in s imula ing pa icula senso y channels (Mølle 44 and Pomiankowski 1993, Candolin 2003, Hebe s and Papaj 2005). Hebe s and Papaj 45 (2005) de eloped a amewo k o es able hypo heses o explaining he e olu ion and 46 unc ioning o mul iple signals. They highligh ed (i) he impo ance o conside ing 47 complex signals and he uni o cha ac e selec ion; (ii) ha complex signals include 48 se e al cha ac e s ha unc ion oge he , ei he acili a ing he ansmission (e.g., using 49 di e en senso y channels) o ein o cing ansmi ed in o ma ion o ecei e s (i.e., 50 edundan in o ma ion); and (iii) ha indi idual signals o componen s o complex signals 51 do no necessa ily unc ion independen ly, bu may in e ac in a unc ional way. 52 Mos esea ch on complex signalling has ocussed on explo ing he associa ion 53 be ween se e al signals (Pe ie e al. 2002, B o-Jo gensen and Dabels een 2008, Mason 54 e al. 2014, Chaine and Lyon 2015, Gi a d e al. 2015), o be ween signals and di e en 55 p oxies o i ness including pheno ypic quali y (Balm o d e al. 1992, Ma in and Lopez 56 2009), ma ing success (Mølle and Pomiankowski 1993) and e icacy o signal 57 4 ansmission in di e en en i onmen s (Endle and Houde 1995). E en hough he s udy 58 o he in e ac ions (i.e., associa ions) be ween di e en signalling cha ac e s is essen ial 59 o know indi idual o complex signals unc ioning, i is one o he leas explo ed a eas 60 wi hin he ield o signal e olu ion. The s udy o signal in e ac ions has he po en ial o 61 shed ligh on signal unc ioning because, o ins ance, de ec ing a posi i e associa ion 62 would sugges ha ansmi ed in o ma ion is edundan o complemen a y. Mo eo e , a 63 nega i e associa ion would indica e ha a ade-o be ween signalling cha ac e s exis s, 64 while he absence o associa ion be ween di e en signals would sugges ha hey con ey 65 di e en in o ma ion o ecei e s (Candolin 2003, Hebe s and Papaj 2005). In mos 66 ins ances, in e -signal in e ac ion occu s when he p esence o one signal o a signal 67 componen al e s he esponse o he ecei e o a second signal o componen by 68 ampli ying o condi ioning he in o ma ion p o ided by each o he . 69 In e ac ions be ween signals may also occu when he p oduc ion o one signal 70 in luences he cos o p oduc ion o ano he signal (Johns one 1996, Candolin 2003). In 71 his case, independen ly o he ansmi ed in o ma ion, he pheno ypic exp ession o one 72 signal impinges on he esul ing pheno ype o he o he signal. Signals a e ypically cos ly 73 o p oduce (Hasson 1994, Sal ado e al. 1996), o main ain (Ruiz-Rod íguez e al. 2015), 74 o o show (i.e. social cos ; Tibbe s and Dale 2004), and he exp ession o signals o 75 signal componen s may be aded-o agains each o he . On he one hand, he e could be 76 a ade-o be ween wo signals (e.g. by using he same esou ces as ca o enes), so ha a 77 lesse exp ession o one inc eases he exp ession o he o he (Ande sson e al. 2002). On 78 he o he hand, i is also possible ha he exp ession o one signal educes he a e age 79 cos s due o social in e ac ions (Mo ales and Velando 2018) and, hus, acili a es o 80 enhances he exp ession o o he signals. Fo ins ance, o namen s ha de elop be o e 81 ep oduc ion and unc ion in social con ex s o he s han sexual (e.g. in a-sexual), and 82 5 could se e o s ablish social hie a chy, may educe agonis ic social in e ac ions and 83 mi iga e subsequen ene ge ic cos s. Sa ed ene gy could hus enhance he p oduc ion o 84 o he sexual o namen s du ing cou ship o ep oduc ion and, he e o e, he exp ession o 85 o namen s de eloped be o e and du ing ep oduc ion could be posi i ely ela ed. This 86 migh be he case o ce ain plumage cha ac e is ics o bi ds ha educe social cos s 87 be o e ep oduc ion (Sena e al. 2000), and hus, could boos he exp ession o o he 88 sexually selec ed ai s, such as song o o he simila lexible dynamics ai s, ha a e 89 exclusi ely exp essed du ing ep oduc ion (Badyae e al. 2002, Mason e al. 2014). 90 De ec ing e idence suppo ing he hypo hesis ha he exp ession o one signal is 91 condi ioned by he exp ession o o he signals can be challenging. A main eason is ha 92 sexual signals a e ypically condi ion-dependen (e.g. Saino e al. 1997, Velando e al. 93 2006, Sole e al. 2008). Thus, de ec ing posi i e o nega i e associa ions be ween he 94 exp ession o di e en signals is no enough o in e causa ion. Ra he , his hypo hesis 95 should be es ed in expe imen al amewo ks whe e he modi ica ion o one signal causes 96 o explains he pheno ypic exp ession o o he signals. As a as we know, his hypo hesis 97 has been es ed expe imen ally only once by Hende son e al. (2018), who manipula ed 98 plumage colou a ion o house inch (Haemo hous mexicanus) males be o e ep oduc ion 99 and de ec ed an e ec on male in es men in song unde cap i i y condi ions. Howe e , 100 he e ec was dependen on expe imen ally modi ied social con ex ( ea he colou a ion 101 o neighbou s) and, hus, i is no comple ely clea ha he de ec ed e ec s we e 102 exclusi ely caused by cos s associa ed o plumage colou a ion. He e, we go a s ep u he 103 and look o expe imen al e idence suppo ing he hypo hesis in he wild in spo less 104 s a lings (he ea e s a lings, S u nus unicolo ). 105 S a lings a e semi colonial and sexually dimo phic bi ds, wi h males showing 106 elonga ed h oa ea he s (Hi aldo and He e a 1974, Lezana e al. 2000) and conspicuous 107 6 yellow beak wi h blue colou ed basal pa (Na a o e al. 2010). These wo sexually 108 dimo phicseconda y sexual ai s could ha bou di e en kinds o in o ma ion o , a leas , 109 in o ma ion a di e en ime scales. The apical pa o hese ea he s is qui e lexible, and 110 males exhibi hem e y conspicuously du ing he en i e yea in social in e ac ions, 111 including cou ship (Apa icio e al. 2001, Ruiz-Rod íguez e al. 2015). In addi ion, hese 112 ea he s hones ly e lec he pheno ypic quali y o indi iduals (Lezana e al. 2000, López-113 Rull e al. 2007, Gil and Cul e 2011, Ruiz-Rod íguez e al. 2015). On he o he hand, 114 du ing ma ing and ep oduc ion ( om Feb ua y o July in ou s udy a ea), he o he wise 115 black colou ed beak o s a lings u n o yellow colou a ion in bo h sexes, while i s basal 116 pa u n o blue in males and o pink in emales (C amp 1998) (Fig.1). Beak colou in 117 s a lings is a sexually dimo phic and dynamic ai ha likely e lec s an ioxidan capaci y 118 (Na a o e al. 2010) and, acco dingly, p e ious s udies ound ha he yellow colou o 119 he beak is ela ed o he le el o ca o enoids and i amin A in he plasma in bo h sexes 120 (Na a o e al. 2010). The moul o h oa ea he s occu s in Sep embe -Oc obe (Veiga 121 and Polo 2016), hus a be o e he ep oduc i e pe iod. The e o e, i is likely ha hese 122 ea he s se e o s ablish social hie a chies wi hin he popula ion du ing he whole yea , 123 allowing o educe agonis ic in e ac ions and o mi iga e i s associa ed cos s (Ande sson 124 1994). I ha was he case, he leng h o he h oa ea he s could play an impo an ole 125 du ing he non-b eeding season by a ec ing he acquisi ion and alloca ion o esou ces, 126 which could be e lec ed in he in ensi y o beak colou a ion in s a ling males. Leng h o 127 h oa ea he s can be easily manipula ed (see Ma e ial and Me hods), so he hypo hesis 128 ha he exp ession o one signal (leng h o h oa ea he s) de e mines he exp ession o 129 he o he (beak colou a ion) can be expe imen ally es ed. 130 We manipula ed he leng h o he h oa ea he s o males by cu ing-o 131 app oxima ely he hal -dis al po ion o he ea he s be o e ep oduc ion, and explo ed i s 132 13 ac o (i.e. dependen a iable), and expe imen al ea men as he ca ego ical p edic o . 286 The da e o i s and las cap u es, as well as numbe o days be ween cap u es we e 287 included as con inuous independen a iables in he s a is ical models. In addi ion, we 288 checked whe he he expe imen did a ec leng h o h oa ea he s a e moul , by 289 ca ying ou epea ed-measu es ANOVAs. In his model, he ea he leng h, a i s and 290 las cap u es, was he dependen a iables ( epea ed measu es), he expe imen al 291 ea men was he ca ego ical p edic o , and he numbe o days be ween cap u es was he 292 con inuous independen a iable. Residuals o all s a is ical models we e plo ed and 293 isually checked o no mali y. All analyses we e pe o med wi h S a is ica V13 (Dell-294 Inc. 2015). 295 296 Resul s 297 The blue ch oma and he yellow- ed ch oma o he s a lings’ beak-base a e 298 nega i ely ela ed in bo h males (Be a(SE) = -0.68(0.08), F1,54 = 74.64, p < 0.001 ) and 299 emales (Be a(SE) = -0.92(0.03), F1,100 = 933.58, p < 0.001). Mo eo e , he base (400-300 475 nm and 570-700 nm) and ip (450-570 nm) beak colou s we e no signi ican ly 301 associa ed in males (blue400-475: Be a(SE) = 0.02(0.13), F1,54 = 0.03, p = 0.867; yellow- ed 302 570-700: Be a(SE) = -0.08(0.11), F1,54 = 0.49, p = 0.487), bu a endency (posi i e o blue 303 and nega i e o yellow- ed ch oma) was de ec ed in emales (blue400-475: Be a(SE) = 304 0.14(0.07), F1,100 = 3.92, p = 0.050; yellow- ed570-700: Be a(SE) = -0.14(0.07), F1,100 = 305 3.77, p = 0.055). 306 Body condi ion was posi i ely and nega i ely ela ed o in ensi y o blue and 307 yellow- ed colou a ion o males’ beak, espec i ely (Table 1, Fig. 3). Nei he he 308 b igh ness o males’ beak no he leng h o hei h oa ea he s we e ela ed o body 309 condi ion (Table 1). In emales, none o hese a iables p edic ed body condi ion (Table 310 1). Simila ly, leng h o h oa ea he s o males, bu no ha o emales, was posi i ely 311 14 and nega i ely ela ed o blue and yellow- ed colou in ensi y o males’ beak, espec i ely 312 (Table 1, Fig. 3). Beak b igh ness did no p edic leng h o h oa ea he s o males o 313 emales (Table 1). Thus, he blue colou in ensi y o males’ beak co- a ied wi h he leng h 314 o h oa ea he s, which migh in o m emales on he pheno ypic quali y (body condi ion) 315 o males. 316 Impo an ly, he expe imen al sho ening o h oa ea he s in males p o oked a 317 educ ion in he in ensi y o he blue, bu no o he , colou a ion o hei beaks (measu ed 318 one o wo yea s a e manipula ion o h oa ea he s) (Table 2, Fig. 4). Mo eo e , he 319 expe imen al manipula ion did no a ec body condi ion o he leng h o h oa ea he s 320 in subsequen cap u es (Table 2). Nei he da e o i s and second cap u e no ime 321 be ween he wo cap u es did explain addi ional signi ican p opo ion o a iance ( esul s 322 no shown). These esul s sugges a di ec link be ween leng h o h oa ea he s and beak 323 colou a ion o males, which is independen o he associa ion o bo h cha ac e s wi h 324 pheno ypic condi ion o males. 325 326 Discussion 327 Ou main esul s a e ha (i) in ensi y o colou a ion o he beak base o spo less 328 s a ling males, bu no ha o emales, was posi i ely ela ed o body condi ion; (ii) beak 329 colou a ion o males was posi i ely ela ed o he leng h o hei o namen al h oa 330 ea he s, and (iii) he expe imen al sho ening o h oa ea he s in males had a nega i e 331 e ec on he blue ch oma in ensi y o he beak o males one yea a e manipula ion. 332 Leng h o h oa ea he s and beak colou a ion a e wo sexually dimo phic ai s ha 333 e lec pheno ypic quali y o males (Apa icio e al. 2001, Na a o e al. 2010) and, hus, 334 ou esul s demons a e a di ec connec ion be ween hese wo ai s sugges ing ha hey 335 may unc ion as a whole in a mul iple signalling amewo k. 336 15 S a lings ha e se e al known sexually dimo phic ai s and a e an app op ia e 337 model sys em o explo e unc ional in e ac ions be ween sexual signals. Mos s udies on 338 sexual signals in his species a e ocussed on he leng h o h oa ea he s o males, which 339 p edic s ma ing success (Apa icio e al. 2001), gene ic he e ozigosi y (Apa icio e al. 340 2001), immune esponse (Gil and Cul e 2011) and elome e leng h (Azcá a e-Ga cía e 341 al. 2020). Bill colou a ion o he dis al yellow pa has also been s udied as a sexually 342 selec ed ai o he species because i is ela ed o ca o enoid and i amin A concen a ion 343 in he blood o males and emales, bu only du ing he ma ing pe iod (Na a o e al. 2010). 344 Sexual di e ences a e howe e mo e appa en a he basal pa o he beak (Fig 1), and 345 we concen a ed on his ai o expe imen ally explo e he possible associa ion wi h he 346 leng h o h oa ea he s. In ag eemen wi h he assump ion ha he blue colou a ion o 347 he basal pa o he beak has a sexual-signalling unc ion, we ound ha i s blue-colou 348 in ensi y was posi i ely ela ed wi h bo h body condi ion and he leng h o he h oa 349 ea he s. Thus, explo ing he in e ac ion be ween hese wo ai s is jus i ied. 350 Leng h o h oa ea he s and beak colo a ion o s a ling males p o ide 351 in o ma ion a di e en ime scales. Black ea he s a e ela i ely s a ic and would p o ide 352 in o ma ion o he pheno ypic condi ion and quali y o males a he ime o moul ing 353 (Badyae and Hill 2000, Hebe s and Papaj 2005). Mo eo e , ea he de e io a ion would 354 also p o ide in o ma ion on ea he quali y and on abili y o males educing ea he 355 deg ada ion (Shawkey e al. 2007, Shawkey e al. 2009, Ruiz-de-Cas añeda e al. 2012, 356 Ruiz-Rod íguez e al. 2015). Thus, leng h o h oa ea he s migh e en include di e en 357 kinds o in o ma ion a a long- e m scale. The beak colou a ion, howe e , should unc ion 358 a a sho ime scale. Like o he colou a ion o o he ba e pa s o bi ds, beak colou a ion 359 has he po en ial o change wi hin weeks, days, hou s, o e en seconds (I e son and 360 Ka ubian 2017). Thus, his kind o dynamic cha ac e s should be con inuously e alua ed 361 16 by ecei e s (Velando e al. 2006, Simons and Ve huls 2011, Dey e al. 2015). As a as 362 we know, associa ions be ween hese wo ypes o sexually dimo phic ai s ha e ne e 363 been assessed. 364 Ou esul s showed ha male body condi ion a he ime o ma ing was ela ed o 365 blue colou a ion o he beak, bu no o he leng h o h oa ea he s, sugges ing ha bo h 366 signals do no p o ide iden ical bu , pe haps, complemen a y in o ma ion. This could be 367 due o he ac ha h oa ea he leng h would explain body condi ion o males a he 368 ime o moul ing, while beak colou a ion would be a mo e dynamic cha ac e ha , 369 simila ly o he colou o he legs o blue- oo ed boobies (Sula nebouxii) (To es and 370 Velando 2007), shows indi idual condi ion a he ime o cap u e. Howe e , leng h o 371 h oa ea he s was posi i ely ela ed o he in ensi y o blue colou a ion o he beak and, 372 hus, i is possible ha bo h ai s con ey edundan in o ma ion o emales. In ag eemen 373 wi h he possibili y ha hese wo ai s ans e complemen a y in o ma ion o emales, 374 we expe imen ally showed a nega i e e ec o leng h o h oa ea he s on he in ensi y 375 o he blue ch oma o he beak base o males se e al mon hs a e he manipula ion. We 376 know ha colou a ion o he ip o he beak e lec s he an ioxidan capaci y o s a lings 377 (Na a o e al. 2010). The associa ion be ween beak colo a ion and ca o enoids’ 378 concen a ion in he blood has also been de ec ed in o he species (Fai e e al. 2003). We 379 did no measu e concen a ion o ca o enoids in he blood in his s udy and, hus, we 380 canno explo e whe he his associa ion exis s o he blue colo a ion o he beak base o 381 males. Mo eo e , colou e lec ance o he beak ip a he ca o enoid wa eleng h, which 382 esul ed posi i ely ela ed o ca o enoid le el in s a lings (Na a o e al. 2010), was no 383 ela ed o colou a ion o he base o he beak o males. Consequen ly, he colou a ion o 384 he beak base is unlikely con eying in o ma ion on an ioxidan capaci y o emales. Thus, 385 ou expe imen al esul s should be in e p e ed as leng h o h oa ea he s unc ioning 386 17 du ing he non- ep oduc i e pe iod and de e mining pheno ypic condi ion o males du ing 387 ma ing. 388 Like o he signals ope a ing in non-sexual scena ios such as pa en -o sp ing 389 communica ion (Mo ales and Velando 2018), o sibling nego ia ion (Johns one and 390 Roulin 2003, Sole and A ilés 2010), including hose media ed by ea he colou a ions 391 (Sena 2006), he leng h o h oa ea he s o males migh se e o s ablish some kind o 392 social hie a chy be ween males ha educe he p obabili y o agonis ic in e ac ions 393 among indi iduals o di e en s a us (Rohwe 1975, Sena 1999, McG aw and Hill 2000). 394 S a lings moul h oa ea he s se e al mon hs be o e ep oduc ion, and males equen ly 395 display hese ea he s while singing in high isible places du ing non- ep oduc i e 396 pe iods (pe s. obs.), which migh ha e a unc ional signi icance in a con ex o social 397 in e ac ions. In some bi d species, p obabili y o social agg ession by conspeci ics is 398 ela ed o ea he cha ac e is ics signalling bi d s a us (Sena 1990, McG aw e al. 2007, 399 Chaine and Lyon 2008)). Mo eo e , agg essions a e mo e common among indi iduals 400 showing simila s a us (Midamegbe e al. 2011), wi h indi iduals ha bou ing signals o 401 highe quali y elici ing lowe le el o agg essi eness (Lopez-Idiaquez e al. 2016). The 402 expe imen al educ ion o h oa ea he s las s un il he nex moul pe iod in au umn and, 403 hus, i is possible ha s a ling males wi h longe h oa ea he s expe ienced lowe a es 404 o social agg essions du ing he non- ep oduc i e pe iod. These cos s can a ec he 405 exp ession o o he ai s ela ed o pheno ypic condi ion, including immune esponses 406 (Hawley e al. 2006), oxida i e s a us (Gal án and Alonso-Al a ez 2009) o he 407 exp ession o sexual signals (Mølle e al. 2000). Al hough we ha e no da a on p obabili y 408 o agg ession o social in e ac ions in gene al in ela ion o leng h o h oa ea he s in 409 s a lings, we hink ha social cos s associa ed o he expe imen al educ ion o leng h o 410 h oa ea he s du ing he non-b eeding pe iod is he mos likely explana ion o he 411 18 de ec ed expe imen al e ec s on beak colou a ion du ing ep oduc ion. Howe e , his 412 mechanis ic explana ion dese es u he esea ch explo ing o ins ance he expec ed 413 associa ion be ween ea he leng h and agg ession du ing he non- ep oduc i e pe iod. 414 Wha e e he mechanis ic explana ions, ou expe imen al esul s s ongly sugges 415 a causal link be ween exp ession o wo sexually dimo phic ai s in spo less s a lings. As 416 a as we know, causal links be ween wo sexually selec ed ai s ha e only been de ec ed 417 in ano he bi d species, he house inch, a highly social species in which head and b eas 418 ea he s o males show g ea a iabili y om ed o yellow colou a ion (Hende son e al. 419 2018). Hende son e al. (2018) ound ha ed- ea he ed males a e mo e a ac i e and 420 sing mo e han yellow- ea he ed males bu , when yellow males we e housed wi h ed 421 males, hey sang mo e han when housed wi h equally una ac i e yellow males. Thus, 422 males adap ed hei singing e o o he social en i onmen (a ac i eness) de e mined 423 by he plumage colo a ion o he social g oups. The e o e, he de ec ed link was 424 explained, no as a di ec consequence o one o he ai s, bu indi ec ly by he social 425 en i onmen in e ms o le el o a ac i eness o neighbou s, which was also 426 manipula ed. Ou expe imen al esul s he e o e show a di ec causal e ec o leng h o 427 h oa ea he s on he exp ession o he colou a ion o he base o he beak o spo less 428 s a ling males, a ai ha is only exp essed du ing he ep oduc i e pe iod. 429 To conclude, we demons a e o he i s ime a causal link be ween he 430 exp ession o wo sexually dimo phic cha ac e s, which is essen ial o unde s and hei 431 unc ionali y in a mul iple signalling amewo k. This ype o in e ac ions be ween 432 sexually selec ed signals migh be widesp ead in na u e and could be mo e easily de ec ed 433 when conside ing signals ha , like ea he colo a ion o mo phological ai s, ha e 434 signalling unc ions in non- ep oduc i e con ex s. 435 436 19 Re e ences 437 438 Ande sson, M. 1994. Sexual selec ion, P ince on, P ince on Uni e si y P ess. 439 Ande sson, M. and Simmons, L. W. 2006. Sexual selec ion and ma e choice. -T ends in ecology 440 & e olu ion, 21: 296-302. 441 Ande sson, S., P yke, S. R., O nbo g, J., Lawes, M. J. and Ande sson, M. 2002. Mul iple 442 ecei e s, mul iple o namen s, and a ade-o be ween agonis ic and epigamic 443 signaling in a widowbi d. -Ame ican Na u alis , 160: 683-691. 444 Apa icio, J. M., Co de o, P. J. and Veiga, J. P. 2001. A es o he hypo hesis o ma e choice 445 based on he e ozygosi y in he spo less s a ling. -Animal Beha iou , 62: 1001-1006. 446 Azcá a e-Ga cía, M., Ruiz-Rod íguez, M., Díaz-Lo a, S., Ruiz-Cas ellano, C., Ma ín-Vi aldi, M., 447 Figue ola, J., Ma ínez-De La Puen e, J., Tomás, G., Pé ez-Con e as, T. and Sole , J. J. 448 2020. O namen al Th oa Fea he s P edic Telome e Dynamic and Ha ching Success in 449 Spo less S a ling (S u nus unicolo ) Males. -F on ie s in Ecology and E olu ion, 7. 450 Badyae , A. V. and Hill, G. E. 2000. E olu ion o sexual dich oma ism: con ibu ion o 451 ca o enoid- e sus melanin-based colo a ion. -Biological Jou nal o he Linnean 452 Socie y, 69: 153-172. 453 Badyae , A. V., Hill, G. E. and Weckwo h, B. V. 2002. Species di e gence in sexually selec ed 454 ai s: Inc ease in song elabo a ion is ela ed o dec ease in plumage o namen a ion in 455 inches. -E olu ion, 56: 412-419. 456 Balm o d, A., Rosse , A. M. and Albon, S. D. 1992. Co ela es o emale choice in esou ce-457 de ending an elope. -Beha io al Ecology and Sociobiology, 31: 107-114. 458 B o-Jo gensen, J. and Dabels een, T. 2008. Knee-clicks and isual ai s indica e igh ing abili y 459 in eland an elopes: mul iple messages and back-up signals. -Bmc Biology, 6: 9. 460 Candolin, U. 2003. The use o mul iple cues in ma e choice. -Biological Re iews, 78: 575-595. 461 Chaine, A. S. and Lyon, B. E. 2008. In asexual selec ion on mul iple plumage o namen s in he 462 la k bun ing. -Animal Beha iou , 76: 657-667. 463 Chaine, A. S. and Lyon, B. E. 2015. Signal a chi ec u e: empo al a iabili y and indi idual 464 consis ency o mul iple sexually selec ed signals. -Func ional Ecology: n/a-n/a. 465 C amp, S. 1998. C amp's he comple e bi ds o he Wes e n Palea c ic, Ox o d, Op imedia, 466 Ox o d Uni e si y P ess. 467 Dey, C. J., Valcu, M., Kempenae s, B. and Dale, J. 2015. Ca o enoid-based bill colo a ion 468 unc ions as a social, no sexual, signal in songbi ds (A es: Passe i o mes). -Jou nal o 469 E olu iona y Biology, 28: 250-258. 470 Edwa d, D. A. 2015. The desc ip ion o ma e choice. -Beha io al Ecology, 26: 301-310. 471 Endle , J. A. and Houde, A. E. 1995. Geog aphic- a ia ion in emale p e e ences o male ai s 472 in poecilia- e icula a. -E olu ion, 49: 456-468. 473 Fai e, B., G egoi e, A., P eaul , M., Cezilly, F. and So ci, G. 2003. Immune ac i a ion apidly 474 mi o ed in a seconda y sexual ai . -Science, 300: 103-103. 475 Gal án, I. and Alonso-Al a ez, C. 2009. The exp ession o melanin-based plumage is sepa a ely 476 modula ed by exogenous oxida i e s ess and a melanoco in. -P oc. R. Soc. Lond. B. 477 Gil, D. and Cul e , R. 2011. Male o namen size in a passe ine p edic s he inhibi o y e ec o 478 es os e one on mac ophage phagocy osis. -Func ional Ecology, 25: 1278-1283. 479 Gi a d, M. B., Elias, D. O. and Kasumo ic, M. M. 2015. Female p e e ence o mul i-modal 480 cou ship: mul iple signals a e impo an o male ma ing success in peacock spide s. -481 P oceedings o he Royal Socie y B-Biological Sciences, 282: 10. 482 Gomez, D. 2006. AVICOL, a p og am o analyse spec ome ic da a, ee p og am a ailable 483 om he au ho a dodogo[email p o ec ed]. 484 G een, A. J. 2001. Mass/leng h esiduals: measu es o body condi ion o gene a o s o spu ious 485 esul s? -Ecology, 82: 1473-1483. 486 20 Hasson, O. 1994. Chea ing signals. -Jou nal o Theo e ical Biology, 167: 223-238. 487 Hawley, D. M., Linds om, K. and Wikelski, M. 2006. Expe imen ally inc eased social 488 compe i ion comp omises humo al immune esponses in house inches. -Ho mones 489 and Beha io , 49: 417-424. 490 Hebe s, E. A. and Papaj, D. R. 2005. Complex signal unc ion: de eloping a amewo k o 491 es able hypo heses. -Beha io al Ecology and Sociobiology, 57: 197-214. 492 Hende son, L. J., B azeal, K. R. and Hahn, T. P. 2018. Plumage colo a ion and social con ex 493 in luence male in es men in song. -Biology Le e s, 14: 4. 494 Hi aldo, F. and He e a, C. M. 1974. Dimo ismo sexual y di e enciación de edades en S u nus 495 unicolo Temm. -Doñana, Ac a Ve , 1: 149-170. 496 I e son, E. N. K. and Ka ubian, J. 2017. The ole o ba e pa s in a ian signaling. -Auk, 134: 587-497 611. 498 Johns one, R. A. 1996. Mul iple displays in animal communica ion: 'Backup signals' and 499 'mul iple messages'. -Philosophical T ansac ions o he Royal Socie y B-Biological 500 Sciences, 351: 329-338. 501 Johns one, R. A. and Roulin, A. 2003. Sibling nego ia ion. -Beha io al Ecology, 14: 780-786. 502 Kokko, H. 2003. The e olu ion o ma e choice and ma ing biases. -P oceedings o he Royal 503 Socie y o London B, Biological Sciences, 270: 653-664. 504 K aaije eld, K., K aaije eld-Smi , F. J. L. and Komdeu , J. 2007. The e olu ion o mu ual 505 o namen a ion. -Animal Beha iou , 74: 657-677. 506 Lezana, L., Mi anda, R., Campos, F. and Pe is, S. J. 2000. Sex di e en ia ion in he spo less 507 s a ling (S u nus unicolo , Temminck 1820). -Belgian Jou nal o Zoology, 130: 139-142. 508 Lopez-Idiaquez, D., Ve ga a, P., Fa gallo, J. A. and Ma inez-Padilla, J. 2016. Female plumage 509 colo a ion signals s a us o conspeci ics. -Animal Beha iou , 121: 101-106. 510 López-Rull, I., Celis, P. and Gil, D. 2007. Egg colou co a ies wi h emale exp ession o a male 511 o namen in he spo less s a ling (S u nus unicolo ). -E hology, 113: 926-933. 512 Lyon, B. E. and Mon gome ie, R. 2012. Sexual selec ion is a o m o social selec ion. -513 Philosophical T ansac ions o he Royal Socie y B-Biological Sciences, 367: 2266-2273. 514 Ma in, J. and Lopez, P. 2009. Mul iple colo signals may e eal mul iple messages in male 515 Sch eibe 's g een liza ds, Lace a sch eibe i. -Beha io al Ecology and Sociobiology, 63: 516 1743-1755. 517 Mason, N. A., Shul z, A. J. and Bu ns, K. J. 2014. Elabo a e isual and acous ic signals e ol e 518 independen ly in a la ge, pheno ypically di e se adia ion o songbi ds. -P oceedings o 519 he Royal Socie y B-Biological Sciences, 281: 9. 520 Mcg aw, K. J. and Hill, G. E. 2000. Ca o enoid-based o namen a ion and s a us signaling in he 521 house inch. -Beha io al Ecology, 11: 520-527. 522 Mcg aw, K. J., Medina-Je ez, W. and Adams, H. 2007. Ca o enoid-based plumage colo a ion 523 and agg ession du ing mol in male house inches. -Beha iou , 144: 165-178. 524 Midamegbe, A., G egoi e, A., Pe e , P. and Dou elan , C. 2011. Female- emale agg essi eness 525 is in luenced by emale colo a ion in blue i s. -Animal Beha iou , 82: 245-253. 526 Mølle , A. P., Bia d, C., Bloun , J. D., Hous on, D. C., Ninni, P., Saino, N. and Su ai, P. F. 2000. 527 Ca o enoid-dependen signals: Indica o s o o aging e iciency, immunocompe ence 528 o de oxi ica ion abili y? -A ian and Poul y Biology Re iews, 11: 137-159. 529 Mølle , A. P. and Pomiankowski, A. 1993. Why ha e bi ds go mul iple sexual o namen s. -530 Beha io al Ecology and Sociobiology, 32: 167-176. 531 Mo ales, J. and Velando, A. 2018. Colo a ion o chicks modula es cos ly in e ac ions among 532 amily membe s. -Beha io al Ecology, 29: 894-903. 533 Na a o, C., Pé ez-Con e as, T., A ilés, J. M., Mcg aw, K. J. and Sole , J. J. 2010. Beak colou 534 e lec s ci cula ing ca o enoid and i amin A le els in spo less s a lings (S u nus 535 unicolo ). -Beha io al Ecology and Sociobiology, 64: 1057-1067. 536 Pe ie , C., De Lope, F., Mølle , A. P. and Ninni, P. 2002. S uc u al colo a ion and sexual 537 selec ion in he ba n swallow Hi undo us ica. -Beha io al Ecology, 13: 728-736. 538 21 Rohwe , S. 1975. Social signi icance o a ian win e plumage a iabili y. -E olu ion, 29: 593-539 610. 540 Ruiz-Cas ellano, C., Ruiz-Rod iguez, M., Tomas, G. and Sole , J. J. 2019. An imic obial ac i i y o 541 nes -lining ea he s is enhanced by b eeding ac i i y in a ian nes s. -FEMS 542 mic obiology ecology, 95. 543 Ruiz-Cas ellano, C., Tomás, G., Ruiz-Rod íguez, M., Ma ín-Gal ez, D. and Sole , J. J. 2016. Nes 544 Ma e ial Shapes Eggs Bac e ial En i onmen . -PLoS One, 11: e0148894. 545 Ruiz-De-Cas añeda, R., Bu J , E., González--B aojos, S. and Mo eno, J. 2012. Bac e ial 546 deg adabili y o an in a ea he unmelanized o namen : a ole o ea he -deg ading 547 bac e ia in sexual selec ion? -Biological Jou nal o he Linnean Socie y, 105: 409-419. 548 Ruiz-Rod íguez, M., Tomás, G., Ma ín-Gál ez, D., Ruiz-Cas ellano, C. and Sole , J. J. 2015. 549 Bac e ia and he e olu ion o hones signals. The case o o namen al h oa ea he s in 550 spo less s a lings. -Func ional Ecology, 29: 701-709. 551 Saino, N., Bolze n, A. M. and Mølle , A. P. 1997. Immunocompe ence, o namen a ion, and 552 iabili y o male ba n swallows (Hi undo us ica). -P oceedings o he Na ional 553 Academy o Sciences o he USA, 94: 549-552. 554 Sal ado , A., Veiga, P., Ma ín, J., López, P., Abelenda, M. and Pue a, M. 1996. The cos o 555 p oducing a sexual signal: es o e one inc eases ha suscep ibili y o male liza ds o 556 ec opa asi ic in es a ion. -Beha . Ecol, 7: 145-150. 557 Sena , J. 1990. Agonis ic communica ion in social species: wha is communica ed? -Beha iou , 558 112: 270-283. 559 Sena , J. 1999. Plumage colo a ion as a signal o social s a us. P oceedings o he In e na ional 560 O ni hological Cong ess, 1999. Bi dLi e Sou h A ica, pp. 1669-1686. 561 Sena , J. 2006. Colo displays as in asexual signal o ag ession and dominance In: HILL, G. E. 562 and MCGRAW, K. J. (eds.) Bi d colo a ion: Func ion and e olu ion. Ha a d Uni e si y 563 P ess, Camb idge, Massachuse s, pp. 87-136. 564 Sena , J. C., Polo, V., U ibe, F. and Came ino, M. 2000. S a us signalling, me abolic a e and 565 body mass in he siskin: he cos o being a subo dina e. -Animal Beha iou , 59: 103-566 110. 567 Shawkey, M. D., Pillai, S. R. and Hill, G. E. 2009. Do ea he -deg ading bac e ia a ec sexually 568 selec ed plumage colo ? -Na u wissenscha en, 96: 123-128. 569 Shawkey, M. D., Pillai, S. R., Hill, G. E., Sie e man, L. M. and Robe s, S. R. 2007. Bac e ia as an 570 agen o change in s uc u al plumage colo : Co ela ional and expe imen al e idence. 571 -Ame ican Na u alis , 169: S112-S121. 572 Simons, M. J. P. and Ve huls , S. 2011. Zeb a inch emales p e e males wi h edde bills 573 independen o song a e-a me a-analysis. -Beha io al Ecology, 22: 755-762. 574 Sole , J. J. and A ilés, J. M. 2010. Sibling compe i ion and conspicuousness o nes ling gapes in 575 al icial bi ds: a compa a i e s udy. -PLoS ONE, 5: e10509. 576 Sole , J. J., Na a o, C., Pé ez-Con e as, T., A ilés, J. M. and Cue o, J. J. 2008. Sexually 577 selec ed egg colo a ion in spo less s a lings. -Ame ican Na u alis , 171: 183-194. 578 Sole , J. J., Ruiz-Cas ellano, C., Figue ola, J., Ma in-Vi aldi, M., Ma inez-De La Puen e, J., Ruiz-579 Rod iguez, M. and Tomas, G. 2017. Telome e leng h and dynamics o spo less s a ling 580 nes lings depend on nes -building ma e ials used by pa en s. -Animal Beha iou , 126: 581 89-100. 582 Tibbe s, E. A. and Dale, J. 2004. A socially en o ced signal o quali y in a pape wasp. -Na u e, 583 432: 218-222. 584 Veiga, J. and Polo, P. 2016. Es o nino Neg o–S u nus unicolo . En: Enciclopedia Vi ual de los 585 Ve eb ados Españoles. Museo Nacional de Ciencias Na u ales, Mad id, Sal ado , A., 586 Mo ales, M. B. (Eds.). 587 Velando, A., Beamon e-Ba ien os, R. and To es, R. 2006. Pigmen -based skin colou in he 588 blue- oo ed booby: an hones signal o cu en condi ion used by emales o adjus 589 ep oduc i e in es men . -Oecologia, 149: 535-542. 590 22 591 29 Fig.4 643 644 Fi s Second Cap u e 0.18 0.20 0.22 0.24 0.26 Blue In ensi y Expe imen al Con ol 645 646 647 30 Annex 1 648 649 Table A1: Numbe o cap u es made o each o he ecap u ed spo less s a ling males 650 du ing he s udy and he numbe o imes he ea men was applied o each male. 651 652 Ring T ea men Yea 2015 Yea 2016 Yea 2017 Numbe o ea men s 3256561 Con ol 2 1 3256564 Con ol 1 1 1 3256565 Con ol 1 1 1 3256567 Con ol 2 1 1 3256590 Con ol 2 1 1 3301955 Con ol 3 1 1 3368660 Con ol 2 1 3368681 Con ol 1 1 1 2 3369509 Con ol 1 1 1 2 3406027 Con ol 1 1 1 3418841 Con ol 1 1 1 3256556 Expe imen al 2 1 3285646 Expe imen al 2 1 3387759 Expe imen al 1 2 1 2 3387764 Expe imen al 2 1 1 3387774 Expe imen al 1 1 1 3387838 Expe imen al 1 1 1 3392095 Expe imen al 1 1 1 3428304 Expe imen al 2 1 3256564b Expe imen al 1 2 1 3256565b Expe imen al 1 1 1 3256567b Expe imen al 1 1 1 653