Mitochondrial COII Gene Sequences Provide New Insights into the Phylogeny of Marmoset Species Groups (Callitrichidae, Primates)

Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com

�2002 S. Karger AG, Basel 0015�5713/02/0735�0240$18.50/0 Accessible online at: www.karger.com/fpr

Leonardo Sena, Departamento de Genética Universidade Federal do Pará, Caixa Postal 8607 BR�66075-900 Belém, Pará (Brazil) Tel. +55 91 211 1558, Fax +55 91 211 1568 E-Mail [email protected]

Reviewed Article

Folia Primatol 2002;73:240–251 Received: November 21, 2001 DOI: 10.1159/000067456 Accepted after revision: June 28, 2002

Mitochondrial COII Gene Sequences Provide New Insights into the Phylogeny of Marmoset Species Groups (Callitrichidae, Primates)

Leonardo Senaa Marcelo Vallinotoa Iracilda Sampaiob

Horacio Schneiderb Stephen F. Ferraria

Maria Paula Cruz Schneidera a DNA Polymorphism Laboratory, Department of Genetics, Universidade Federal do Pará, Belém, and b Molecular Biology Laboratory, Universidade Federal do Pará, Bragança, Brazil

Key Words Callithrix � Cebuella � Mico � Marmoset � Molecular phylogeny� Mitochondrial DNA � Cytochrome oxidase II

Abstract Mitochondrial cytochrome oxidase II (COII) gene sequences (549 base

pairs) were used to investigate the taxonomic relationships among 12 marmoset (Callithrix, Cebuella and Mico) taxa. A large number of substitutions were found in the third base codon positions, providing a strong phylogenetic signal in a gene coding a conserved protein. Despite the significant affinity between the 2 Amazonian genera Cebuella and Mico, found in recent molecular studies, the analysis presented here did not resolve convincingly the phylogenetic relation-ships between the 3 genera. Mico nevertheless formed 3 distinct clades, reflect-ing a basic division of species groups based on geographic distribution (east or west of the Rio Tapajós) rather than morphology (presence or absence of auricular hair). This supports the taxonomic distinction of the allopatric emiliae forms. In Callithrix, Callithrix aurita forms a distinct clade, but the remaining morphotypes form a somewhat contradictory cluster, possibly resulting from an extremely rapid radiation.

Copyright © 2002 S. Karger AG, Basel

Molecular Insights into Marmoset Species Group

241 Folia Primatol 2002;73:240�251

Introduction

The marmosets, Callithrix, Cebuella and Mico (nomenclature follows Rylands et al. [1]), are the smallest living simians and are among the most flexible ecologi-cally of Neotropical monkeys [2]. This classification is a relatively recent consen-sus, in which Mico refers to the Amazonian marmosets of Hershkovitz�s [3] argen-tata species group, previously synonymous with Callithrix.

Marmosets are found throughout most of tropical South America south of the Amazon/Japurá and east of the Andes, including both forest biomes and intervening scrub and savanna. Marmosets exhibit a diversity of body size and pelage colour-ation and are traditionally divided into 2 distinct groups on the basis of morphol-ogy, geographic distribution and karyology [3]. The Amazonian group (Cebuella and Mico: 2n = 44 chromosomes) occurs in the southern Amazon basin, south of the Amazonas/Solimões/Caquetá river system and west of the Rio Tocantins, ex-tending southwards to the chaco scrublands of northern Paraguay. The genus Cal-lithrix (2n = 46) occurs in the Atlantic forest, and the neighbouring caatinga and cerrado biomes of central and north-eastern Brazil.

While these 2 groups are relatively well defined, the taxonomic status of the different morphological forms remains controversial. In his classic review, Her-shkovitz [3] identified only 4 marmoset species, each with a number of subspecies, whereas de Vivo [4] attributed species status to 5 forms of Callithrix and 7 of the genus Mico. The division of the marmosets into 3 distinct genera has recently been upheld [1] in order to avoid paraphyly of Callithrix with Cebuella [5�7], which is more closely related to Mico.

Previous molecular studies of marmoset phylogeny were based on sequences of the D-loop [5, 6] and β2-microglobulin [6, 8]. In this paper, we analysed se-quences of the mitochondrial cytochrome oxidase II (COII) gene in order to help define the phylogenetic relationships between the marmosets. The COII gene codi-fies a subunit of the cytochrome c oxidase enzymatic complex, which plays an es-sential role in aerobic respiration [9]. COII has a high evolutionary rate of non-synonymous substitutions in hominoids, but has a less rapid turnover in other mam-mals, including other primates [9�11]. In addition to complementing available mo-lecular studies, the present paper includes a number of taxa not included in previ-ous analyses, and provides some particularly important insights into the phyloge-netic relationships within Mico.

Methods

DNA Samples, PCR Conditions and Sequencing Blood samples were collected from 27 marmosets representing 12 of the 21 currently

recognised [1] morphospecies (table 1) and 1 Goeldi�s monkey, Callimico goeldii (a second sequence was obtained from the GenBank [12]), as the outgroup. DNA samples were ex-tracted from leucocytes following the protocol suggested by Sambrook et al. [13]. Poly-merase chain reaction (PCR) was carried out using primers specific for the tRNAAsp (5�-AAC CAT TCA TAA CTT TGT CAA-3�) and tRNALys (5�-CTC TTA ATC TTT AAC TTA AAA G-3�) genes [14], which flank the COII gene, as follows: initial denaturation at 94oC for 3 min, 30 cycles of denaturation at 94oC for 1 min, annealing at 45oC for 1 min and ex-tension at 72oC for 2 min, and final extension at 72oC for 10 min. The amplified products were electrophoresed in 1% agarose gel, excised and purified with Qiaex II (Qiagen, Valen-

242 Folia Primatol 2002;73:240�251 Sena/Vallinoto/Sampaio/Schneider/Ferrari/ Schneider

Table 1. The callitrichid species analysed in the present study

Species Code1 Origin (Brazilian state) Accession No.

Callimico goeldii CGO-1 Brookfield Zoo U36847 Callimico goeldii CGO-2 Acre AY118175 Cebuella pygmaea CPY National Primate Centre AY118176 Mico argentatus (n = 4) MAR-1, MAR-2 Rio Anauera (Pará) AY118182,

AY118183 Mico cf. emiliae (n = 4) MEM-RO Rio Jamari (Rondônia) AY118177 Mico emiliae (n = 1) MEM-PA Alta Floresta (Mato Grosso) AY118178 Mico saterei (n = 2) MSA-1 Rio Canumã (Amazonas) AY118180 Mico saterei (n = 1) MSA-2 Rio Mari-Mari (Amazonas) AY118181 Mico humeralifer (n = 2) MHU-1, MHU-2 National Primate Centre AY118184,

AY118185 Mico mauesi (n = 2) MMA-1, MMA-2 Rio Abacaxis (Amazonas) AY118186,

AY118187 Mico melanurus (n = 1) MME Serra dos Pacaás Novos (Rondônia) AY118179 Callithrix aurita (n = 2) CAU-1, CAU-2 Rio de Janeiro Primate Centre AY118188,

AY118189 Callithrix jacchus (n = 2) CJA-1, CJA-2 Extremós (Rio Grande do Norte) AY118190,

AY118191 Callithrix geoffroyi (n = 1) CGE Rio Doce (Minas Gerais) AY118192 Callithrix kuhli (n = 1) CKU Rio de Janeiro Primate Centre AY118193 Callithrix penicillata (n = 1) CPE-1 Rio Bagagem (Goiás) AY118194 Callithrix penicillata (n = 2) CPE-2, CPE-3 University of Brasília AY118195,

AY118196 1The same code was used for individuals with identical sequences.

cia, Calif., USA) as described in the kit protocol. Purified fragments were sequenced by the dideoxy chain termination method using a SequiTerm Cycle sequencing kit (Epicentre, Madison, Wisc., USA) and the protocol suggested by the vendor.

Phylogenetic Analyses DNA sequences were aligned by hand using the ESEE200b sequence editor [15]. Pair-

wise divergence values between operational taxonomic units were obtained using PAUP [16] using the maximum likelihood model. PAUP was also used to generate a maximum parsi-mony tree. Bootstrap values [17] were obtained using 2,000 replicates of the original DNA sequence data. A maximum likelihood tree was also constructed using the settings provided by the Modeltest [18]. The synonymous/non-synonymous substitution ratio in the COII se-quences of the different species was estimated with the MEGA2 program [19] using the modified Nei-Gojobori method (Jukes and Cantor model). MEGA2 was also used to esti-mate the transition/transversion ratio.

Results

A total of 549 base pairs (bp) of the COII gene were sequenced and aligned for the 13 morphospecies studied. In common with most mtDNA sequences, there was generally a marked imbalance in nucleotide proportions, especially at the second and third codon positions. The overall deviation resulted primarily from a relative


243 Folia Primatol 2002;73:240�251

lack of G at the second and third codon positions, where it contributed only 9.9 and 5.7% of the nucleotides, respectively. The transition/transversion ratio was also relatively high (5.6), and transitions were proportionally more frequent at the third codon position, although the sequences were not saturated by multiple hits, allow-ing the inclusion of all sites in the phylogenetic analysis (fig. 1).

The putative COII amino acid sequences were well conserved (fig. 2), exhibit-ing little variation compared with that of Callimico, with a synonymous/non-synonymous substitution ratio of 57.55. Conserved regions, such as the stretch of aromatic residues (104�106, 108�113) and the copper ligand site (His-161), were unchanged when compared with other mammals [9, 11]. The consensus maximum parsimony tree (fig. 3) divided marmosets into 3 clades, corresponding to the 3 genera. The sister-grouping of Callithrix and Mico was supported by a bootstrap value of 76%.

Mico formed 3 clusters. The first encompassed Mico argentatus and Mico emiliae from Pará, which were separated from the remaining species, with a boot-strap value of 76%. Mico humeralifer, Mico mauesi, Mico saterei and Mico melanurus also formed a cluster supported by a significant bootstrap value of 86%. Mico cf. emiliae from Rondônia was peripheral to this cluster, but with an incon-clusive bootstrap value of 57%. With the exception of the 2 populations of M. emiliae (14 bp different) and the Callithrix penicillata specimens (9�11 bp different in pairwise comparisons), all individuals belonging to the same species had virtu-ally identical sequences (1�3 bp different). Callithrix formed a strong monophyletic clade, supported by a bootstrap value of 100%. Callithrix aurita was very distinct from the remaining species, but while they form a significantly monophyletic clade, the internal arrangements were not resolved satisfactorily.

Fig. 1. Relationship between substitutions at third base positions and sequence divergencein 549 bp of the COII gene in marmosets. Divergence values higher than 15% refer tocomparisons with Callimico goeldii (outgroup).



245 Folia Primatol 2002;73:240�251

Fig. 2. Amino acid sequences inferred from 549 bp of the mitochrondrial COII gene inmarmosets and Callimico. The sequence starts at the 10th amino acid. Dots indicate aminoacids identical to Callimico (CGO-1), and ? indicates missing data. The two trans-membrane domains (25�48 and 63�83) are contained by the rectangles, the conservedaromatic residues (104�106, 108�113) are indicated by bars and the copper ligand site(161) is marked by an asterisk [10].


A variety of likelihood distance values were found between species of the same genus (table 2), although many values between members of different species were much lower than that of 2.84% found between M. emiliae and M. cf. emiliae (table 2). The maximum likelihood tree (fig. 4) also supports the distinction be-tween M. emiliae and M. cf. emiliae.

Discussion

In hominoids, COII is a variable protein subunit with an amino acid substitu-tion rate 5 times higher than those observed in other mammals. A correspondingly high rate is seen in cytochrome c, a subunit codified by a nuclear gene, with which COII interacts in the cytochrome c oxidase enzymatic complex [10]. In the marmo-sets, there was very little variation in the putative amino acids codified by the COII gene (although the amino terminal region was not sequenced fully), suggesting that this gene is subject to purifying selection (fig. 2).

The phylogenetic arrangement of the present study (fig. 3) is broadly similar to previous molecular phylogenies [5, 6], the main difference being the exclusion of Cebuella as the sister group of Mico. This finding is incongruent with most other genetic evidence [20�22] and cranial morphology [23]. It seems likely that COII

Fig. 3. Maximum parsimony tree of the 549 bp of the mitochondrial COII gene inmarmosets and Callimico, obtained with PAUP. Bootstrap percentages from 2,000replications are shown above each node.

Table 2. Maximum-likelihood values among Mico and Callithrix species. Abbreviations as in Table 1

CGO-1 CGO-2 CPY MEM-RO MEM-PA MME MSA-1 MSA-2 MAR-1 MAR-2 MHU-1 MHU-2 MMA-1 MMA-2 CAU-1 CAU-2 CJA-1 CJA-2 CGE CKU CPE-1 CPE-2

CGO-1 CGO-2 0.00567 CPY 0.29901 0.30508 MEM-RO 0.34880 0.34874 0.14184 MEM-PA 0.33882 0.34487 0.10941 0.02843 MME 0.31507 0.31502 0.13603 0.02622 0.03048 MSA-1 0.31982 0.31977 0.13969 0.02837 0.03262 0.01340 MSA-2 0.33157 0.33151 0.13969 0.02838 0.03268 0.01341 0.00368 MAR-1 0.36132 0.36126 0.14202 0.03317 0.01354 0.04007 0.04238 0.04239 MAR-2 0.36421 0.36414 0.14069 0.02910 0.01373 0.03826 0.04067 0.04069 0.00374 MHU-1 0.34263 0.34257 0.13343 0.02868 0.02849 0.01767 0.01973 0.01973 0.03803 0.03628 MHU-2 0.33768 0.33762 0.13383 0.02875 0.02855 0.01771 0.01977 0.01978 0.03812 0.03637 0.00000 MMA-1 0.32722 0.32716 0.13933 0.03160 0.03405 0.01127 0.01844 0.01845 0.03996 0.04081 0.01868 0.01868 MMA-2 0.33462 0.33456 0.14139 0.03349 0.03326 0.01774 0.01986 0.01987 0.04318 0.04114 0.01579 0.01579 0.00220 CAU-1 0.36453 0.35813 0.15301 0.16586 0.17274 0.17708 0.18973 0.18970 0.19243 0.18773 0.18326 0.18382 0.16489 0.18088 CAU-2 0.36568 0.35904 0.16585 0.16562 0.17276 0.18181 0.19039 0.19036 0.19320 0.18833 0.18368 0.18426 0.17212 0.18121 0.00577 CJA-1 0.33357 0.32748 0.15254 0.18081 0.17903 0.18339 0.19610 0.19607 0.18983 0.18966 0.18966 0.19023 0.18522 0.18731 0.06768 0.06760 CJA-2 0.32753 0.32148 0.14831 0.17647 0.17473 0.17905 0.19164 0.19161 0.18540 0.18516 0.18522 0.18578 0.18005 0.18287 0.06488 0.06470 0.00183 CGE 0.35191 0.34567 0.16556 0.17643 0.18334 0.17901 0.20057 0.20054 0.19429 0.19419 0.18517 0.18573 0.17488 0.18728 0.07652 0.07676 0.02172 0.01961 CKU 0.36299 0.35655 0.16127 0.19493 0.19294 0.18842 0.20151 0.20148 0.20442 0.20448 0.19492 0.19552 0.18592 0.18795 0.06488 0.06470 0.01160 0.00960 0.02213 CPE-1 0.34504 0.33864 0.16483 0.18515 0.18326 0.19706 0.21059 0.20061 0.19458 0.19448 0.19440 0.19501 0.20176 0.20146 0.06745 0.06549 0.01384 0.01178 0.02027 0.01840 CPE-2 0.35197 0.34573 0.16117 0.18964 0.18776 0.18337 0.19607 0.19604 0.19884 0.19881 0.18964 0.19021 0.18003 0.18729 0.07343 0.07355 0.01141 0.00944 0.02173 0.01571 0.01807 CPE-3 0.36351 0.35712 0.16821 0.20176 0.19971 0.19517 0.20831 0.20828 0.21130 0.21147 0.20176 0.20237 0.19394 0.19941 0.06694 0.06684 0.01560 0.01355 0.02635 0.00758 0.02267 0.01979


may have been unable to resolve this phylogenetic question because of the temporal proximity of the separation events [5].

Perhaps the most interesting results of this analysis are the redefinition of the Mico species groups and its support of the re-evaluation of M. cf. emiliae from Rondônia of Ferrari et al. [24, in prep.], in which the morphotype is assigned to a new species. The two classic species groups [3], argentatus and humeralifer, were based on the presence or absence of auricular hair. Members of the former group are known as the �bare-eared� marmosets, whereas those of the latter are �tassel-eared�. Despite changes in their taxonomic composition, the 2 groups were main-tained in subsequent classifications [4, 25]. In the present study, however, the ge-nus�s principal division (fig. 3) is based on a geographic rather than a morphologi-cal characteristic, given that the M. argentatus/M. emiliae clade is separated from that of the remaining species by the Rio Tapajós, one of the largest Amazon tribu-taries (fig. 5). The second main clade, located to the west of the Tapajós, includes both bare-eared (M. melanurus) and tassel-eared (M. humeralifer, M. mauesi and M. saterei) marmosets, as supported by previous studies [5, 6].

This geographic division is especially important to the evaluation of the geo-graphically distinct emiliae forms (fig. 3, 4). Following the original description of M. emiliae from Pará [26], the form was subsequently regarded as either a subspe-cies of M. argentatus [27, 28] or synonymous with M. argentatus argentatus [3].

Fig. 4. Maximum likelihood tree of the 549 bp of the mitochondrial COII gene inmarmosets and Callimico, obtained with PAUP [16] using the settings provided byModeltest [18].


249 Folia Primatol 2002;73:240�251

Fig. 5. Hypothetical radiation of the marmosets based on COII phylogeny and the present-day distribution of morphotypes based on the maximum parsimony tree obtained in thisstudy. The dark diamond represents the hypothetical point of origin for the marmosetradiation, while the circles represent the main dichotomies, supported by high bootstrapvalues. The line linked to the present-day distribution of Cebuella pygmaea is dotted in order to emphasise its occurrence in the relatively very distant past. Abbreviations are as intable 1.

However, de Vivo [4] not only reinstated M. emiliae, but also included specimens from Rondônia, on the basis of their morphological similarities. Nevertheless, the 2 populations are almost certainly allopatric [24] and, while the M. emiliae COII gene sequence is almost 2.5 times more similar to M. argentatus than to M. cf. emiliae, the latter is less distinct from M. saterei specimen 1 or M. melanurus, for example, than it is from M. emiliae (table 2).

The present study also adds to the growing consensus on the position of C. aurita as an early offshoot of the Callithrix radiation (fig. 3). This is consistent with conclusions based on morphological [29, 30], ecological [31], vocal [32] and mo-lecular [5, 6] characteristics. The results throw little light onto the relationship be-tween the remaining taxa, however. This, together with recent molecular [5], bio-chemical [33] and cytogenetic [34] data indicates a very rapid and recent radiation of Callithrix throughout central and eastern Brazil, and tends to support Hershkovitz�s [3] scheme, in which the different Callithrix morphotypes are distin-guished at no more than the subspecific level. Nevertheless, while phylogenetic


reconstruction may resolve adequately the status of distantly related taxa, such as the geographically distinct emiliae forms presented above, reliable evaluation of closely related taxa should include all available information on morphological, eco-logical and even behavioural traits [1]. When all these variables are taken into ac-count, the present consensus [1, 4, 32] is that all the forms presented here are valid species.

Overall, it may be difficult to construct reliable phylogenetic trees for some marmoset species using small fragments of mtDNA because of the short internode lengths, which reduce the effect of genetic drift that may lead to reciprocal mono-phyly. Given this, a more conclusive analysis of marmoset phylogeny will require a more systematic sample of a larger number of populations and gene markers.

Acknowledgments

We thank Arlindo Júnior and Helena Santos for technical support, and José Augusto P. Muniz (Centro Nacional de Primatas) and Alcides Pissinati (Centro de Primatologia do Rio de Janeiro) for providing specimens. This study was supported by CNPq, CAPES, FINEP and UFPa.

References

1 Rylands AB, Schneider H, Langguth A, Mittermeier RA, Groves CP, Rodríguez-Luna E: An assessment of the diversity of New World primates. Neotrop Primates 2000;8:61�93.

2 Ferrari SF: Ecological differentiation in the Callitrichidae; in Rylands AB (ed): Marmosets and Tamarins: Systematics, Ecology and Behaviour. Oxford, Oxford University Press, 1993, pp 314�328.

3 Hershkovitz P: Living New World Monkeys, Part 1 (Platyrrhini), with an Introduction to Pri-mates. Chicago, Chicago University Press, 1977.

4 de Vivo M: Taxonomia de Callithrix Erxleben, 1777 (Callitrichidae, Primates). Belo Horizonte, Fundação Biodiversitas, 1991.

5 Tagliaro CH, Schneider MPC, Schneider H, Sampaio I, Stanhope MJ: Marmoset phylogenetics, conservation perspectives, and evolution of the mtDNA control region. Mol Biol Evol 1997;14:674�684.

6 van Roosmalen MGM, van Roosmalen T, Mittermeier RA, Rylands AB: Two new species of marmoset, genus Callithrix Erxleben 1777 (Callitrichidae, Primates), from the Tapajós/Madeira interfluvium, South Central Amazonia, Brazil. Neotrop Primates 2000;8:2�18.

7 Tagliaro CH, Schneider MPC, Schneider H, Sampaio I, Stanhope MJ: Molecular studies of Cal-lithrix pygmaea (Primates, Platyrrhini) based on transferrin intronic and ND1 regions: Implica-tions for taxonomy and conservation. Gen Mol Biol 2000;23:729�737.

8 Canavez FC, Moreira MAM, Ladasky JJ, Pissinatti A, Parham P, Seuanez HN: Molecular phylog-eny of New World primates (Platyrrhini) based on β2-microglobulin DNA sequences. Mol Phyl Evol 1999;12:74�82.

9 Adkins RM, Honeycutt RL, Disotell TR: Evolution of eutherian cytochrome c oxidase subunit II: Heterogeneous rates of protein evolution and altered interaction with cytochrome c. Mol Biol Evol 1996;13:1393�1404.

10 Cann RI, Brown WM, Wilson AC: Polymorphic sites and the mechanism of evolution in human mitochondrial DNA. Genetics 1984;106:479�499.

11 Adkins RM, Honeycutt RL: Evolution of the primate cytochrome c oxidase subunit II gene. J Mol Evol 1994;38:215�231.

12 Ashley MV, Vaughn JL: Owl monkeys (Aotus) are highly divergent in mitochondrial cytochrome c oxidase (COII) sequences. Int J Primatol 1995;16:793�806.

13 Sambrook J, Fritch EF, Maniatis T: Molecular Cloning: A Laboratory Manual, ed 2. Cold Spring Harbor, Cold Spring Harbor Laboratory Press, 1989.

14 Ruvolo M, Disotell TR, Allard MW, Brown WM, Honeycutt RL: Resolution of the African homi-noid trichotomy by use of a mitochondrial gene sequence. Proc Natl Acad Sci USA 1991;88:1570�1574.


251 Folia Primatol 2002;73:240�251

15 Cabot EL, Beckenbach AT: Simultaneous editing of multiple nucleic acid and protein sequences with ESEE. Comput Appl Biosci 1989;5:233�234.

16 Swofford DL: PAUP*: Phylogenetic Analysis using Parsimony (and Other Methods). Sunderland, Sinauer Associates, 1998.

17 Felsenstein J: Phylogenies and the comparative method. Am Nat 1985;125:1�15. 18 Posada D, Crandall KA: Modeltest: Testing the model of DNA substitution. Bioinformatics

1998;14:814�818. 19 Kumar S, Tamura K, Jacobsen I, Nei M: MEGA2: Molecular Evolutionary Genetics Analysis,

Version 2.0. University Park and Tempe, Pennsylvania and Arizona State Universities, 2000. 20 Nagamachi CY, Pieczarka JC, Barros RMS: Karyotypic comparison among Cebuella pygmaea,

Callithrix jacchus and C. emiliae (Callitrichidae, Primates) and its taxonomic implications. Ge-netica 1992;85:249�257.

21 Canavez F, Alves G, Fanning TG, Seuánez HN: Comparative karyology and evolution of the Amazonian Callithrix (Platyrrhini, Primates). Chromosoma 1996;104:348�357.

22 Porter CA, Czelusniak J, Schneider H, Schneider MPC, Sampaio I, Goodman M: Sequences of the primate epsilon-globin gene: Implications for systematics of the marmosets and other New World primates. Gene 1997;205:59�71.

23 Natori M: Craniometrical variations among eastern Brazilian marmosets and their systematic relationships. Primates 1994;35:167�176.

24 Ferrari SF, Sena L, Schneider MPC: Definition of a new species of marmoset (Primates, Callitri-chinae) from southwestern Amazonia based on molecular, ecological, and zoogeographic evi-dence. Resumos do Congresso Brasileiro de Primatologia 1999;9:80�81.

25 Rylands AB, Coimbra-Filho AF, Mittermeier RA: Systematics, geographic distribution, and some notes on the conservation status of the Callitrichidae; in Rylands AB (ed): Marmosets and Tama-rins: Systematics, Ecology, and Behaviour. Oxford, Oxford University Press, 1993, pp 11�77.

26 Thomas O: On mammals from the lower Amazonas in the Goeldi Museum, Pará. Ann Mag Nat Hist 1920;6:226�283.

27 Cruz Lima E: Mamíferos da Amazônia. Contribuições do Museu Paraense Emílio Goeldi de História Natural e Etnografia, Belém, Pará. Belém, Museu Paraense Emílio Goeldi, vol 1: Intro-dução Geral e Primatas, 1944.

28 Ávila-Pires FD: Sobre a validade de Callithrix argentata emiliae Thomas, 1920, e sua distribui-ção geográfica. Resumos do Congresso Brasileiro de Zoologia 1985;12:318.

29 Natori M: Interspecific relationships of Callithrix based on the dental characters. Primates 1986;27:321�326.

30 Coimbra-Filho AF: Sistemática, distribuição geográfica e situação atual dos símios brasileiros (Platyrrhini-Primates). Rev Brasil Biol 1990;50:1063�1079.

31 Ferrari SF, Corrêa HKM, Coutinho PEG: Ecology of the �southern� marmosets (Callithrix aurita and Callithrix flaviceps): How different, how similar? in Norconk MA, Rosenberger AL, Garber P (eds): Adaptive Radiations of Neotropical Primates. New York, Plenum Press, 1996, pp 157�171.

32 Mendes SL: Padrões Biogeográficos e Vocais em Callithrix do Grupo jacchus (Primates, Callitri-chidae); unpublished PhD thesis, Universidade Estadual de Campinas, 1997.

33 Meireles CMM, Sampaio MIC, Schneider H, Schneider MPC: Protein variation, taxonomy and differentiation in five species of marmosets (genus Callithrix Erxleben, 1777). Primates 1992;33:227�238.

34 Nagamachi CY, Pieczarka JC, Schwarz M, Barros RMS, Mattevi MS: Comparative chromosomal study of five taxa of genus Callithrix, group jacchus (Platyrrhini, Primates). Am J Primatol 1997;41:53�60.

Documents

Mitochondrial COII Gene Sequences Provide New Insights into the Phylogeny of Marmoset Species Groups (Callitrichidae, Primates)