A reappraisal of the systematic afï¬nities of Socotran, Arabian

A reappraisal of the systematic affinities of Socotran,Arabian and East African scops owls (Otus, Strigidae)

using a combination of molecular, biometric andacoustic data

JEAN-MARC PONS,1,2* GUY M. KIRWAN,3 RICHARD F. PORTER4 & J�EROME FUCHS5

1UMR7205 ‘Origine, Structure et Evolution de la Biodiversit�e’, D�epartement Syst�ematique et Evolution, Mus�eumNational d’Histoire, Naturelle, 55 Rue Buffon, Paris, 75005, France

2Service de Syst�ematique Mol�eculaire, UMS 2700, Mus�eum National d’Histoire Naturelle, 43 rue Cuvier, Paris,75005, France

3Field Museum of Natural History, 1400 South Lakeshore Drive, Chicago, IL 60605, USA4BirdLife International, Wellbrook Court, Girton Road, Cambridge CB3 0NA, UK

5Department of Ornithology and Mammalogy, California Academy of Sciences, 55 Music Concourse Drive,San Francisco, CA 94118, USA

We investigated phylogenetic relationships among Otus scops owls from Socotra Island,the Arabian Peninsula and East Africa using molecular, vocalization and biometric data.The Socotra Scops Owl Otus senegalensis socotranus, currently treated as a subspecies ofthe African Scops Owl Otus senegalensis, is more closely related to the Oriental ScopsOwl Otus sunia and to the endemic Seychelles Scops Owl Otus insularis. Considerablemitochondrial genetic distance and significant morphological differentiation from its twoclosest relatives, as well as its distinctive vocalizations compared with O. insularis,strongly support recognition of Socotra Scops Owl as a full species. Unexpectedly, twotaxa from the Arabian Peninsula, Pallid Scops Owl Otus brucei and African Scops OwlOtus senegalensis pamelae, represent very distinct lineages; O. brucei is basal to a cladethat includes taxa found in the Indo-Malayan region and on Indian Ocean islands. Incontrast, O. s. pamelae occupies a well-supported basal position within a clade of conti-nental Afro-Palaearctic taxa. The uncorrected-p genetic distance between O. s. pamelaeand its closest relatives (other populations of senegalensis from mainland Africa) is c. 4%.As O. s. pamelae is also well differentiated phylogenetically, morphologically and vocallyfrom O. s. senegalensis, we recommend its elevation to species status, as Otus pamelae.Among mainland African O. senegalensis subspecies, Ethiopian populations appear torepresent the most divergent lineage, whereas other lineages from Somalia, Kenya andSouth Africa are poorly differentiated. The large genetic distance between the Ethiopianhaplotype and other African haplotypes (3.2%) suggests that the Ethiopian Otus mayrepresent a cryptic taxon, and we recommend that more individuals be sampled to assessthe taxonomic status of this population.

Keywords: biogeography, Indian Ocean Islands, Indo-Malaya, phylogeny, Socotra Island, taxonomy.

Scops owls (Otus, Strigidae) represent a genus ofsmall to medium-sized nocturnal birds with strongcolonizing capability, suggested by their presence

on small and remote islands. Despite high speciesrichness, the morphological diversity encounteredwithin the genus is limited, with the occurrence ofonly two primary plumage colour types: rufous orgrey. This lack of major plumage differences iscommon among members of the Strigidae and is

*Corresponding author.Email: [email protected]

© 2013 British Ornithologists’ Union

Ibis (2013), doi: 10.1111/ibi.12041

usually explained by the need to remain inconspic-uous during the day as a defence against predatorsor from being mobbed by other birds (Marks et al.1999). As a consequence of their highly conservedmorphology, it is unsurprising that the systematicsof the genus Otus, as traditionally defined, havebeen challenged by molecular data (Fuchs et al.2008, Wink et al. 2009, Miranda et al. 2011).These studies suggested not only that the genusOtus is polyphyletic, because the White-facedScops Owl Otus leucotis is sister to the genus Asio,whereas the New World Screech Owls have affini-ties with the genera Bubo, Strix and the Asio–Otusleucotis clade (Fuchs et al. 2008), but also thatspecies limits in some groups must be reconsidereddue to either divergent mtDNA sequences(Miranda et al. 2011) or paraphyly of traditionallyrecognized species (e.g. Collared Scops Owl O. let-tia; Fuchs et al. 2008).

As currently understood, the genus Otus has itscentre of diversity in the Palaearctic and Oriental(Indo-Malayan) regions (26 species). Secondaryradiations have occurred on the Indian Oceanislands, which are occupied by six species, and inAfrica with four mainland species (Eurasian ScopsOwl Otus scops, African Scops-Owl Otus senegal-ensis, Sokoke Scops Owl Otus ireneae, SandyScops Owl Otus icterorhynchus) and two insularspecies (S~ao Tom�e Scops Owl Otus hartlaubi,Pemba Scops Owl Otus pembaensis) (Marks et al.1999).

Fuchs et al. (2008) suggested that the rapiddivergence following the initial colonization of theIndian Ocean islands (Madagascar and the Como-ros archipelago) led to distinctive genetic lineages,and this result has favoured the recognition ofeach insular Otus taxon as a distinct species(Fig. 1). Intriguingly, molecular DNA sequencedata suggest that several species have strikinglydifferent affinities than recognized under tradi-tional hypotheses, suggesting that morphology andvocalizations are weak characters with which todefine phylogenetic relationships in scops owls.For example, O. pembaensis proved not to be con-specific with Madagascar Scops Owl Otus rutilus,contrary to previous hypotheses (Marks et al.1999), but instead is closely related to mainlandO. senegalensis (Fuchs et al. 2008). Likewise, theS~ao Tom�e endemic Otus hartlaubi was found tohave strong affinities with O. senegalensis. Evenmore surprisingly, the Seychelles Scops Owl Otus

insularis formed a well-supported clade withOriental Scops Owl Otus sunia (Fuchs et al.2008).

Samples of Socotra Scops Owl, currently con-sidered to be a subspecies of O. senegalensis, wereunavailable to Fuchs et al. (2008), and the samewas true of several taxa found on the Arabian Pen-insula (Pallid Scops Owl Otus brucei, and Otus sen-egalensis pamelae) and East Africa (northernSomalia, Ethiopia, Kenya), where several subspe-cies of O. senegalensis have been described on thebasis of differences in voice, plumage colorationand size (Chapin 1930, Meinertzhagen 1948,Keith & Twomey 1968) but are not generally rec-ognized in the modern literature (Kemp 1988,Marks et al. 1999, Dickinson 2003, K€onig et al.2008). The affinities of these taxa are still poorlyknown and this uncertainty is reflected by frequentchanges in their taxonomic treatment (Marks et al.1999, K€onig et al. 2008). Since Sclater (1924), theSocotran population has usually been treated as asubspecies of O. senegalensis (Otus senegalensis soco-tranus), but it is sometimes placed together withPallid Scops Owl O. brucei (e.g. Kemp 1988,K€onig et al. 1999), whose range otherwise encom-passes the eastern part of the Arabian Peninsulaextending east to western Pakistan and north tosoutheastern Turkey and Central Asia (Marks et al.1999). This taxon’s treatment as a subspecies ofO. senegalensis appears to have been influenced bythe report ‘that the song on Socotra is identical tothat in Kenya’ (A. Forbes-Watson in Marshall1978). However, recently K€onig et al. (2008) andPorter and Aspinall (2010) considered socotranusto represent a separate species.

Similarly, scops owls from southern Arabia, cur-rently recognized as Otus senegalensis pamelae,have also been suggested to be closely related toO. brucei (Marks et al. 1999), although Marshall(1978) already noted that this taxon’s vocalizationsare ‘intermediate between the stutter of Otus sen-egalensis senegalensis and the purr of O. sunia di-stans of Thailand.’ Furthermore, the systematicaffinities of O. brucei remain unclear. It has beentreated either as a subspecies of Otus scops (Mein-ertzhagen 1948) or Otus sunia prior to vocal stud-ies that supported its specific status (Marshall1978, Roberts & King 1986).

We here examine further the patterns of pheno-typic differentiation among scops owl taxa,and discuss these with respect to phylogenetic


2 J.-M. Pons et al.

relationship highlighted by our analyses of mito-chondrial and nuclear sequence data.

METHODS

We collected blood samples for the Socotra ScopsOwl and used toe-pads from museum specimensheld at the Natural History Museum, Tring, UK,for O. brucei, O. s. pamelae and O. s. senegalensiscollected in Ethiopia, Somalia and Kenya. In addi-tion, our taxonomic sampling was completed withseveral species included in Fuchs et al. (2008;Table 1).

Laboratory procedures

Total genomic DNA was extracted from blood pre-served in EDTA using a CTAB-based protocol(Winnepenninckx et al. 1993). DNA was extractedfrom toe-pads using a DNeasy Tissue Kit (QiagenInc., Valencia, CA, USA) adding 20 lL of dith-iothreitol (DTT; 0.1 M). Genomic extractions oftoe-pads were performed separately from the DNAextractions of fresh tissues of Socotra Scops Owl.Prior to each extraction, the workstation and allequipment were cleaned using a bleach solution andUV light was used to help prevent contamination.

Two mitochondrial markers, nicotinamidedehydrogenase subunits 2 and 3 (ND2, ND3),and two nuclear introns (myoglobin intron-2(MB) and TGFb2 intron-5 (TGFb2)) wereamplified via the polymerase chain reaction(PCR) and sequenced. ND2 and ND3 werePCR-amplified and sequenced using primer pairsL5219/H6313 (Sorenson et al. 1999) andL10755/H11151 (Chesser 1999), respectively.The MB nuclear intron was PCR-amplified andsequenced using MYO2/MYO3F (Slade et al.1993, Heslewood et al. 1998). TGFb2 was PCR-amplified and sequenced using primer pairs TG5/TG6 (Bures et al. 2002). The following internalprimers were designed to amplify and sequenceDNA obtained from the toe-pads: ND2intL1 5′-CATYAARTRYTTCCTAGTACAAGC-3′; ND2intL2 5′-CAGCYATTGCAATAAAACTAGGAC-3′;ND2intR1 5′-GGACAATGRGACATCACACAA-3′; ND2intR2 5′-CCTAAAYCAAACACAAATCCGA-3′; ND3intL 5′-CTVCCATTCTCARTMC-GATTCTTC-3′; ND3intR 5′-CCTATTCCTYCTRTTY GAYCTRG-3′; MYO2intL 5′-TTATTTACTGWTGGCTAGTTGG-3′; MYO2intR 5′-GGTATGTGAATATCCAAGTTTAGA-3′. We obtainedDNA sequences of 500 and 320 bp from toe-padsamples for ND2 and MB, respectively. To enable

Afro-Palaearctic cladeIndo-Malayan/Indian Ocean clade

Figure 1. Map depicting the geographical range of Otus taxa in the region of interest. Afro-Palaearctic clade: A1 = Otus senegalen-sis pamelae; A2 = Otus scops; A3 = Otus pembaensis; A4 = Otus hartlaubi; A5 = Otus senegalensis. Indo-Malayan/Indian Oceanclade: i1 = Otus brucei; i2 = Otus moheliensis; i3 = Otus pauliani; i4 = Otus capnodes; i5 = Otus mayottensis; i6 = Otus rutilus;i7 = Otus madagascariensis; i8 = Otus senegalensis socotranus; i9 = Otus insularis; i10 = Otus sunia.


Systematics of Otus scops owls 3

Table1.

List

ofsa

mples

used

andGen

Ban

kac

cessionnu

mbe

rsforthefour

loci

analys

ed.Tax

onom

yfollowsMarks

etal.(199

9).New

sequ

ence

s:(T)individu

alsforwhich

theso

urce

ofDNA

was

atoe-pa

dfrom

ahistorical

mus

eum

spec

imen

;(B)individu

alsforwhich

theso

urce

ofDNAwas

bloo

d.

Spe

cies

Vou

cher/Tissu

enu

mbe

rGeo

grap

hica

lorig

inMyo

glob

inTGFB2

ND2

ND3

Bub

obu

boMNHN

24-55

Franc

eEU60

1069

EU60

0949

EU60

1029

EU60

0992

Otusca

pnod

esMNHN

30-10J

Anjou

anEU60

1078

EU60

0957

EU60

1038

EU60

1000

Otusha

rtlaub

iMNHH

32-04G

S~ ao

Tom

� eEU60

1072

EU60

0952

EU60

1032

EU60

0995

Otusinsu

laris

D.Currie

5H21

863

Mah

� e,Sey

chelles

EU60

1059

EU60

0940

EU60

1022

EU60

0983

Otusinsu

laris

D.Currie

5H21

866

Mah

� e,Sey

chelles

EU60

1060

EU60

0941

EU60

1023

EU60

0984

Otusire

neae

MNHN

32-06J

(M.Vira

niC30

577)

Ken

yaEU60

1077

EU60

0956

EU60

1037

EU60

0999

Otuslettialettia

MNHN

JF14

2La

osEU60

1073

EU60

0953

EU60

1033

EU60

0996

Otuslettiaus

surie

nsis

UWBM

7537

9Rus

sia

EU60

1075

EU60

0955

EU60

1035

EU60

0998

Otuslong

icornis

FMNH

4330

20Lu

zon,

Philippine

sEU60

1084

EU60

0952

EU60

1043

EU60

1005

Otuslong

icornis

ZMUC

1142

06Isab

ela,

Philippine

sEU60

1063

Nose

quen

ceEU60

1052

6EU60

0987

Otusmay

ottens

isMNHN

R22

May

otte

EU60

1087

EU60

0965

EU60

1046

EU60

1008

Otusmeg

alotis

FMNH

4330

19Lu

zon,

Philippine

sEU60

1083

EU60

0961

EU60

1041

EU60

1004

Otusmeg

alotis

ZMUC

1142

08Isab

ela,

Philippine

sEU60

1064

EU60

0944

EU60

1027

EU60

0988

Otusmiru

sFMNH

3574

29Minda

nao,

Philippine

sEU60

1099

EU60

0978

EU60

1057

EU60

1020

Otusmoh

eliens

isMNHN

E-135

Moh

� eli

EU60

1086

EU60

0964

EU60

1045

EU60

1007

Otuspa

uliani

MNHN

R24

Grand

eCom

ore

EU60

1100

EU60

0979

EU60

1058

EU60

1021

Otuspe

mba

ensis

MNHN

unca

talogu

edPem

baIsland

EU60

1090

EU60

0967

EU60

1048

EU60

1010

Otuspe

mba

ensis

MNHN

unca

talogu

edPem

baIsland

EU60

1091

EU60

0968

EU60

1049

EU60

1011

Otusmad

agas

carie

nsis

FMNH

3931

49Mad

agas

car

EU60

1082

EU60

0960

EF19

8290

EU60

1003

Otusrutilus

FMNH

4311

50Mad

agas

car

EU60

1068

EU60

0948

EF19

8307

EU60

0991

Otussc

ops

MNHN

23-5F

Franc

eEU60

1079

EU60

0958

EU60

1039

EU60

1001

Otusse

nega

lens

isMVZun

catalogu

edSou

thAfrica

EU60

1098

EU60

0976

,EU60

0977

EU60

1053

EU60

1019

Otussp

iloce

phalus

MNHN

15-58

China

EU60

1080

EU60

0980

EU60

1040

Nose

quen

ceOtussu

nia

MNHN

6-98

Tha

iland

EU60

1081

EU60

0959

EU60

1041

EU60

1002

Strixaluc

oMNHN

24-27

Franc

eEU60

1070

EU60

0950

EU60

1030

EU60

0993

New

sequ

ence

sOtusbruc

eiBMNH,19

79.2.22(T)

Oman

KC13

8811

Nose

quen

ceKC13

8817

KC13

8825

Otusbruc

eiBMNH,19

79.11.06

(T)

Oman

KC13

8811

Nose

quen

ceKC13

8818

KC13

8826

Otuss.

pamelae

BMNH,19

77.21.11

(T)

Sau

diArabia

KC13

8812

Nose

quen

ceKC13

8819

KC13

8827

Otuss.

pamelae

BMNH,19

37.4.17(T)

Sau

diArabia

KC13

8812

Nose

quen

ceKC13

8820

KC13

8828

Otuss.

sene

galens

isBMNH,19

12.12.23

.44(T)

OrrValley,

Ken

yaKC13

8813

Nose

quen

ceKC13

8821

KC13

8829

Otuss.

sene

galens

isBMNH,19

65.M

.501

3(T)

North

Som

alia

KC13

8814

Nose

quen

ceKC13

8822

KC13

8830

Otuss.

sene

galens

isBMNH,19

23.8.7.755

3(T)

North

Som

alia

KC13

8815

Nose

quen

ceKC13

8822

KC13

8830

Otuss.

sene

galens

isBMNH,18

87.11.11

.44(T)

Aby

ssinia,Ethiopia

Nose

quen

ceNose

quen

ceKC13

8823

KC13

8831

Otuss.

soco

tran

usMNHN

Unc

atalog

ued(B)

Mer

Koh

,Soc

otra

Island

KC13

8816

KC13

8810

KC13

8824

KC13

8832

Otuss.

soco

tran

usMNHN

Unc

atalog

ued(B)

Mer

Koh

,Soc

otra

Island

KC13

8816

KC13

8810

KC13

8824

KC13

8832

Otuss.

soco

tran

usMNHN

Unc

atalog

ued(B)

Mer

Koh

,Soc

otra

Island

KC13

8816

KC13

8810

KC13

8824

KC13

8832

BMNH,Natural

History

Mus

eum,Trin

g;FMNH,Field

Mus

eum

ofNatural

History,Chica

go;MNHN,Mus

� eum

Nationa

ld’Histoire

Naturelle,Paris;MVZ,Mus

eum

ofVertebrate

Zoo

logy

,Berke

ley;

UWBM,Unive

rsity

ofWas

hing

ton,

Burke

Mus

eum,Sea

ttle;

ZMUC,Zoo

logica

lMus

eum,Unive

rsity

ofCop

enha

gen.


4 J.-M. Pons et al.

detection of possible contamination, negative con-trols were included during PCR-amplification ofDNA extracts, which utilized the Illustra Hot StartMix RTG (Buckinghamshire, UK).

Determining the phase of alleles

We used PHASE V2.1.1 (Stephens et al. 2001,Stephens & Donnelly 2003) to infer the associa-tion among heterozygous sites for each nuclearlocus and individual. Three runs using differentseed values were performed and results werecompared across runs. We used the recombina-tion model and ran the iterations of the finalrun 10 times longer than for the other two runs.The best haplotype estimate was used for thephylogenetic analyses.

Phylogenetic analyses

Phylogenetic analyses including estimation of indi-vidual gene trees and a concatenated approachwere conducted using maximum likelihood (ML)and Bayesian inference (BI), as implemented inRAXML V7.0.4 (Stamatakis 2006, Stamatakis et al.2008) and MRBAYES 3.1.2 (Ronquist & Huelsen-beck 2003), respectively. The most appropriatemodels of nucleotide substitution were determinedusing TOPALI v2.5 (Milne et al. 2009) and theBayesian information criterion (BIC). Maximumlikelihood and Bayesian analyses under the concate-nated approach were performed, allowing substitu-tion model parameters to differ among loci(Nylander et al. 2004). For the gene tree analyses,we ran several preliminary analyses by changingthe branch-length prior, from unconstrained:exp(10) to unconstrained:exp(200), and the heatingtemperature from 0.2 to 0.1. Better mixing wasalways achieved using a temperature of 0.1 insteadof the 0.2 default value. The best-fit brlens priorwas selected using the Bayes factor (BF; Nylanderet al. 2004). A value > 4.6 for lnBF was consideredto be strong evidence against the simpler model(Jeffreys 1961). For MRBAYES 3.1.2, we useddefault priors for the base frequency and substitu-tions models. For the gene tree analyses, weobserved better mixing of the chains and higherlikelihood using the unconstrained:exp(10) branch-length prior for the mtDNA data and uncon-strained:exp(150) or unconstrained:exp(200) fornuclear data. Analyses of the concatenated datawere performed using an unconstrained:exp(100)

prior for estimation of the branch-lengths. In allMRBAYES analyses, four Metropolis-coupled Markovchain Monte Carlo runs, one cold and threeheated, were conducted for 20 million iterationseach, with trees sampled every 1000 iterations.Two independent Bayesian runs initiated from ran-dom starting trees were performed for each dataset,and the log-likelihood values and posterior proba-bilities were checked to ascertain that the chainshad reached stationarity. We ensured that thepotential scale reduction factor (PSRF) approached1.0 for all parameters and that the average standarddeviation of split frequencies converged towardszero. For the concatenated data, we use the con-sensus sequences from the two alleles for the twonuclear loci, using the IUPAC code for the hetero-zygous sites. The two independent Bayesian runswere combined to estimate the consensus topologyand posterior probabilities of clades.

We used TRACER v1.5 (Rambaut & Drummond2007) to determine that our sampling of the pos-terior distribution had reached a sufficient effectivesample size (ESS > 200) for meaningful parameterestimation.

Genetic distance

Uncorrected-p genetic distances among taxa werecalculated using the concatenated mitochondrialsequences (ND2 and ND3). According to evolu-tionary rates estimated from complete mitochon-drial genomes for Hawaiian honeycreepers (Lerneret al. 2011), the ‘standard’ cytochrome-b geneevolves at a slower rate (0.014 substitutions persite per million years) than ND2 (0.029 s/s/my)and ND3 (0.024 s/s/my) used in the present studyto calculate genetic distances among taxa.

Morphological data

For museum specimens of O. senegalensis, O. s.pamelae, O. s. socotranus and O. sunia (SupportingInformation Table S1), we obtained the followingbiometric measurements: wing length (flattened)and tail length, using a standard metal wing-rulewith a perpendicular stop at zero (measured to0.5 mm), and culmen length (to skull), culmenwidth (at the feathers) and tarsus length (mea-sured from the tarsal joint to the last completescute before the toes diverge), using digital calli-pers (to a precision of 0.01 mm). All measure-ments were taken by G.M.K.



All measurements were log-transformed tonormalize the distribution and to allow forallometry. However, morphological variableswere not size-standardized because size is animportant taxonomic criterion (Helbig 2000,Kaboli et al. 2007) and must be considered withshape in studies of morphological characters(Mosimann & James 1979). Sexual dimorphismwas tested using analysis of variance (ANOVA).Principal component analyses (PCA) using corre-lation matrices were performed with five biomet-ric variables to visualize patterns of covariationin a multivariate space. In addition, discriminantfunction analyses (DFA) were performed using apriori defined groups. Groups included in eachDFA were chosen on the basis of our phyloge-netic results: O. s. socotranus was compared withits closest relatives (O. sunia, O. insularis) andO. s. pamelae was compared with five O. sen-egalensis populations from eastern Africa. All sta-tistical analyses were conducted using STATISTICAversion 6 (StatSoft Inc, Tulsa, OK, USA).

Vocalization

We qualitatively compared sound recordings(n = 38) of O. senegalensis, O. s. pamelae and O.insularis. Localities and sample sizes are pre-sented in Supporting Information Table S2.Recordings were obtained from two online data-bases, Xeno-canto (XC; www.xeno-canto.org)and the Cornell Lab of Ornithology MacaulayLibrary (ML; http://macaulaylibrary.com/), com-mercially available recordings (Chappuis 2000),from colleagues or from personal recordings.Sonograms were created using WAVESURFER

1.8.8p4 (http://sourceforge.net/projects/wavesur-fer/) under the following parameters: FFT win-dow length: 2048 points; window type:Hamming; analysis bandwidth: 57 Hz; and win-dow length: 768 points. Vocalizations are com-mented upon in the Discussion with respect tophylogenetic relationships.

RESULTS

Phylogeny

The length of the alignment, number of individu-als sequenced, DNA substitution models, likeli-hood scores and clock model selected for eachindividual locus are indicated in Table 2.

Total evidenceThe analyses of the concatenated dataset resultedin a 50% majority consensus rule tree that was verysimilar to the mtDNA topology. However, mostnodes received better support values (Fig. 2). TheAfrican species Otus ireneae, endemic to the coastalforests of Kenya and Tanzania, was found to be thesister-taxon to all of the remaining Otus scops owls(Fig. 2). The phylogenetic tree was then dividedinto two clades. One Indo-Malayan clade includedMountain Scops Owl Otus spilocephalus, Otus lettiaand Philippine Scops Owl Otus megalotis, whereasthe other clade consisted of all the western IndianOcean taxa as well as O. scops and all Africanmainland and insular taxa. This clade can bedivided further into two highly supported sub-clades, which comprised Indo-Malayan/IndianOcean taxa and Afro-Palaearctic taxa (Fig. 2).

The two Arabian taxa belonged to differentclades: O. brucei was recovered with moderatesupport to be the first offshoot in the Indo-Mala-yan/Indian Ocean (ML = 80, BI = 0.74), whereasO. s. pamelae grouped with the Afro-Palaearctictaxa in the second clade. Socotra Scops OwlO. s. socotranus grouped with Oriental ScopsOwl and Seychelles Scops Owl (ML = 100,BI = 1.0), but the relationships among thesethree species were not resolved. The latter cladewas closely related to the other Indian Oceantaxa (ML = 88, BI = 0.96). Two species endemicto the Philippines (Luzon Scops Owl Otus longi-cornis and Mindanao Scops Owl Otus mirus)were the sister-group to this mostly Indo-Mala-yan clade (ML = 100, BI = 1.0). The AfricanScops Owl was not monophyletic because theArabian taxon O. s. pamelae was sister to theclade formed by O. scops, O. s. senegalensis, O.hartlaubi and O. pembaensis (ML = 99, BI = 1.0).There was substantial mitochondrial geneticdivergence among samples of O. s. senegalensis;the individual collected in Ethiopia differed byup to 3.2% from other O. s. senegalensis individ-uals included in our study.

mtDNAThe final mtDNA data retained for the analysesincluded 30 individuals and 1392 bp (ND2:1041 bp, ND3: 351 bp). Four individuals werenot included in the analyses because they had thesame haplotype as another conspecific individual(the two Seychelles Scops Owls, the three SocotraScops Owls and the two Pemba Scops Owls). The


6 J.-M. Pons et al.

tree resulting from the concatenated mtDNA datahad uneven support, with most of the relation-ships among species of recent origin having lowsupport (BI = 0.50) whereas the backbone rela-tionships of the Otus clade often received strongsupport (ML = 100, BI = 1.0) (Supporting Infor-mation Fig. S1).

Nuclear dataThe two nuclear loci were phased prior to thephylogenetic analyses and only unique alleles wereretained for tree reconstruction. Sharing of allelesequences was widespread among closely relatedspecies. Three alleles were shared among speciesfor MB, the first being found among several

Table 2. Length of the alignment and DNA substitution models selected for each individual locus.

Locus ND2 ND2 ND3 mtDNA MB TGFb2 Concatenated

Length (bp) 1041 500 351 1392 723 593 2708No. of individuals (haplotypes) 34 (29) 34 (29) 34 (29) 34 (30) 33 (26) 27 (33) 32Substitution model TrN+Γ HKY +Γ K81uf +Γ +I partitioned TrnEf K81 + Γ partitionedML score �6208.53 �2958.01 �2066.58 �8276.17 �1459.86 �1474.96 �10943.43BI harmonic mean �6256.55 �3019.23 �2138.1 �8351.17 �1514.25 �1565.65 �11098.72

Figure 2. Bayesian phylogenetic tree obtained with the concatenated data set (ND2, ND3, MB, TGFb2). Values close to nodesrepresent maximum likelihood bootstrap percentages (ML) and Bayesian posterior probabilities (PP). As outgroups, we usedEurasian Eagle Owl Bubo bubo and Tawny Owl Strix aluco (not shown on the figure).



members of the Indo-Malayan/Indian Ocean clade(O. insularis, O. mirus, O. rutilus, O. longicornis,Karthala Scops Owl O. pauliani and O. s. socotr-anus). The second and third alleles were sharedamong species of the Afro-Palaearctic clade(O. pembaensis, O. s. pamelae and O. s. senegalen-sis; and O. scops and O. s. senegalensis, respec-tively). For TGFb2, alleles were only sharedbetween O. rutilus and Mayotte Scops Owl Otusmayottensis. Overall, there were a greater numberof alleles for TGFb2 than for MB (33 vs. 26).

The tree resulting from the BI of the MB allelesrecovered the same five primary lineages high-lighted by the mtDNA (O. irenenae, O. brucei,Indo-Malayan/Indian Ocean clade, Afro-Palaearcticclade and Indo-Malayan clade) but relationshipsamong these five lineages were not supported.Relationships within these clades formed a polyto-my (Supporting Information Fig. S2).

The TGFb2 tree exhibited greater structure.Otus ireneae diverged at the base of the Otusclade (ML = 80, BI = 1.0) and members of the‘Indo-Malayan clade’ were paraphyletic withrespect to the ‘Indo-Malayan/Indian Ocean’ and‘Afro-Palaearctic’ clades. Members of the two lat-ter clades were recovered in a general polytomy(Supporting Information Fig. S3). Yet, as forother loci, O. s. socotranus clustered with O. su-nia and O. insularis (ML = 43, BI = 0.93)whereas O. s. senegalensis clustered with O. pem-baensis (ML = 75, BI = 0.96). Sequences fromO. s. pamelae and O. brucei could not beobtained for this locus.

The concatenation of the nuclear data usingunphased sequences yielded a 50% majority ruleconsensus tree that was very similar to the MBand TGFb2 gene trees. A major discrepancyinvolved the relationships of Pallid Scops Owl,which clustered with Sokoke Scops Owl at thebase of the Otus clade (ML = 65, PP = 0.89), aresult that was not recovered in any of the sin-gle gene tree analyses. We attribute this result tothe joint effect of MB having limited variation toresolve its relationships and the fact that wecould not sequence TGFb2 for O. brucei. Thethree primary clades within Otus (Indo-Malayan,Afro-Palaearctic, Indian Ocean/Indo-Malayan)were all recovered as monophyletic, althoughsupport was variable (ML = 58, BI = 0.93;ML = 62, BI = 0.94; and ML = 90, BI = 1.0,respectively), but no resolution was achievedwithin each of the clades.

Vocalization

Otus s. socotranusAlthough our sample size was of only three differ-ent recordings, G.M.K. and R.F.P. have considerablefield experience with O. s. socotranus and its vocal-izations (G.M.K. during 1 week of intensive generalavifaunal fieldwork in spring 1993 and R.F.P. during11 visits to the island between 1993 and 2011, for atotal of 6 months). R.F.P. has heard a total of 68 dif-ferent socotranus individuals singing as part offocused fieldwork on the taxon designed to eluci-date its general biology, natural history and popula-tion size. The song is a repeated, low-pitched woopwoop da-pwoorp…woop woop da-pwoorp…woop woopda-pwoorp, each phrase lasting c. 2 s with similar-length intervals between them (Supporting Infor-mation Audio File S1). The two woop notes aremuch less audible, except at close range. This is verysimilar to at least some songs of O. sunia, but less soto O. senegalensis over most of mainland Africa. Wefound no evidence of close vocal similaritiesbetween O. s. socotranus and other populations ofO. senegalensis, including those in Arabia. Acrossmuch of eastern and southern mainland Africa, sen-egalensis gives a soft, purring trill, kjurrrr, lasting upto 1 s (but often just 0.5 s). Inter-phrase intervals aretypically 4–8 s. Sonograms of typical songs are pre-sented in Figure 3.

Otus pamelaeSouthern Arabian O. s. pamelae differs from otherpopulations of senegalensis. The sound is obviouslyhigher pitched than that of eastern and southernAfrican populations, each phrase lasts c. 1 s, andthe notes are harsher sounding, with a morenoticeable tremolo. By contrast, inter-phrase inter-vals are similar to those in eastern and southernAfrican populations. The sole recording availableto us from the Republic of Somaliland appears topresent a similar vocal type to those in southernArabia and is the focus of ongoing research (G.M. Kirwan & R. F. Porter unpubl. data). We donot agree with Marshall (1978) that pamelae is‘intermediate’ between senegalensis and sunia in itssong; it is clearly similar to the former in mostrespects.

Morphology

We did not observe significant sexual dimorphismin morphometric traits (ANOVA, all comparisons


8 J.-M. Pons et al.

P > 0.05, results not shown) except for the billlength of socotranus (F1,17 = 13.92, P = 0.0001).All individuals were thus pooled to perform themultivariate analyses. Descriptive statistics for eachtaxon are shown in Table 3.

PCATable 4 summarizes the eigenvalues, the propor-tion of variance captured by the first two PCs and

factor loadings for each biometric variable andeach PCA analysis. The first two components(PC1, PC2) together accounted for a large amountof the variance (> 70%).

Morphological divergence of O. s. socotranus. PC1explained 54.5% of the variation and described sizedifferences among taxa, each standardized biomet-ric variable, except bill width, being roughly equally

Figure 3. Examples of sonograms from typical songs of Otus taxa: Otus senegalensis socotranus, Socotra Island, Yemen, March1993 (recorder: Peter Davidson/Ornithological Society of the Middle East; Supporting Information Audio File S1); Otus insularis, nearLa Mission, Mah�e, Seychelles, December 2008 (recorder: Daniel F. Lane) (XC26204); Otus sunia, Orullathany, near Thattekkad,Kerala, India, December 2009 (recorder: Yong Ding Li) (XC42319); Otus senegalensis pamelae, Ain Hamran, Oman, January 2010(recorder: Herman van Oosten) (XC44491); Otus senegalensis, Awash National Park, Ethiopia, February 2011 (recorder: David Mar-ques) (XC82660); Otus senegalensis, South Luangwa National Park, Zambia, October 2007 (recorder: Derek Solomon) (XC43146);Otus brucei, Khatmat Milalah, Oman, January 2010 (recorder: Herman van Oosten) (XC44545); Otus scops, H€ameenlinna, Mustila,Finland, June 2011 (recorder: Juha Honkala) (XC82047).



and positively correlated with PC1. PC2 captured acomparatively minor amount of variation (31.5%)and described differences in shape among taxa, billwidth having the highest factor loading. There wasclear separation of PC1 and PC2 scores among so-cotranus, insularis and sunia individuals (ANOVA

PC1, F2,22 = 101.46, P < 0.0001; ANOVA PC2,F2,22 = 31.28, P < 0.0001). Socotra Scops Owlshad shorter wings and a shorter tail than O. insular-is. In addition, PC2 clearly separated O. sunia fromsocotranus, which possessed a relatively larger andwider bill. PC3 (9% of the total variation,F2,22 = 0.17, P = 0.84) and PC4 (4% of the totalvariation, F2,22 = 0.29, P = 0.75) did not provideany clear separation among these three taxa.

Morphological divergence of O. s. pamelae. As for theprevious PCA, each standardized morphometricvariable was positively correlated with PC1 andcontributed roughly equally to PC1, whichexplained most of the variation (49.4%). PC2 cap-tured 27.94% of the variance and was positively cor-related with bill length and bill width. Otus s.pamelae was partially separated from its Africancounterparts along PC1 (ANOVA PC1, F5,27 = 6.74,P = 0.0003). Most O. s. pamelae individuals tendedto be larger in wing and tail length than their Afri-can counterparts. There was no clear separation ofPC2 scores among scops owls whatever their geo-graphical origin. PC3 and PC4, which explained 12and 8% of the total variation, respectively, did notpermit further separation of O. s. pamelae fromAfrican taxa (PC3, F5,27 = 1.66, P = 0.18; PC4,F5,27 = 1.38, P = 0.36).

DFAMorphological divergence of O. s. socotranus (Fig. 4).A multivariate ANOVA was strongly significantTa

ble3.

Des

criptivestatisticsof

morph

olog

ical

varia

bles

includ

edin

themultivariate

analyses

.Sam

plesize

,mea

n(sd).

Afro

-Palae

arctic

clad

eIndo

-Malay

an/In

dian

Oce

anclad

e

Otusse

nega

lens

ispa

melae

Otusse

nega

lens

isse

nega

lens

isOtusse

nega

lens

isso

cotra

nus

Otussu

nia

Otus

insu

laris

Arabian

Pen

insu

laSom

alia

Ethiopia

Ken

yaSud

anUga

nda

Soc

otra

India

Sey

chelles

Wing

leng

th7,

137.6(5.6)

5,12

1.4(2.9)

3,13

0.3(3.5)

7,13

2.1(6.8)

7,13

4.1(4.8)

6,13

1.7(6.1)

21,12

9.3(3.8)

17,14

5.7(3.8)

3,16

4.7(5.9)

Taillen

gth

7,60

.2(3.4)

5,51

.8(0.4)

3,57

(3)

7,57

.2(4.1)

7,60

.1(3.4)

6,60

.2(4)

20,54

.1(4.3)

17,64

.9(3.3)

3,77

(4.3)

Tarsu

sleng

th7,

30.3

(1.9)

4,22

.7(1.5)

3,24

.3(1)

7,24

.5(1.48)

6,25

(2.1)

6,25

(1.7)

13,24

.4(2.3)

9,25

(2)

3,32

.1(0.7)

Billleng

th7,

18.4

(1.4)

5,16

.9(0.6)

3,17

.6(0.6)

7,18

.3(0.8)

7,17

.4(1.2)

6,16

.7(1.3)

21,19

.4(0.9)

16,18

.5(1.1)

3,23

.5(1.3)

Billwidth

7,7.6(0.5)

5,7.5(0.7)

3,7.4(0.6)

7,8.3(0.4)

7,7.8(0.4)

6,7(0.5)

13,9.1(0.8)

16,6.7(0.7)

3,9.9(1.3) Table 4. Eigenvalues, percentage variance explained and fac-

tor loadings of morphometric traits for two principal compo-nents (PC1, PC2): (A) Otus senegalensis socotranus, Otussunia, Otus insularis. (B) Otus senegalensis pamelae, Otussenegalensis senegalensis from eastern Africa.

A – PC1 A – PC2 B – PC1 B – PC2

Eigenvalue 2.73 1.58 2.47 1.39Variance explained 54.54% 31.54% 49.41% 27.94%Factor loadingsWing 0.88 �0.37 0.90 �0.13Tail 0.82 �0.46 0.86 �0.34Tarsus 0.84 0.20 0.79 �0.14Bill length 0.74 0.59 0.51 0.68Bill width 0.12 0.91 0.20 0.87


10 J.-M. Pons et al.

(Wilks’ lambda = 0.011, F10,36 = 30.1, P < 0.0001)and fully separated the three taxa. The variablesthat contributed most to the discriminationbetween groups are wing length (root 1: standard-ized coefficient = 1.06) and bill length (root 2: stan-dardized coefficient = �0.71). Socotran birds wereshorter in wing length than sunia and insularis, andhad a wider and larger bill than sunia. The classifiedmatrix derived from the DFA correctly classified100% of socotranus (n = 7) with high posteriorprobabilities (PP = 1.0).

Morphological divergence of O. s. pamelae (Fig. 5).The first two discriminant functions explained87% of the variability in the dataset. DFA signifi-cantly differentiated biometric characteristicsamong groups (MANOVA, Wilks’ lambda = 0.080,F25,86 = 3.50, P < 0.0001). However, there was noclear pattern of geographical structure amongmainland African Scops Owls. In contrast, O. s.pamelae was clearly separated from the othergroups: O. s. pamelae is larger in wing and tarsus

length than its African counterparts. The classifiedmatrix derived from the DFA correctly classified100% of O. s. pamelae with high posterior proba-bilities (PP > 0.99) for five individuals and lowposterior probabilities for only two individuals(PP < 0.50).

DISCUSSION

Species limits

We recovered the same general topology as that ofFuchs et al. (2008), which was expected given thatthe taxonomic and gene sampling was very similar,with only a few nodes showing less support. Otusbrucei was recovered with moderate support tooccupy a basal position within the Indo-Malayansub-clade, whereas all senegalensis taxa belonged tothe Afro-Palaearctic sub-clade.

The Pallid Scops Owl is considered by someauthorities to be a subspecies of either O. scops or

Figure 4. Discriminant function analysis performed on five biometric variables of 25 scops owls from Socotra (Otus senegalensissocotranus), the Seychelles (Otus insularis) and India (Otus sunia). The variables that most contributed to the discrimination betweengroups are wing length and bill width (standardized coefficients, respectively, 1.06 and �0.40, root 1) and bill length and bill width(standardized coefficients, respectively, �0.71 and �0.65, root 2).



O. sunia, or regarded as distinct, forming a super-species with O. senegalensis, O. scops and O. sunia(Marks et al. 1999). Our molecular results do notsupport such systematic arrangements among thesefour species. Indeed, the Pallid Scops Owl is notclosely related to either O. scops or O. senegalensis,but appears to be the most ancient lineage thatdiverged from the Indo-Malayan and Comoriansub-clade (Fig. 1). Otus brucei is in widespreadsympatry with O. scops without evidence of inter-breeding, and also exhibits unique vocal (for sono-grams see Rasmussen & Anderton 2005: 238) andmorphological characters (Shirihai 1993) that fullysupport specific taxonomic treatment according tothe guidelines proposed by Helbig et al. (2002).

The Socotra Scops Owl is currently treated as asubspecies of O. senegalensis in most of the litera-ture (e.g. Marks et al. 1999, Dickinson 2003,Clements 2007). This taxonomic treatment is notsupported by our molecular results and our qualita-tive analyses of vocalization data. Together with O.

insularis and O. sunia, the Socotra Scops Owl formsa well-supported monophyletic group nested withinthe Indo-Malayan/Indian Ocean clade. SocotraScops Owl vocalizations are reminiscent of O. su-nia, in accordance with the close relationship to thelatter taxon highlighted in our phylogenetic tree.

Mitochondrial uncorrected-p distances from itssister species O. sunia and O. insularis are 4.7 and3.5%, respectively, indicating that the two taxa areclearly genetically distinct. These genetic distancesbased on ND2 and ND3 genes are of the same orderof magnitude as genetic distances usually found inthe cytochrome-b gene between sister-species ofraptors (Wink & Sauer-G€urth 2000). Socotra ScopsOwl is clearly distinguishable morphologically fromits closest relatives; it is significantly smaller than O.sunia and O. insularis in wing and tail length. TheSocotra Scops Owl is further differentiated from O.sunia with respect to its larger bill length and billwidth. Otus senegalensis socotranus is definitivelymonomorphic (occurring solely in a grey-brown

Figure 5. Discriminant function analysis performed on five biometric variables of 33 scops owls from the Arabian Peninsula (Otussenegalensis pamelae) and eastern Africa (Otus senegalensis senegalensis). The variables that most contributed to the discrimina-tion between groups are tarsus length and wing length (standardized coefficients, respectively, 1.02, and 0.89, root 1) and tail lengthand bill width (standardized coefficients, respectively, �0.77 and 0.61, root 2).



morph; G.M. Kirwan & R.F. Porter pers. obs.),unlike O. sunia, which possess two colour morphs(rufous and grey-brown) in most or all populations,even for most or all insular subspecies, e.g. Otussunia japonicus (Japan) and Otus sunia leggei (SriLanka), as well as mainland taxa, e.g. O. s. sunia(northern Indian subcontinent) and Otus s. stictono-tus (Far East Asia) (G.M. Kirwan pers. obs.;Rasmussen & Anderton 2005, K€onig et al. 2008).Overall, our results suggest that this insular taxonrepresents an independent evolutionary lineage.Socotra Scops Owl possesses several independentdiagnostic characters (sensu Helbig et al. 2002) thatstrongly favour its status as a species Otus socotr-anus, supporting the treatment adopted by K€oniget al. (2008), Jennings (2010) and Porter andAspinall (2010).

Another taxon that deserves further attention isO s. pamelae and its position within the Afro-Pal-aearctic clade (ML = 99, BI = 1.0). This southernArabian taxon is highly divergent from African sen-egalensis (uncorrected-p mitochondrial genetic dis-tance = 4%), making O. senegalensis, as currentlydefined, polyphyletic. The song of pamelae is verydifferent from that of O. scops and O. brucei butmore similar to that of O. senegalensis. It neverthe-less differs from the latter’s song in being higherpitched, sounding ‘scratchier’ and having moreprolonged notes; the song sounds two-parted, dueto the much quieter first note (G.M. Kirwan & R.F. Porter pers. obs., K€onig et al. 2008). In terms ofbiometrics, our results clearly suggest that pamelaeis longer winged and longer legged than mainlandAfrican populations of senegalensis. In comparisonwith populations of O. senegalensis in continentalAfrica, Arabian pamelae is distinguished in beingpaler overall, with less distinct streaking over theunderparts and a less obvious whitish line on thescapulars (K€onig et al. 2008). Such plumage char-acteristics might result from adaptation to aridhabitats, as has been demonstrated for other des-ert-dwelling species (Guillaumet et al. 2008). Ourstudy reveals that Arabian Scops Owls possess sev-eral diagnostic genetic and phenotypic characters.We therefore consider the most appropriate taxo-nomic treatment is to recognize Arabian ScopsOwl as a species, Otus pamelae, and not as a sub-species of O. senegalensis as it was originallydescribed based solely on morphological data(Bates 1937).

Among O. s. senegalensis (sensu stricto), theEthiopian lineage is genetically distinct (uncor-

rected-p mitochondrial distance = 3.2%) fromother lineages found in eastern (northern Somaliaand Kenya) Africa and in South Africa. Weincluded only three Ethiopian scops owls in ourbiometric analyses and these were not differenti-ated from their African counterparts. The level ofphenotypic distinctness of the Ethiopian popula-tion remains to be established with a larger samplesize. Furthermore, as our phylogenetic analysesonly included one individual, collected in the late19th century without precise data concerning itsgeographical provenance, we advocate additionalgenetic studies prior to any taxonomic change.

The two Somaliland scops owls sampled weregenetically poorly differentiated from their Kenyanand South African counterparts. However, popula-tions from the highlands of Somaliland are vocallydistinct from other eastern African populationsand possess a similar song to the phylogeneticallydistantly related Arabian O. pamelae. In accor-dance with Fuchs et al. (2008) concerning theaffinities of the Indian Ocean scops owls, the pres-ent study affirmed that vocal differences amongOtus taxa do not always reflect evolutionary affini-ties among lineages.

Biogeographical aspects

That Otus socotranus is most closely related toIndo-Malayan (O. sunia) and Seychelles (O. insu-laris) Scops Owls might appear surprising given itsgeographical location in the Gulf of Aden, closestto the African mainland (Fig. 1). Although Socotrais very distant from other Indian Ocean islands,Socotra Scops Owls are more closely related toother insular Otus taxa than to their Arabian orAfrican counterparts. Surprisingly, of all westernIndian Ocean Islands only Pemba Island, located50 km off mainland Tanzania, was colonized byscops owls from Africa (Fuchs et al. 2008, thisstudy). Having been above sea level consistentlysince 31 mya (Braithwaite 1987), Socotra has mostprobably served as a ‘stepping stone’ betweenAfrica and Arabia for several unrelated organisms(Wranik 2003). However, in the case of Otus, theSocotra archipelago represents the westernmostpoint reached by this Indo-Malayan clade, as itseemingly has never successfully colonized easternAfrica. Furthermore, although usually consideredpart of the Afrotropical zoogeographical realm,Socotra lies close to this realm’s junction with thePalaearctic and Oriental realms, making it unsur-



prising that flora and fauna representative of boththese regions occur on the archipelago (Cheung &DeVantier 2006, Porter & Kirwan 2010). Socotra’smodern diversity is the result of complexbiogeographical patterns in which vicariant as wellas long-distance transoceanic dispersal events haveplayed a major role (G�omez-D�ıaz et al. 2012).Given its long isolation from the Arabian Peninsulaand its ecological characteristics, the archipelagosupports high levels of endemism among severalzoological and botanical groups, leading to theislands being often referred to as the ‘Galapagos ofthe Indian Ocean’. For example, among reptiles,90% of species are endemic (G�omez-D�ıaz et al.2012).

The finding that O. socotranus represents adistinctive species brings to 10 the number ofavian taxa now recognized at the species level asbeing endemic to the Socotran archipelago,thereby significantly enhancing its importance asan Endemic Bird Area (Stattersfield et al. 1998).However, further molecular work on endemicSocotran birds is required to elucidate their phylo-geography. Although most of these taxa appear tobe Afrotropical in origin, the limited available dataare as yet largely inconclusive in determining this.For example, Socotra Sunbird Nectarinia balfourilies outside both clades of Indian Ocean Nectari-niidae and taxon sampling was insufficientlydetailed to resolve its position and closest relatives(Warren et al. 2003). Furthermore, although thetwo endemic sparrows, Socotra Sparrow Passer insu-laris and Abd Al Kuri Sparrow Passer hemileucushave been sampled (Ryan et al. 2010), most oftheir presumed closest relatives in the Passer motit-ensis complex (Kirwan 2008) have not. On theother hand, the endemic Socotra Buzzard Buteo so-cotranus does indeed appear to be most closelyrelated to Afrotropical Buteo (Riesing et al. 2003,Porter & Kirwan 2010).

The two Arabian taxa, O. brucei and O. pamelae,were the first taxa to diverge within their respectiveIndo-Malayan/Indian Ocean and Afro-Palaearcticclades. This unexpected finding advocates the needfor fresh studies to understand the role played bythe Arabian Peninsula in the evolution of Otus lin-eages, and birds in general.

Many studies of mammalian and reptilian taxawith disjunct distributions in the Horn of Africa andsouthwest Arabia have demonstrated that coloniza-tion most frequently occurred from Africa into theArabian Peninsula (Portik & Papenfuss 2012). Such

work has also revealed the role played by the RedSea as a more or less effective biogeographical bar-rier. Our results clearly eliminate a possible Africanorigin of Arabian pamelae and underline the impor-tant biogeographical role of the Red Sea in prevent-ing any gene flow between African senegalensis andtheir Arabian counterparts.

Our limited genetic data on the Ethiopian sen-egalensis lineage imply a long period of genetic iso-lation from other eastern and southern AfricanScops Owl populations. Additional studies areclearly needed to investigate relationships amongOtus scops owls from the Horn of Africa. How-ever, our preliminary results fully agree with previ-ous observations that the Rift Valley stronglyimpacted the phylogeographical structure of sev-eral species distributed in this region (e.g. Galeridalarks, Guillaumet et al. 2008, Ostrich Struthiocamelus, Freitag & Robinson 1993, Lanius shrikes,Fuchs et al. 2011). The Horn of Africa is animportant hotspot of avian endemism in Africa,with several Endemic Bird Areas having been iden-tified (Stattersfield et al. 1998). Thirty bird speciesendemic to Ethiopia and Eritrea alone are knownto date (Ash & Atkins 2009, Redman et al. 2009)and additional cryptic species probably remain tobe identified in this part of Africa.

We thank Mark Adams at the Natural History Museum,Tring, for expediting the toe-pad samples of variousscops owl populations. For access to relevant specimenmaterial at the same institution, G.M.K. is indebted toRobert Prys-Jones and Mark Adams; at the SmithsonianInstitution, National Museum of Natural History, inWashington, DC, James P. Dean and Christina A. Geb-hard afforded similar assistance, and at the FieldMuseum of Natural History, Chicago, G.M.K. wasassisted by John Bates, Mary Hennen and David Willard.Fieldwork by G.M.K. in Yemen and Socotra was under-taken within the framework of the 1993 OrnithologicalSociety of the Middle East (OSME) South Yemen expe-dition, while his museum studies in North America werepartly supported by a grant from OSME and furtherenabled by a scholarship award from the Field Museumof Natural History (Chicago). Additional vocal materialwas kindly supplied by Nik Borrow, Peter Davidson, JonHornbuckle, Michael Mills and Dave Showler. Willem-Pier Vellinga (of Xeno-canto) very kindly assisted in pre-paring the sonograms for this paper. We are also gratefulto C�eline Bonillo and to the technical staff of the ‘Ser-vice de Syst�ematique Mol�eculaire’ for helping with labo-ratory work. We thank Daniel Pons who helped to drawFigures 1 and 3 and two anonymous referees, GaryVoelker and Rauri Bowie for helpful comments on anearlier version of the manuscript. This project was



supported by the network ‘Biblioth�eque du Vivant’funded by the CNRS, the Mus�eum National d’HistoireNaturelle, the INRA and the CNS (Centre National deS�equenc�age).

REFERENCES

Ash, J. & Atkins, J. 2009. Birds of Ethiopia and Eritrea: AnAtlas of Distribution. London: Christopher Helm.

Bates, G.L. 1937. Descriptions of two new races of Arabianbirds: Otus senegalensis and Chrysococcyx klaasi arabicus.Bull. Brit. Orn. Club 57: 150–151.

Braithwaite, C.J.R. 1987. Geology and palaeogeography of theRed Sea Region. In Edwards, A.J. & Head, S.M. (eds) KeyEnvironments: Red Sea: 22–44. Oxford: Pergamon Press.

Bures, S., Dvornik, P. & Saetre, G.-P. 2002. Hybridizationand apparent hybridization between meadow pipit (Anthuspratensis) and water pipit (A. spinoletta). Hereditas 136:254–256.

Chapin, J.P. 1930. Geographic variation in the African ScopsOwl. Am. Mus. Novit. 412: 1–11.

Chappuis, C. 2000. African Bird Sounds, 2: West and CentralAfrica. CDs. Paris: Soci�et�e d’�Etudes Ornithologiques deFrance & London: British Library National Sound Archive.

Chesser, R.T. 1999. Molecular systematics of theRhinocryptid genus Pteroptochos. Condor 101: 439–446.

Cheung, C. & DeVantier, L. 2006. Socotra: A Natural Historyof the Islands and their People. Hong Kong: OdysseyBooks.

Clements, J.F. 2007. The Clements Checklist of the Birds ofthe World. Ithaca, NY: Cornell University Press.

Dickinson, E.C. (ed.) 2003. The Howard and Moore CompleteChecklist of the Birds of the World, 3rd edn. London:Christopher Helm.

Freitag, S. & Robinson, T.J. 1993. Phylogeographic patternsin mitochondrial-DNA of the Ostrich (Struthio camelus). Auk110: 614–622.

Fuchs, J., Pons, J.-M., Goodman, S.M., Bretagnolle, V.,Melo, M., Bowie, R.C.K., Currie, D., Safford, R., Virani, M.Z., Thomsett, S., Hija, A., Cruaud, C. & Pasquet, E. 2008.Tracing the colonization history of the Indian Ocean scops-owls (Strigiformes: Otus) with further insight into the spatio-temporal origin of the Malagasy avifauna. BMC Evol. Biol. 8:197.

Fuchs, J., Crowe, T.M. & Bowie, R.C.K. 2011.Phylogeography of the Fiscal Shrike (Lanius collaris): anovel pattern of genetic structure across the arid zones andsavannas of Africa. J. Biogeogr. 38: 2210–2222.

G�omez-D�ıaz, E., Sindaco, R., Pupin, F., Fasola, M. &Carranza, S. 2012. Origin and in situ diversification inHemidactylus geckos of the Socotra Archipelago. Mol. Ecol.21: 4074–4092.

Guillaumet, A., Crochet, P.-A. & Pons, J.-M. 2008. Climate-driven diversification in two widespread Galerida larks. BMCEvol. Biol. 8: 32.

Helbig, A. 2000. Was ist eine Vogelart? Ein Beitrag zuraktuellen Diskussion um Artkonzepte in der OrnithologieLimicola 14: 57–79.

Helbig, A.J., Knox, A.G., Parkin, D.T., Sangster, G. &Collinson, M. 2002. Guidelines for assigning species rank.Ibis 144: 518–525.

Heslewood, M.M., Elphinstone, M.S., Tidemann, S.C. &Baverstock, P.R. 1998. Myoglobin intron variation in theGouldian Finch Erythrura gouldiae assessed by temperaturegradient gel electrophoresis. Electrophoresis 19: 142–151.

Jeffreys, H. 1961. The Theory of Probability, 3rd edn. Oxford:Oxford University Press.

Jennings, M.C. 2010. Atlas of the Breeding Birds of Arabia.Fauna of Arabia, Vol. 25. Riyadh & Frankfurt: KingAbdulaziz City for Science & Technology & Fauna of Arabia.

Kaboli, M., Aliabadian, M., Guillaumet, A., Roselaar, C.S. &Prodon, R. 2007. Ecomorphology of the wheatears (genusOenanthe). Ibis 149: 792–805.

Keith, S. & Twomey, A. 1968. New distributional records ofEast African birds. Ibis 110: 537–548.

Kemp, A.C. 1988. Strigidae, typical owls. In Fry, C.H., Keith,S. & Urban, E.K. (eds) The Birds of Africa, Vol. 3: 110–155.London: Academic Press.

Kirwan, G.M. 2008. Studies of Socotran birds III.Morphological and mensural evidence for a ‘new’ species inthe Rufous Sparrow Passer motitensis complex endemic tothe island of Abd ‘Al-Kuri, with the validation of Passerinsularis Sclater & Hartlaub, 1881. Bull. Brit. Orn. Club 128:83–93.

K€onig, C., Weick, F. & Becking, J.-H. 1999. Owls: A Guideto the Owls of the World. Robertsbridge: Pica Press.

K€onig, C., Weick, F. & Becking, J.-H. 2008. Owls: A Guideto the Owls of the World, 2nd edn. London: ChristopherHelm.

Lerner, H.R.L., Meyer, M., James, H.F., Hofreiter, M. &Fleischer, R.C. 2011. Multilocus resolution of phylogenyand timescale in the extant adaptive radiation of HawaiianHoneycreepers. Curr. Biol. 21: 1838–1844.

Marks, J.S., Cannings, R.J. & Mikkola, H. 1999. FamilyStrigidae (typical owls). In del Hoyo, J., Elliott, A. & Sargatal,J. (eds) Handbook of the Birds of the World, Vol. 5: 76–243.Barcelona: Lynx Edicions.

Marshall, J.T. 1978. Systematics of smaller Asian night birdsbased on voice. Orn. Monogr. 25: 1–57.

Meinertzhagen, R. 1948. On the Otus scops (Linnaeus)group, and allied groups, with special reference to Otusbrucei (Hume). Bull. Brit. Orn. Club 69: 162–165.

Milne, I., Lindner, D., Bayer, M., Husmeier, D., McGuire, G.,Marshall, D.F. & Wright, F. 2009. TOPALIv2: a richgraphical interface for evolutionary analyses of multiplealignments on HPC clusters and multi-core desktops.Bioinformatics 25: 126–127.

Miranda, H.C. Jr, Brooks, D.M. & Kennedy, R.S. 2011.Phylogeny and taxonomic review of Philippine lowlandscops owls (Strigiformes): parallel diversification of highlandand lowland clades. Wilson J. Orn. 123: 441–453.

Mosimann, J.E. & James, F.C. 1979. New statistical methodsfor allometry with application to Florida Red-wingedBlackbirds. Ecology 33: 444–459.

Nylander, J.A.A., Ronquist, F., Huelsenbeck, J.P. &Nieves-Aldrey, J.L. 2004. Bayesian phylogenetic analysisof combined data. Syst. Biol. 53: 47–67.

Porter, R.F. & Aspinall, S. 2010. Field Guide to the Birds ofthe Middle East, 2nd edn. London: Christopher Helm.

Porter, R.F. & Kirwan, G.M. 2010. Studies of Socotran birdsVI. The taxonomic status of the Socotra Buzzard. Bull. Br.Orn. Club 130: 116–131.



Portik, D.M. & Papenfuss, T.J. 2012. Monitors cross the RedSea: the biogeographic history of Varanus yemenensis. Mol.Phylogenet. Evol. 62: 561–565.

Rambaut, A. & Drummond, A.J. 2007. Tracer v1.5. Availablefrom http://tree.bio.ed.ac.uk/software/tracer/

Rasmussen, P.C. & Anderton, J.C. 2005. Birds of SouthAsia: The Ripley Guide. Barcelona: Lynx Edicions &Washington, DC: Smithsonian Institution.

Redman, N., Stevenson, T. & Fanshawe, J. 2009. Birds ofthe Horn of Africa: Ethiopia, Eritrea, Djibouti, Somalia, andSocotra. Princeton: Princeton University Press.

Riesing, M.J., Kruckenhauser, L., Gamauf, A. & Haring, E.2003. Molecular phylogeny of the genus Buteo (Aves:Accipitridae) based on mitochondrial marker sequences.Mol. Phylogenet. Evol. 27: 328–342.

Roberts, T.J. & King, B. 1986. Vocalizations in the owls ofthe genus Otus in Pakistan. Ornis Scand. 17: 299–305.

Ronquist, F. & Huelsenbeck, J.P. 2003. MrBayes 3:Bayesian phylogenetic inference under mixed models.Bioinformatics 19: 1572–1574.

Ryan, P.G., Nupen, L., Rose, B. & Suleiman, A.S. 2010.Geographic variation in Socotra Sparrows Passer insularis.Bull. Brit, Orn. Club 130: 75–82.

Sclater, W.L. 1924. Systema Avium Aethopicarum. Part 1.London: British Ornithologists’ Union.

Shirihai, H. 1993. Separation of Striated Scops Owl fromEurasian Scops Owl. Br. Birds 86: 286–287.

Slade, R.W., Moritz, C., Heideman, A. & Hale, P.T. 1993.Rapid assessment of single-copy nuclear DNA variation indiverse species. Mol. Ecol. 2: 359–373.

Sorenson, M.D., Ast, J.C., Dimcheff, D.E., Yuri, T. &Mindell, D.P. 1999. Primers for a PCR-based approach tomitochondrial genome sequencing in birds and othervertebrates. Mol. Phylogenet. Evol. 12: 105–114.

Stamatakis, A. 2006. RAxML-VI-HPC: Maximum likelihood-based phylogenetic analyses with thousands of taxa andmixed models. Bioinformatics 22: 2688–2690.

Stamatakis, A., Hoover, P. & Rougemont, J. 2008. A rapidbootstrap algorithm for the RAxML web-servers. Syst. Biol.75: 758–771.

Stattersfield, A.J., Crosby, M.J., Long, A.J. & Wege, D.C.1998. Endemic Bird Areas of the World: Priorities forBiodiversity Conservation. Cambridge: BirdLife International.

Stephens, M. & Donnelly, P. 2003. Comparison of Bayesianmethods for haplotype reconstruction from populationgenotype data. Am. J. Hum. Gen. 73: 1162–1169.

Stephens, M., Smith, N.J. & Donnelly, P. 2001. A newstatistical method for haplotype reconstruction frompopulation data. Am. J. Hum. Gen. 68: 978–989.

Warren, B.H., Bermingham, E., Bowie, R.C.K., Prys-Jones,R.P. & Th�ebaud, C. 2003. Molecular phylogeographyreveals island colonization history and diversification ofwestern Indian Ocean sunbirds (Nectarinia: Nectariniidae).Mol. Phylogenet. Evol. 29: 67–85.

Wink, M. & Sauer-G€urth, H. 2000. Advances in the molecularsystematics of African raptors. In Chancellor, R.D. &Meyburg, B.-U. (eds) Raptors at Risk, Proceedings of the VWorld Conference on Birds of Prey and Owls: 135–147.Berlin: Hancock House Publishers.

Wink, M., El-Sayed, A.-A., Sauer-G€urth, H. & Gonzalez, J.2009. Molecular phylogeny of owls (Strigiformes) inferredfrom DNA sequences of the mitochondrial cytochrome b andthe nuclear RAG-1 gene. Ardea 97: 581–591.

Winnepenninckx, B., Backeljau, T. & de Wachter, R. 1993.Extraction of high molecular weight DNA from mollusks.Trends Genet. 9: 407.

Wranik, W. 2003. Fauna of the Socotra Archipelago. Rostock:Universit€at Rostock.

Received 14 July 2012;revision accepted 12 January 2013.Associate Editor: Gary Voelker.

SUPPORTING INFORMATION

Additional Supporting Information may be foundin the online version of this article:

Audio S1. A typical song of Otus socotranusrecorded by Peter Davidson, Ornithological Soci-ety of the Middle East

Figure S1. 50% majority rule consensus treeresulting from the Bayesian analyses of the concat-enated mtDNA dataset.

Figure S2. 50% majority rule consensus treeresulting from the Bayesian analyses of myoglobinintron-2.

Figure S3. 50% majority rule consensus treeresulting from the Bayesian analyses of TGFb2intron-5.

Table S1. Information on museum specimensincluded in the biometric analyses.

Table S2. Information on sound recordingsincluded in this study.



A reappraisal of the systematic afï¬nities of Socotran, Arabian

Documents

Transcript of A reappraisal of the systematic afï¬nities of Socotran, Arabian