ACCEPTED VERSION This Is the Peer Reviewed Version of the Following Article: Anna M

ACCEPTED VERSION This is the peer reviewed version of the following article: Anna M. Kearns, Leo Joseph, Jeremy J. Austin, Amy C. Driskell and Kevin E. Omland Complex mosaic of sexual dichromatism and monochromatism in Pacific robins results from both gains and losses of elaborate coloration Journal of Avian Biology, 2020; 51(4):e02404-1-e02404-19

This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions.

PERMISSIONS

https://authorservices.wiley.com/author-resources/Journal-Authors/licensing/self-archiving.html

Wiley's Self-Archiving Policy

Accepted (peer-reviewed) Version

The accepted version of an article is the version that incorporates all amendments made during the peer review process, but prior to the final published version (the Version of Record, which includes; copy and stylistic edits, online and print formatting, citation and other linking, deposit in abstracting and indexing services, and the addition of bibliographic and other material.

Self-archiving of the accepted version is subject to an embargo period of 12-24 months. The standard embargo period is 12 months for scientific, technical, medical, and psychology (STM) journals and 24 months for social science and humanities (SSH) journals following publication of the final article. Use our Author Compliance Tool to check the embargo period for individual journals or check their copyright policy on Wiley Online Library.

The accepted version may be placed on:

 the author's personal website  the author's company/institutional repository or archive  not for profit subject-based repositories such as PubMed Central Articles may be deposited into repositories on acceptance, but access to the article is subject to the embargo period. The version posted must include the following notice on the first page:

"This is the peer reviewed version of the following article: [FULL CITE], which has been published in final form at [Link to final article using the DOI]. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions."

The version posted may not be updated or replaced with the final published version (the Version of Record). Authors may transmit, print and share copies of the accepted version with colleagues, provided that there is no systematic distribution, e.g. a posting on a listserve, network or automated delivery.

There is no obligation upon authors to remove preprints posted to not for profit preprint servers prior to submission.

17 May 2021

http://hdl.handle.net/2440/126038 Accepted Article Corresponding author: Corresponding USA D.C., 6 5 Australia SA, Adelaide, 4 Australia 3 USA Washington, 2 1 Kearns M. Anna coloration elaborate losses of and C ifrne bten hs eso ad h Vrin f eod Pes ct ti atce s doi: as article this cite Please Record. of Version [10.1111/jav.0 the and version this between differences been not has th but through review peer full undergone and publication for accepted been has article This date: Decision E USA. Maryland, Laboratories of Analytical Biology, National Museum of Natural History, Smithsonian Inst Smithsonian ofHistory, Natural Museum National Biology, ofAnalytical Laboratories Gardens, Carlton Victoria, Museum Dept, Sciences Sciences,Univ Biological of School DNA, Ancient for Centre Australian CSIR Collection, Wildlife National Australian Inst Biology Conservation Smithsonian Genomics, Conservation for Center Univ omplex mosaic of sexual dichromatism and monochromatism in Pacific robins results from both gains both from robins results in Pacific monochromatism and dichromatism sexual of mosaic omplex .

of Maryland, Dept of Biological Sciences, Baltimore, Maryland, USA Maryland, Baltimore, Sciences, Biological of Dept Maryland, of

cpeiig tpstig pgnto ad roraig rcs, hc my ed to lead may which process, proofreading and pagination typesetting, copyediting, e

31 2404 1 ,2 -

, Leo Joseph , Leo - Jan mail: mail: ]. - ‘This article is protected by

2020

[email protected] Anna M. Kearns, Univ. of Maryland, Dept Maryland, of Univ. Kearns, M. Anna

3 , Jeremy J. Austin J. Jeremy ,

O National Research Collections Australia, Canberra, ACT, ACT, Canberra, Australia, Collections Research National O 4 ,5 , Amy C.Driskell Amy ,

of Biological Sciences, Baltimore, Sciences,Baltimore, ofBiological 6

and .

of Adelaide, North Terrace, Terrace, North ofAdelaide, Kevin E. Omland Kevin . , National Zoological Park, Park, Zoological National ,

. , Washington, Washington, , 1

Accepted Article color arose via arosevia color was radiation robin on populations r r r which lineages, distinct four dichromatism island reduced and evolution plumage infer from morphometrics and color plumage DNA, nuclear mtDNA, relationships is robins in Pacific variation plumage sexual Concept. Species Biological the and speciation it and ornithologists, from interest standing monochromatism ofthe most one exhibit robins Pacific ABSTRACT island dichromatism reduced reconstruction, state ancestral Keywords islands. separate obin obin obin ” ” taxon taxon P. multicolor P.multicolor

P.pusilla P.polymorpha :

boundaries sexual dichromatism, island speciation, southwest Pacific, southwest Pacific, speciation, island dichromatism, sexual

in the group the in

repeated repeated Vanuatu, Fiji and Samoa Fiji and Vanuatu,

Peale, 1848 Peale, . The and and most most

Mayr, 1934 1934 Mayr, ‘This article is protected by and relationships and evolutionary origin evolutionary r losses of elaborate plumage elaborate of losses ed li . Here, w Here, kely sexually dichromatic and that the radiation that and the dichromatic sexually kely - capped capped

warrant recognition a recognition warrant ( in honor of Ernst Mayr’s detailed work on the southwest Pacific robins southwest Pacific the on work detailed Mayr’s ofErnst honor in e for the populations on the for populations r

use obin obin complex complex

. We use these data to data these Weuse . . primarily historical museum specimens to obtain to specimens museum historical primarily

Our data suggest Ourdata P. goodenovii P. s s

of this geographic mosaic geographic ofthis influential role in the development of Ernst Mayr’s theories on theories Ernst Mayr’s of development in the role influential a lack of lack a r One factor limiting our understanding of the evolution of theof of evolution understanding limitingour factor One ange s separate species separate

. Our Our data . in males in -

- that the common ancestor of the entire Pacific Pacific the entire of ancestor common that the

mosaic resolved

and gains of elaborate plumage plumage ofelaborate gains and

test show that that show all named taxa in the radiation in taxa named all s hypotheses about about hypotheses

taxon boundaries taxon –

the previously recognized recognized previously the remain sexual sexual plumage color, spectrophotometry, spectrophotometry, color, plumage the Pacific robin radiation comprises comprises radiation robin Pacific the - wide mosaic of sexual plumage plumage of sexual widemosaic

dichromatism we propose to name to propose we poorly understood poorly

colonization history colonization Island

and

phylogenetic phylogenetic s

, and and , and and nse nse in females females in

in order order in sampling of sampling

Norfolk Norfolk long despite “ : Mayr's “ Solomon Solomon ) for thefor to , on

- Accepted Article drive N Hill 2003) and Badyaev elaboration plumage sexual and dichromatism toexplore experiment robin flycatchers on gained and lost radiations 2016) al. et Doutrelant congeners being al. et 201 Matysioková 2005, al. et Heinsohn dichromatism Sexual INTRODUCTION evolutionary origins evolutionary ancestor. a from dichromatic founded initially were mosaic geographic itsAt extreme, 1997) in result effects Ba 2000, Griffith al. et 1999, McLain Omland 1997, 1995, al. et (McLain sizes pair lossof dichromatism gradual 2013) Burns and Shultz 2012, atural selec atural - bonds in sedentary sedentary in bonds

of . radiation

lesscolorful the loss of colorful plumage in both sexes in preference for more cryptic plumage cryptic more for preference in sexes both in colorful plumage lossof the 4

populations populations ) Several Here G

where where (Omland 1997, Grant 2001, Badyaev and Hill 2003, Baker et al. 2006, Roulin and Salamin 2010, 2010, Salamin and Roulin 2006, al. et Baker Hill 2003, and Badyaev Grant 2001, 1997, (Omland

enetic drift drift enetic 2019) al. et 2009, (Uy tion in response to increased predation risk and/or different environmental conditions on islands islands on conditions environmental different and/or risk predation to increased response in tion ,

we use we different ( random random Petroica multicolor Petroica

non sexual dichromatism and e and dichromatism sexual , less sexually dimorphic less , sexually s multiple isolated islands isolated multiple

in with of the complex geographic mosaic of sexual plumage coloration in the coloration plumage ofsexual mosaic geographic the complex of - mutually exclusive mutually evolve

the interplay between natural and sexual selection in driving driving in selection sexualand natural between interplay the multilocus DNA multilocus

‘This article is protected by sexual dichromatism sexual results in results (but see Avery et al. al. et 2014) Avery see (but .

island island changes in changes in

selectivepressures

costly s

– . via populations

2 e.g. because of because e.g. elaborate plumage elaborate

) more losses of costly elaborate plumage than gains gains than elaborateplumage costly losses of more

, an interplay between interplay an R g sexual sexual educed educed

olden olden species complex species a nd hypotheses hypotheses w plumage color plumage and

sexual

– on isolated islands isolated on laborate plumage coloration have apparently been differentially differentially been apparently have coloration plumage laborate produc and plumage plumage histlers histlers in 7, Shultz and Burns 2017) Burns and Shultz 7,

e.g.

/or (Peterson 1996, Grant 2001, Bennett and Owens 2002, and 2002, Owens Bennett 2001, Grant 1996, (Peterson Pacific and other other and Pacific A) A)

Turdus Turdus . B selection the the

(owing to (owing Across the southwest Pacific there are s are there Pacific southwest Across the ) Pachycephala

loss of migratory drive and and drive lossofmigratory (Mayr (Mayr 1963)

ing natural and sexual selection sexual and natural

depending on the on depending spp spp

less complex song compared to their mainland mainland to their compared song lesscomplex

n island n only a small handful of males of handful small a only (Peterson 2007) (Peterson

could result in the gradual appearance of appearance the gradual in result could mu

spp . lti T s explain explain -

hese radiations offer a valuable natural natural offer valuable a radiations hese compared to to compared island island (Andersen . I to to sland populations are are populations sland dyaev and Hill 2003) and dyaev test these hypotheses hypotheses testthese genes of the the founders genes of reduced island island dichromatism reduced radiations radiations

, Solomon Islands Solomon , and small effective population population effective small and

establishment of of establishment et al. 2014b) al. et

reduced island island reduced (Bennett and Owens 2002, 2002, Owens and (Bennett mainland (Badyaev and Hill 2003) and (Badyaev

even if all islands islands all if even

(Soler and Moreno Moreno and (Soler everal polytypic polytypic everal

siring Pacific robin robin Pacific

cause

. known for known and Pacific Pacific and

3 Monarcha Monarcha in

)

(Omland (Omland offspring F longer s the the

ounder ounder t he he the . 1 )

) . Accepted Article monochromatic monochromatic color modes with sexes both sexes both where from dichromatism described subspecies goodenovii P. 2016)2015, al. et Kearns 1937, species distinct four, orpossibly three, identified have studies molecular recent and color flycatchers, Islands Solomon Grant 2001) 1997, Futuyma repeated radiation instead form a species complex with mainland mainland with complex species a form instead s molecular robins Pacific 1976) Mayr and Diamond ( perhaps and morphs a with Furthermore, dennisi ultimately le ultimately carlet carlet ation ation

more more r of sexual of

have obin obin reflect birds of different ages ( ofdifferent ages reflectbirds

Despite having Despite Each gains and losses of sexual dichromatism in island island in dichromatism sexual losses and of gains (

Petroica multicolor Petroica studies have have studies ( of Fig. 1 Fig.

melanized linked to different reproductive strategies strategies differentreproductive to linked

P. P. boodang elaborate a Pacific robin Pacific P. P. p.

owing to a lack of w to lack of a owing d endemic to mainland Australia are monotypic, while the Pacific Pacific the monotypic,while are Australia to mainland endemic distinct distinct

ing ‘ plumage color masculin ). are near monochromatic monochromatic near are

polymorpha polymorpha Two Two to taxa taxa the from ation the development of the the of development the

species/subspecies in species/subspecies ‘This article is protected by

‘mascul version of of version established

with marked with marked influenced Ernst Mayr’s theories on allopatric speciation allopatric speciation on theories Ernst Mayr’s influenced modes of sexual coloration are present in the are in present coloration sexual modes of

(Mayr 1934, Schodde and Mason 1999) Mason and Schodde 1934, (Mayr differs from that seen in Samoa and Vanuatu. Both Vanuatu. and in Samoa seen that from differs , ized s

, while , little is known of littleisknown (Fig . None of the other of None g

’ species complex species Solomon Islands Solomon olden olden i

nized’ nized’ elaborate elaborate has two male male two has . –

Norfolk Norfolk P. p. P.p. Norfolk Island, Island, Norfolk ell that that . Sixteen w - differences between the sexes between differences res females females however, however, histlers) histlers) polymorpha polymorpha despite long being treated conspecific with with conspecific treated being long despite olved olved black and red red and black r hav obin obin

the Pacific robin species complex species complex robin Pacific the distinct distinct the evolutionary origins of origins evolutionary the color ‘ ing either ingeither Biological Species Concept Biological but but Australia ) sexually dichromatic dichromatic sexually taxon

approach the striking geographic mosaic mosaic geographic the striking approach . ( P. multicolor P. all are confirmed adults confirmed are all and Pacific robins Pacific are Fiji and Samoa each have a single single a have each Samoa Fiji and

with morphs females

taxa

plumage ( plumage brown or russet head or brown n both sexes sexes both

are r

ed taxa –

(

recognized - one with one Fig capped capped endemic to endemic

(Mayr 1963, Peterson 1996, Omland 1997, Omland 1996, Peterson 1963, (Mayr

. ,robin Pacific and

Fi 1)

a textbook example oft example textbook a .

with with g.

(marked sexual dichromatism) dichromatism) sexual (marked –

avian avian phylogenetic phylogenetic

Solomon Islands Solomon 1). it is unclear whether it isunclear obin typical

variation in sexual dichromatism sexualdichromatism in variation

‘ based on gonads on based in the Pacific robin species complex species Pacific robin the in

femin Vanuatu has Vanuatu , Norfolk Island Island Norfolk ’, Vanuatu, Fiji Vanuatu, radiatio s

Mayr 1942, 1963, Diamond 1970, 1970, Diamond 1963, 1942, Mayr

P. goodenovii P. s P. p. P.

elaborate rather than black heads heads black than rather varies in their degree of sexual sexual degree of their in varies r s ized obin obin

in the southwest Pacific southwest Pacific in the and plumage evolution plumage and polymorpha polymorpha ns (e.g. ns relationships. R relationships. the mainland Australian Australian the mainland

’ (Mayr 1934, Mayr et al.et Mayr 1934, (Mayr

P. pusilla P. mode mode dull

taxa taxa with , however, elaborate elaborate however, ,

male and and plumage ofsexual

and ) or if it is if ) or brown brown

Turdus Turdus ) he phenomenon of phenomenon he of sexual sexual of . Furthermore . th Samoa ( Samoa plumage ese two male male two ese r ed has fourteen fourteen has and and

all three three all plumage - capped capped spp ecent ecent P. P. p. to to herit Fig .

taxa taxa (Fig , and one and able .

, the ,

r in in 1) or or obin .

1) .

Accepted Article robins across the Pacific. Pacific. the across robins of origins evolutionary testthe to data these subspecies Second, flow. gene and ofdivergence their history modeling of capable methods coalescent multilocus using lineages Islands Solomon the of distinctiveness species testthe to data these 1934 (Mayr radiation this on studies in seminal his Ernst Mayr by examined specimens museum and historical modern of series extensive lineage 1934 Mayr, Samoa Fiji and species distinct MtDNA Pacific southwest on taxon

2 1) Here populations with the same mode of sexual coloration are eachother are sexualcoloration of mode same the with populations of monophyly of lack taxon monochromatic sexeseach in between ofdichromatism/monochromatism degree the in diffe plumage ancestor dichromatic sexually a of a) find reconstruction to expect to ascribed currently complex species robin of the Pacific members all connecting the node for predicted ancestor dichromatic sexually a find suggest suggest )

( Did l t Was we see Kearns et al. 2016 for 2016for al. et Kearns see and nuclear loci loci nuclear and Norfolk Island Norfolk ,

in the the in

we in has precedence for the the for precedence has osses of sexual dichromatism occur multiple times multiple occur dichromatism sexual osses of vestigate the the vestigate he ancestor of the ofthe ancestor he that these islands were founded by by founded were islands these that

— examine (VFS)

SOL one from the Solomon Islands and and Islands the Solomon from one taxa rences between monochromatic monochromatic between rences

(Kearns et al. 2016, 2019b, a) 2019b, 2016, al. (Kearnset ‘This article is protected by and and (

“ – variation in nuclear and mtDNA loci, morphometrics and plumage color across an an color across plumage and morphometrics loci, mtDNA and in nuclear variation ( Pacific Pacific

Norfolk appear to further appearto W i.e. VFS relationships between and and between relationships P.pusilla e focus on on focus e populations with the same the with populations connecting connecting

lineages r obin r entire obin obin further SOL . ”

two P.pusilla P. multicolor P.

. lineage, and lineage, Pacific robin Pacific P. goodenovii, P. multicolor, P. multicolor, goodenovii, P.

Having established taxon boundaries and relationships and boundaries taxon established Having

divide u key questions key s the complex geographic mosaic of sexual plumage coloration in in coloration plumage sexual mosaicof geographic thecomplex tification)

P. P. pusilla with the same the . Bougainville Island taxa ) represents a species distinct from the rest of the the the of rest from distinct species a represents ) Petroica If copyright. All rights reserved.’ quantif

fourteen subspecies) subspecies) fourteen . radiation

mode species status is warranted, warranted, speciesis status :

between the SOL and VFS lineages, VFS lineages, and SOL the between

ancestor. into two intotwo y

of Pacific Pacific of o

pusilla pusilla the the f sexual f

sexually dichromatic sexually distinctiveness of the fourteen named named the fourteen of distinctiveness

in the SOL and VFS lineages VFS SOLand in the (

SOL distinct distinct Peale, 1848 has precedence for the for precedence has 1848 Peale, and and coloration r

obins (SOL) ) for

’ and Vanuatu/Fiji/Samoa and the the s closest relatives relatives sclosest (Kearns et al. 2016, 2019a) 2016, al. et (Kearns ,

lineages

the SOL and VFS lineages, b) b) VFS lineages, and theSOL Ma

SOL – , and the other from Vanuatu, otherVanuatu, from the and ,

many of which were were of which many yr et al. yret . Alt

Petroica and and

that might that might ernatively, if island island if ernatively,

? VFS

1937 Ifso

polymorpha polymorpha this would would this lineages lineages ). , c) differences c) differences

we expect to to expect we

We first use use Wefirst , represent represent

we ?

( If so, we so, If VFS and and then .

d ) VFS

use use

) a

Accepted Article ambrynensis polymorpha included samples) contemporary historical examine we Here, examine with Omland Vanuat Previously, robins. Pacific of subspecies all of distinctiveness and genetic structuring of phylogeographic patterns wide th use we Here S variation Genetic methods and Material tissues were tissues were Fiji 3; = n ( lineage from introns nuclear Wesample distributions. fragmented and remote widespread, with species Pacific for istypical been have tissues that contemporary the few on exclusively relies intron dataset therefore Ournuclear introns. nuclear to sequence degraded was too specimens historical CLOCK in were examined and Seas Expedition South Table kleinschmidti ampling r ed A u, u, 1 - ) an ) polymorpha 201 In addition to mtDNA, we examine examine we mtDNA, to addition In the the capped ND2 ND2

) which which

museum . approach kleinschmidti

Most of the historical specimens were collected between 1912 and 1953 as part of the Whitney Whitney as partthe 1953 of and 1912 between collected were historical specimens the of Most 1)

d 5

n = 3,= n available from Samoa or NorfolkIsland. or Samoa from available phylogeographic structuring within within structuring phylogeographic 9 n = 3, 3, = n

) n = 3, 3, = n

two , sequences for all described subspecies of Pacific of subspecies all described for sequences Kearns Kearns r mtDNA e while while ed result r obins and obins

Z ND2 capped capped toepad toepad kulambangrae feminina feminina -

taveunensis

and sequencing and linked n = 3, 3, = n Kearns ed et al et ‘This article is protected by

( sequences f sequences

n = 3,= n

GenBank GenBank in the description of the description in ND2 ND2 samples . r

obins from Australia and and Australia from obins kulambangrae sexchromosome (2015) (2015) SOL

n = 3, n3, = et al et dataset previously published in Kearns et al. (2015, 2016) (2015, al. et in Kearns published previously dataset taveunensis taveunensis

n = 2; Samoa 2; = n and and

. n = 3, 3, = n ) accessions: (2016) (2016) used these these used , cognata cognata rom 5

VFS

Norfolk Norfolk

36 Pacific Pacific 36 dennisi

showed that the endangered Norfolk Norfolk endangered the that showed lineages lineages n = 4) and seven were from the from were seven 4) and = n n=1) ( n=1) n = n 3,

five five intron a previous previous

ND2

(n = 3) (n= KP203816 r new obins (all historical samples), and and historicalsamples), (all obins

n = n 2, the SOL and VFS and theSOL Supple autosomal introns (GAPDH, CLTC, PCBD, CSDE PCBD, (GAPDH, CLTC, introns autosomal similis

R s data to test differentiation across test to differentiation data of Pacific of

subspecies obins (BRM 14 Pacific Pacific 14

KP203833; ( , ACO1 ,

14 collected for this radiation for collected

r n = n obins. Critically obins. contemporary contemporary

( r - robins tann P obins studies based Material Material etroica pusilla etroica 3

) lineage ) . Trials showed that the DNA from the from the DNA that showed Trials .

e ( Supple nsis

n = 2 = n MG676914 of which seven were from the were seven from ofwhich –

Solomon Islands Solomon n = 2; Fiji 2; = n Appendix 1, Appendix s, which is t which s, VFS ; Vanuat mentary frozen (Mayr 1934, LeCroy 2008) LeCroy 1934, (Mayr ,

r lineage (Vanuatu (Vanuatu lineage Kearns et al. (2016) did not did et (2016) al. Kearns obins form a species complex complex formspecies a obins

tannensis 4 -

r MG676937 u

tissue robin robin ed

(n = 7) (n

Material (n= 17) he focus of our study our of focus he

– - Table

capped capped

such sparse sampling sampling sparse such to examine range to examine

(n = 10) (n=

s subspecies in in subspecies Kearns Kearns amples : becki becki A : Appendix 1, 1, Appendix soror soror 1 ) r . obins (all (all obins ). ). This This ambrynensis : ,

N and n = 2,= n 22 , o fresh o SOL n = 3,= n

Accepted Article significant evidence of recombination of evidence significant sums of difference the using analyses. polymorphisms method’ ‘subtraction codes. ambiguity IUPAC with sitescoded 5 using were phased introns Nuclear allele Unrooted MK248661 MK127556 sequence temperatures annealing and five in published phylogenies failed to offer strong support for the reciprocal monophyly of all the populations currently currently the all populations of the reciprocal monophyly support for strong to offer failed phylogenies of monophyly testthe to able are we s taveunensis kulambangrae nine from sampled introns nuclear Seven of tests Multilocus in mtDNA net divergence the measure network all unrooted estimate to used pecies pecies

independent runs independent introns t ree analyses. ree (Clement et al. al. et 2000) (Clement d for this study this for d TOPALi TOPALi

- - were were n MK248678 MK127575 (Kearns et al. 2016) al. et (Kearns

n = 4) and seven and = n 4)

that that 1) networks produced for this study this for produced were used to test for taxon boundaries boundaries taxon to testfor used were , v2.5 v2.5 taxon boundaries taxon

repeating the final run run final repeating the were

(Dolman and Moritz 2006) Moritz and (Dolman By using subspecies as the asthe subspecies using By ‘This article is protected by ). ). MK127576 ; PCBD: (Milne et al. 2004) al. et (Milne

were deposited in GenBank (accessions: (accessions: GenBank in were deposited

too comple following following - ele networks for elenetworks of

calculated using default settings. default using calculated - squares (DSS) method (slidin method (DSS) squares VFS

(GenBank accessions: KT372722 accessions: (GenBank PHASE v2.1 v2.1 PHASE

each x to resolve, and thus these sequences were omitted from further further from omitted were sequences thusand these to resolve, x lineage (Vanuatu (Vanuatu lineage

and gene flow gene and was found in the seven introns seven in the was found studies previous ND2

using the protocol described in Kearns in the described protocol using

r Pacific Robin Robin Pacific ed Sequences with l with Sequences

10 was used to test for signals of recombination of recombination signals test to for was used

between taxa, islands and archipelagos using theusing archipelagos and islands taxa, between - - MK127598

mtDNA and intron nuclear and mtDNA capped capped times . (Stephens and Donnelly 2003) Donnelly and (Stephens a priori a Several sequences Several , 70% 70% ,

lin ; taxonomic unit for species tree analyses in this study study this in analyses tree species for unit taxonomic and introgression introgression and probabilit CSDE:MK248639 eage, which is important given that mtDNA mtDNA given that is important which eage, ength polymorphisms were resolved using the the using resolved were polymorphisms ength ,

seven g window: 100 bp; step size: 10 bp). bp). 10 size: step bp; 100 window: g

DnaSP 5.10 5.10 DnaSP BRM:MK121750

– SOL y threshold with uncertain heterozygous heterozygous uncertain with threshold y

KT372779). KT372779). n = 3; Fiji 3; = n in GAPDH, CLTC and CSDE CLTC and GAPDH, in . Pop

s lineage ( lineage

using an unrooted TCS unrooted an using ART using STRUCTURE using (Rozas et al. 2003) al. et (Rozas - MK248660; CLTC: MK248660; kleinschmidti

(Leigh and Bryant 2015) and Bryant (Leigh under the following settings the following under et al. et New s New polymorpha - MK121772; GAPDH: GAPDH: MK121772;

( 2016 equences

in the nuclear introns introns nuclear in the ) and with primers primers with and )

n = 3, 3, = n

Dxy n = n 3, .

All was used to was used , IMa2 ,

for the other the other for

haplotype haplotype statistic. No

had introns introns

and and length length

was

— Accepted Article each model each 500 with generations million of one runs – analyses STRUCTURE STRUCTURE for files output the create to designate needing VFS lineage. entire the than Fiji and rather Vanuatu skins Island Norfolk 2019b) al. et (Kearns the (2019b) al. Kearnset for performed analyses 0.68) = probability (posterior lineages l as circumscribed reached station reached ( 5000 generations. every taken of1 runs independent ofinvariants. Two proportion and substitu HKY+I+G S 5.0, 1.2) = = (M mean population and S 1.25) = = 4.0, (M birthrate on lognormal a prior clock.rate, for prior exponential an with introns v2.4.8 BEAST in implemented (J metrics Ln(P)K and DeltaK ineage <

akobsson and Rosenberg 2007) Rosenberg and akobsson admixture with correlated allele frequencies correlated with admixture http://tree.bio.ed.ac.uk/software/tracer/ a priori a .

Thus, our nuclear analyses are restricted to testing the the testing to restricted analyses are nuclear our Thus, , while Bayesian trees only offered weak support for a sister relationship between the the between for sisterrelationship a support weak offered only trees Bayesian while , Species trees were estimated estimated were trees Species STRUCTURE

. taxonomic units, which effectively enforced the monophyly of samples designated to each lineage each to designated samples of the monophyly enforced effectively units,which taxonomic CLUMPAK

arity and that ESSs were all above 100. Finally 100. ESSsabove all that and were arity owing to th to owing P. P. pusilla tion model with empirical base frequencies and estimated values for kappa, gamma, shape shape gamma, kappa, for values estimated and frequencies base empirical with model tion . a priori

Critically, we were unable to sample nuclear introns from specimens from from specimens from introns sample nuclear to unable were we Critically, ‘This article is protected by

(Pritchard et al. 2000) al. et (Pritchard were performed with and without without and with performed were (Kopelman et al. 2015) al. et (Kopelman e lack of freshtissue lackof e

(

and and

maximum likelihood trees place Norfolk and placeNorfolk trees likelihood maximum taxon After omitting a burnin of 1 of1 a burnin omitting After (Bouckaert et al. et 2014) (Bouckaert to combine results for each K across independent runs using runs independent across K each for results combine to

and and

boundaries. boundaries. , and subspecies as subspecies ,and DISTRUCT ; from the seven nuclear loci loci nuclear seven the from Kearns et al. 2016, 2019b, a) 2019b, 2016, al. et Kearns

analyses > 000 generations discarded as burnin were performed for K =K 1 for wereperformed as burnin discarded generations 000 ) to assess whether the two independent runs had converged, converged, had runs independent two the assess to whether ) across all of acrossof all ,

was used to test population structuring in nuclear loci without loci without nuclear in structuring testpopulation to was used and and s PGDSpider v2.1.0.0 PGDSpider

and the degraded quality of the DNA obtained from museum museum from obtained DNA ofthe quality degraded the and was then used to test for the best value of of K value the best testfor used to was then using using (Rosenberg 2003) (Rosenberg no admixture admixture no

the a priori a the using a Yule speciat Yule a using u × nique haplotypes nique Petroica

r 10 × ed

monophyly and relationships of samples from from samples of relationships and monophyly 10 8 - ,

capped capped

generations were performed with samples samples performedwith were generations we used used we 7

generations we used TRACER v1.6 TRACER v1.6 used we generations with with

in taxonomic unit taxonomic where where

. (Lischer and Excoffier 2011) Excoffier and (Lischer *BEAST

This approach differs from s differs from Thisapproach algorithms r independent allele frequencies. frequencies. allele independent obins and obins LOGCOMBINER to combine the two two the combine to LOGCOMBINER SOL r

ed from the seven nuclear introns seven the nuclear from ion prior, a strict clock on all on a clock prior, strict ion -

capped capped (Heled and Drummond 2010) Drummond and (Heled

and and .

under under

. All introns used a used All introns . VFS r obins sister to the the sisterto obins two two

lineages alternative models alternative

SOL

the using both both using

CLUMPP Samoa

were used were pecies tree tree pecies was used to to was used and and Ten Ten VFS – and SOL 4 for 4 .

Accepted Article and divergence time time ( divergence and 2005) Hey prior of the constrain indefinitely plateaued then densities and in posterior peak ( we that genera a assumed of1.35 rate mutation and model HKY a used 100 of u mixing, parameter convergence, (IMa2 twothan populations’ more for model analytic sizes ( population D TREEANNOTATOR. runs independent Pacific Pacific b VFS these, and subspecies described for was assessed variation Phenotypic Phenot runs three the replicate between consistency used seeds, starting 10 ESSvaluesabove plots and parameter 3 (at reached been had mixing chain and t etween etween Markov independent 80 sing ) emographic parameters for the SOL and VFS lineages and thefor SOL parameters emographic

with our data our with

lineage 49

000 steps 000 yp r we obin obin Norfolk ( ic va ic 24 males 24 . re re Final

(Vanuatu: 79, Fiji: 32, Samoa: 31) Samoa: Fiji: 32, 79, (Vanuatu: unable to obtain closed posterior densities for ancestral population size ( size ancestralpopulation for densities posterior to obtain closed unable lineages were were not examined lineages riation

–

and and r Ne runs used the following the following used runs of tiontime and then we estimated a maximum a we then estimated and

obins q2 , 25 females)25 ,

the same settings above and and above settings same the ) ) selecting an ‘infinite’ run duration recording output every30 recording output duration run ‘infinite’ selectingan t t maintained a flat curve indefinitely at indefinitely curve flat a maintained

) = ) were estimated estimated were ‘This article is protected by and and ,

6 ed 2.

q2 additional additional - The first final Thefirst capped capped one - to biologically meani biologically to coupled chains with a geometric heating scheme (g1 = 0.96, g2 = 0.5) = = 0.96,g2 (g1 scheme geometricheating a with chains coupled

priors were from the from were

year r obins obins

island island from the seven nuclear introns nuclear introns the seven from

186 Pacific Pacific 186 , , burnin in order to convert parameters to demographic units demographic to parameters to convert order in

010 .

parameter

000.

Specimens were measured were measured Specimens run was monitored was monitored run and Pacific Pacific and

(following Hey 2005) (following populations

008 steps) as determined by a lack of trends in the L[P] and t and L[P] the in trends lack of a by as determined steps) 008 ×

The final two replicate runs were were runs two replicate Thefinal and run length, three final replicate replicate final three length, run and . SOL 10^9 10^9 were were Kearns R ngful values values ngful

r (Hey2009)

obins, however, differences between between differences obins,however, et al. et

( in subspecies that occur on multiple islands on occur subspecies that in gene flow (2 flow gene for 3 for — without returning to zer to returning without - credibility tree (MCC) with mean heights using heights mean with (MCC) tree credibility a population size ( size population

low posterior density, while while density, posterior low (2016) previously examined differences differences examined previously (2016) and then stopped when adequate conve adequate when stopped then and

010 (following the recommendations of Won and and Won of the recommendations (following 142 (93 males, 49 females)49 males, (93 142 . .

008 steps. 008 Following initial test runs to optimize optimize to testruns initial Following using t using at NM

the American Museum of Natural Natural of Museum American the ), divergence time ( time divergence ), he ‘Isolation he q

Finally, we checked for checked we Finally, ) = 5, migration rate ( rate 5,migration = ) initiated o. Accordingly, w o.Accordingly,

(Ellegren2007) min IMa2 representing all fourteen fourteen all representing q2 .

Each nuclear intron nuclear Each ) and divergence time time )divergence and - runs were performed performed were runs

with with different withdifferent t . showed a distinct distinct a showed

Trial runs showed showed runs Trial were from the the from were and and t - ) and and ) Migration within .

We e opted to e opted m . , burnin ,burnin

rgence rgence Of Of ) 10, = -

Accepted Article (MANOVA with Pillai test for significance) test for Pillai with (MANOVA Kearns hue and data these from eachindividua for patch plumage each PX a with three Material tests ( specimens all for taken revision taxonomic and 1953 1912 ( in History History HSD was used to further explore ANOVA results among subspecies within each island group island each within subspecies among results exploreANOVA further to was used HSD sizes ( sample small to owing subspecies plumage all using standardize the in discriminant most females and males with versus subspecies Vanuatu) all from subspecies monochromatism dull with subspecies versusall Vanuatu) and su coloration plumage sexual of mode each within subspecies Supple bspecies with marked dichromatism dichromatism marked with bspecies

were available for males and females of all subspecies except except subspecies all of females and males for available were throat) using back, (crown, patches plumage MANOVA level of sexual dichromatism dichromatism ofsexual level

et al. et mentary in New York, Delaware Museum of Natural ofNatural Museum Delaware York, New in ‘ We measured three morphometric variables (bill length, tail length, wing length) and reflectance of of reflectance and length) wing tail length, length, (bill variables morphometric three Wemeasured Appendix 1, Appendix H4a - Gain 2 pulsed xenon light source (O light xenon source pulsed 2 d

analyses

2015, 2016 2015, mean differences mean ’ )

as part of the Whitney South Seas Expedition and were and Seas Expedition South the of Whitney as part e (Hill and McGraw2006) and (Hill

were used to calculate tristimulus tristimulus calculate to wereused Material Material sville .

; The The variable variable

5 subspecies from Solomon Islands, Vanuatu and Samoa) and Vanuatu Islands, Solomon subspecies from 5 of Mayr (1934). ofMayr .

We did not statistically test the distinctiveness of plumage color color plumage of distinctiveness test statisticallythe not Wedid , Table Table

‘This article is protected by two male plumage morphs of subspecies subspecies morphsof male plumage two and Academy of Natural Sciences at Drexel University in Philadelphia in University at Sciences Drexel ofNatural Academy and ) Supple . Appendix 1, Appendix W between groups using linear discriminant analysis (LDA). analysis lineardiscriminant using groups between A2 e

tested tested using ). ). mentary

across

Owing for significance in significance for Cohen’s D and we tested we and D Cohen’s l Table (elaborate males, dull males, females (elaborate

and Material Material

Supple using CLR v1.05 CLRusing v1.05 cean Optics, cean

subspecies

to the condition of some specimens, not all measurements were were measurements not all specimens, some theof condition to then between A2 mentary ) . m Appendix 1, Appendix color Most of the of Most an Ocean Optics USB2000 reflectance spectrophotometer spectrophotometer reflectance Ocean USB2000 an Optics ean spectral reflectance curves (320 curves reflectance eanspectral

History in Wilmington Historyin elaborate monochromatism monochromatism elaborate

with different plumage modes plumage different with Dunedin, FA, USA) FA, Dunedin,

variance variables

archipelagos Material Material

(Montgomerie 2008) (Montgomerie inplumage

specimens examined examined specimens T for for (dull ‘feminized’ males and females and males ‘feminized’ (dull able able P. p. polymorpha p. P.

using multivariate analysis of variance variance analysis of multivariate using for each each for Appendix 1, Appendix significant significant P. A2 ;

in morphometric variables and 2) and variables morphometric in p 6 subspecies from Solomon Islands, Fiji SolomonIslands, from subspecies 6

. taveunensis . color crucial to crucial ). ). Sufficient samples for quantitative quantitative samples for Sufficient (brightness (brightness .

, Florida Museum of Natural of Museum Florida , Three readings were taken for taken were readings Three

variables differences between between differences

( Table Fig

(

(elaborate ‘masculinized’ ‘masculinized’ (elaborate the seminal plumage theseminal for details of protocols see protocols of details for

. were were

1) were collected between between werecollected

, of Fiji ( of A

we . To To ‘

between between – 2 We then B1 –

treated separately treated

). 700 nm) calculated calculated nm) 700

i.e. calculated explore explore dif ’ However, However, , saturation ,saturation

comparing all comparing Supple

with

each named named each identified the identified the sexes sexes the

mentary ; ferences ferences Tukey 3 - ‘ based based S5a

in in ’ -

Accepted Article studies in mtDNA from inferred their relationships on based phylogeny the to subspecies 2019b) al. et Kearns see discussion further bivittata introns nuclear species the at focused an use 2003) A coloration plumage sexual ancestral of Reconstruction the 0.0167) = the comparisons control multiple for to order in correction usingBonferroni adjusted levels significance th notdo approach differences these however, sexes, the between differences minor some show monochromatic dull and elaborate species/subspe many that Note monochromatic. sexually as elaborate classified Guinea’s New and surfaces dorsal the on feathers melanized lightly having sexes as both plumage monochromatic sexually dull and surfaces dorsal the on feathers melanized heavily having sexes plumage elaborate plumage ancestral ncestral state reconstruction state ncestral

MASS, ggplot2, ggplot2, MASS, no

n using using

equal rate of character character change of rate equal - (Miller and Lambert 2006, Kearns et al. 2019 al. Kearnset 2006, Lambert and (Miller

independent tristimulus color variables (hue, saturation, brightness) brightness) saturation, (hue, variables tristimuluscolor independent (both sexes: brown or brown sexes: (both inferred solely from their position in phylogenies based on mtDNA mtDNA on based in phylogenies their position from solely inferred For ancestral state reconstructions, w reconstructions, state ancestral For

(analyses performed in performed (analyses both both (from Kearns et 2 Kearns al. (from parsimony and likelihood ( likelihood and parsimony

examining examining either in at least one sex least one at in P. archboldi P. archboldi dplyr, dplyr, - level ‘This article is protected by ggord and lsr and ggord ,

which is primarily based on a species tree analysis of two mtDNA loci and five five and loci mtDNA two of analysis species a on tree based iswhich primarily s os

light grey) light of plumage of as dull sexually monochromatic, while New Guinea’s Guinea’s New while monochromatic, sexually as dull e

JMPv10) of of : 019b)

1) 1) . Weconside . the marked sexually dichromatic mode dichromatic sexually marked the (i.e., (i.e., the three modes of sexualcoloration of threemodes the

packages . As such, we classified New Zealand’s Zealand’s New classified we As such, . , but with the position of the position with , but irrespective of dichromatism level ofdichromatism irrespective .

were performed in Mesquite v3.51 in Mesquite were performed . T

Markov k Markov All other statistics All other he second phylogeny focused atsubsp the focused phylogeny second he e defined elaborate sexually monochromatic monochromatic sexually elaborate defined e red two phylogenies of of two phylogenies red .

) .

(males: black; females: black or dark grey) or dark black females: black; (males: and plots and We performed t Weperformed

missing

were performed performed were Petroica

P. multicolor, archboldi P. multicolor, P.

. at or2

) ND2 .

each (Maddison and Maddison Maddison and (Maddison

cies classified as sexually as sexually cies classified wo

) the presence or absence of of absence or the presence )

; Lewis 2001) ; Lewis P. P. australis this . O (Kearns et al. 2019a) al. et (Kearns

reconstructions of of reconstructions plumage ne

(Fig.2) P. bivittata P.bivittata ecies

phylogeny was phylogeny in plumage as both plumageas both

- R v3.5.1 Rv3.5.1 patch added level, and and

and and , which both both ,which P. longipes, longipes, P.

was (p = 0.05/3 0.05/3 (p= other other across across using using

and and (for (for

, P. Accepted Article Models using admixture and correlated frequencies frequencies correlated and admixture using Models of tests Multilocus VF Pacific all from divergent steps mutational several flow gene than alleles rather ofancestral retention Fig. 1, four other between haplotypes shared no by isrepresented lineage and Vanuatu from dataset the nuclear for sampling taxon complete The m at Differentiation Species RESULTS r included best frequencies independentallele and admixture no and 3 = K at Islands shown (not Islands Solomon VF and SOL but Red including models admixture method Ln(P)K the best using K fit as the = (K 4) four and populations method, DeltaK the obins were excluded. were obins (

placed subspecies subspecies placed S - VFS fit ) /SOL that was internal in the network the in was internal that /SOL

1 K kulambangrae

d )

, and

and and istinctiveness of SOL and VFS linea VFS SOLand of istinctiveness

– however, however, nuclear introns had shared haplotypes between the two lineages the two between haplotypes had introns shared nuclear )

diffe

inferred similar structuring to those including thoseincluding to structuring similar inferred ( polymorpha differentiating all Solomon Islands robins from robins in Vanuatua in from robins Islands Solomon all differentiating S SOL )

lineages formed distinct clusters at K = 3, and 3, = K at clusters distinct formed lineages rentiating rentiating

formed distinct clus distinct formed

kleinschmidti lineages lineage all tDNA and nuclear loci nuclear tDNAand

polymorpha polymorpha ‘This article is protected by

vs shared alleles were internal in the network the in internal shared alleles were

kulambangrae kulambangrae the rest of the Pacific Pacific the rest the of vs vs either /species

were were kulambangrae

the the

- and and r capped Robins capped 3.85% divergent 3.85% ed VF

from the Solomon Islands intermediate to these two clusters two ( these to Islands intermediate the Solomon from boundaries - taveunensis ters at ters capped capped ( S ) and

and and

vs vs K = 4 ( 4 = K

( polymorpha r Supple SOL obins from all Pacific from obins Vanuatu Vanuatu

r , and thus the VFS lineage is represented by isrepresented thus the VFS lineage and , ges obins ,

(not shown) (not from Fiji from K = 2 differentiated Pacific Pacific 2= differentiated K

(Omland et 2006) al. (Omland in mtDNA in lineages ( lineages robins robins Fig

in nuclear introns in nuclear Three nuclear introns (CLOCK, PCBD, CSD PCBD, (CLOCK, introns nuclear Three

Fiji r ( Admixture models excluding excluding models Admixture ed henceforward

Supple

two populations (K = 2) as the best fit K using using K fit best 2)as the = (K populations two Material Material

both selected both - ND2

( capped capped both

with some ambiguity some with polymorpha

suggesting suggesting

( mentary

r polymorpha polymorpha Fig obins obins Appendix 1, Fig. Fig. 1, Appendix r obins . .

2 Red

, the VF(S) lineage) VF(S) the

( when when

Table 1) Table two ( two populations Supple or Material Material vs vs

- – r capped capped sharing is likely caused by the the by is likelycaused sharing obins only shared a single allele with with allele shared single a only

differentiating differentiating kulambangrae

and and r ed mentary . - and and ) We did not have have Wedidnot capped kulambangrae kulambangrae Fiji when Fiji when

r 1, Fig. Appendix at K = 4 = K at obins were either either obinswere

1 r r ). ). ed ed

Material Material

- - ( ambrynensis ambrynensis r , while the SOL the while , capped capped capped capped Fig K = 2 = K obins were were obins

. Solomon Solomon vs vs

r Models using using Models ed . Fig

Vanuatu/Fiji 3) - ) capped capped .

. from the from as

Appendix Appendix r r For For 3). obins obins

1 the the E)

). T he he had had T ,

he he

Accepted Article intervals ( size population Effective ( flow gene zero of VF and SOL the low between very 235 occurred likely most years aftermillion 1 point the high than lower for densities the posterior although Notably, 3 377 0.69) = probability by (represented = probability (posterior the of monophyly of exclusion the to Pacific the southwest in robins all of monophyly bodied birds from Vanuatu from birds bodied divided smaller variation morphometric the divide clearly not did Morphometric variation Phenotypic the of estimates individual from stem doubt VF as the range similar = intervals credibility HPD 790

758

915 848)

467 ya) 467 I M )

. We were unable to get an accurate estimate of ancestral population size (mean (mean size = population ancestral of estimate accurate an to get unable Wewere . Multilocus = ( a2 Fig sampling 31

estimated estimated .

728 likely likely

3 variables were were variables ambrynensis ambrynensis ) historical historical

. Broad . our small samplesizes small (both our SOL – m = m (

)

Fig coalescent species trees estimated from all seven nuclear introns nuclear seven all from estimated trees species coalescent 243 . stemming

Given these issues, our results therefore need to be interpreted as broad interpreted to be need therefore results theseGiven our issues, ‘This article is protected by that the SOL and VF SOL and the that

( . lineage (represented by by (represented lineage 0) could not be rejected by likelihood ratio (LLR) tests performed within IMa2. IMa2. within tests performed (LLR) ratio likelihood by berejected not 0)could 0.95 S

247) 95% highest posterior density posterior 95% highest ) 022 SOL dynamics between these two lineages two these between dynamics

Ne . lineage.

( 0

) Supple , whereas there was weak ,there whereas ) from Vanuatu, and and Vanuatu, from

– – significant than for VF for than

estimates estimates

and and

from is from 2 1

071 000 ( VFS mentary Our issues achieving closed posterior densities for several several densities for posterior closed issues achieving Our S )

716

000 000 years sues lineages ( lineages - lineages lineages were sma were bodied birds from the Solomon Islands, Fiji and Samoa from larger from Samoa Fiji and the Islands, Solomon birdsfrom bodied ly ) ( , however, however, ,

S obtaining closed posterior densities for divergence time ( time divergence densities for posterior closed obtaining differen

( ) Material Material S t

(mean = (mean

) fail the small number of nuclear loci used and our limited taxon taxon limited and our used loci ofnuclear number small the (

Fig lineages diverged around 400 around diverged lineages polymorpha polymorpha

Fig ago kleinschmidti ( ed Supple ller for SOL for ller . tiated .

. 3

to return to zero, posterior densities posterior tozero, return to

). Posterior densities included zero and an assum an and zero densities included ).Posterior . 449 er

Thus, the divergence of the SOL and VF and SOL ofthe divergence the Thus, (HPD)

mentary high point of posterior densities ap posterior of point high support across the archipelagos (Table 2), however, LDA LDA however, 2), (Table archipelagos across the

631; 95% HPD credibility intervals = 165 = intervals credibility HPD 95% 631; and and

and and credibility interval credibility ( . mean = mean

kul

for the monophyly of of monophyly the for Material Material taveunensis ambangrae r ed

- 127 ) capped capped . Males from Vanuatu had had Vanuatu from Males . Appendix 1, Fig. 5A Fig. 1, Appendix

273; 273;

000 years ago (High point = point (High yearsago 000

) from Fiji) from

r was strongly supported was strongly obins ( obins 95% HPD credibility credibility HPD 95% s were estimated swere NM

) was estimated to be be to was estimated ) supported the reciprocal the reciprocal supported the

Fig we

(posterior (posterior

VF . 863 re re

peared withina peared 3). 3). preliminary preliminary substantially substantially parameters no parameters (

S 163 ) The ( ) . Instead, . Instead, S

lineage lineage

) (

; 95% 235 lineages lineages t

). ). 690 ption ption

022 – and and

- –

Accepted Article ML topological septentrionalis kulambangrae southernmost ( introns nuclear the in Subspecies distinctiveness Subspecies VFS lineages and SOL between it is complex species this in variation plumage of complexity VFS and SOL from subspecies correction. Bonferroni after morphometrics in distinguishable the Solo from however, 0.0001), < p HSD: all Tukey (pairwise archipelagos other to lengths compared wing longer significantly ( power kulambangrae ofrusset exception the with plumage of analyses LDA p 0.50, = = p differences significant kulambangrae CLOCK. and CSDE both at subspecies and locus, each at polymorpha = 3,17; male: F = 14.50, < p 14.50, F = male: 3,17;

0.0003, 0.0003, bootstrap Supple

= after Bonferroni correction, correction, afterBonferroni

support for support for df

0.69, 0.69, ( Supple

mon Islands (p < 0.02) (p Islands mon polymorpha <

ment = , and and , and and (2.2%) and and

68 Supple 3,17; male: F = 12.92, p < p F 12.92, male: = 3,17; df SOL there were no shared haplotypes between between haplotypes wereshared there no ) septentrionalis polymorpha ary

(Kearns et 2016) al. (Kearns dennisi mentary = ,

3,17) 3,17) (

‘This article is protected by

in in septentrionalis mentary lineage Material mtDNA mtDNA wing length and tail length, but not bill length (Wing: ANOVA ANOVA (Wing: length but not bill tail length, and length wing

versus

within ranged ranged

0.0001, 0.0001, Supple Material Material were were

with ( Material Material relationships relationships per se per Appendix 1, Fig. Fig. 1, Appendix Fig each of the of each , however, these were based on small sample sizes and thus have limited limited thus have sizes and sample small on based these were however, ,

the the between 1.8 between in each mode of sexual plumage coloration sexual of plumage mode each in . mentary . deeply deeply

df 3) Birds from Fiji, Samoa and the Solomon Islands were not statistically werenot SolomonIslands the and Fiji,Samoa from Birds

only only Appendix 1, Fig. Fig. 1, Appendix

(2.4%)

(details below) (details . SOL = STRUCTURE analyses analyses STRUCTURE . Subspecies in the Solomon Islands also showed showed also SolomonIslands the in Subspecies Appendix 1, Fig. 1, Fig. Appendix Nuclear introns introns Nuclear 3,19; Bill: ANOVA Bill: ANOVA 3,19;

females from Va from females 0.0001, 0.0001, divergent

Material Material lineage other subspecies other

among the the among ( Fig – 1.9% 5 .

). 2 df

= - (Fi (Table 1) (Table headed headed

1 3, ANOVA 3, 19; Tail: Solomon Solomon Islands ) also also g.

– 1

nuatu were statistically distinguishable from those from statisticallydistinguishable were nuatu

4 ). kulambangrae at least one of these these of least one at , challenging mtDNA ( mtDNA )

– The deepest deepest The showed structuring showed We found significant differences differences significant Wefound in the Solomon Islands Solomon the in . MtDNA d MtDNA . Supple P. p. polymorpha P.

female: F = 0.33, 0.33, p F = female: . showed population structuring between between structuring population showed Despite their distinctiveness t distinctiveness their Despite mentary Dxy 2 ). ).

to to ivergence subspecies subspecies

mtDNA mtDNA SOL subspecies overlapped in in subspeciesoverlapped SOL

and and = associate these as these associate

–

1.8 Material Material

polymorpha female (Table (Table

sub – males, and also females from from females also and males,

between between 2.5% divergences were between between were divergences = s species had unique alleles unique specieshad

for (Bayesian (Bayesian

0.80, – : F = 7.79, p 0.0017, 0.0017, p 7.79, = : F 3

; Fig. 1, Appendix

dennisi )

Table1 kulambangrae , but , but statistically

– kulambangrae df

female: F = 11.25, 11.25, F = female: or any other other any or

= owing to the owing differences differences here was poor was poor here

be ( 3,16; F male: 3,16; PP , 2.5%

Fig tween tween < . 0.76

) 2 , , )

and and 3

, and and – df 5 )

Accepted Article Bonferroni correction ( correction Bonferroni (Fig other each from divergent 0.21% and traits found subspecies (2015) that al. et Kearns by Fig. (Fig haplotypes show however did DNA Fiji and radiat introns nuclear of analyses STRUCTURE ( Dxy VFS lineage the in Subspecies distinctiveness Subspecies Evolution of plumage color plumage of Evolution Fig. however, n intr ( samples historical Notably, and lineage paraphyletic each within Subspecies Fijiand Samoa ( Vanuatu, of the archipelagos between comparisons kleinschmidti Fig o differences in morphometric in differences ons ons

. taveunensis tannensis 2 1 = 0.16 =

2 – ed – ). ,

even among some subspecies with the same same the with subspecies some among even do not differentiate differentiate not do

Kearns Kearns

are at least at are ). from a from In In mtDNA mtDNA these tests had limited power owing to owing power limited thesetests had

– Vanuatu 0.96% ) into two paraphyletic lineages that were more divergent (0.87%) than most most than (0.87%) more divergent were that lineages paraphyletic two into .

. al. et In Samoa, s Samoa, In

mtDNA

were were 2) paraphyly was apparent in samples originating from both both from originating samples in was apparent paraphyly )

Supple and many unique alleles in the other nuclear introns ( introns nuclear the other alleles in many and unique . 3 ,

Vanuatu, Fiji and Samoa archipelagos archipelagos Samoa Vanuatu, Fiji and 2016 subspecies substit 0.16% ‘This article is protected by Supple

haplotype sampled from a singlea from sampled haplotype some evidence of someevidence ) mentary

kleinschmidti amples from Upolu and Savai’i Savai’i and Upolu amples from n ut within the the within

divergent, ions ions mentary

showed were representatives from Vanuatu and Fiji and Vanuatu from representatives were

or plumage plumage or showed mtDNA divergence ( divergence mtDNA showed

different from from different Material Material .

Material Material 2

shallower divergences compared to those in the SOL lineage (mtDNA (mtDNA lineage the SOL in to those compared divergences shallower VFS kleinsch ,

from from Table 1), but d but 1), Table

(Fig early stages of of stages early measurements across subspecies and and across subspecies measurements

Appendix 1,A Table Appendix had unique unique had lineage also differed also . Vanua Levu Vanua 3) midti Appendix 1,Fig. Appendix small sample sizes ( sample small this central Vanuatu haplotype Vanuatu central this . We found that Wefound mode mode

copyright. All rights reserved.’ from from id not differ in plumage, wing, tail or bill length after after length bill tail or wing, in plumage, idnot differ mtDNA feminina differentiation since there were no shared mtDNA mtDNA shared no were there since differentiation of sexual plumage plumage of sexual

across several morphometric and plumage color color plumage and morphometric acrossseveral

Viti Levu and and Viti Levu i we (n = 1) and Viti Levu (n = 2). (n Viti Levu and 1) (n= slands had unique unique had slands Fig re not not re haplotypes 2 . 2, . 2

). ). – individual l l 5 ND2 Dxy ND2

Though sample size is small, n issizesmall, sample Though reciprocal Supple Table 1) and 1) Table ). ). Vanuatu InFiji,m Supple becki

inferred to form distinct clusters in in clusters form distinct to inferred contemporary contemporary mentary

– color , while s while ,

= 0.66−0.75% = subspecies subspecies kleinschmidti kleinschmidti ND2 ly

were were mentary

(Fig tDNA divided divided tDNA

monophyl isolated isolated ation previous previous

haplotypes that were were that haplotypes

Material . 0.32 equences from Samoa from equences 2).

(e.g. elaborate elaborate (e.g.

Material Material

pairwise MtDNA % Tukey except except islands in Fiji in islands tissues and

etic in mtDNA in etic detailed detailed divergent from from ,

Table 1). Table Appendix 1, Appendix subspecies subspecies - soror soror HSD showed showed HSD Appendix 1, Appendix

Vanua Levu Levu Vanua and nuclear nuclear and uclear uclear analyses analyses .

similis ,

Accepted Article Within the elaborate monochromatic monochromatic elaborate the Within between differences plumage Arethere (Fig into the diverged that ancestor fo plumage elaborate with of reconstructions ancestor monochromatic elaborate or (Fig phylogeny level the subspecies model for the likelihood under was equivocal coloration plumage sexual of modes three the of the into robin the Pacific of ancestor the for dichromatism sexual species the on reconstructions and parsimony Likelihood dichromatic? sexually radiation robin Pacific the entire of the ancestor Was more overlap more 5 ( hue and saturation crown with mostly associated axis LD1 the the differentiated ( another direction in hue back and in direction one hue throat and crown especially involving elaborate females among archipelagos Fig. 1, Appendix ( patch plumage back the in but also 1), (Fig. patches plumage F only: headed the russet orexcluding including either males of subsets females in subspecies between differences significant found D ig. ig. ). Among the Among ). . 3

5 – , 4

Supple SOL ).

Individual ANOVAs show that differences are not only present in the variable crown and throat and crown the variable in present are notonly differences that show ANOVAs Individual

p0.00018 < and and between subspecies ( between mentary

black 3 russet VFS the the – 4 . ). LDA showed discrete differen discrete showed LDA ).

5B presence or absence of elaborate p elaborate of absence presenceor - ‘This article is protected by lineages

headed elaborate males, patterns were similar to elaborate females, but with much much with but females, elaborate to similar were patterns males, headedelaborate - Material headed male morph of morph male headed ) Supple 4 , parsimony reconstructions gave equal probability to eith to probability equal gave reconstructions , parsimony r all r of ; all males: p < 3.874e < ; all p males: SOL

(Fig mentary Appendix 1, Fig. Fig. 1, Appendix Petroica, Supple

. and and

5 plumage mode plumage ( A Supple ,

VFS and within within and

Material Material Supple mentary

for the ancestor of the entire Pacific robin radiation, for the the radiation,for robin Pacific entire ofthe the ancestor for

mentary lineages, and for the ancestor of both of both ancestor the for and lineages, P. p. polymorpha polymorpha P. p. mentary - 09 Fig. 1, Appendix 6 Material Material – , ) monochromatic taxa taxa monochromatic 7

tiation of subspecies between, but not within, butwithin, not between, subspecies tiation of 3 (Table ). Material

lumage

- headed male morph of of morph male headed Material species complex level phylogeny phylogeny level Appendix Fig. 1, Appendix ( MANOVA Pillai test Pillai MANOVA , Table 3 Table

Appendix 1, Fig. Fig. 1, Appendix Fig 4 Fig Supple supported an ancestor with at least one sex sex with least one at an ancestor supported from all black all from Appendix 1, Fig. Fig. 1, Appendix 5D ; , Supple

mentary ). Unsurprisingly, LDA strongly LDA strongly ).Unsurprisingly, Fig 4 Fig on all on

from support an ancestor with marked ancestormarked with an support and for the ancestor that diverged diverged that ancestor the for and ,

mentary

5 Supple patches andcolo patches

Material Material - P. p. polymorpha P.p. polymorpha the SOL and VFS lineages? and the SOL headed elaborate males along along males elaborate headed ). 7A

p SOL er a marked dichromatic dichromatic marked a er 6 ). ). mentary

< 3.894e < ) Material Material . In contrast, all Incontrast,

While

Appendix 1, Fig. 1, Appendix and

VFS Material Material

- reconstruction reconstruction r variables variables r Appendix 1, 1, Appendix 07 )

(black lineages lineages

and in both both in and -

- Accepted Article Tests of the level of sexual dichromatism in plumage revealed that revealedthat plumage in dichromatism sexual of level the Testsof taxon? monochromatic ineach sexes between dichromatism of degree inthe differences Arethere groups between differentiation differentiation some showed females dichromatic whereas males, dichromatic among subspecies of differentiation discrete no showed (Table < 0.07) (p females in between differences significant found variables colour and w subspecies between 5C males monochromatic mo indull subspecies between overlap 3 (Table 0.053) < females (p and < 0.15) (p males both in subspecies between differences significant find to failed and in particular in particular and Our DISCUSSION 4). (Table different significantly not was subspecies monochromatic ofdull plumage back elabor all in in plumage back were found differences ( plumage back s subspecies, differentiated significant subspecies monochromatic than subspecies dichromatic in larger substantially Supple taxa dichromatic sexually marked than degree lesser to much though a dichromatic, sexually fact ).

goal was to was to goal For comparison to the monochromatic plumage modes, we also evaluated the degree of the degree evaluated also we modes, plumage the monochromatic to comparison For colour varia and patches all on based MANOVA mode, plumage monochromatic dull the Within mentary ,

Fig Fig 4 exual dichromatism exual ,

Table4

Material Material Supple resolve test hypotheses about the the testhypotheses about ithin the marked sexual dichromatism plumage mode plumage dichromatism sexual the marked ithin

‘This article is protected by where larger sample sizes were available sizes were sample larger where , mentary

Fig species boundaries and and boundaries species Appendix 1, Table A3 1, Table Appendix in monochromatic subspecies across several variables acrossseveral subspecies monochromatic in .

3 4 albeit , ,

Fig 4 Fig

Supple ( Material Material Supple

was most accentuated was most based on based ,

Supple mentary nochromatic females nochromatic mentary Appendix 1, Fig. 1, Fig. Appendix origins of origins

mentary small sample sizes sample small

Material Material )

Material Material . explore the explore Standardized mean differences between sexes were were sexes between differences mean Standardized ate subspecies except except subspecies ate

in crown and throat plumage patches compared to compared patches throat plumage and crown in

Appendix 1, Fig. 1, Fig. Appendix complex complex Appendix 1, Fig. 1, Fig. Appendix 3 dichromatic dichromatic , but d , but – evolutionary evolutionary 4 ). LDA showed discrete differentiation with no with differentiation discrete showed LDA ). Appendix 1, Fig. Fig. 1, Appendix , which could enhanc could which ,

( Supple ifferentiation was less strong for dull for was less strong ifferentiation geographic mosaic of sexual plumage plumage ofsexual mosaic geographic even ‘monochromatic’ ‘monochromatic’ even subspecies in males (p < 0.04), but not not but 0.04), < males (p in subspecies mentary history history 3 5 P. p. polymorpha. p. polymorpha. P. – . B 4 MANOVA based on all patches patches all on based MANOVA ). ). , however, the were sexes however, ,

Even so, significant so, Even significant (Table 4). In monochrom In 4). (Table 3

of Material Material – 4

the ). Conversely, LDA LDA Conversely, ). e the perception of e the perception Pacific robin radiation robin Pacific

Appendix 1, Fig. Fig. 1, Appendix subspecies

In contrast, the the Incontrast,

differentiation differentiation

(Table 4 (Table

are in in are bles bles atic ,

, Accepted Article 2019a radiation robin Pacific entire Pacific following diverged VFS lineages and SOL the that likely most it appears Overall, history Colonization below) justification further (see polymorpha Petroica (Fig the of in males. plumage independent repeated via originated has coloration plumage sexual of mosaic the geographic that and ancestor dichromatic sexually a from founded reconstruction of and coloration ornithologists coloration Pacific radiations and and radiations Pacific (Fig polymorphisms ancestral unsorted stillretain to owing colonization of wave archipelagos these Vanuatu in Fiji (paraphyletic lineages mtDNA divergent islands and Australia. Pacific single a following lineages differentiated SOL VFS and the testwhether ultimately to required willbe VFS lineages and SOL .

) SOL C . .

However, However, Denser taxonomic sampling of nuclear loci including the Norfolk Robin and all subspecies within the the within subspecies all and Robin Norfolk the including loci ofnuclear sampling taxonomic Denser –

T subspecies), which suggest that some gene flow or secondary colonizations flow or secondary gene some suggest that which subspecies), differences in genetic drift and founder effects on isolated islands. islands. isolated on effects founder and indrift genetic differences D

Irrespective of of Irrespective he star he in the Pacific robin robin radiation Pacific in the

) and and f . ollowing the ollowing Wefollow

for decades for VFS - shaped polytomy shaped there is some uncertainty stemming issomeuncertainty there In addition,o In after the initial radiation. Alternatively, such patterns could result after a single founding founding a single after result could suchpatterns Alternatively, the initial after radiation.

lineages if

Mayr, 1934 for the the for 1934 Mayr, have been argued to to argued been have ‘This article is protected by

there was there

Kearns

this, - initial wide radiation or wideradiation 1934) (Mayr when Norfolk Norfolk when , which are estimated to have to estimated are which ,

we find little molecular evidence for dispersal evidencefor littlemolecular find we .

colonization of the southwest the Pacific of colonization ur ur et al. et ancestral ancestral

incomplete lineage sorting lineage incomplete molecular molecular

in in mtDNA

(2016) in (2016) .

– gains Based on on Based

SOL a pattern that apattern plumage, we hypothesize that the Pacific robins robins Pacific the that hypothesize we plumage, reflect a process wherein colonization of archipelagos occurred occurred ofarchipelagos colonization wherein process a reflect r originated from two separate dispersals from New Guinea or or Guinea New from dispersals separate from two originated obins are included are obins

analyses

of of

lineage and and lineage in

recommend kleinschmidti

elaborate the VFS lineage VFS lineage the . nuclear loci, mtDNA, quantitative measures of plumage plumage of measures quantitative loci, mtDNA, nuclear

Similar mtDNA polytomies have been found in other in other found been have polytomies mtDNA Similar offer clear support for the evolutionary distinctiveness distinctiveness the evolutionary for support clear offer from from

has intrigued evolutionary biologists evolutionary intrigued has

plumage in females and losses of elaborate losses ofelaborate and in females plumage Petroica diverged around around diverged ing their recognition as two distinct species as two ingtheir recognition conflicting conflicting

or or ) and Vanuatu ( Vanuatu and ) in mtDNA analyses analyses mtDNA in differential fixation of divergent haplotypes haplotypes ofdivergent fixation differential is indicative of a recent radiation that could could radiationthat a recent of isindicative

pusilla pusilla . T support for support for wo possible wo Peale, 1848 for the 1848 Peale, 400

/geneflow soror ,000 ,000 ya a

versus the five other other the five versus rela

single (Kearns et al. 2016, 2016, al. et (Kearns

exception could

among archipelagos archipelagos among tionship

with no gene flow gene no with

radiation across the the across radiation were most likely likely most were have have

and and s VFS

s within the the swithin are occurred

the the

lineage

–

in in

Accepted Article Ancestral Ancestral dichromatic? sexually radiation robin Pacific the entire of the ancestor Was color plumage of Evolution 2011) al. et (Cibois scale al. et 2014b) (Andersen honeyeaters Pacific e.g. the diverged genes easily exchanged more have would robins and were connected islands northern the levelssea when lower of times during even isolated remained Island sea of models Monarcha polymorpha and differentiated al. et 2014b) Andersen 2001, (Grant colo of wave initial this islands following between flow little gene no with or simultaneously near polytypic polytypic species 1 Pacific: species; 2 Zealand: New Petroica taxa that show measures spectrophotometry monochromatic even since likely highly is also as well ancestor less straightforward are species. subspecies and plumaged dull entirely radiation robin Pacific plumage

( Table 4 Table

the ancestor of SOL of ancestor the

, of

(in either or both sexes) both (inor either species species and have state flycatchers both

was most divergent (Fig divergent was most Pacific robin radiation is radiation robin Pacific , -

Fig level change during the Pleistocene glacial cycles glacial Pleistocene the change during level reconstructions reconstructions

the with marked with . appear to have been isolated for a longer period of time than oftime longerperiod a isolated for tobeen have appear

4 both sexually dichromatic and monochromatic coloration monochromatic and dichromatic sexually both , entire entire Supple (Uy et al.et 2009) (Uy ‘This article is protected by (Fig , h

– (Andersen et al. 2014a) al. et (Andersen .

owever, al

Pacific robin Pacificrobin .

though the radiation of Pacific ofPacific the radiation though and 5 mentary

)

. This strongly This . sexually dichromatic dichromatic sexually of sexual plumage plumage sexual of

VFS

as an ancestral state throughout the the throughout state ancestral as an

on balance there is most support for a for support istheremost balance on .

Material Material In contrast, the s Incontrast,

( 2 compared to the deeper divergences seen in other Pacific other radiations in seen divergences the deeper to compared

Supple and Golden Whistlers Whistlers Golden and – females are lighter than males in in males than lighter are females spe 3), which mirrors patterns in in patterns mirrors which 3), )

cies complex (i.e. including (i.e.including cies complex and and

mentary suggests that suggeststhat Appendix Fig. 1, Appendix Reconstructions of the three modes of sexual plumage color color sexual plumage of threemodes the of Reconstructions taxa taxa , , kingfishers three of six six of three (Hope 1996) (Hope color plumage have copyright. All rights reserved.’

ubspecies within within ubspecies Material Material ation minor minor r elaborate plumage has has been plumage elaborate eed species species

(Andersen et al. al. et 2013) (Andersen in in (Andersen et al. al. et 2014b) (Andersen . dominate thephylogeny dominate Overall, however, Overall,

Appendix 1, Fig. Fig. 1, Appendix Petroica Petroica warblers is warblers 3 level (18 ka ka (18 – 4 ) with with (Eaton 2005) (Eaton s Petroica other Solomon Island radiations such as such radiations Island Solomon other

of of the the – P. goodenovii goodenovii P. all sexual marked overwhelmingly support elaborate support elaborate overwhelmingly monochromatic monochromatic

2 mya), which predict that Makira Makira that predict which mya), 2 sexual SOL

likely on a similar recent time recent similar a on likely elaborate and dull monochromatic dull monochromatic and elaborate

(Boles 2007) (Boles

phylogeny, including for the the for including phylogeny,

those

lineage are more aremore lineage dichromatism

6 it is striking how recently recently how it isstriking . Additionally, Additionally, . –

7 and and

). ). . This is consistent with . with This isconsistent in in

and and (

A dichromatic ancestor ancestor dichromatic A Australia: 5 species; species; Australia:5 ly dichromatic secondarily

VFS g olden olden plumage plumage . P. multicolor P.

. Makira Island Island Makira

– w

i.e. our our i.e. histlers histlers across all across nization nization are are

lost ) a ) in -

– s

Accepted Article have evolved independently owing to similar selective pressures or it could result from common ancestry ancestry common from result it could or pressures selective similar to owing independently evolved have monochromatic in heads elaborate feminized distinctive the example, For independently. monochromatism elaborate evolved archipelagos these within taxa whether monochromatism elaborate of 4 coloration taxon monochromatic Fig. 1, Appendix archipelagos; different from subspecies differentiated discretely females elaborate betw find we Namely, VFS lineages o robins, Pacific Within lineages? VFS and SOL the in times multiple occur dichromatism sexual losses of Did What ambrynensis cognata/feminina ancestor monochromatic elaborate the for coloration sexual of ancestral an mode favor clearly not do reconstructions state ancestral Similarly, in females. backs black plumage monochromatic (Fig reconstructions state ancestral by analyses phylogenetic previous and study our in equivocal Islandswere Solomon in subspecies among relationships that given , Supple een monochromatic taxa between taxa between monochromatic een

ha s driven the evolution of plumage coloration in Pacific robins inPacific coloration plumage of the evolution sdriven Our finding of high variability of elaborate monochromatic plumage on differen on plumage monochromatic elaborate of variability ofhigh finding Our mentary

(Fig. 2 (Fig. (dichromatic)

and likely even likely and

a – ) or 2) a dull monochromatic ancestor with independent gains of elaborate plumage in plumage ofelaborate gains with ancestor independent dull monochromatic or2) a ) 5D)

Material Material ) support for a sexually dichromatic ancestor dichromatic forsexually a support ) 3) . ,

(Table 4) (Kearns et al. 2016, 2019 2016, al. (Kearnset c) differences in the degree degree the in c) differences ‘This article is protected by urdata

VFS and that dichromatic thatand dichromatic

Appendix 1, Fig. Fig. 1, Appendix and and

multiple multiple

lineage – support

tannensis/similis

and on

e with at least two independent losses of elaborate plumage ( plumage elaborate lossesof independent least two at with

d each

) a lack of monophyly of populations with the same mode of sexual ofsexual mode the same with populations of lackofmonophyly a ) . –

losses and within and 5B) is that the ancestor of the SOL lineage had elaborate elaborate had lineage the SOL of ancestor the is that 5B)

multiple two scenarios seem plausible: either 1) a marked dichromatic or dichromatic a marked 1) either plausible: seem scenarios two on

3 Vanuatu, Samoa and Solomon Islands. However, it is unclear it isunclear However, Islands. Solomon and Samoa Vanuatu, within archipelagos within – kulambangrae 5

independent , (elaborate monochromatic) (elaborate ,

Table 4 Table the SOL and VFS lineages and SOL the 2019b of dichromatism/monochromatism between sexes in each each in sexes between ofdichromatism/monochromatism copyright. All rights reserved.’ ) dennisi . A third alternative, which appears to be supported supported be appears which to alternative, A . third ) is consistent with at least three independent origins origins independent least three at with is consistent )

losses of and and

(as discussed above), b) above), discussed (as

and and

– septentrionalis septentrionalis

all polymorpha

sexual dichromatism dichromatism sexual over a relatively recent time scale. scale. time recent relatively a over ?

(Table 3) (Table

in the SOL lin the SOL in could have second have could Supple (

Furthermore, Furthermore, plumage differences differences plumage

t archipelagos (Fig t archipelagos mentary in the SOL and and SOL the in soror soror eage could could eage

Material Material arily lost lost arily versus versus LDA

of of .

Accepted Article females and males and females pressures selection selected negatively prevalence likely pluma monochromatic gainsthan plumage elaborate costly losses of more robins Pacific in dichromatism losses sexual of of driver dominating 2013) Burns Shultz and 2012, Moreno and Soler and effects processes selective exclusive 2016) al. et Doutrelant 2010, Salamin and Roulin sexual reduced of species) 2 Pacific: six I dichromatism dichromatism radiation robin 3 Table with subspecies between diversity genetic low t which island taxa, in quicklyfix can differences genetic Hill 2003) and Badyaev 2002, Owens Bennett and Grant 2001, 1997, Omland drift genetic selectionand/or of the direction in differences random purely by driven be 2002) Owens and (Bennett elaborate gain or plumage male elaborate radiation robin Pacific sland colonization has clearly played played clearly has colonization sland monochromatic species within within species monochromatic

that that ) suggests that suggests that ) Little is known about the underlying genetic and mechanistic causes of changes in sexual sexual in changes of causes mechanistic and genetic underlying the about isLittleknown a mixture of founder effects, effects, of founder mixture a

of of 4) Pacific robin taxa with taxawith robin Pacific random random (Owen

(Peterson 1996, Omland 1997, Grant 2001, Bennett and Owens 2002, Badyaev and Hill 2003) and Badyaev 2002, Owens and Bennett 2001, Grant 1997, Omland 1996, (Peterson dichromatism on islands on dichromatism ) 2019 al. et (Kearns

(e.g. (e.g. under these processes under on different islands could create a geographic mosaic in sexual dichromatism sexual in mosaic a geographic create could differenton islands s and Short 1995, Kimball and Ligon 1999, Badyaev and Hill 2003) and Badyaev Ligon and 1999, Kimball Short 1995, s and and

g such such ge than dull monochromatic plumage than monochromatic ge dull ‘This article is protected by by enetic enetic reduced marked high rates of inbreeding high of rates causing some causing , and random and diverse and random

on islands on drift drift as such as such

predation dichromatic plumage ( plumage dichromatic Petroica Petroica (Omland 1997, McLain et al. 1999, Griffith 2000, Badyaev and Hill 2003, Hill 2003, and Badyaev 2000, Griffith etMcLain 1999, al. 1997, (Omland elaborate mono elaborate a a ,

: natural selection and/or reduced sexual selection selection have sexual and/or reduced selection natural role in shaping the evolution of plumage coloration in in coloration plumage theof evolution shaping in role 2019b differential losses or gains of sexual dichromatism dichromatism sexual of gains lossesor differential island 1)

fe (Badyaev and Hill 2003) and (Badyaev

on islands on ( increased n increased male plumage. male occur on islands (New Guinea: 2 species; New Zealand: 2 species; 2 Zealand: New species; 2 Guinea: (New islands on occur Omland 1997 Omland ) taxa taxa . processes are influencing plumage evolution in evolution plumage influencing processes are This This .

, which is hypothesize which ,

Overall, it appears it appears Overall, (Grant 2001) (Grant

and in in and to chromatic ,

pattern pattern copyright. All rights reserved.’ ke and/ atural selection Fig e ypically are isolated, have small population sizes population small have isolated, are ypically pancestral , Pacific robins more taxa have elaborate elaborate taxa have more Pacificrobins

or Figuerola and Green 2000, Badyaev and Hill 2003, Hill 2003, and Badyaev 2000, Green and Figuerola S .

(Badyaev and Hill 2003) and (Badyaev election and and election 4 fits with global observations of the phenomena the phenomena of observations global fitswith increased increased , Supple . plumage

That we also found plumage variation variation plumage found also Thatwe .

R unlikely that unlikely since since sexual sexual elaxation elaxation mentary , 2) 2) territor d to result from four non from result to d

since since genetic genetic reduced reduced genetic dichromatism dichromatism .

ial

elaborate plumage is not isnot plumage elaborate Material Material This is because phenotypic and and phenotypic is because This or strengthening strengthening or

drift together act drift often

competition genetic drift has been the been drift has genetic drift drift sexual selection sexual .

Instead, it is more it ismore Instead,

is predicted to drive drive to ispredicted

(Peterson 1996, 1996, (Peterson Appendix 1, Fig. Fig. 1, Appendix . However, it is . However, and others to lo to others and in

island taxa taxa island that affects both both affects that led to the to led Petroica

the Pacific - of mutually mutually

across across

, different different

founder founder

– the the

can can 3 a s ,

ll – e 5 , .

Accepted Article 4) Fig. shared few Namely, O implications Taxonomic 2019) al. et al. et Uy 2017, Charmantier al. et 2016, Uy 2015, al. et Tringali evolution insights into novel provide undoubtedly multiple comparing approach multipl evolved seemingly have and phylogeny Petroica way independent be re it to potentially allowing the genome, in retained traita if even that possible r Norfolk united long that taxonomy the of basis the were traits these that the between traits morphometric and in plumage differences and epithets: species and names common concepts. species Phylogenetic with consistent species separate as treatment history evolutionary we DNA, nuclear Material females e

obins ( obins laborate laborate ur genetic and phenotypic data phenotypic and genetic ur

IMa2 IMa2 “ 1 Mayr's ) , P. , while in in while , 3

estimates of of estimates Appendix 1, Fig. Fig. 1, Appendix ) 1)VFS have and SOL and the Pacific robinsparticular, in Pacific the and

mono and sexual dichromatismsexual and boodang support for for support nuclear nuclear r obin chromatic (Raikow et al. 1979, Marshall et al. 1994, al. et 2011) Wiens Marshall al.et 1979, (Raikow VFS ” still

) in a single polytypic and polyphyletic species ( species polyphyletic and insingle polytypic ) a alleles Petroica and likely reproductive isolation isolation likelyreproductive and divergence with a lack of of post lack a with divergence two distinct lineages/clusters distinct lineages/clusters two

believe that e that believe ‘This article is protected by all elaborate subspecies subspecies elaborate all

plumage plumage and 3 like elaborate plumage plumage elaborate like –

pusilla 5 Petroica

). ). two nuclear loci loci nuclear two

Following Kearns Kearns Following reciprocally monophyletic and deeply divergent mtDNA (Fig mtDNA divergent deeply and monophyletic reciprocally A are consistent with treating the treating with consistent are cknowledging our small sample sizes and incomplete taxon samplin taxon incomplete sample sizessmall and our cknowledging

( (Husby et al. 2012, Roulin and Ducrest Roulinand 2012, al. et (Husby ach ach i.e. in i.e. Peale, 1848 for the for 1848 Peale, “

Solomon Solomon lineages with different modes of sexual plumage coloration would would coloration plumage ofsexual modes different with lineages of these lines of evidence demonstrate the long the demonstrate evidence these of lines of our developing understanding of understanding developing our SOL all elaborate subspecies subspecies elaborate all SOL

since all three modes of sexual coloration are scattered across the the across are scattered coloration sexual of three all since modes (Gill 2014) (Gill have have e times in in times parallele with with r obin et al. et was lost in a distant ancestor, its genetic basis can be be can basisits genetic ancestor, distant a in was lost - in in divergence gene flow ( flow gene divergence fixed differences differences fixed females with females with ” copyright. All rights reserved.’ of - STRUCTURE and species tree analyses (Fig analyses tree species and STRUCTURE

expre Petroica polymorpha Petroicapolymorpha (2016), we recommend the following English English following the recommend we (2016), VFS

the SOL and VFS theand SOL and under the Generalized Lineage, Evolutionary and Lineage, Evolutionary the Generalized and under

ssed in descendant taxa in a seemingly inseemingly a taxa descendant in ssed SOL and VFS lineages VFS lineages and SOL lineage. lineage.

SOL and VFS and SOL

but slightly different ways different slightly but black heads black P. multicolor multicolor P. Notably, a lack lack of a Notably, ( Supple have brown have r . This is certainly a possibility for for possibility a This . is certainly obins and Pacific robins with with Pacific robins and obins the genomic basisgenomic the

lineages and support their support and lineages Fig 2013, Huang and Rabosky 2014, 2014, Rabosky and Huang 2013, .

Mayr, 1934 for the the 1934 for Mayr, ) mentary .

(Fig lineages 3 D sensu sensu ), and 5) and ), . -

is not surprising given given surprising isnot

4 or russet or ,

Material Material - Supple

Mayr 1934). 1934). Mayr term, independent term, as distinct species. species. as distinct other diagnosable diagnosable other differences in in differences of of genomic A . . -

mentary headed headed 2 pl Appendix 1, 1, Appendix ,

SOL umage umage Table 1), 1), Table .

A lineage lineage

g in in g s – carlet carlet B , )

2) 2) ,

Accepted Article sequences (Fig sequences nuclear tissues and fresh had we for which fivesubspecies the in differentiation ofnuclear evidence However differences Supple was color plumage on paraphyletic regarding caveat possible a 19 al. et Mayr ( flycatchers sp. species both at seen 2000) Lanyon taxa ofavian range wide a for documented are well patterns plumage in stasis radiation robin the Pacific from divergent is deeply which and Australia, endemicto lineage a to belong Robins Scarlet that shown subsequently have analyses Molecular Conclusion song and behavior their distinctive on based distinct as a taxon this recognizing with consistent is not radiation robin Pacific subspecies the Samoan of divergence of level ( Vanuatu in subspecies other ( divergence mtDNA their since subspecies i from result these could or patterns histories evolutionary disparate with distinctsubspecies represent could islands Levu mtDNA li two paraphyletic level.The subspecific the at recognized be that should taxa distinct represent Samoa Fiji and in islands isolated on populations testwhether to n the Pacific (Fig Pacific the n ) (Andersen et al. 2014b) al. et (Andersen mentary W , all fourteen subspecies had distinct mtDNA haplotypes (Fig haplotypes mtDNA distinct had subspecies fourteen all , e incomplete lineage sorting or sorting lineage incomplete in plumage or morphometrics morphometrics or in plumage s Monarcha

find support for the distinctiveness of the thirteen subspecies that were described by Mayr (1934 by Mayr described subspecies were that thirteen the of distinctiveness the for support find 37 . Furthermore, similar findings of discordance between morphology and genetics have been been geneticsand have morphology between ofdiscordance similar findings Furthermore, . .

Material Material 3 ), as as well , .

Su 2). Likewise, Samoan populations on Upolu and Savai’i islands could represent distinct distinct represent could islands Savai’i and Upolu on populations Samoan Likewise, 2). - level and higher taxonomic levels in th in levels taxonomic and higher level pple too too ‘This article is protected by sp. Appendix 1, Fig. 1, Fig. Appendix mentary ) difficult to disentangle disentangle to difficult (Uy et al. 2019) al. et (Uy , Pacific Meliphagidae honeyeaters honeyeaters Pacific , Meliphagidae tannensis Dxy

= 0.18 = Material Material

that we recently described from Vanuatu in Kearns in Vanuatu from described recently we that – kleinschmidti gene flow following secondary colonization of Fiji from elsewhere elsewhere ofFiji from colonization secondary following flow gene 0.64%) and Fiji and ( 0.64%) ( 3 Supple Dxy Appendix 1, Fig. 1, Fig. Appendix . –

(Pratt and Mittermeier 2016) Mittermeier and (Pratt 5 P. p. pusilla p. pusilla P. statistical tests for allow not sizessmall did and sample ), our

from Fiji (below). from Fiji (below). sexual

Material compared to other subspecies and species in the the in species and tosubspecies other compared Dxy ePacific 1

plumage coloration mode mode coloration plumage ). (Kearns et al. 2016) al. et (Kearns

= 0.16 = (Andersen et al. al. et 2014a) (Andersen

Denser taxonomic sampling will be required required sampling will be taxonomic Denser neages found on Vanua Levu and Viti and Vanua Levu on found neages Appendix T 1, Appendix — – . e.g., Golden Whistlers( Golden e.g.,

0.96%) (Fig 0.96%) 2; D .

ND2 Dxy ND2 istinctiveness istinctiveness (reviewed (reviewed species as recently proposed proposed asrecently species able A2 ableA2 . Convergence, reversal and and reversal Convergence, . .

= 0.16 = 2). Critically, the low low Critically, the 2). ,

(Fig and by of subspecies subspecies of – et al. et for sample sizes). sizes). sample for

2.5) and there was there and 2.5) Omland and Omland . Solomon Islands Islands Solomon

4 Pachycephala Pachycephala ,

(2015)

based based –

with with ,

Accepted Article (Garnett et al. 2010, BirdLife International 2012) BirdLife International 2010, al. et (Garnett robin Pacific to help and taxa mainland island versus in coloration plumage sexual general in speciation 2009) al. are Pacific the in species to responses be is t island isolated D Anders 2010, Phillimore and al.et Clegg 2009, Moyle 2009, al. et Uy 2008, al. et Jønsson 2008, Kennedy and Jones 2007, Franklin and Filardi Kirchman 2007, and Smith 2005, ‘great techniques molecular Modern University, University, Museum, Burke Washington of University Collection, Wildlife National N of Museum atAmerican Filardi Chris the and Capainolo skins tissues and to access provided Acknowledgements request. upon MK127556 MK121772; (MK121750 numbers accession the following under GenBank isdepositedin data sequence All new Data hat one of the most pronounced signals from from signals most pronounced ofthe one hat ifferences in in ifferences clear clear

unable t unable -

speciators’ speciators’ availability statement that that , but also because these radiations hold historical importance to the fields of ornithology and and of ornithology fields to the importance historical hold these radiations because but also , o determine the extent to which speciation in these insular radiations inradiations these insular speciation which to extent the determine o A more more local local

radiation . Townsend Peterson, Michael Andersen and Mark Robbins at the Kansas University Natural Natural Kansas University the at Robbins and Mark Andersen Michael Peterson, Townsend . how have have robust phylogenetic frameworks frameworks phylogenetic robust of the Pacific ofthe selective pressures selective

natural natural – led to led

(Mayr 1934, 1934, (Mayr Wethank , which ultimately will help to conserve threatened p robin threatened conserve to will help which , ultimately ‘This article is protected by - MK127598; MK248639 MK127598; increasingly threatened by habitat loss, climate change and predation and change loss, climate habitat by threatened increasingly

discordan and sexual Jean Woods at the Delaware Museum of Natural History, Natural of Museum Delaware the at Woods Jean

now show that that show now

Allen Kearns who helped measure specimens, and specimens, measure helped who Kearns Allen have occurred occurred have 1942)

t –

or to random drift and chance. drift and chance. random to or concordan and in particular particular in selection . Our Nate Rice at the Academy of Natural Sciences of Drexel Drexel of Sciences ofNatural the at Academy Rice Nate

the the findings on

and and

. r - Pacific avifauna is one of isof one avifauna Pacific

multiple temporal and spatial scales spatial and temporal multiple adiations

need to to need Paul Sweet, Thomas Trombone, Lydia Garetano, Peter Peter Lydia Garetano, Trombone, Thomas Sweet, Paul available are measurements phenotypic Raw MK248678). t genetic drift drift genetic

here add to add here be developed. bedeveloped. atural History, aturalHistory,

and archipelagic archipelagic interplay interplay

en et al. al. et 2013 en our understanding of the evolution of of the evolution of our understanding phenotypic phenotypic This should be a priority, not just because because just a priority,not be should This

resolve Without such new such data Without Rob Palmer at the Australian theat Australian Palmer Rob in each species and on each each on and species each in radiation

colonizations the the and a, 2013b, a, has has opulations across the Pacific Pacific the opulationsacross taxonom

all institutions that institutionsall that genetic divergence. divergence. genetic occurred via concerted concerted via occurred - specific

(Filardi and Moyle Moyle and (Filardi

Sharon Birks at the the at Birks Sharon

2015a, 2015b) 2015a, y

of the so the of

of idiosyncrasy

the iconic theiconic (Kingsford et (Kingsford sets -

called - we Given Given . will

it Accepted Article Permits interest of Conflicts the paper wrote and specimens; museum historical and intron nuclear data, phenotypic the Author toKEO grant Science Foundation National a and AMK, Funding ofNatural Sciences. Museum University Dittmann Donna and Brumfield Robb Historyand ofNatural Museum Florida the of Ornithology Division the at Webber Tom and Kratter Andrew Steadman, David Museum, History contributions

–

All tissues obtained from overseas institutions were imported under appropriate appropriate permits. under imported were institutions overseas from obtained All tissues

This research was funded by an Australian Biological Resources Study Churchill Fellowship to to Fellowship Churchill Study Resources Biological an Australian by was funded Thisresearch

; A –

–

The auth The ‘This article is protected by

ll authors comment ll authors AMK, KEO and LJ conceived the study and formulated the questions; AMK collected collected the AMK questions; formulated and study the conceived LJ and AMK, KEO ACD facilitated the collection of nuclear intron data; data; intron nuclear of collection the ACDfacilitated ors declare no conflicts of interest. of conflicts no declare ors

the modern the modern

(DEB - 1119506)

. .

at the Louisiana State Louisiana the at AMK analyzed the the data AMK analyzed

Accepted Article Andersen, M. J., Shult, H. T., Cibois, A., Thibault, J. C., Filardi, C. E. and C.E. Filardi, C., J. Thibault, A., T., Cibois, Shult,H. J., M. Andersen, M and C.E. Filardi, C.H., Oliveros, J., M. Andersen, R.G. Moyle, and C.E. Filardi, P., J. Dumbacher, L., Joseph, I., S., Á. Mason, Nyári, J., M. Andersen, phylo molecular A 2014a. R.G. Moyle, A. and Naikatini, J., M. Andersen, Moyle, and C.E. A.,P. Filardi, Hosner, J., M. Andersen, REFEREN Bennett, P. M. and Owens, I. 2002. Evolutionary ecol I. Evolutionary Owens, 2002. and M. P. Bennett, A. S. M. Baker, C., M. Baker, ecology. and phylogeny to relation in dichromatism sexual Avian E. G. 2003. Hill, and V. A. Badyaev, of divergence Contemporary 2014. L. J. Lockwood, and P. Cassey, D., J. Avery, Linnean Society 89: 331 89: Society Linnean island an from Examples songs: 27 34: Systematics and Evolution, Ecology, of Review Annual Biology Avian of Todiramphus Australo in sympatry secondary and diversification sequences. Dwarf complex. species Pachycephalidae) the bird: polytypic the most world's of systematics Molecular 71: 308 Evol. radiation. Polynesian anand isolated paraphyly revealsextensive Meliphagidae) 118 Evol. 83: Phylogenet. Australo major within a relationships novel and paraphyly extensive reveals - C Kingfisher Kingfisher ES -

The Auk 130: 118130: TheAuk ). ). – 315. -

Royal Society Open Science 140375 Science 2: Open RoyalSociety ‘This article is protected by 45: 45: 291 Ceyx lepidus Ceyx

– and Tilghman, L. M. 2006. Differing effects of isolation on evolution of bird of evolution on isolation effects of Differing 2006. L. M. Tilghman, and – 342. – 295. 136.

–

- (Aves: Alcedinidae) inferred from mitochondrial and nuclear DNA nuclear and mitochondrial from inferred Alcedinidae) (Aves: 131. mainland comparison of three species. ofspecies. three comparison mainland -

Zoological Journal of the Linnean Society 170: 566 170: Society the Linnean of Journal Zoological copyright. All rights reserved.’ Variable ofthe Phylogeography 2013. R.G. oyle, ogy of birds: life histories, mating systems and systems and mating histories, of life ogy birds: -

Pacific kingfishers (Aves: Alcedinidae: Alcedinidae: (Aves: kingfishers Pacific R. G. 2015a. Phylogeny of the monarch flycatchers flycatchers the monarch of Phylogeny R. 2015a. G. – 140375. Pachycephala pectoralis/melanura Pachycephala – 49.

geny of Pacific honeyeaters (Aves: ofPacific honeyeaters geny Moyle, R. G. 2015b. Rapid Rapid 2015b. R.G. Moyle, -

Biological Journal of the the of Journal Biological

- island bird plumage. plumage. bird island Pacific radiation. radiation. Pacific -

Mol. Phylogenet. Mol.Phylogenet.

– 588. -

Mol. Mol.

Journal Journal (Aves: (Aves: 2014b. 2014b. Accepted Article Charmantier, A., Wolak, M. E., Grégoire, A., Fargevieille, A. and Doutrelant, C. 2017. Colour ornamentatio Colour C.2017. Doutrelant, and A. Fargevieille, A., M. E., Grégoire, Wolak, A., Charmantier, C.T., Wu, Vaughan, D., Kühnert, J., Heled, R., Bouckaert, Sae and G. Andersson, M., Webster, T., Borge, Robins). (Australasian Petroicidae Family 2007. W.E. Boles, 2012. International BirdLife Diamond, J. 1970. Ecological consequences of island colonization by southwest Pacific birds, I. Types of of Types I. birds, southwest Pacific by colonization ofisland Ecological consequences 1970. J. Diamond, 2000. A. K. and D. Crandall, Posada, M., Clement, phenotypic and genetic drift on and flow gene of influence The Phillimore, B.2010. A. and M. S. Clegg, J. A., S. Thibault, Sonsthagen, B., Slikas, E., Pasquet, R., G. Graves, S., J. Beadell, A., Cibois, niche shifts. shifts. niche 9: 1 Ecol. Mol. 1077 of species two in divergence 1963 R.C.2011. Fleischer, sexes. across variances (co) genetic tit:Quantitative blue the in Ed.). AnalysisPrlic, (A Evolutionary Bayesian for Platform Software A 2: BEAST2014. J. A. Drummond, 1861 two in autosomes and thechromosome Z on divergence 438 ppp. Edicions, Lynx Chickadees. Titsand Picathartes to 12., vol the world, of birds March 19 on Downloaded www.iucnredlist.org. 2013.2. Version extinction. – – – - 1975. 1975. 1861. 1092.

PLoS Comput. Biol. 10:Biol. e1003537 Comput. PLoS -

- Oxford Univ. Press Univ. Oxford

Proc. Natl. Acad. Sci. U.S.A 67: 529 67: U.S.A Sci. Acad. Natl. Proc. – 3. ‘This article is protected by

Charting the course of reed of course the Charting Petroica multicolor Petroica Zosterops . Oxford, UK Oxford, .

in Vanuatu. Vanuatu. in – tre, G. 2005. Contrasting patterns of polymorphism and and of polymorphism patterns G. Contrasting 2005. tre, 6. . In: IUCN 2013. IUCN Red List of Threatened Species. Species. Threatened List of Red IUCN 2013. IUCN In: .

Philos. Trans. R. Soc. Lond., B,Biol. Lond., R.Soc. Philos. Trans. - bl H., Xie, D., Suchard, M. A., Ramba A., M. Suchard, D., Xie, H., ers across the Pacific islands. islands. Pacific ersthe across Ficedula Ficedula

In: del Hoyo, J. et al. (eds), Handbook of the ofthe (eds), al. et J.Handbook Hoyo, In: del -

Heredity 118: 125 118: Heredity 2015. flycatcher species. species. flycatcher

– 134. -

- Genetics 171: Genetics 171:

J. Biog. 38 Biog. J. – 489.

- C., and and C.,

ut, A. and A. ut, Sci. 365: 365: Sci.

. - n

Accepted Article Filardi, C. E. and Moyle, R. G. 2005. Single origin of a paa origin of Single 2005. R.G. Moyle, and C.E. Filardi, cavity topatterns, mating relation in sexual of dimorphism evolution The A. 2000. J. Green, and J. Figuerola, of birds. genomics Molecular evolutionary H. 2007. Ellegren, 200 M. Eaton, P. Crochet, A., Grégoire, W., J. M., Streicher, Paquet, C., Doutrelant, d and isolation refugial on perspective multilocus A C.2006. Moritz, and G. Dolman, Species E. Mayr, 1976. and M. J. Diamond, Griffith, S. C. 2000. A tradeA C.2000. S. Griffith, of evolution the Reconstructing R.2001. P. Grant, hypothesis? null birds: Which of taxonomy Species B.2014. F. Gill, In: Garnet Island). (Norfolk Robin Scarlet G. 2010. Dutson, and K. J. Szabo, T., S. Garnett, Biology. Evolutionary 1997. D. Futuyma, in the house sparrow sparrow house the in 403. (2011) G Dutson Australasia. Anseriformes. the in sympatry and insularity nesting, 130. birds. “monomorphic” islands. on colouration bird of patterns ( skinks 262 –

266. Carlia 5. Human vision fails to distinguish widespread sexual dichromatism among sexually sexually among dichromatism sexual widespread distinguish to fails vision 5.Human

Nature 438: 216 438: Nature ). ). -

The Action Plan for Australian Birds 2010 Birds Australian for Plan The Action ‘This article is protected by Evolution 60: 573 Evolution Passer domesticus - -

off between reproduction and a and condition reproduction between off Proc. Natl. Acad. Sci. U.S.A 102: 10942 102: U.S.A Sci. Acad. Natl. Proc. – 219.

– 582. -

- Sinauer Associates, Sunderland, MA. Sunderland, Associates, Sinauer - area relation for birds of the Solomon archipelago. archipelago. the Solomon of birds for relation area .

Ecology Letters 19: 19: 537 Letters Ecology -

Proc. R. Soc. B 267: 1115 B 267: R.Soc. Proc.

Funct Ecology 14: 1 14: Funct Ecology Pacific bird group and upstream colonization of of and colonization upstream group bird Pacific -

Cytogenetic and Genome Research 117: 120 117: Research Genome and Cytogenetic . CSIRO Publishing, Melbourne. in press. in Melbourne. .CSIRO Publishing, - - –

– A. and Covas, R. 2016. Worldwide Worldwide R.2016. Covas, and A. - The Auk 131: 150 131: Auk The 545. dependent sexually selected ornament ornament sexually selected dependent 10946. – 1119.

–

10.

ivergence in rainforest rainforest in ivergence

– 161. -

t ST, Szabo JK, Szabo t ST, Oikos 92: 385 92: Oikos

PNAS 73: PNAS 73: –

– Accepted Article Hope, G. S. 1996. Quaternary change and historical biogeography of Pacific Islands. Islands. Pacific of biogeography historical and change Quaternary 1996. S. G. Hope, Measurements. and Mechanisms I. Volume Coloration. Bird J. K. 2006. McGraw, and E. G. Hill, Populations. Two Than More for Models Migration with Isolation 2009. J. Hey, the of the peopling of portrait genetic population A founders: World New theof number On 2005. J. Hey, data. multilocus species from trees of inference Bayesian A. 2010. Drummond, and J. Heled, R Heinsohn, Jønsson, K. A., Bowie, R. C. K., Moyle, R. G., Christidis, L., Filardi, C. E., Norman, J. A. and Fjeldså, J. Fjeldså, A.J. and Norman, C.E., Filardi, L., Christidis, R.G., Moyle, R.C.K., Bowie, A., K. Jønsson, evol and convergence Plumage R. 2008. S. Kennedy, W.and A. Jones, Rose and M. Jakobsson, Linked Chromosome Sex A. 2012. Qvarnstr, and L. Gustafsson, W., H., Forstmeier, Schielzeth, A., Husby, the Correlation Reexamining Diversification: and Selection Sexual L. D. 2014. Rabosky, and H. Huang, the Philippines. Philippines. the 1801 structure. population analysis of in multimodality and switching label with dealing ofSexual. the Evolution and Variance Genetic Dich between 165 ppp. Publishing, SPB Academic processes. and patterns The (eds), E. S. Miller, Press. University 9 Americas. Evol. reversal. role sex without 20.

– 27: 570 1806. ., Legge, S. and Endler, J. A. 2005. Evolution: Extreme reversed sexual dichromatism in a bird bird a in dichromatism sexual reversed Extreme Evolution: 2005. A. J. Endler, and S. Legge, ., -

PLoS Biology 3: 3: e193. Biology PLoS – romatism and Speciation Rate in Birds. Birds. in Rate Speciation and romatism 580. -

The Condor 110: 35 TheCondor 110: ‘This article is protected by

nberg, N. A. 2007. CLUMPP: a cluster matching and permutation program for permutation and program matching cluster CLUMPP: a 2007. A. N. nberg, origin and evolution of Pacific island biotas, New Guinea to eastern Polynesia: easternPolynesia: to Guinea New islandbiotas, Pacific of evolution origin and -

Science 309: 617 Science309:

– 44.

– 619. -

Am. Nat. 184: E101 184: Am. Nat. – – 190. 11.

utionary history of the island thrush in thrush the island of history utionary – -

E114. Mol. Biol. Evol. 27: 905 Biol. Evol. Mol. -

In: Keast, A. J. and and J. A. In: Keast,

Bioinformatics 23: 23: Bioinformatics -

Mol. Biol. Mol. -

Harvard Harvard – Accepted Article Kearns, A. M., White, L. C., Austin, J. J. and Omland, K. E. 2015. Distinctiveness of Pacific Robin Robin Pacific of Distinctiveness E. K. 2015. Omland, and J.J. C., L. Austin, White, M., A. Kearns, E. K. Omland, C.and Driskell, A. L., A., Joseph, K., Thierry, M. F., J. Gobbert, Malloy, M., A. Kearns, K. Omland, and F. J. Malloy, C., A. Driskell, C., Baker, J., J. L.,Austin, White,C., L. Joseph, M., A. Kearns, Cortes F., J. Malloy, A., L., Thierry, Joseph, M., A. Kearns, Kingsford, R.T., Watson Kingsford, Bowie K., F. E. L., Barker, Braun, R.T., Kimball, dichro plumage avian of Evolution D.J. 1999. Ligon, and R.T. Kimball, Biology 23: 23: 834 Biology Biodiversity for Issues Policy Conservation Major 2009. A. K. Wilson, and F. H. Recher, B., Raynor, P., H. Possingham, C., Morley, B.G., Mackey, A., D. Keith, M., Gawel, genome. avian the across spread K. Han, perspec DNA.ancient and morphometrics sexualdichromatis of patterns disparate from revealed Vanuatu in subspecies 48131: Evol. Phylogenet. Mol. histor the diversification unravel help introns Nuclear 2019b. 335. p and relationships surprising unlocks DNA species: ancient endangered distinct a Robins are Island Norfolk 2016. E. sp robin highland Guinea’s New of affinities of Diversification complexes species of withinone diversification and phylogenetics Molecular 2008.

t - i L., Harshman, J. and and J. Heimer Harshman, L., ve. ve. -

Am. Nat. 154: 182 154: Am.Nat. – 840. ‘This article is protected by Petroica henotypic discordance within the Australo the within discordance henotypic Pachycephala pectoralis/ melanura pectoralis/ Pachycephala , J. E. M., Lundquist, C. J., Venter, O., Hughes, L., Johnston, E. L., Atherton, J., J., L., Atherton, E. L., Johnston, Hughes, O., Venter, C.J., Lundquist, M., E. J. ,

robins across the Australo the robins across – 54. – -

193. Mol. Phylogenet. Evol. 50: 654 Evol. Phylogenet. Mol. -

Emu 115: 89115: Emu - Torres, V. 2009. V. 2009. Torres, ecies. ecies. , R. C. K., Braun, M. J., Chojnowski, J. L., Hackett, S. J., J., S. J. Hackett, Chojnowski, L., J., M. Braun, R.C.K., , copyright. All rights reserved.’ – 98. -

Emu 119:Emu 205

- Pacific region: first insights into the phylogenetic the phylogenetic into insights first region: Pacific - . A wellA Rodriguez, M. N. and Omland, K. E. 2019a. 2019a. E. Omland, K. N. and M. Rodriguez, -

Journal of Avian Biology 39: 39: 473 Biology ofAvian Journal y of the Australo of y - Pacific Robins. Robins. Pacific - tested set of primers to amplify regions regions amplify to primers setof tested the most geographically variable bird bird variable most geographically the – – 660. 217. matism from a proximate proximate a from matism

in Oceania. Oceania. in

m, plumage colouration, colouration, plumage m, - -

Pacific Pacific Conserv. Genet. 17: 17: 321 Genet. Conserv.

Conservation Conservation Petroica – 478.

robins. robins.

- –

Accepted Article Maddison, D. and Maddison, W. 2003. MACCLADE. W.2003. Maddison, and D. Maddison, connecting for tool conversion data automated an PGDSpider: 2011. L. Excoffier, and L. E. H. Lischer, Leigh specimens Type 2008. M. LeCroy, Ko Vanuatu of structure and genetic phylogeography Comparative 2007. D. J. Franklin, and J.J. Kirchman, Mayr, E. 1963. Animal Species and Evolution. Evolution. Species and Animal E. 1963. Mayr, Species. Of Origin The And Systematics E. 1942. Mayr, Whitne the during collected Birds E. 1934. Mayr, Remeš B., Matysioková, ofgenes. resurrection and death the and Dollo's law 1994. R.A. Raff, C.and E. Raff, C.R., Marshall, pelman, N. M., Mayzel, J., Jakobsson, M., Rosenberg, N. A. and Mayrose, I. 2015. Clumpak: a program program a Clumpak: I. 2015. Mayrose, N. and A. Rosenberg, M., J., Jakobsson, M., Mayzel, N. pelman, American Museum Novitates 714: 1 Novitates 714: Museum American 48: 1322 Biology Avian incubation. during to exposure predators in differences sex by explained is not songbirds 12283. 91: Sci USA Acad Natl g and genetics population 1110 6: Evol Ecol Methods Petroicidae. and Rhipiduridae, Monarchidae, Acanthizidae, Maluridae, Platysteiridae, Muscicapidae, Sylviidae, Passeriformes: 1179 15: Resour. K. across inferences structure population packaging modes clustering and identifying for a and passerine. dove, a rail, a variation in region Control birds: , J. W. and Bryant, D. 2015. PopART: Full 2015. D. Bryant, W.and J. , – ‘This article is protected by 1191. , V. and Cockburn, A. 2017. Broad A. 2017. Cockburn, and V. ,

– 1330. enomics programs. enomics programs. – 1116.

of birds in the American Museum of Natural History. Part 7. 7. Part History. Natural of AmericanMuseum the birdsin of

Bulletin of the American Museum of Natural History 313: 1 History Natural of Museum American the of Bulletin

– 19.

- Harvard University Press, Camb Press, University Harvard feature software for haplotype network construction. construction. network haplotype for software feature y South Sea Expedition. 29, Notes on the on genus Notes 29, Sea Expedition. y South -

Columbia University Press. University Columbia - scale variation in sexual dichromatism in in dichromatism sexual in variation scale -

Mol. Phylogenet. Evol. 43: 14 Evol. Phylogenet. Mol. – 299.

ridge, Massachusetts. ridge,

- -

Journal of Journal Mol. Ecol. Ecol. Mol. – Petroica 287. – 23. - -

Proc Proc

. -

Accepted Article Montgomerie, R.2 Montgomerie, for software TOPALi: G. 2004. McGuire, D. and Husmeier, D., G., Marshall, Rowe, F., Wright, I., Milne, Zealand’s of New phylogeny molecular A D. 2006. Lambert, and H. Miller, sel Sexual P. T. 1995. and P. Redfearn, M. Moulton, McLain, Moulton, K., D. McLain, Sea South Whitney the during G.J. collected Birds 1937. Correia, and Beck, R.H. L., Macmillan, E., Mayr, Omland, K. E., Baker, J. M. and Peters, J. L. 2006. Genetic signatures of Geneticsignatures 2006. L. J. Peters, and M. J. E., Baker, K. Omland, ( in orioles evolution plumage Reconstructing S. 2000. Lanyon, and M. E. K. Omland, of Loss Repeated Reconstructions: Ancestral of Assumptions Standard Two Examining E. K. 1997. Omland, Pleistocene Explosive 2009. J. Diamond, and Smith, C.E. C.E., Filardi, R.G., Moyle, history of Old and New World Holarctic ravens ( ravens WorldHolarctic New Old and of history patterns. in reversal and convergence (Anatini). Ducks Dabbling in Dichromatism speciator.” “great ofa expansion hemispheric http://post.queensu.ca/~mont/color/analyze.html). 1806 20: alignments. DNA multiple within sequences recombinant of identification automatic sequences. DNA mitochondrial on based species islands. oceanic on birds 549 1: Research plumage press. in 912. no. novitates; Museum American Hebrides. New Tanna, from collection a On 32, Expedition.

- –

1807. plumage and dimorphic 008. CLR, version 1.05. Queen’s University, Kingston, Canada. (available at (available Canada. Kingston, University, Queen’s 1.05. CLR, version 008.

– 565. ‘This article is protected by

M. P. and Sanderson, J. G. 1999. Sexual selection and extinction: The fate of fate extinction: The and selection Sexual G. J. 1999. Sanderson, P. M. and

Oikos 74: 27 Oikos74: - monomorphic birds introduced onto islands. islands. onto introduced birds monomorphic -

Evolution 54: 2119 Evolution – 34.

51: 1636 -

- Mol. Phylogene Mol.

IJOES: 14 IJOES: – 1646. ection and the risk of extinction of introduced introduced extinction of of the risk and ection – 2133.

– ). ). 16. -

Mol. Ecol. 15: 795 15: Mol.Ecol.

t. Evol. 40: 844 Evol. t.

intermediate divergence: population population divergence: intermediate Petroica Petroica -

(Aves: Petroicidae) (Aves: Petroicidae) Evolutionary Ecology Ecology Evolutionary Icterus – diversification and and diversification

855. – -

808. Bioinformatics Bioinformatics –

1868. ): repeated repeated ):

Accepted Article Rosenberg, N. A. 2003. distruct: a program for the graphical display of population structure. structure. of population display the for graphical program distruct:a A. N. 2003. Rosenberg, re Theevolutionary L.S. 1979. Berman, R. S. and Borecky, R.J., Raikow, multilocus using structure of population Inference 2000. P. Donnelly, M. and Stephens, K., J. Pritchard, J. C.2016. Mittermeier, and D. H. Pratt, in variation Geographic 1996. T. A. Peterson, the in coloration and size in variation Geographic 2007. A. Peterson, theories new for implications birds: in dimorphism sexual of basis Hormonal 1995. R.V. Short, and I. Owens, Shultz, A. J. and Burns, K. J. 2013. Plumage evolution in relation to light environment in a novel clade of cladeof novel in a environment light to relation in evolution Plumage 2013. J. Burns,K. and J. A. Shultz, of TheDirectory 1999. I. Mason, R.and Schodde, Sanchez J., Rozas, body and dichromatism sexual melanism, of evolution the and Insularity 2010. N. Salamin, and A. Roulin, A. Ducrest, and A. Roulin, the coalescent and other methods. other and coalescent the worldwide the in size 24: 594 Biology Notes 137 4: Ecology assemblage. bowerbird the in muscle data. genotype Archipelago. Samoan ofthe avifauna endemic 156 116: Club Ornithologists' limits. species of review selection. sexual of - DelBarrio, J., Messeguer, X. and Rozas, R. 2003. DnaSP, DNA polymorphism analyses by analyses by polymorphism DnaSP, DNA Rozas, R.2003. and X. Messeguer, J., DelBarrio, -

– Genetics 155: 945 155: Genetics 608. ‘This article is protected by -

TREE 10: 44 TREE - – -

distributed barn owl. owl. barn distributed L. 138.

- 2013. Genetics of colouration in birds. birds. colouration in of Genetics 2013.

Natural History Museum The University of Kansas 40: 1 40: Kansas of TheUniversity Museum History Natural

– 172. - – –

47. Notes on the natural history, taxonomy, and conservation of the ofthe conservation and taxonomy, history, the natural on Notes Bioinformatics 19: 2496 19: Bioinformatics 959. -

The Condor 81: 203 Condor 81: The

sexual dichromatism in birds. birds. in dichromatism sexual -

Journal

Australian Birds: Passerines. Birds: Passerines. Australian -

of Evolutionary Biology 23: 925 Biology ofEvolutionary – – 206. 2497. ournal of Ornithology 128: 217 128: ofOrnithology ournal Turdus poliocephalus poliocephalus Turdus -

Seminars in Cell and Developmental Developmental Cell and Seminarsin

- e - stablishment of a lost ancestral ancestral lost ofa stablishment

Bulletin ofBritish Bulletin The -

CSIRO Publishing. CSIRO

– 17. – complex: A first first A complex: 934.

Molecular Molecular

– 241.

Accepted Article Uy, J. A. C., Cooper, E. A. and Chaves, J. A. 2019. Convergent melanism in populations of a Solomon Island Island Solomon ofa populations in melanism Convergent 2019. A. J. Chaves, and A. E. Cooper, C., A. J. Uy, juveni dimorphic ofsexually inheritance and Selection 2015. A. Husby, R.and Bowman, A., Tringali, S European habits in nesting to relation in dichromatism ofsexual Evolution 2012. J. Moreno, and J.J. Soler, in Variation Morphological and Molecular of Patterns C.E. 2007. Filardi, and C.E. Smith, 2017. J. Burns,K. and J. A. Shultz, Chimpanzees. Genetics of Population Divergence 2005. J.Hey, and Y. Won, Re 2011. J.J. Wiens, C., A. J. Uy, C C., A. J. Uy, tephens, M. and Donnelly, P. 2003. A comparison of bayesian methods for haplotype reconstruction from from reconstruction haplotype for methods ofbayesian comparison A 2003. P. Donnelly, and M. tephens, evaluating Dollo's law. Dollo's law. evaluating Islands. Solomon the species of flycatcher incipient between flycatchers. island in melanism parallel with associated genes are different pigmentation in Mutations 2016. mechanisms. genetic unique by is mediated flycatcher coloration. plumage data. genotype population hypothesis. Wallace's testof A passerines: Birds. Land Island (Aves: Thraupidae). songbirds of family the largest in dichromatism tanagers. Neotropical

Moyle, R. G. and Filardi, C. E. 2009. Plumage and song differences mediate species recognition recognition species mediate differences song Plumage and 2009. C.E. and R.G. Filardi, Moyle, ooper, E. A., Cutie, S., Concannon, M. R., Poelstra, J. W., Moyle, R. G. and Filardi, C. E. C.E. Filardi, and R.G. Moyle, Poelstra,W., R., J. M. Cutie, Concannon, A., S., E. ooper, -

Proc. R. Soc. B 283: 20160731 B 283: R.Soc. Proc. - evolution of lost mandibular teeth in frogs in teeth mandibular lost of evolution ‘This article is protected by -

- 479. Auk 124: The

- Ecology and Evolution 5413 5: Evolution and Ecology

Mol. Phylogenet. Evol. 66: 66: 112 Evol. Phylogenet. Mol. -

Evolution 65: 1283 Evolution -

Am. J. Hum. 73: 1162 Am.Hum. J. The role of sexual and natural selection in shaping patterns of sexual sexual of patterns shaping in selection natural and roleofsexual The

1169. – 5422. - –

125. Emu 119: 242 119: Emu

after more than 200 million years, and re and years, million 200 than more after

Evolution 63: 153 Evolution -

Evolution 71: 1061 71: Evolution – 250. -

Mol. Biol. Evol. 22: 297 Biol. Evol. Mol.

– 164.

– Some Solomon Solomon Some 1622. – 1074.

le – 307. -

Accepted Article JAV (Appendix material Supplementary

‘This article is protected by -

0 2404

- 0 2404 >). Appendix Appendix >). 1 .

Accepted Article except for for except Cartoon (elaborate monochromatism elaborate or females) and males ‘feminized’ (dull dull monochromatism females), males,dull (elaborate dichromatism (S Samoa and divided (NI different three Figure Legends Figure :

pink), pink),

1. Geographic mosaic of sexual dichromatism across the Pacific robin radiation. robin Pacific across the dichromatism sexual mosaic of Geographic 1.

into birds polymorpha r ed recognized subspecies in the Solomon Islands ( Islands Solomon the in subspecies recognized

illustrate illustrate -

mode green capped capped

s )

of sexual of archipelagos ‘This article is protected by

r where the two where two the the male plumage form followed by the female plumage form for all subspecies all form subspecies for plumage the female by form followed maleplumage the obin obin P.goodenovii color .

Sex symbols indicate whether the species/subspecies has marked sexual marked has species/subspecies the whether indicate symbols Sex ation types of of types ‘masculinized’ males and females) males and ‘masculinized’

across the Pacific rob Pacific across the

(RC male male :

blue and

), Pacific Pacific

Vanuatu

(black = elaborate; white = dull) = elaborate;white = (black

–

i.e. i.e. r obin obin

( Norfolk Norfolk V P.pusilla :

orange

Insets illustrate the Insets illustrate r obin obin ), ), , which is which , Fiji (F plumage P. multicolor P. :

purple

)

Accepted Article

‘This article is protected by

Accepted Article hatched hatched cir black by arerepresented haplotypes (Sa). for except subspecies, and size sample by scaled Figure

See inset for key. key. for inset See

2. Phylogeographic structuring of mtDNA ND2 across Paci ND2 the mtDNA of structuring Phylogeographic 2. lines. lines.

soror soror ‘This article is protected by Subspecies with multiple islands sampled are indicated. Predicted unsampled unsampled Predicted indicated. are sampled islands multiple Subspecies with color (VSo) and and (VSo) ed and labeled by subspecies. Labels use the the use Labels subspecies. by labeled and ed

cles and the number of mutations between haplotypes are indicated by indicated are haplotypes between ofmutations the number cles and similis (VSi), and subspecies from the the from subspecies and (VSi), copyright. All rights reserved.’ fi c robin species group. species robin c

fi rst letter of archi of the rst letter SOL

lineage (So) and Samoa Samoa and (So) lineage

Haplotypes are are Haplotypes pelago and and pelago Accepted Article lineages estimated by IMa2. by estimated lineages the as the robin of Pacific subspecies using *BEAST kelinschmidti Fiji, F = Vanuatu, = V (archipelago: bar each below isindicated sample each for origin subspecies and archipelago was K of value predictedthat K Delta K. of eachvalue at a to population a bar) vertical by (represented individual Figure

SOL SOL

3. 3.

and and = Nuclear m Nuclear

Solomon Islands, RCIslands, = Solomon VFS , FT = FT , two

lineages lineages

population ultilocus t ultilocus taveunensis ‘This article is protected by estimated by estimated

est s , , SoK = SoK , s whereas whereas

of of r

ed taxon - capped capped

kulamb IMa2. (D) (D) IMa2. Ln(P)K selected four populations as the best value of K. The K. of value best as the populations four selected Ln(P)K

boundaries. r angrae obin from Australia; subspecies: VA = VA subspecies: from obin Australia; a priori M igration rates (gene flow) flow) igration (gene rates copyright. All rights reserved.’ , SoP SoP , =

(

taxonomic unit taxonomic ) STRUCTURE plots showing assignment of each each of assignment showing plots ) STRUCTURE polymorpha s . ). ( ). (C) (C) B D ) Species tree estimated in in treeestimated Species ) between the between ivergence time time ivergence ambrynensis

SOL estimates estimates

and and the best best the , FK = FK = , VFS

of of

Accepted Article in variation in plumage in plumage variation in D females plumage in ofvariation range Figure

iffering russet iffering 4

(yellow) . Variation in in Variation - headed

from sexually dichromatic, dichromatic, sexually from patterns of of patterns ‘This article is protected by

brightness and hue hue brightnessand (r)

and black and saturation sexual sexual

- headed headed

monochromatic and and monochromatic in Supple

the Pacific robin radiation. therobin Pacific P. p. polymorpha polymorpha P.p. mentary

Material Material

dull are plotted separ areplotted

monochromatic monochromatic Appendix 1, Fig. Fig. 1, Appendix

Boxplots

in in males ately. The range range The ately.

depict the the depict taxa (green) (green) 4 – 5 . .

and Accepted Article 5 Figure

‘This article is protected by

Accepted Article Vanuatu/Fiji/Samoa lineage. Vanuatu/Fiji/Samoa as follows letters three first use the Abbreviations percentage. a represent Values radiation. robin Pacific in the southwest archipelago each within mtDNA ofnet Measures 1. Table Legends Table Fiji sim tan cog fem amb sor Vanuatu den sep kul pol Islands Solomon S F V / Regions Archipelagos VFS SOL NI RC /Lineages Species kle(vanua) kle(viti)

–

RC,

r ed

(viti) - RC pol kle sor

capped capped V - - - - -

‘This article is protected by

(vanua)

r obin; NI, Norfolk NI, obin; 0.85 0.53 2.20 0.66 1.81 amb kul kle NI F - - - - -

divergence between species/lineages, archipelagos/regions and subspecies subspecies and archipelagos/regions species/lineages, between divergence

SOL 0.96 0.32 0.18 0.50 1.81 2.38 0.75 0.66 3.71 3.64 fem bec sep S - - - -

SOL VFS 3.77 3.85 3.11 3.33 0.16 0.80 0.18 0.21 0.53 1.92 1.92 2.52 3.95 3.97 cog den tav r - - -

obin; S obin;

OL 3.04 0.21 0.18 0.21 0.53 3.24 3.22 tan NI , Solomon Islands lineage; VFS, lineage; Islands Solomon , - - -

ames, the first letter for archipelagos and and archipelagos for letter first the ames, kle vanua kle

tav & tav 3.26 0.32 0.32 0.28 0.32 0.64 3.45 3.43 sim RC - - -

Dxy estimates expressed as estimates expressed

Accepted Article upolu savaii vs upolu Samoa tav&vanua bec&viti tav bec

‘This article is protected by 0.21

0.87 - - -

Accepted Article islands Samoa subspecies Fiji subspecies Vanuatu subspecies Solomon Archipelagos ‘***’ codes: symbol using and bold, in is denoted Significance variables. all across the MANOVA for is reported (>F) ( ThePillai statistic variable. morphometric each for reported archipelago. each within islands or subspecies among and archipelagos, 2. Table

MANOVA of morphometric variables calculated between Solomon, Vanuatu, Fiji and Samoa Samoa Fiji and Vanuatu, Solomon, between calculated variables morphometric of MANOVA

< 0.001 ales Fem s Male ales Fem s Male ales Fem s Male ales Fem s Male ales Fem s Male

‘**’ p=0.3584 aF=1.4295, P=0.5174, p=0.6279 aF=0.59604, P=0.11326, p=0.3998 aF=2.9323,P=0.89793, p=0.6049 aF=0.76079, P=0.20486, p=0.005251** aF=2.536, P=1.1294, p=1.113e aF=4.4904, P=1.0785, p=0.0001337*** aF=4.8201, P=1.4242, p=0.003409** aF=3.2518, P=1.0938, p=5.015e aF=4.2849,P=0.56008, p=1.631e aF=8.5844,P=0.59088, Variables All < 0.01

, - - - 06*** 05*** 11***

‘*’

0.05 .

p=0.03533* F=2.9648, . p=0.06802 F=2.2487, p=0.8022 F=0.3321, p=0.6893 F=0.50, p=0.2514 F=1.40, p=0.01355* F=3.73, Length Bill p=0.06828 . p=0.06828 F=4.9247, p=0.7768 F=0.0831, p=0.1225 F=4.5542, p=0.5328 F=0.6489, P), approximate F statistic (F) and probability probability and F statistic(F) approximate P),

p=0.09316 . p=0.09316 F=2.1969, p=0.02773* F=2.8355, p=0.000112*** F=13.655, p=0.0003717*** F=10.561, p=0.0047** F=4.83, p=3.416e F=22.157, Length Wing p=0.8481 F=0.04, p=0.207 F=1.7297, p=0.5157 F=0.54, p=0.9025 F=0.1031,

F statistic and probability (>F) is (>F) probability F statisticand

- 11***

p=0.02692* F=3.1885, p=3.205e F=12.219, p=0.001447** F=8.3379, p=0.0001478*** F=12.471, p=1.428e F=13.046, p=9.336e F=21.057, Tail p=0.1057 F=3.6214, p=0.7559F=0.1, p=0.2006 F=2.6727, p=0.3474 F=1.1122, Length

- - - 07*** 06*** 11***

Monochromatic Dull Monochromatic Elaborate Dichromatic symbol codes: using and bold, in is denoted Significance variables. each red thedividing and other Islands, Solomon the red tests: excluding two into one divided are mode,males monochromatic elaborate modes plumage dull and elaborate dichromatic, within subspecies between calculated variables color plumage of MANOVA 3. Table

plumage colour plumage

Females Males Females (black) Males (all) Males Females Males

variable. The Pillai statistic (P), approximate F statistic (F) and probability (>F) is reported for the MANOVA across all across the for MANOVA is reported (>F) probability and (F) F statistic approximate (P), Pillaistatistic variable.The p=0.05267 . P=1.9243, aF=5.649, p=0.1522 P=1.3788, aF=1.7262, p=3.894e P=2.7959, aF=4.1278, p=0.0001838*** P=2.2924, aF=2.685, p=3.874e P=3.1412, aF=3.9433, p=0.06925 . P=2.9133, aF=1.7873, p=0.04252* P=1.5725, aF=1.5834, Variables All Accepted Article - - 07*** 09***

p=0.1865 F=2.0352, p=0.04972* F=3.7476, p=2.309e F=22.639, p=0.1543 F=1.8466, p=0.2982 F=1.2935, p=0.5617 F=0.7788, p=0.5777 F=0.7327, Brightness

- headed and black and headed - 07***

p=0.06815 . F=3.674, p=0.08177 . F=3.0103, p=0.01092* F=4.2799, p=0.00143** F=6.2883, p=0.0003831*** F=6.835, p=0.238 F=1.6198, p=0.02138* F=3.4403, Saturation Back

- headed morphs of morphs headed

p=0.1835 F=2.0594, p=0.06549 . F=3.3328, p=7.926e F=10.479, p=0.5267 F=0.8183, p=0.3948 F=1.081, p=0.5045 F=0.8845, p=0.07752 . F=2.3705, Hue ‘***’

- 05***

< 0.001

polymorpha , p=0.01067* F=7.8413, p=0.0627 . F=3.3973, p=4.891e F=35.25, p=0.002276** F=5.7757, p=4.998e F=12.113, p=0.02628* F=4.2053, p=0.9121 F=0.2418, Brightness

‘**’

- - headed morph of subspecies subspecies of morph headed 09*** 06***

0.01

. F statistic and probability (>F) is reported for (>F) is reported probability F statistic . and p=0.06309 . F=3.8155, p=0.5774 F=0.5713, p=2.964e F=47.801, p=0.0004898*** F=7.5459, p=3.147e F=61.384, p=0.007161** F=6.2339, p=0.01584* F=3.6991, Saturation ,

‘*’ Crown - - 10*** 13*** <

0.05

p=0.02157* F=6.0553, p=0.0499* F=3.7421, p=1.105e F=18.695, p=0.303 F=1.2901, p=0.3343 F=1.2079, F=4, p=0.03051* p=0.1663 F=1.7591, Hue

- 06***

polymorp

p=0.3662 F=1.1257, p=0.9421 F=0.0599, 05*** F=13.9, p=1.083e p=3.232e F=11.291, p=0.0001039*** F=8.2415, p=0.5448 F=0.809, p=0.2595 F=1.4035, Brightness

05***

from the the from

- . For the the For p=0.2007 F=1.9296, p=0.4415 F=0.8673, *** p=0.0007692 F=7.2755, *** p=0.0002175 F=8.5761, *** p=0.0007906 F=6.1114, p=0.592 F=0.7265, p=0.01232 * F=3.9197, Saturation

Throat

p=0.175 F=2.1287, p=0.3122 F=1.2665, p=0.0003171*** F=8.4457, p=0.002169** F=5.828, p=0.002244** F=5.1352, p=0.2044 F=1.7734, p=0.07079 F=2.444, Hue

Accepted Article Monochromatic Elaborate Dichromatic A3) 1, Table Appendix material (Supplementary s morph headed subspecies. and patch. plumage each for subspecies ofeach sexual of dichromatism the degree in differences mean Standardized 4. Table Monochromatic Dull

ignifican tly different means between the sexes as determined from ANOVA are ANOVA from determined as sexes the between means different tly

males

Subspecies within each plumage mode are denoted by the first three letters of the archipelago archipelago the of letters first the by three aredenoted mode plumage withineach Subspecies For

subspecies , and the other the and other , vansor vanfem vancog sampus van van M) (red solpol M) (black solpol solden fijkle fijbec vanamb solsep solkul

tan sim

polymorpha polymorpha comparing females and the the and females comparing Brightn ess 10.85 18.11 2.00 0.42 0.10 1.65 2.61 3.26 0.02 1.38 0.69 7.44 5.30 6.93

( solpol), solpol), Saturati Back 10.91 16.35 13.79 on . 8.58 6.35 0.62 0.58 0.43 2.59 3.70 2.30 1.38 1.31 2.12

two tests two Hu 4.7 4.7 3.8 5.9 3.2 0.1 0.5 0.7 1.2 0.0 4.3 1.0 0.4 0.5 e

4 3 4 3 4 1 0 7 9 9 5 7 9 5

Brightn

are performed, one comparing females and females comparing one performed, are black ess 9.42 7.65 7.64 5.76 0.68 1.82 0.16 1.24 3.16 1.13 1.79 7.33 4.62 5.74

headed morph headed Saturati Crown 13.38 13.68 on 6.63 9.28 4.35 1.18 1.92 1.24 1.83 1.94 1.04 1.13 5.56 8.93

denoted denoted Hu 2.4 0.5 3.2 2.1 1.1 0.7 1.1 0.6 1.2 0.1 1.1 3.1 3.1 3.4 e

0 0 7 0 1 4 4 6 8 1 9 3 1 8

males

Brightn

ess 20.74 12.78 inbol . 7.63 7.94 4.65 0.69 2.04 2.07 4.45 7.14 6.55 2.96 1.82 4.48

Variables with with Variables

d Saturati Throat on 7.83 7.49 7.34 3.74 2.00 1.22 0.62 3.00 4.39 1.25 2.44 1.87 7.60 5.02

red Hu 2.0 1.9 2.3 3.1 4.7 5.7 1.8 0.9 3.4 6.2 9.1 6.3 5.0 5.3 e

7 2 4 8 8 8 8 2 0 4 7 1 6 2