Molecular Phylogenetics and Evolution 82 (2015) 426–435

Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier .com/locate /ympev

Morphology and mitochondrial phylogenetics reveal that the AmazonRiver separates two eastern squirrel monkey species: Saimiri sciureus andS. collinsi

http://dx.doi.org/10.1016/j.ympev.2014.09.0201055-7903/� 2014 Elsevier Inc. All rights reserved.

⇑ Corresponding author at: Museu Paraense Emílio Goeldi, Coordenação deZoologia, Setor de Mastozoologia, CZO, Av. Perimetral n� 1901, 66077-530 Belém,PA, Brazil

E-mail addresses: michellemercesmasto@gmail.com (M.P. Mercês), jlynchalfaro@ucla.edu (J.W. Lynch Alfaro), wallaxaugusto@gmail.com (W.A.S. Ferreira),harada@ufpa.br (M.L. Harada), cazuza.junior@gmail.com (J.S. Silva Júnior).

Michelle P. Mercês a,b,⇑, Jessica W. Lynch Alfaro c,d, Wallax A.S. Ferreira b, Maria L. Harada b,José S. Silva Júnior a

a Museu Paraense Emílio Goeldi, Belém, PA, Brazilb Universidade Federal do Pará, Belém, PA, Brazilc Institute for Society and Genetics, University of California, Los Angeles, CA 90095, United Statesd Department of Anthropology, University of California, Los Angeles, CA, United States

a r t i c l e i n f o a b s t r a c t

Article history:Received 1 October 2013Revised 18 September 2014Accepted 19 September 2014Available online 20 October 2014

Keywords:AmazonCebidaePlatyrrhiniTaxonomyPhylogenetic Species ConceptPrimates

Saimiri has a complicated taxonomic history, and there is continuing disagreement about the number ofvalid taxa. Despite these controversies, one point of consensus among morphologists has been that theeastern Amazonian populations of squirrel monkeys form a single terminal taxon, Saimiri sciureus sciureus(Linnaeus, 1758). This group is distributed to both the north and south of the middle to lower AmazonRiver and in the Marajó Archipelago. However, a recent molecular study by Lavergne and colleagues sug-gested that the Saimiri sciureus complex (comprised of S. s. sciureus sensu lato, S. s. albigena, S. s. macrodon,and S. s. cassiquiarensis) was paraphyletic. The discordance between morphological and molecular studiesprompted us to conduct a new multidisciplinary analysis, employing a combination of morphological,morphometric, and molecular markers. Our results suggest the currently recognized taxon S. s. sciureuscontains two distinct species, recognized by the Phylogenetic Species Concept: Saimiri sciureus (Linnaeus,1758) and Saimiri collinsi Osgood, 1916. East Amazonian squirrel monkeys north of the Amazon have agray crown (S. sciureus), and south of the Amazon, the crown is yellow (S. collinsi). Morphometric mea-surements also clearly distinguish between the two species, with the most important contributing factorsincluding width across upper canines for both sexes. For males, the mean zygomatic breadth was signif-icantly wider in S. sciureus compared to S. collinsi, and for females, the width across the upper molars waswider in S. sciureus compared to S. collinsi. Mitochondrial phylogenetic analyses support this separation ofthe eastern Amazonian squirrel monkeys into two distinct taxa, recovering one clade (S. sciureus) distrib-uted to the north of the Amazon River, from the Negro River and Branco River to the Guiana coast and theBrazilian state of Amapá, and another clade (S. collinsi) south of the Amazon River, from the region of theTapajós River to the state of Maranhão, as well as within the Marajó Archipelago. The revalidation of thespecies S. collinsi was corroborated by all of the methods in the study, as the clades recovered in ourmolecular study are congruent with the pattern of morphological variation. We confirm both the para-phyly of the Saimiri sciureus complex and the paraphyly of the subspecies S. s. sciureus as defined inthe current literature.

� 2014 Elsevier Inc. All rights reserved.

1. Introduction come to the ground or use higher canopy levels (Boinski, 1989;

Squirrel monkeys (genus Saimiri Voigt, 1831) are arboreal Neo-tropical primates that use the midlevel of the forest and occasionally

Stone, 2006). They are found throughout the Amazon Basin, includ-ing some areas in transition to the Cerrado biome, as well as in asmall part of Central America (Hershkovitz, 1984; Silva Júnior,1992).

Recent systematic revisions of Saimiri encompass many dis-agreements about the number of valid taxa and the status of thesetaxa (i.e., species vs. subspecies), as well as about their taxonomicrelationships with one another (Hershkovitz, 1984; Thorington,1985; Silva Júnior, 1992; Costello et al., 1993; Boinski and Cropp,

M.P. Mercês et al. / Molecular Phylogenetics and Evolution 82 (2015) 426–435 427

1999; Lavergne et al., 2010; Rylands et al., 2012, 2013; Paglia et al.,2012). Two factors which may contribute to discrepancies acrossstudies include the existence of significant intrapopulational vari-ation in Saimiri and the use of different geographical samples indifferent morphological analyses. Additional long-standing issueshave been the paucity of samples available for analysis from someareas within the Amazon and the fact that hybridization is knownto occur between some species of squirrel monkeys (for example, S.macrodon and S. boliviensis: Hershkovitz, 1984; Thorington, 1985;Silva et al., 1992; Costello et al., 1993).

According to the taxonomic arrangement proposed by Rylandset al. (2000), Groves (2005) and Rylands and Mittermeier (2009),five species are included within Saimiri: S. sciureus, S. oerstedii, S.ustus, S. vanzolinii and S. boliviensis. Under this taxonomy, S. s. sciu-reus sensu lato, S. s. macrodon, S. s. cassiquiarensis, and S. s. albigenaare all subspecies of Saimiri sciureus, or what we call the ‘S. sciureuscomplex’. However, Rylands and Mittermeier (2009) and Rylandset al. (2012) emphasize that this list might change, both throughnew discoveries and through more complete studies with bettersampling and improved methods. In fact, in the most recentreviews, Paglia et al. (2012) recognize seven species (Saimiri sciure-us, S. collinsi, S. cassiquiarensis, S. macrodon, S. ustus, S. boliviensisand S. vanzolinii) within Brazil, and Rylands et al. (2013) recognizeseven total species with 11 terminal taxa: S. oerstedii oerstedii, S. o.citrinellus, S. cassiquiarensis cassiquiarensis, S. c. albigena, S. macr-odon, S. ustus, S. sciureus sciureus, S. s. collinsi, S. boliviensis bolivien-sis, S. b. peruviensis, and S. vanzolinii.

Despite taxonomic disagreements about Saimiri as a whole,there has been a general consensus among morphologists withrespect to the eastern Amazonian populations of Saimiri(Hershkovitz, 1984; Thorington, 1985; Silva Júnior, 1992;Costello et al., 1993; Boinski and Cropp, 1999): they are consideredone terminal taxon, Saimiri sciureus sciureus (Linnaeus, 1758). Thisgroup occurs throughout the middle to lower Amazon on bothsides of the Amazon River, as well as in the Marajó Archipelago.To the north of the Amazon River, the distribution extends fromthe forests on the coast of the Guianas and the Brazilian state ofAmapá to the left bank of the lower Negro River and the BrancoRiver. South of the Amazon River, the western, southern and south-eastern distributional limits are still unknown, but this group canbe observed at least from the right bank of the Tapajós River tothe Zona dos Cocais in Maranhão state, Brazil (Silva Júnior,1992). Field observations in São Roberto, Maranhão (M. G. M. Lima,personal communication) indicate that the distribution of S. collin-si extends into the edges of the Cerrado biome, in gallery forests.

The taxon Saimiri collinsi was described by Osgood in 1916,based on a male, from a series of four specimens collected on Mar-ajó Archipelago by Emília Snethlage, deposited in the FieldMuseum of Natural History (specimen number 19534). The otherthree specimens are housed in the Museu Paraense Emílio Goeldi(MPEG 1378, 1379, and 1380) and were not cited in Osgood’sdescription. The type locality is Fazenda Teso, Soure municipality,Marajó Island, Pará state, Brazil. Cruz Lima (1945), Cabrera(1958) and Hill (1960) later included this taxon within Saimiri sciu-reus as a different subspecies from S. s. sciureus named S. s. collinsi.While Hershkovitz (1984) did not describe any geographical varia-tion in morphology for the S. s. sciureus populations north andsouth of the Amazon River or on Marajó Island, Thorington(1985) found geographical variation within S. s. sciureus in that ani-mals from the Guianan populations had a different coloration tothe crown region and the back when compared to animals frompopulations south of the Amazon River. Silva Júnior (1992) alsodescribed some individual morphological variation for S. s. sciureus,such as a black coloration for some parts of the fur for some indi-viduals, but noted no consistent geographical variation within thissubspecies.

In a study that included behavioral, morphological and genetic(mitochondrial DNA sequence data from the D-Loop) information,Boinski and Cropp (1999) found S. sciureus to have little variationand thus considered it a single taxon. However, their S. sciureusspecimens were limited to Guyana and Suriname, both north ofthe Amazon River; these Guianan S. sciureus formed a clade withS. oerstedii in their genetic analyses (Boinski and Cropp, 1999). Ina second molecular analysis, Cropp and Boinski (2000) addressedthe issue of Saimiri phylogenetic relationships using sequence datafrom two mitochondrial regions (portions of the D-Loop and thecytochrome b gene, or Cyt b) and two nuclear genes (ZFX and IRBP),and they inferred a trichotomy of S. sciureus, S. oerstedii and S. ustusin their topology. However, as before, they did not use any S. sciu-reus samples from south of the Amazon River in their study.

In a subsequent phylogenetic study by Lavergne et al. (2010)using sequence data from the Cyt b gene, the S. sciureus species com-plex, including S. s. sciureus, S. s. collinsi (from Marajó Island), S. s. cas-siquiarensis, S. s. albigena and S. s. macrodon, was found to be aparaphyletic group. S. s. sciureus was sister to S. oerstedii, while thesubspecies S. s. cassiquiarensis, S. s. albigena and S. s. macrodon formeda different clade. The populations from the Marajó Archipelago wereconsidered a separate valid taxon (S. s. collinsi) that was sister to S.ustus. Based on these results, Lavergne et al. (2010) suggested thatS. s. collinsi, S. s. cassiquiarensis, S. s. albigena and S. s. macrodon shouldbe reallocated into S. ustus (S. u. cassiquiarensis, S. u. albigena, S. u.macrodon and S. u. collinsi). Subsequently Chiou et al. (2011)reported that according to the Principle of Priority in the Interna-tional Code for Zoological Nomenclature, the name S. cassiquiarensiswould have priority in relation to S. ustus Geoffroy Saint-Hilaire1843, because it was published three years earlier by Lesson (1840).

These disagreements point to the need for a new evaluation ofthe taxon S. s. sciureus. Our aims were to test for the presence ofmore than one phylogenetic species among eastern Amazoniansquirrel monkeys (north of the Amazon River, south of the AmazonRiver, and on the Marajó Archipelago), as determined by fixed mor-phological differences between geographic populations, using thePhylogenetic Species Concept. We also employed a molecular phy-logenetic analysis to test the relationships between morphologi-cally identified species, and cranial morphometric analyses toexplore how squirrel monkeys differ morphologically within thisbroad geographic area.

2. Material and methods

2.1. Morphology and morphometrics

Here our nomenclature follows that in Paglia et al. (2012),which recognizes seven species of South American squirrel mon-keys – Saimiri sciureus, S. collinsi, S. cassiquiarensis, S. macrodon, S.ustus, S. boliviensis and S. vanzolinii – in addition to S. oerstedii fromCentral America. We used morphology and morphometry to char-acterize the differences across individuals. For the recognition ofvalid taxa, we incorporated the Phylogenetic Species Concept(sensu Cracraft, 1983), in which we looked for consistent, fixedmorphological differences across geographically defined groups.

We analyzed 250 Saimiri sciureus and S. collinsi specimens(Appendix I), including taxidermied skins and/or crania depositedin the mammal collections in the following institutions: MuseuParaense Emílio Goeldi (MPEG); Museu Nacional, UniversidadeFederal do Rio de Janeiro (MN); Museu de Zoologia da Universidadede São Paulo (MZUSP); and Instituto de Pesquisas Científicas e Tec-nológicas do Estado do Amapá (IEPA). We also requested and ana-lyzed photographs of the holotypes from Saimiri sciureus from thecurators of the Museu Adolphi Friderici, and of S. collinsi from the

428 M.P. Mercês et al. / Molecular Phylogenetics and Evolution 82 (2015) 426–435

Field Museum of Natural History. We also examined the otherspecimens from the Snethlage S. collinsi series at MPEG.

Only adult specimens were used in the morphological and mor-phometric analyses. Adults were identified using the classificationproposed in Silva Júnior (1992) and Muniz (2005). As squirrel mon-keys show significant sexual dimorphism (Muniz, 2005;Goldschmidt et al., 2009), the sexes were analyzed separately.

For the morphological analysis, we characterized the type andcolor of pelage in specific regions of the body. We chose ten char-acters, defined in the works of Silva Júnior (1992, 2001, 2002): 1.Hair coloration in the central crown area; 2. Hair coloration onthe nape of the neck; 3. Hair coloration on the shoulders; 4. Haircoloration in the middle of the back (the ‘saddle’); 5. Hair colora-tion on the posterior region of the back (the ‘hip’); 6. Colorationof the distal and median portions of the forelimbs; 7. Colorationof the distal and median portions of the hindlimbs; 8. Extensionof coloration on the forelimbs; 9. Extension of coloration on thehindlimbs; and 10. Presence of the pre-auricular patch.

For traditional morphometry, we obtained 18 measurements,following Silva Júnior (1992) and Muniz (2005), using the nomen-clature of Hershkovitz (1977): total length – from prostion to opis-thocranium (TL), basion-vertex distance (BVD), length of uppermolar series (UMS), braincase width – euryon to euryon (BW),orbital breadth (OB), nasal width (NW), width across upper canines(AUC), zygomatic breadth (ZIG), width of the foramen magnum(WFM), palate width (PW), across upper molars (AUM), mandibu-lar length (ML), infradentale-gnathion distance (IGD), across lowermolars (ALM), across lower canines (ALC), length of lower molarseries (LMS), canine length (CAL) and width of mandibular condyl-ion (WCM). Measurements were taken with digital calipers with aprecision to 0.01 mm, always from the right side of the cranium,unless there was damage to that region.

We used the Kolmogorov–Smirnoff test to test for deviationsfrom normality in the data. Discriminant Function Analyses(DFA) were performed to test for the existence of significant statis-tical differences in cranial morphometry between the definedgroups. Results were analyzed in the program Statistica 8.0, witha significance level set at 5%.

2.2. DNA extraction, amplification, sequencing, and alignment

For our molecular study, almost all of our samples were of driedmuscle tissue recovered from museum specimens collected manyyears ago. Due the degraded nature of the tissue, the nuclearDNA was likely to be fragmented and the amplification ofnuclear-mitochondrial insertions (numts) less probable. This alsomade it difficult to amplify nuclear markers in general, so our focusfor this study was on mitochondrial DNA, which is available athigher copy number and so easier to amplify in degraded tissue.This study includes the most intensive sampling of squirrel mon-key DNA from the eastern Amazon to date.

DNA was isolated from museum tissue samples using the phe-nol-chloroform protocol as described by Sambrook and Russell(2001). Genomic DNA was used as template for polymerase chainreaction (PCR) in order to amplify the portions of the Cyt b andD-Loop regions. Each PCR had a final volume of 25 ll and contained100 ng of DNA, 2.5 ll of 1 � reaction buffer, 0.5 mM of each primer,0.03 U/lL Taq DNA polymerase, 1.5 mM MgCl2, 0.1 lg/lL bovineserum albumin and 1.25 mM of each dNTP. For PCR, we used theprimers described in Kocher et al. (1989) and Cropp and Boinski(2000), as well as internal primers specific to squirrel monkeys,designed for this study. Primer sequences and annealing tempera-tures are listed in Supplementary Material Table 1. The PCR profilesfor amplifying the Cyt b gene included an initial denaturation stepat 95 �C for 1 min, followed by 40 cycles of denaturation at 95 �Cfor 1 min, annealing of each primer cited above for 1 min, polymer-

ization at 72 �C for 2 min, and a final extension at 72 �C for 10 min.The PCR profile for amplifying the D-Loop region included an initialdenaturation step at 94 �C for 3 min, followed by 30 cycles of dena-turation at 94 �C for 35 s, annealing of each primer cited above for35 s, polymerization at 70 �C for 2 min and 30 s, and a final exten-sion at 70 �C for 3 min.

All PCR products were purified using the EZ-10 Spin ColumnPCR Products Purification KIT (Ludwig Biotec, Porto Alegre–RS, Bra-zil), and all PCR products presented unique bands. The internalprimers were designed specifically based on squirrel monkey Cytb and D-Loop (see Supplementary Table 1), decreasing the likeli-hood of amplification of ‘‘numts’’ or mitochondrial pseudogenespresent in nuclear DNA (Zhang and Hewitt, 1996). Areas of ampli-con overlap showed consistency in base pair sequence. While thereremains the possibility that some of the sequences generated inthis study are numts, our sequences align with the expected spe-cies identity found for the Cytb and D-Loop regions of whole mito-chondrial genome sequences of Saimiri from Chiou et al. (2011),and our Cyt b sequences from individuals south of the AmazonRiver and on the Marajó Island form a clade with the Marajó IslandCyt b sequences from Lavergne et al. (2010). Purified PCR productswere sequenced with a DYEnamic ET Dye Terminator CycleSequencing Kit in the MegaBACE 1000 System (Amersham Phar-macia Biotech – UK). We deposited all sequences of both loci inGenBank. The accession numbers with their respective localitiesare included in Supplementary Material Table 2.

2.3. Phylogenetic analyses

The complete data set for Cyt b gene included 55 sequences,while the complete data set of the D-Loop region included 43sequences. The combined mitochondrial data set (Cyt b + D-Loop)comprised 32 sequences. We first performed a Bayesian phyloge-netic analysis using Mr. Bayes 3.2 for each marker separately,and in the absence of topological incongruences between the sep-arate analyses we performed Bayesian inference (BI) phylogeneticanalyses on the concatenated sequences to improve tree resolu-tion. As Cyt b and D-Loop are both part of the mitochondrial gen-ome, they are inherited together and are both fragments of thesame marker, although they are expected to have different ratesof evolution, because Cyt b is a protein-coding gene and D-Loop isa non-coding region. Our phylogenetic analyses additionallyincluded sequences of S. ustus, S. macrodon, S. albigena (fromColombia, not assessed in Paglia et al., 2012), S. cassiquiarensis, S.collinsi and S. oerstedii from GenBank and/or our laboratory whenavailable (Supplementary Material, Table 2). We rooted the treeswith Saimiri boliviensis as the sister group to all other squirrel mon-keys, as found for whole mitochondrial genome studies (e.g., Chiouet al., 2011), studies using several nuclear markers (e.g., Perelmanet al., 2011), and studies using a combination of nuclear and mito-chondrial markers (e.g., Springer et al., 2012).

Sequences were edited and aligned using the default settings ofClustalW implemented in the software BioEdit Version 7.0.9 (Hall,1999) and then checked manually in Geneious 5.3.4. We parti-tioned our mitochondrial data into D-Loop and Cyt b, taking intoaccount the three codon positions of Cyt b. Using PartitionFinder1.1.1 (Lanfear et al., 2014), we recovered the best model for D-Loopas K80+I+G, and for the three codon positions of Cyt b as follows:codon position 1, K80+I; codon position 2, HKY+I; codon position3, GTR+G. MrBayes 3.2 (Ronquist and Huelsenbeck, 2003) was usedfor Bayesian inference using the corresponding models. We ran aMetropolis-coupled MCMC with five heated chains for 4 � 106 gen-erations, with trees sampled every 1000 generations. Stationaritywas assessed by examining the standard deviation of split frequen-cies and by plotting the �ln L per generation using Tracer version

M.P. Mercês et al. / Molecular Phylogenetics and Evolution 82 (2015) 426–435 429

1.5 (Rambaut and Drummond, 2009). Trees generated before sta-tionarity were discarded as ‘‘burn-in’’.

We used MEGA 5.1 (Tamura et al., 2011) to calculate interspe-cific sequence divergence with a Kimura 2-parameter (K2P) modelfor both Cyt b and D-Loop.

3. Results

3.1. Morphology

In our analysis of pelage color, we detected geographic and indi-vidual variation in several characters. Most importantly, we foundtwo distinct geographically distributed color states of the crownregion, delimited by the Amazon River, with a yellower cap in squir-rel monkeys south of the Amazon and on the Marajó Archipelago,and a grayer cap north of the Amazon (Fig. 1, Table 1). The first state,

Fig. 1. Map with Saimiri specimens used for morphological and molecular analyses. mopoints to individuals with S. collinsi morphology found in Faro, north of the Amazon RiveJuruti. (For interpretation of the references to color in this figure legend, the reader is re

a yellow crown with little gray, occurred south of the Amazon Riverin Brazil, from the Tapajós River to the state of Maranhão, and in theMarajó Archipelago, considered here as S. collinsi. The second state,a predominantly gray crown washed in yellow, occurred north ofthe Amazon River, extending from the Negro and Branco rivers toAmapá, Brazil, and reaching the forests in Guyana, French Guiana,and Suriname, considered here as S. sciureus.

In the S. collinsi specimens, the shoulder region was orange-yel-low, washed with gray. This orange color is most bright in speci-mens to the extreme west of the distribution of S. collinsi(Santarém, Pará state). The hip and saddle regions are most com-monly orange-chestnut with black patches, or less frequently purechestnut brown. The yellow on hands and feet extends beyondwrists and the ankles.

The specimens identified as S. sciureus showed individual varia-tion in the amount of black in the shoulder region. Melanizationwas also seen in the medial region of the back. The predominant

lec = molecular genetic evaluation; morpho = morphological evaluation. Blue arrowr; red arrow points to morphological S. sciureus found south of the Amazon River, inferred to the web version of this article.)

Tabl

e1

Mor

phol

ogic

alco

mpa

riso

nac

ross

Saim

iris

peci

es.T

heta

ble

form

atis

mod

ified

from

Paim

etal

.(20

13),

and

the

data

from

S.m

acro

don

1,S.

cass

iqui

aren

sis,

S.oe

rste

diia

ndS.

ustu

sba

sed

onPa

imet

al.(

2013

),G

rove

s(2

001)

,and

Her

shko

vitz

(198

4).D

ata

for

S.sc

iure

usan

dS.

colli

nsi

wer

ege

nera

ted

inth

isst

udy.

Ch

arac

ters

/tax

aSt

ates

S.co

llins

iS.

sciu

reus

S.m

acro

don

1S.

cass

iqui

aren

sis

S.us

tus

S.oe

rste

dii

1.H

air

colo

rati

onin

the

cen

tral

crow

nar

eaY

ello

ww

ith

litt

legr

ayG

ray

was

hed

inye

llow

Gra

yish

-oli

vace

ous

Dar

k-ch

estn

ut

Blu

ish

-gra

yB

lack

,bla

ckis

hbr

own

orgr

ay2.

Hai

rco

lora

tion

onth

en

ape

ofth

en

eck

Yel

low

wit

hli

ttle

gray

Gra

yw

ash

edin

yell

owG

rayi

sh-o

liva

ceou

sLi

ght-

ches

tnu

tC

hes

tnu

tsp

eckl

edw

ith

blac

kG

ray

was

hed

wit

hor

ange

3.H

air

colo

rati

onin

the

shou

lder

sO

ran

ge-y

ello

ww

ash

edw

ith

gray

Gra

yw

ash

edin

yell

owG

rayi

sh-o

liva

ceou

sLi

ght-

ches

tnu

tC

hes

tnu

tsp

eckl

edw

ith

blac

kG

ray

was

hed

wit

hor

ange

4.H

air

colo

rati

onin

the

mid

dle

ofth

eba

ck(t

he

‘sad

dle’

)O

ran

gish

-ch

estn

ut

wit

hbl

ack

patc

hes

Pure

ches

tnu

tR

eddi

sh-b

row

nLi

ght-

ches

tnu

tC

hes

tnu

tsp

eckl

edw

ith

blac

kO

ran

gesp

eckl

edw

ith

blac

k5.

Hai

rco

lora

tion

onth

epo

ster

ior

regi

onof

the

back

(th

e‘h

ip’)

Ora

ngi

sh-c

hes

tnu

tw

ith

blac

kpa

tch

esPu

rech

estn

ut

Red

dish

-bro

wn

Red

dish

-ora

nge

Ch

estn

ut

spec

kled

wit

hbl

ack

Ora

nge

spec

kled

wit

hbl

ack

6.C

olor

atio

nof

the

dist

alan

dm

edia

npo

rtio

ns

ofth

efo

reli

mbs

Yel

low

Bu

rnt

yell

owto

redd

ish

Yel

low

ish

-ora

nge

Red

dish

-ora

nge

Red

dish

-ora

nge

Ora

nge

7.C

olor

atio

nof

the

dist

alan

dm

edia

npo

rtio

ns

ofth

eh

indl

imbs

Yel

low

Bu

rnt

yell

owto

redd

ish

Yel

low

ish

-ora

nge

Red

dish

-ora

nge

Red

dish

-ora

nge

Ora

nge

8.Ex

ten

sion

ofco

lora

tion

onth

efo

reli

mbs

Bey

ond

wri

stB

eyon

dw

rist

From

han

dsto

elbo

ws

From

han

dsto

elbo

ws

Res

tric

ted

toth

eh

ands

Bey

ond

wri

st

9.Ex

ten

sion

ofco

lora

tion

onth

eh

indl

imbs

Bey

ond

ankl

esB

eyon

dan

kles

Bey

ond

ankl

esB

eyon

dan

kles

Bey

ond

ankl

esB

eyon

dan

kles

10.P

rese

nce

ofth

epr

e-au

ricu

lar

patc

hPr

esen

tPr

esen

tPr

esen

tPr

esen

tA

bsen

tPr

esen

t

Fig. 2. Discriminant function analyses for morphometric measurements for Saimirimales from eastern Amazonia.

430 M.P. Mercês et al. / Molecular Phylogenetics and Evolution 82 (2015) 426–435

color of the hip and saddle was pure chestnut, or, in some individ-uals, blackish chestnut or with patches of black and light chestnutbrown. Hands and forelimbs varied from burnt yellow to reddish,with the forearm containing some black hairs. In the hindlimbs,colors also ranged from burnt yellow to reddish, and this colorationextended from the foot past the ankles.

In terms of sex-specific variation, squirrel monkey malesshowed little individual variation, but females showed individualvariation in both the crown coloration and the presence or absenceof the pre-auricular patch.

While in general the Amazon River was a clear barrier to thetwo squirrel monkey types, we identified two S. collinsi individuals,based on morphology, from the northern bank of the Amazon Riverin the Faro region (see map in Fig. 1). We also identified morpho-logical S. sciureus in the region of Juruti, immediately south of Ama-zon River (see Fig. 1). These areas appear to be two localities whereboth species of squirrel monkeys are in contact, suggesting theopportunity for hybridization.

3.2. Morphometric analysis

We analyzed 133 crania from adult squirrel monkeys. Analyseswere separated for males and females because of sexual dimor-phism. All the measurements presented a normal distributionusing the Kolmogorov–Smirnoff test.

For males, we performed a discriminant function analysis on 61crania, including 36 of S. collinsi and 25 of S. sciureus. With regardto all variables, means were significantly different between S. coll-insi and S. sciureus (Wilk’s Lambda = 0.30; F(19,41) = 4.97; p < 0.05).The most important variables separating Saimiri sciureus from S.collinsi were: AUC, ZIG, ML and TL (Fig. 2).

The analysis of 72 female crania, 54 of S. collinsi and 18 of S. sciu-reus, revealed statistically significant morphological differencesbetween S. sciureus and S. collinsi (Wilk’s Lambda = 0.44;F(19,52) = 3.47; p < 0.05). The measurements most influential in sep-arating the species were: AUC, AUM, LMS and BW (Fig. 3).

3.3. Molecular analysis

Phylogenetic trees derived from the independent analyses ofCyt b and D-Loop region differed in topology, but there was littlesupport for the conflicting nodes among them (Supplementary

Fig. 3. Discriminant function analyses for morphometric measurements for Saimirifemales from eastern Amazonia.

M.P. Mercês et al. / Molecular Phylogenetics and Evolution 82 (2015) 426–435 431

Figs. 1 and 2). Thus, we report results from our analysis based onthe combined mitochondrial dataset (Cyt b + D-Loop).

In our Bayesian phylogenetic tree inferred from the combineddataset of Cyt b and D-Loop sequences (Fig. 4), our central resultwas that Saimiri sciureus (from north of the lower Amazon River)and Saimiri collinsi (from south of the lower Amazon River andon the Marajó Archipelago) formed two distinct clades and werenot sister taxa. There was weak support (0.59 posterior probabil-ity) for S. oerstedii as the sister clade to S. sciureus, and strong sup-port (1.00 posterior probability) for S. macrodon as the sister taxonto S. collinsi. There was no apparent geographic structure withinthe S. sciureus clade, but the S. collinsi clade comprised two

Fig. 4. Bayesian tree based on two mitochondrial regions (1957 bp) with posterior probaall other squirrel monkeys, per Chiou et al. (2011), Perelman et al. (2011) and Springer

geographic subclades. Additionally, the sample from Juruti (southbank of Amazon River) that was identified morphologically as S.sciureus nested within the S. sciureus clade from north of the Ama-zon, while that from Faro (north bank of Amazon River), identifiedas S. collinsi on the basis of morphology, was nested within thesouth Amazonian S. collinsi clade. In other words, there was concor-dance between genetically and morphologically identified species,and limited evidence for local movement of squirrel monkey spe-cies across the Amazon River.

Kimura 2-parameter genetic distances across squirrel monkeyspecies ranged from 3.7%–0.5% for Cyt b and 6.7%–3.3% for D-Loop(see Supplementary Table 3 for Cyt b and Supplementary Table 4for D-Loop values). The genetic distance between S. collinsi and S.sciureus was 4.9% for D-Loop and 3.4% for Cyt b.

4. Discussion

Previous reviews of the genus Saimiri (Hershkovitz, 1984;Thorington, 1985; Silva Júnior, 1992; Costello et al., 1993;Boinski and Cropp, 1999; Cropp and Boinski, 2000) took into con-sideration not only external morphology, but sometimes also mor-phometric, molecular, cytogenetic and behavioral data todifferentiate among taxa. One of the great difficulties in distin-guishing across species of squirrel monkeys is that they have highintraspecific morphological variation. Hershkovitz (1984) did notspecify which species concept he adopted, but it was most likelythe Biological Species Concept of Ernst Mayr. Thorington (1985)mentioned the use of the Biological Species Concept in the Materialand Methods section of his article, while other relevant authors donot discuss this issue. Here we employed the Phylogenetic SpeciesConcept to assess geographic differentiation among eastern Ama-zonian squirrel monkeys.

Thorington (1985) pointed out that squirrel monkeys from theGuianas had a grayish crown region and that the rest of the back

bilities indicated for all nodes in the tree. Tree is rooted with S. boliviensis as sister toet al. (2012).

Fig. 5. Distribution of Saimiri sciureus (red) and Saimiri collinsi (blue) based on museum records and data from Hershkovitz (1984) as well as our findings in this study. (Forinterpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

432 M.P. Mercês et al. / Molecular Phylogenetics and Evolution 82 (2015) 426–435

was orange and yellow. The monkeys south of the Amazon River, incontrast, had a lighter yellow back and a bright tinge of yellow onthe crown. In our study, we confirmed these two morphotypesdescribed by Thorington (1985). Thorington (1985) considered Sai-miri collinsi Osgood, 1916 to be a junior synonym of S. sciureus,despite the geographic differences in morphology. Hershkovitz(1984) did not report any morphological difference related to geog-raphy between S. collinsi and S. sciureus, although he did reportindividual variation within this group. In his identification keyfor squirrel monkey taxa, Hershkovitz (1984) indicated that S. s.sciureus had yellow or orange hands, wrists and feet; a gray orbuffy agouti crown that could be bordered or mixed with blackishin females; and a grayish agouti, olivaceous agouti or reddishbrown agouti back.

Osgood (1916) observed that Saimiri collinsi was very similar toS. sciureus, but that S. collinsi had darker and more richly coloredhands and feet, and its back was lighter and less tawny. In the cur-rent study, in our examination of the specimens collected in thetype locality of S. collinsi, we detected this pattern of lighter colorwith less patchiness in both the type specimen and all the otherspecimens from this area, as well as in all the specimens collectedto the south of the Amazon River (with the exception of the S. sciu-reus individual in Juruti).

Our findings indicate clear morphological and morphometricdifferences between the populations on the two sides of the Ama-zon River. This has allowed us, following the Phylogenetic SpeciesConcept, to make the morphological identification of two phyloge-netic species with unique characters. The principal morphologicaldifference is in the pelage color on the crown (with S. sciureus northof the Amazon being predominantly gray on the crown, and S. coll-

insi from the south predominantly yellow), as well as on the hipand saddle (S. sciureus—pure chestnut with blackish tinge; S. collin-si—orange chestnut). Additionally, the width across upper caninesin both sexes, the zygomatic breadth for males, and the widthacross upper molars for females differed significantly between spe-cies. In males the zygomatic breadth is wider in S. sciureus than S.collinsi, and similarly, the distance between the upper molars iswider for S. sciureus females in comparison to S. collinsi females.

We also want to point out the clear phenotypic differencesbetween these eastern Amazonian taxa and other squirrel monkeypopulations, including those that formerly have been consideredsubspecies of S. sciureus. The taxon S. macrodon can be distin-guished from both S. collinsi and S. sciureus in the back and crownregion: S. macrodon has an olivaceous gray crown and a chestnutgray back (Paim et al., 2013). S. cassiquiarensis can be distinguishedby its intense yellow crown and its reddish-chestnut back, as wellas its nuchal band which forms a weakly contrasting ’collar’(Hershkovitz, 1984; Paim et al., 2013). S. collinsi and S. sciureus bothhave hairy ears, not the bare ears found in S. ustus. S. ustus is alsodistinct in terms of coloration, with a bluish gray crown. S. sciureusdiffers morphologically from its sister species S. oerstedii in variouscharacters; for example S. sciureus has a grayish crown region,while S. oerstedii crown color is more geographically variable, butcan be jet black, and the back of S. oerstedii is orangish-black, quitedifferent from the chestnut brown back of S. sciureus (Table 1).

The molecular analyses corroborate and reinforce our morpho-logical results, indicating the existence of two distinct phylogroupsoccurring on opposite sides of the middle to lower Amazon River.In fact, the two clades fall on two different branches of the Saimiriphylogeny and do not form a monophyletic clade to the exclusion

M.P. Mercês et al. / Molecular Phylogenetics and Evolution 82 (2015) 426–435 433

of other squirrel monkey taxa. Instead, they show genetic diver-gence similar to that between other well-established Saimiri spe-cies. These results extend the findings of Lavergne et al. (2010),which indicated a distinct taxon on the Marajó Archipelago.Lavergne et al.’s (2010) study did not contain samples from otherlocalities south of the lower Amazon River, but in our study wefound that the Marajó Island samples – including those fromLavergne et al. (2010) – are nested within a broader S. collinsi cladefrom south of the Amazon.

In summary, within Saimiri sciureus sciureus (sensu Hershkovitz,1984) there was a clear separation between the taxa on the twosides of the Amazon River, with the north and south groups form-ing independent phylogroups. The genetic divergence between thenorth and south clades was as high as that between other speciesof squirrel monkeys, and the northern and southern populations donot form a clade. Thus, not only is the taxon currently referred to asS. sciureus paraphyletic, but the currently recognized eastern sub-species, S. s. sciureus, is also paraphyletic. However, it is necessaryto note that mitochondrial paraphyly does not always indicate spe-cies paraphyly. Mitochondrial paraphyly could be the result ofincomplete lineage sorting, or of introgression through recent orpast hybridization with other species. Hybridization is known tooccur in both captivity and the wild between some Saimiri species(Silva et al., 1992; Hershkovitz, 1984; Thorington, 1985; Costelloet al., 1993; Lavergne et al., 2003), and the opportunity for hybrid-ization between S. sciureus and S. collinsi exists at least in Juruti andFaro regions, where some squirrel monkeys were found on theopposite bank of the Amazon River than expected. Incomplete line-age sorting in mitochondrial markers has been found in severalother recent radiations of primates, such as macaques (Tosi et al.,2003) and baboons (Zinner et al., 2009), and should be consideredas a possibility in squirrel monkeys, as they also diversified quiterecently, in the Pleistocene (Chiou et al., 2011).

An important next step will be an analysis of eastern Amazoniansquirrel monkey diversity using nuclear markers. Limitations tomitochondrial markers include that they are inherited throughthe maternal line only, and that they serve as a single locus inher-ited as a whole without recombination. Other primate studies, suchas Tosi et al. (2003), show that a combination of mitochondrial andautosomal markers can illuminate different aspects of a radiation’sevolutionary history. This is especially important for squirrel mon-keys, known to have variable sex-biased dispersal patterns acrosspopulations and species (Boinski and Cropp, 1999), a characteristicthat may affect the genetic signature in relation to geography.

We confirm the existence of more than one taxon within therange of what was previously considered Saimiri sciureus sciureus(sensu Hershkovitz, 1984). Based on our analysis of morphologicaland morphometric characters, we restrict the species Saimiri sciure-us to north of the Amazon River, from the Guianas and Amapá, Bra-zil to as far west as the Negro and Branco rivers in Amazonas andRoraima states, Brazil, and we revalidate the species Saimiri collinsiOsgood, 1916, as proposed by Lavergne et al. (2010) based on mito-chondrial phylogenetics. We also extend the geographic distribu-tion of S. collinsi to include the entire region south of the AmazonRiver, from the eastern border of the Zona dos Cocais in Maranhãoto the left bank of the Tapajós River (Fig. 5). The southern limit to S.collinsi is still unknown. In terms of the northwestern distributionof S. collinsi, we observed the existence of a population in themunicipality of Faro, Pará State, Brazil, that forms a cluster withspecimens from Juruti, Pará. These individuals showed a geneticdivergence from the other S. collinsi populations sampled. At pres-ent, we have only a few samples from this region, and we suggestthe need for more intensive study here to better understand thetaxonomy and phylogenetic relationships of this population.

In fact, the situation in Juriti and Faro is particularly complex.One specimen with the morphological characteristics of S. sciureus

(MPEG39469) was collected in Juruti township, Pará State, south ofthe Amazon River, in the same locality where several S. collinsisamples were collected (MPEG39472, MPEG39473, LB798). Thisspecimen had a mtDNA haplotype within the S. sciureus phylo-group despite being collected south of the Amazon River. Similarly,two specimens with morphological characteristics of Saimiri collin-si (MPEG39848, MPEG39862) were collected on the north bank ofthe Amazon River, in the State Forest of Faro in the municipality ofFaro, Pará State. The mtDNA haplotype for these Faro specimenswas S. collinsi.

The water dynamics of this particular region of the AmazonRiver, near the confluence of the Tapajós, may facilitate the cross-ing or passive transport of primates, because during the drier peri-ods many islands surface and then disappear when the water levelrises (Ayres and Clutton-Brock, 1992). In both cases (Juruti andFaro), it is not clear whether the movement across the river wasdue to human-mediated transport, regional water dynamics, oreven a change in recent demographic dynamics. In any case,hybridization may now occur between these two species in theseareas, and there may be other, not yet discovered areas wheresquirrel monkeys have crossed the lower Amazon. Paim et al.(2013) also observed a possible case of hybridization betweenthe two taxa S. cassiquiarensis and S. macrodon in the Mamirauáregion. While the occurrence of hybridization in zones of contactis possible, careful studies with field observations are needed tounderstand cross-species relationships. Further morphological sur-veys and studies with nuclear markers would help us to betterunderstand the time frame of these dispersal events and the pres-ent inter-species dynamics for squirrel monkeys along the lowerAmazon River.

Acknowledgments

This project was funded by CNPq through a master’s degree fel-lowship to MPM (protocol number 134647/2011-4). The authorswish to thank the staff of Museu Paraense Emílio Goeldi (MPEG),the Museu Nacional at the Universidade Federal do Rio de Janeiro(MN), Museu de Zoologia da Universidade de São Paulo (MZUSP),Laboratório de Biologia Molecular Francisco Mauro Salzano at theUniversidade Federal do Pará (UFPA) and Instituto de PesquisasCientíficas e Tecnológicas do Estado do Amapá (IEPA). We alsowant to thank to the curators of the Museu Adolphi Friderici andthe Field Museum of Natural History for the photographs of keyspecimens. We give special thanks to Anthony Di Fiore and theanonymous reviewers for their excellent input which greatlyimproved the manuscript.

Appendix A

List of specimens analyzedSaimiri collinsi: Brazil, Maranhão: Bacabal, Fazenda Lagoa Nova

(MPEG 23174); Bacabal, BR-316, left bank of Mearim River (MPEG21989); Imperatriz (MPEG 1904, 2448); Lago da Pedra, Pedra Preta,Alto da Liberdade (MPEG 23029, 23032); Pará: 126 km south ofTucuruí, right bank of Tocantins River (MPEG 12180-81); 170 kmsouth of Tucuruí, left bank of Tocantins River (MPEG 12192-97);20 km south of Jacundá, left bank of Tocantins River (MPEG12191); Acampamento Saúde IV (MPEG 12198-99); Ananindeua,CENP (MPEG 21452, 21454); Ananindeua, mouth of TraquateuaRiver (MPEG 707-8); Belém, Utinga forest, Água Preta (MPEG21460); Bragança (MPEG 130); BR 010, km 93 (MZUSP 8921-22);Cachoeira do Espelho, Xingu River (MZUSP 25435); Cametá(MZUSP 5531); Chaves, igarapé Taperebá (MPEG 1784); Chaves,Fazenda São Vicente (MPEG 782-85); Curionópolis, Projeto Crista-lino (MPEG 38485, 38489); Faro, Floresta Estadual de Faro (MPEG

434 M.P. Mercês et al. / Molecular Phylogenetics and Evolution 82 (2015) 426–435

39848, 39862); mouth of Tapajós River (MPEG 660); CavianaIsland, Fazenda São Luiz (MPEG 22033, 23055, 35284-86, 35296,35299); Island of Gurupá, Casa Santo Antônio, left bank of JaburuRiver (MPEG 21450); Island of Gurupá, Mariony River(MPEG22974); Mexiana Island (MPEG 35295, 35297); Itaituba (MPEG2400); Jacundá, 126 km south of Tucuruí, right bank of TocantinsRiver (MPEG 12182-84); Jacundá, 120 km south of Tucuruí, rightbank of Tocantins River (MPEG 12185-90); Juruti, AcampamentoBarroso (MPEG 39472-73); Marajó, igarapé Taperebá (MZUSP8677-78, 19704-5); Nova Timboteua (MN 33528-46, 33548-63,33565-75); Nova Timboteua, Timboteua Velha, Fazenda São Fran-cisco da Trombeta (MN 23531-32); Porto de Moz (MN 33526-27); Arapiuns River, Tapajós (MPEG 664-65); Santa Bárbara, com-unidade Araci (MPEG 37915, 37939, 38175); Santarém (MPEG21451, 21453, 21458, MN 11942, 11943-50, 11952-53, 11956-60,11962-63, 11966-73); Santarém, Arapiuns (MPEG 21461); Santa-rém, Taperinha (MPEG 36); Santo Antônio, Tocantins River (MZUSP13473); Soure (MPEG 1367, 1379-80); Soure, Fazenda Teso (MPEG1378, 1387, 1388, FMNH 19534 – holotype); Tucuruí, OperaçãoCurupira (MPEG 21456); Tucuruí, UHE Tucuruí (MPEG 21455,21462-64); Utinga (MN 2882).

Saimiri sciureus: Brazil, Amapá: Cachoeira de Santo Antônio(MPEG 21817-19); Amapá city (MPEG 1196-98, 1200, 1202,8482); Fazenda Nova (MPEG 1204); Iracema, Aporema River(MPEG 1218-19); Itaubal (Macacoari), Fazenda Lago Novo (MPEG24027, 24029, IEPA 7); Itaubal (Macacoari), Fazenda Veja (MPEG24028, IEPA 37-39); Macapá (MPEG 1265); left bank of Jari River(MPEG 21962-64, 21970); Mazagão, Vila Nova River (MPEG 659,666); Mazagão, igarapé Novo, afluente esquerdo do rio Iratapuru(MPEG 1538-39); Mazagão, igarapé Rio Branco, afluente direitodo rio Maracá (MPEG 1902); PARNA Montanhas do Tumucumaque,Anacui River (IEPA 1306); PARNA Montanhas do Tumucumaque,Araguari River (IEPA 1460); Posto DNERu, Tracajatuba River(MZUSP 19709-10); REBIO do Lago Pirituba (IEPA 3563); RDS rioIratapuru, on the bank of Cupixi River (IEPA 1299, 2855); RDS rioIratapuru, igarapé Baliza (IEPA 2853-54); Vila Nova River (MPEG662); Tracajatuba River (MN 20543, 20560, 20562-66); Vila Velhade Caciporé, Oiapoque (MPEG 2288, 2290-99, MN 20592); Pará:Alenquer, ESEC Grão-Pará (MPEG 39857-58); Almeirim, right bankof Paru do Leste River, FLOTA Paru (CN 243-44, 276); Almeirim,Paru do Leste River, Uchizal (MN 23533-35); Juruti, base Capiranga(MPEG 39469); Lago Paru, Trombetas (MZUSP 19135, 19141);Óbidos, ESEC Grão-Pará (CN 297); Oriximiná, Cachoeira Porteira(MPEG 661, 21971, 22050-51, 23256). Cayenne, French Guiana(NRMA 532005 – holotype).

Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.ympev.2014.09.020.

References

Ayres, J.M., Clutton-Brock, T.H., 1992. River boundaries and species range size inAmazonian Primates. Am. Nat. 140 (3), 531–537.

Boinski, S., 1989. The positional behavior and substrate use of squirrel monkeys:ecological implications. J. Hum. Evol. 18, 659–677.

Boinski, S., Cropp, S.J., 1999. Disparate data sets resolve Squirrel Monkey (Saimiri)taxonomy: implications for behavioral, ecology and biomedical usage. Int. J.Primatol. 20 (2), 237–256.

Cabrera, A., 1958. Catálogo de los mamíferos de América del Sur. Revista Argentinode Ciencias Naturales, ‘‘Bernardino Rivadavia’’. Cien. Zool. 4, xvi + iv + 308.

Chiou, K.L., Pozzi, L., Lynch Alfaro, J.W., Di Fiore, A., 2011. Pleistocene diversificationin living squirrel monkeys (Saimiri spp.) inferred from complete mitochondrialsequence. Mol. Phylogenet. Evol. 59, 736–745.

Costello, R.K., Dickinson, C., Rosenberger, A.L., Boinski, S., Szalay, F.S., 1993. Squirrelmonkey (genus Saimiri) taxonomy: a multidisciplinary study of the biology of

the species. In: Kimbel, W., Martin, L. (Eds.), Species, Species Concepts andPrimate Evolution. Plenum Press, New York, pp. 177–210.

Cropp, S., Boinski, S., 2000. The Central American Squirrel Monkey (Saimirioerstedii): introduced hybrid or endemic species? Mol. Phylogenet. Evol. 16(3), 350–365.

Cruz Lima, E. da, 1945. Mammals of Amazonia, Volume 1. General introduction andprimates. Belém, Contribuições do Museu Paraense Emílio Goeldi de HistóriaNatural e Etnografia.

Goldschmidt, B., Mota-Marinho, A., Araújo-Lopes, C., Brück-Gonçalves, M.A., Matos-Fasano, D., Ribeiro-Andrade, M.C., Ferreira-Nascimento, L.W., Simmer-Bravin, J.,Monnerat-Nogueira, D., 2009. Sexual dimorphism in the squirrel monkey,Saimiri sciureus (Linnaeus, 1758) and Saimiri ustus (I. Geoffroy, 1844) (Primates,Cebidae). Braz. J. Biol. 69 (1), 171–174.

Groves, C.P., 2001. Primate Taxonomy. Smithsonian Institution Press, Washington,DC, USA.

Groves, C.P., 2005. Order primates. In: Wilson, D.E., Reeder, D.M. (Eds.), MammalSpecies of the World: A Taxonomic and Geographic Reference, third ed. TheJohns Hopkins University Press, Baltimore, pp. 111–184.

Hall, T.A., 1999. BioEdit, a user friendly biological sequence alignment editor and aanalysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95–98.

Hershkovitz, P., 1977. Living New World Monkeys (Platyrrhini), vol. 1. UniversityChicago Press, Chicago.

Hershkovitz, P., 1984. Taxonomy of the squirrel monkey genus Saimiri (Cebidae,Platyrrhini): a preliminary report with description of a hitherto unnamed form.Am. J. Primatol. 7, 155–210.

Hill, O.C., 1960. Primates. Comparative anatomy and taxonomy. Cebidae, Part A, vol.IV. Edinburgh University Press.

Kocher, T.D., Thomas, W.K., Meyer, A., Edwards, S.V., Paabo, S., Villiblanca, F.X.,Wilson, A.C., 1989. Dynamics of mitochondrial DNA evolution in animals:amplification and sequencing with conserved primers. Proc. Natl. Acad. Sci. USA86, 6196–6200.

Lanfear, R., Calcott, B., Kainer, D., Mayer, C., Stamatakis, A., 2014. Selecting optimalpartitioning schemes for phylogenomic datasets. BMC Evolution. Biol, 14, 82.http://dx.doi.org/10.1186/1471-2148-14-82.

Lavergne, A., Catzeflis, F., Lacôte, S., Barnaud, A., Bordier, M., Mercereau-Puijalon, O.,Contamin, H., 2003. Genetic analysis of the Saimiri breeding colony of thePasteur Institute (French Guiana): development of a molecular typing methodusing a combination of nuclear and mitochondrial DNA markers. J. Med.Primatol. 32, 330–340.

Lavergne, A., Ruiz-García, M., Catzeflis, F., Lacote, S., Contamin, H., Mercereau-Puijalon, O., Lacoste, V., Thoisy, B., 2010. Phylogeny and phylogeography ofsquirrel monkey (Genus Saimiri) based on Cytochrome b genetic analysis. Am. J.Primatol. 72, 242–253.

Muniz, I.C.M. 2005. Desenvolvimento do dimorfismo sexual nos macacos-de-cheiro(Saimiri Voigt, 1831). Dissertação (Mestrado em Zooologia). UniversidadeFederal do Pará e Museu Paraense Emílio Goeldi. Belém, PA. 117p.

Osgood, W.H., 1916. Mammals of the Collins-Day South American expedition. FieldMus. Nat. History: Zool. Ser. 10 (14), 199–216.

Paglia, A.P., Fonseca, G.A.B., Rylands, A.B., Hermann, G., Aguiar, L.M.S., Chiarello,A.G., Leite, Y.R.L., Costa, L.P., Siciliano, S., Kierulff, M.C.M., Mendes, S.L., Tavares,V.C., Mittermeier, R.A., Patton, J.L. 2012. Lista Anotada de Mamíferos do Brasil/Annotated checklist of Brazilian Mammals. 2ª edição/2nd ed. Occasional Papersin Conservation Biology, No. 6. Conserv. Intern., Arlington, VA.

Paim, F.P., Silva Júnior, J.S., Valsecchi, J., Harada, M.L., Queiroz, H.L., 2013. Diversity,geographic distribution and conservation of squirrel monkeys, Saimiri(Primates, Cebidae), in the floodplain forest of Central Amazon. Int. J.Primatol. 34, 1055–1076.

Perelman, P., Johnson, W.E., Ross, C., Seuánez, H.N., Horvath, J.E., Moreira, M.A.M.,Kessing, B., Pontius, J., Roelke, M., Rumpler, Y., Schneider, M.P.C., Silva, A.,O’Brien, J., Pecon-Slattery, J., 2011. A molecular phylogeny of living primates.PLoS Genet. 7 (3), e10011342.

Rambaut, A., Drummund, A.J., 2009. Tracer v1.5. Available from: <http://beast.bio.ed.ac.uk/Tracer>.

Ronquist, F., Huelsenbeck, J.P., 2003. Mr. Bayes 3.2: Bayesian phylogenetic inferenceunder mixed models. Bioinformatics 19, 1572–1574.

Rylands, A.B., Mittermeier, R.A., 2009. The diversity of the new world primates(Platyrrhini): an annotated taxonomy. In: Garber, P.A., Estrada, A., Bicca-Marques, J.C., Heymann, E.W., Strier, K.B. (Eds.), South American Primates:Comparative Perspectives in the Study of Behavior, Ecology, and Conservation.Springer, New York, pp. 23–54.

Rylands, A.B., Schneider, H., Langguth, A., Mittermeier, R.A., Groves, C.P., Rodríguez-Luna, E., 2000. An assessment of the diversity of New World primates.Neotropical Primates 8, 61–93.

Rylands, A.B., Mittermeier, R.A., Silva Júnior, J.S., 2012. Neotropical primates:taxonomy and recently described species and subspecies. Int. Zoo Yearbook 46,1–14.

Rylands, A.B., Mittermeier, R.A., Bezerra, B.M., Paim, F.P., Queiroz, H.L., 2013. Speciesaccounts of Cebidae. In: Mittermeier, R.A., Rylands, A.B., Wilson, D.E. (Eds.),Handbook of the Mammals of the World, Primates, vol. 3. Lynx Edicions,Barcelona, pp. 390–413.

Sambrook, J., Russell, D.W., 2001. Molecular Cloning: A Laboratory Manual, thirded. Cold Spring Harbor Laboratory Press, New York.

Silva, B.T.F., Sampaio, M.I.C., Schneider, H., Schneider, M.P.C., Montoya, E.,Encarnacion, F., Salzano, F.M., 1992. Natural hybridization between Saimiritaxa in the Peruvian Amazonia. Primates 33, 107–113.

M.P. Mercês et al. / Molecular Phylogenetics and Evolution 82 (2015) 426–435 435

Silva Júnior, J.S. 1992. Revisão dos macacos-de-cheiro (Saimiri Voigt, 1831) da BaciaAmazônica (Primatas: Cebidae). Dissertação (Master’s thesis). UniversidadeFederal do Pará e Museu Paraense Emílio Goeldi. Belém, PA. 499p.

Silva Júnior, J.S. 2001. Especiação nos macacos-prego e caiararas, gênero CebusErxleben, 1777 (Primates, Cebidae). Tese (PhD thesis). Universidade Federal doRio de Janeiro. Rio de Janeiro, RJ. 377p.

Silva Júnior, J.S., 2002. Taxonomy of capuchin monkeys, Cebus Erxleben, 1777.Neotropical Primates 10, 29.

Springer, M.S., Meredith, R.W., Gatesy, J., Emerling, C.A., Park, J., et al., 2012.Macroevolutionary dynamics and historical biogeography of primatediversification inferred from a species supermatrix. PLoS ONE 7 (11), e49521.http://dx.doi.org/10.1371/journal.pone.0049521.

Stone, A., 2006. Age and seasonal effects on predator-sensitive foraging inSquirrel Monkeys (Saimiri sciureus): a field experiment. Am. J. Primatol. 69,127–141.

Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5:molecular evolutionary genetics analysis using maximum likelihood,evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28,2731–2739.

Thorington, R.W., 1985. The taxonomy and distribution of squirrel monkey(Saimiri). In: Rosenblum, L.A., Coe, C.L. (Eds.), Handbook of Squirrel MonkeyResearch. Plenum Press, New York, London, pp. 1–33.

Tosi, A.J., Morales, J.C., Melnick, D.J., 2003. Paternal, maternal, and biparentalmolecular markers provide unique windows onto the evolutionary history ofmacaque monkeys. Evolution 57, 1419–1435.

Zhang, D.-X., Hewitt, G.M., 1996. Nuclear integrations: challenges for mitochondrialDNA markers. Trends Ecol. Evol. 11, 247–251.

Zinner, D., Groeneveld, L.F., Keller, C., Roos, C., 2009. Mitochondrial phylogeographyof baboons (Papio spp.) – indication for introgressive hybridization? BMC Evol.Biol. 9, 83. http://dx.doi.org/10.1186/1471-2148-9-83.