Amazonia has the richest primate fauna in the world. Nonetheless, the diversity and distribution of Amazonian primates remain little known and the scarcity of baseline data challenges their conservation. These challenges are especially acute in the Amazonian arc of deforestation, the 2500 km long southern edge of the Amazonian biome that is rapidly being deforested and converted to agricultural and pastoral landscapes. Amazonian marmosets of the genus Mico are little known endemics of this region and therefore a priority for research and conservation efforts. However, even nascent conservation efforts are hampered by taxonomic uncertainties in this group, such as the existence of a potentially new species from the Juruena–Teles Pires interfluve hidden within the M. emiliae epithet. Here we test if these marmosets belong to a distinct species using new morphological, phylogenomic, and geographic distribution data analysed within an integrative taxonomic framework. We discovered a new, pseudo-cryptic Mico species hidden within the epithet M. emiliae, here described and named after Horacio Schneider, the pioneer of molecular phylogenetics of Neotropical primates. We also clarify the distribution, evolutionary and morphological relationships of four other Mico species, bridging Linnean, Wallacean, and Darwinian shortfalls in the conservation of primates in the Amazonian arc of deforestation.
There are 146 primate species and subspecies in Amazonia, representing 20% of the global primate diversity1 and comprising the most diverse primate fauna in the world2. Nonetheless, primate diversity remains understudied in Amazonia, as manifested by regular discoveries of new species3,4,5. This incomplete taxonomic knowledge and the scarcity of basic ecological and distributional data for even well-known species is a major impediment to the design and implementation of effective conservation actions6. Bridging these substantial Linnean, Wallacean, and Darwinian shortfalls is also a pre-requisite to understanding the biotic and abiotic drivers of the evolutionary history of Neotropical primates7.
The Amazonian arc of deforestation concentrates nearly one-third of all global deforestation8,9. It also harbours 52 primate species—over one third of Amazonian primates—of which 42% are threatened with extinction according to the IUCN10 (Supplementary Table S1). Therefore, research and conservation efforts on the primates of the arc of deforestation are priority within the Neotropics.
Endemic to this region are marmosets of the genus Mico3,11. One of the main taxonomic, distributional, and evolutionary uncertainties—the Linnean, Wallacean and Darwinian shortfalls, respectively—in Mico species concern the Snethlage’s marmoset M. emiliae12. The pelage colour of this species has confounded researchers for a century hindering accurate assessments of the taxonomy and distribution of this and other five Mico species, as well as the assessment of species diversity in this genus.
The description of M. emiliae in 1920 was based on the pelage colour of two specimens collected at “Maloca, upper Curuá River”12—an imprecise locality in the Tapajós–Xingu interfluve, southern Amazonia, Brazil. Researchers debated for 70 years if M. emiliae was distinct from M. argentatus13,14,15,16,17,18 until both were accepted as separate species based on patterns of pelage colouration19. Nonetheless, additional taxonomic confusion has arisen concerning the identification of marmosets from three interfluves—Guaporé–Ji-Paraná, Ji-Paraná–Aripuanã, and Juruena–Teles Pires (Fig. 1)—as M. emiliae based on an apparent similarity in pelage colour patterns of these marmosets.
The marmosets from the Ji-Paraná–Aripuanã and from the Guaporé–Ji-Paraná interfluves are currently considered two distinct species, M. marcai20 and M. rondoni21, respectively. However, the eight specimens collected in 1995 from two localities in the Juruena–Teles Pires interfluve and stored in the Museu Paraense Emílio Goeldi (Goeldi’s Museum) have been classified as M. emiliae by some authors21,22, whereas others have considered only the Tapajós–Xingu marmosets as M. emiliae11,23,24,25, casting into doubt the taxonomic identity of the Juruena–Teles Pires marmosets.
To date, M. emiliae, M. marcai, M. rondoni, and the Juruena–Teles Pires were not subjected to morphological or molecular studies with enough specimens to permit the differentiation of population from species-level diversity. Primate species are known to be limited by rivers in Amazonia26 and thus it is highly likely that the Juruena–Teles Pires marmosets represent a distinctive species—and, if that were the case, this taxon would occur in an area of intense deforestation and thus likely to be threatened with extinction4,27.
As part of the findings from an ongoing research on ecology, evolution, and systematics of Amazonian marmosets, which has as one of the main goals to clarify the taxonomy of the Juruena–Teles Pires marmosets, we found that these marmosets present a cohesive pelage colour pattern that is also distinctive from M. emiliae and other geographical neighbour taxa. Here we test whether the Juruena–Teles Pires marmosets represent a species distinct from M. emiliae, as well as M. argentatus, M. marcai, and M. rondoni. We conducted field expeditions to collect new distribution records, tissue samples and specimens, and examined specimens in museums, including all type and voucher specimens of M. emiliae, M. marcai, M. rondoni, and Juruena–Teles Pires marmosets plus numerous M. argentatus specimens. We then generated new morphological, phylogenomic and distribution data, which were analysed and interpreted within an integrative taxonomic framework.
Integrative approach in taxonomic hypothesis-testing and decision-making
We adopted an integrative approach28 to test the hypothesis that Juruena–Teles Pires marmosets belong to a species distinctive from M. emiliae, and to test whether M. marcai, M. rondoni, and M. argentatus are valid species. Integrative taxonomy provides an objective framework to test species hypotheses and accommodates practical constraints of data collection and analyses, resulting in accurate taxonomic decisions and more stable classifications––especially when morphology and nuclear DNA data are used28,29. Here we use pelage colour, nuclear genomic DNA, and distribution datasets, each associated with specific criteria to refute or to accept the null hypothesis (Supplementary Table S3), following the monophyly and diagnosability conceptualization of the phylogenetic species concept30.
Between 2015 and 2018, we conducted ten field expeditions across southern Amazonia to obtain new distribution records, specimens, and samples to overcome the previous scarcity of these materials and data in museums and in literature. The surveys consisted of trekking in the forest or canoeing up streams, while playing long-calls of M. marcai to stimulate vocalization and approximation of marmosets and increase detection probabilities3. For each observation we registered the exact location with a GPS device and the type of habitat. The collection of specimens and samples of muscular tissue were carried out with a permit from the Chico Mendes Institute of Biodiversity Conservation (permit number 50416), the federal institution that regulates biodiversity research in Brazil. Specimen sampling followed the protocols established for studying primates in protected areas of Amazonia31 and the code of best practices for field primatology of the International Primatological Society (http://www.internationalprimatologicalsociety.org/policy.cfm). Specimens were stored in the mammal collections of the National Institute of Amazonian Research and Goeldi’s Museum; tissue samples were preserved in 96% ethanol in the Animal Genetics Tissue Bank of the Federal University of Amazonas. No specimen was subjected to experimental conditions or protocols.
We collected data of pelage colour of 10 chromogenetic fields32 (Supplementary Fig. S4) through the direct examination of 598 skins of specimens obtained in the field and stored in museums encompassing all known species of the genera Callibella and Mico (Supplementary Table S5). The chromogenetic fields are informative morphological characters because they summarize most of the variation in the pelage colouration and provide an objective basis for comparisons in marmosets3,18 and other primates4,18,33. We also examined tegument colour on the face and ears, and hairiness of the ears to delimit supra-specific lineages as these characters are synapomorphies of four species groups of Mico3. Based on these four species groups of Mico and morphological synapomorphies, we expected (i) marmosets from Juruena–Teles Pires interfluve to form a lineage together with M. marcai and M. melanurus; (ii) M. argentatus, M. emiliae, M. intermedius, M. leucippe, M. munduruku, and M. rondoni to form a second lineage; (iii) M. humeralifer and M. mauesi to form a third lineage; and (iv) M. saterei to form a fourth, single-species lineage.
After DNA extraction from muscular tissue samples34, we used a modified protocol of ddRAD sequencing35 optimized for the IonTorrent PGM that permits simultaneous digestion, ligation, and barcoded adapter incorporation (https://github.com/legalLab/protocols-scripts)36. Samples were sequenced on the IonTorrent PGM using the manufacturer’s recommended protocol and the sequencing reads were processed using the pyRAD pipeline37. For the de novo assembly, we used a minimum coverage of 5 × per locus, assembling all fragments of 320–400 base pairs. Nucleotides with PHRED scores < 30 were excluded, as well as loci with more than three low-quality nucleotides. Following demultiplexing and extraction of loci using the above criteria, we proceeded with clustering of alleles within loci and of loci across individuals. We generated a dataset for downstream analyses that included all individuals and loci present in at least 50% of the samples38, which was subjected to PartitionFinder239 to estimate the optimal number of partitions. For Bayesian Inference analysis carried out in BEAST240, site models, clock models and trees of each partition were unlinked. Site models were implemented based on the PartitionFinder2 results, and all partitions were allowed to evolve under an uncorrelated lognormal model with clock rates and standard deviations of clock rates being estimated. We ran the MCMC for 109 generations, collecting 5 × 104 samples. We carried out partitioned Maximum Likelihood analysis in RAxML41 with site models for each partition based on the PartitionFinder2 results. We sampled the genomes of representative species from the four Mico lineages as delimited in the morphology section, to have a representation of all four major lineages of Mico in the analyses. We sampled multiple individuals of the Juruena–Teles Pires marmosets and of the other species of the same lineage, which included M. marcai, as well as the other species previously confounded with M. emiliae (M. argentatus and M. rondoni); we used Callibella humilis, Cebuella niveiventris, and Callithrix jacchus as outgroups (Supplementary Table S6). We consider as strong support for Bayesian inference a posterior probability ≥ 0.95 and the equivalent bootstrap proportion for Maximum Likelihood inference (≥ 70%)42.
We then carried out a path sampling analysis in BEAST2 focusing on M. argentatus, M. emiliae, and M. leucippe, whose phylogenetic relationships are unresolved according to our results and to a previous study3. The objective was to investigate if the observed lack of monophyly is compatible with incomplete lineage sorting, hybridization or the existence of polymorphic species, i.e. a species that encompass others currently recognized as distinctive based on pelage colour patterns. Using DiscoSnp-RAD43, we extracted single nucleotide polymorphisms (SNPs) from our reads using a minimum read depth of 5. The highest quality SNPs were then sampled from each locus and the SNPs were further filtered on quality. We retained only those SNPs with rank > 0.9––a statistic incorporating the discriminant power and read coverage of each SNP––and those that were present in at least 90% of the samples. We then used path sampling in BEAST2, and collected marginal probabilities of alternate taxonomic hypotheses: one species (M. emiliae-leucippe-argentatus), two species (M. argentatus and M. emiliae-leucippe; M. emiliae and M. leucippe-argentatus; M. leucippe and M. emiliae-argentatus) and the current hypothesis of three species (M. emiliae, M. leucippe, and M. argentatus). Marginal probabilities of the competing taxonomic hypotheses were then compared by Bayes factors44.
All the localities of Mico specimens collected in the field and examined in museums were georeferenced and grouped according to their morphotypes on a relief map (vegetation type, altitude, river basins) to explore potential geographical barriers and areas of gene flow using QGIS45.
All the marmosets from the Juruena–Teles Pires interfluve have discrete and objectively identifiable diagnostic states in pelage colour characters: the uniform lead coloration of saddle and rump and the cream-silvery underparts, which present orangish hues in living specimens (Fig. 2), are autapomorphies that readily distinguish these marmosets from M. emiliae and all other Mico species. The Juruena–Teles Pires marmosets, M. argentatus, M. emiliae, M. marcai, M. melanurus, and M. rondoni are clearly diagnosable in terms of pelage colour (Table 1; Fig. 3).
We obtained a molecular dataset consisting of 2081 loci spanning 717,129 base pairs, representing an average sampling effort at the DNA level of 22,410 base pairs per specimen (n = 32). After analyses in PartitionFinder2, the 2081 loci were clustered into 421 partitions. In both Bayesian inference and Maximum Likelihood phylogenetic analyses, species monophyly and species-level relationships were highly supported (pp ≥ 0.99; bp > 70%) with exception of M. emiliae, M. leucippe, and M. argentatus (Fig. 4; Supplementary Fig. S7). As expected, we retrieved four lineages or species groups in the genus Mico and found lineage membership patterns coherent with our predictions based on morphological synapomorphies. The Juruena–Teles Pires marmosets are monophyletic, sister to M. marcai, and both form a strongly supported lineage with M. melanurus. Mico emiliae is monophyletic, but nested within a clade that also includes paraphyletic M. leucippe and M. argentatus. These three species, together with M. intermedius, M. munduruku, and M. rondoni comprise the second major lineage of Mico (Fig. 5). A third lineage is comprised of M. humeralifer and M. mauesi, and Mico saterei was retrieved as an additional monotypic lineage. The clade formed by the Juruena–Teles Pires marmosets does not include M. emiliae specimens, nor is it sister to the M. emiliae clade; actually both taxa belong to separate lineages of Mico.
Our path sampling analysis clearly supports M. argentatus, M. emiliae, and M. leucippe as three separate species, considering that a Bayes factor > 10 is decisive44. Our results reject the hypothesis of only one species––M. emiliae-M. leucippe-M. argentatus (BF = 367.46)––and the hypotheses of two species––M. argentatus and M. emiliae-M. leucippe (BF = 145.08), M. emiliae and M. leucippe-M. argentatus (BF = 99.44), or M. leucippe and M. argentatus-M. emiliae (BF = 348.29). These results support the unambiguous diagnosis of these three species according to our morphological data.
The ranges of M. emiliae and Juruena–Teles Pires marmosets are separated by the Teles Pires River and in the headwaters of the Teles Pires River both are substituted by M. melanurus (Fig. 6; Supplementary Table S8). There is no evidence of range overlap between the Juruena–Teles Pires marmosets and any other Mico species. The distributions of M. emiliae and Juruena–Teles Pires marmosets are allopatric; M. emiliae is parapatric with M. leucippe around the Cachimbo highlands3 and both M. emiliae and the Juruena–Teles Pires marmosets are parapatric with M. melanurus at the extreme south of their distributions. Mico marcai46, M. rondoni21 and M. argentatus25 are allopatric to M. emiliae and the Juruena–Teles Pires marmosets, being separated by rivers and by the ranges of other Mico species.
Integrative approach in taxonomic hypothesis-testing and decision-making
The marmosets from the Juruena–Teles Pires interfluve have unique states of pelage colour characters when compared to all nominal species of Mico species, thus are clearly diagnosable, form a fully supported clade in our phylogenomic trees and are found only in the Juruena–Teles Pires interfluve, without evidence of range overlap with any other congeneric species. The criteria adopted here reject the null hypothesis that marmosets from the Juruena–Teles Pires interfluve are M. emiliae or another known species of Mico. We therefore describe them as a new species.
Order Primates Linnaeus, 1758
Family Callitrichidae Gray, 1821
Genus Mico Lesson, 1840
Mico schneideri sp. n. Costa-Araújo, Silva-Jr., Boubli, Rossi, Hrbek & Farias
Holotype. INPA 7293, tissue CTGA 5934, field number RCA 60, adult female, stuffed skin, skull, skeleton. This specimen was collected on April 2nd, 2016 in an urban forest fragment located in Paranaíta city (09°41′21″ S, 56°29′10″ W), on the left margin of Teles Pires River, Mato Grosso State, Brazil, by Rodrigo Costa Araújo.
Type locality. Paranaíta municipality, left margin of the Teles Pires River, northern Mato Grosso State, Brazil (09°41′21''S, 56°29′10''W).
Paratypes. Urban forest fragment, Paranaíta city, left margin of Teles Pires River (09°41′21″ S, 56°29′10″ W): INPA 7294, tissue CTGA 5935, field number RCA 61, adult female; INPA 7295, tissue CTGA 5936, field number RCA 62, adult male; both preserved in fluid and collected on April 2nd, 2016 by Rodrigo Costa Araújo. Urban forest fragment, Alta Floresta city, left margin of Teles Pires River (09°51′50′′ S, 56º04′20′′ W): MPEG 24595, field number RA 41, subadult male, stuffed skin, skull; MPEG 24596, field number RA 42, subadult male, stuffed skin, skull; both collected on October 16, 1995 by R. Alperin, R. Rodrigues, and N. Silva. Urban forest fragment, Alta Floresta city, left margin of Teles Pires River (09°53′04′′ S, 56°04′21′′ W): UFMT 3851, field number RVR 40, adult male, stuffed skin, skeleton, and tissue, collected on May 6, 2014 by Rogério Rossi. Peri-urban forest fragment, Alta Floresta, left margin of Teles Pires River (09°58′57′′ S, 56°04′21′′ W): UFMT 4833, field number RVR 43, adult male, stuffed skin, skeleton, and tissue; UFMT 3852, field number RVR 44, adult male, stuffed skin, skeleton, and tissue; UFMT 4834, field number RVR 45, adult male, stuffed skin, skeleton, and tissue, collected on May 6, 2014 by Rogério Rossi. Ourolândia, left margin of Teles Pires River (10°23′26″ S, 56°24′28′′ W): MPEG 24606, field number RA 63, male, stuffed skin, collected on October 22 1995; MPEG 24608, field number RA 68, adult male, stuffed skin, skull; MPEG 24609, field number RA 69, adult female, stuffed skin, skull; MPEG 24610, field number RA 70, adult female, stuffed skin, skull; MPEG 24611, field number RA 71, adult male, stuffed skin, skull; all collected on October 23 1995 by R. Alperin, R. Rodrigues, and N. Silva.
Diagnosis. Uniform lead colour on saddle and rump, and underparts cream-silvery with orange hues.
Etymology. The new species is named in honour of Professor Horacio Schneider, a pioneer, and a major contributor to the phylogenetic studies of Neotropical Primates, who humbly accepted to have this species named in his honour.
Description of the holotype. Hairs on the face are short, mostly white but also black or bi-banded black-white, distributed all over the face, denser in circumbucal area, rhinarium and sides of the face, increasing in size towards the sides of the face; eumelanic vibrissae on the rhinarium, supraorbital region and along the zygomatic bone. Face tegument eumelanic in the centre of the supraorbital area, around the eyes, along the sides and the middle of the nose thrill, on the rhinarium and circumbucal area; dark brown eyes. White and long hairs on outer and inner pinnae, longer and denser on the inner surface, partially covering the pinnae; ear tufts absent; eumelanic tegument on a large proportion of each ear, but paler than the eumelanic tegument of the face. White hairs surrounding the face and on the sides of the head, longer than on the face; white hair on the head, covering the lower portion of the pinnae; black crown, separated from facial hairs by a horizontal line of white hairs. Gray mantle. Dorsal forearms greyish cream, blackish golden hairs on hands; ventral forelimb hairs cream, white on ventral neck and chest. Cream hairs on the belly. Saddle and rump of a uniform lead colour. Hairs on underparts cream-silvery on the anterior region, grading to light orange towards the posterior area on the dorsal and ventral hind limbs; ventral hind limb hairs pure orange on the posterior area, whereas pure cream on the anterior area. Goldenish orange hairs on feet. Blacktail with orange hairs on the ventral surface of tail insertion, an inch in length. The tegument is slightly eumelanic on the ventral surface of hands, unpigmented on the ventral surface of feet; claw-like nails, curved dorsoventrally in all digits except the hallux, which bears a flat nail.
Intra-specific morphological variation. There is only minor individual variation among the marmosets from the Juruena–Teles Pires interfluve manifested as the conspicuousness of the orangish tone over the basic cream colour of hairs in the chest, belly, and ventral underparts. In prepared skins, such variation is little perceptible, as well as the orangish colour of the underparts, whereas in fresh specimens the ventral region of fore and hind limbs and belly hairs can show a bright and vivid orange-cream colour. A small amount of cream hair can be observed on and around the ears, chest, and in an even smaller quantity on the mantle in the paratypes INPA 7294 and INPA 7295. There is also a tonal variation in the colour of the hairs on hands, feet, and lower portion of underparts of the paratypes deposited at MPEG, attributable to fading in storage, varying from goldenish orange to light orange on feet and posterior area of underparts, and varying from golden to light yellow on hands (see Fig. 2, Supplementary Fig. S9).
Geographic distribution. Mico schneideri sp. n. is endemic to the Juruena–Teles Pires interfluve, southern Amazonia, Mato Grosso State, Brazil. The species distribution is limited by the Juruena River to the west and by the Teles Pires River to the east, proceeding north to their confluence. The southern portion of the species range is less well-defined, but it extends to the headwaters of the Juruena and Teles Pires rivers, but no further south than the city of Lucas do Rio Verde. In this region, the Amazonia biome transitions to the Cerrado biome, and thus parapatry is expected between M. schneideri sp. n. and M. melanurus, the only species of Mico known to occur in the Cerrado.
Habitat. Primary and secondary terra firme forests, and Amazonia-Cerrado transitional forests.
Suggested vernacular names. Schneider’s marmoset (English); sagui-de-Schneider (Portuguese).
Mico emiliae is morphologically diagnosable, monophyletic, and allopatric along most of its range. Mico argentatus and M. leucippe are morphologically diagnosable and, although they were not retrieved as monophyletic in our phylogenetic inferences, path sampling analysis of genomic data associated with morphology and distribution data provide decisive support for their recognition as distinct species. Mico argentatus is allopatric on the east bank of the Xingu River and on the west bank of the Xingu River, this species is parapatric to M. leucippe––which is, in turn, apparently restricted to a narrow area between the Jamanxim and the Irirí-Curuá Rivers3,25. The distribution of M. leucippe is not limited by any conspicuous geographical barrier that could prevent introgression with M. argentatus or M. emiliae.
In the early twentieth century, the German ornithologist Emilie Snethlage conducted two field expeditions across the forests of the Tapajós–Xingu interfluve, southern Amazonia, Pará State, Brazil. In the second expedition, Snethlage discovered ‘small and enchanting black-headed marmosets, closely related to [M. argentatus], and certainly belonging to a new species’47. She collected two of these marmosets at “Maloca, upper Curuá River”, which became the type specimens and the type locality of M. emiliae12‚. Since then, M. emiliae accumulates a history of taxonomic uncertainties, which permeates the taxonomy of M. argentatus, M. marcai, M. rondoni, and Mico schneideri sp. n., and thus the pattern of species diversity in this genus. None of these five marmoset species had been subjected to robust taxonomic assessments to date due to a scarcity of specimens for morphology-based studies and the lack of tissue samples for molecular phylogenetic analysis.
The two marmosets Snethlage collected at Maloca12 and the three specimens collected at the Cachimbo highlands48 were, until now, the only specimens known from the region of the type locality of M. emiliae in the Tapajós–Xingu interfluve. Following its original description12, M. emiliae was considered as invalid or as subspecies of M. argentatus13,14,15,16,17,18. Decades later, when these two species were considered valid, morphologically divergent marmosets from three other interfluvial regions in Amazonia were identified as M. emiliae19 leading to questions about the taxonomic identity of these populations and of M. emiliae. Moreover, the type locality of M. emiliae is uninformative: maloca is the name of the traditional dwellings of the Chipaya and Curuahy, the native peoples from the Curuá River basin that Snethlage travelled with––and there are several malocas depicted on her expedition map47. Until now, no geographical coordinates had been proposed for the type locality of M. emiliae because no field research was conducted on this species subsequent to Snethlage’s expedition.
To resolve these uncertainties, we first focused on finding M. emiliae type locality. Initially, we studied Snethlage’s narrative, the map of her field expedition, and the labels of M. emiliae type specimens, which she handwrote. Based on these sources we identified only one “Maloca”––which matches exactly the spelling on the map and the labels of type specimens. Based on this deduction and using local relief references we restricted the type locality of M. emiliae and extracted the geographic coordinates of this specific maloca which is located not far south of the mouth of the Curuaés stream on the east bank of the Curuá River within the Tapajós–Xingu interfluve (09°41′21′′ S, 56°29′10′′ W).
We then conducted field expeditions to the type locality of M. emiliae as here defined, where we recorded groups and collected a specimen that had the same morphology and pelage coloration as the type specimens of M. emiliae, confirming the location of this specific maloca where Snethlage was in 1911. We also conducted several field expeditions within the Tapajós–Xingu to collect additional distribution records, specimens, and tissue samples of M. emiliae, as well as within all other main interfluves across southern Amazonia to obtain these materials and data from all Mico species. The surveys included the localities where specimens of Mico schneideri sp. n. were previously collected and the type localities of M. marcai and M. rondoni, so our datasets included topotypic material from these four species.
Combining data collected from specimens obtained in the field and examined in museum collections, we were able to demonstrate that M. emiliae is an evolutionarily independent lineage composed of specimens distinctive in pelage colour, which occur in an area largely restricted by the Rivers Curuá, Irirí, São Benedito and Teles Pires in the west and north, and by the Xingu River at the east. We also identified the type locality of M. emiliae for which we provided exact geographical coordinates, and redefined its distribution––extending it 200 km south. The southern limit of the distribution remains uncertain, but it probably extends to the northern edge of the Cerrado biome––where in further surveys we expect to find a small contact zone between M. emiliae and M. melanurus at the headwaters of the Teles Pires and Xingu rivers. We deem it probable that M. emiliae also occurs in the Cerrado savanna vegetation, as we found groups of this species in Amazonian white-sand savanna ecosystems––scrubland and campinarana49 vegetation––in our surveys of the Amazonian forests of the Tapajós–Xingu interfluve.
The marmosets from the Juruena–Teles Pires interfluve herein described as Mico schneideri sp. n. are, in fact, sister to M. marcai and not to M. argentatus50 nor nested within M. emiliae22. Mico schneideri sp. n. and M. marcai are diagnosable by pelage colour patterns, are reciprocally monophyletic, belong to a different lineage than M. emiliae, and the distributions of both species are allopatric. Therefore, Mico schneideri sp. n. is a valid species. Further field surveys are necessary to better define the southern limit of the geographical distribution of this new species. As there is no evidence that Mico schneideri sp. n. occurs in Amazonian white-sand savanna vegetation, we do not expect to find it in the Cerrado biome.
The inclusion of topotypic specimens of M. emiliae and M. rondoni in a molecular phylogenetic analysis for the first time also confirmed the hypothesis that M. rondoni is a distinct and valid species. Although historically confounded with M. emiliae19,20, M. rondoni is monophyletic, has a distribution restricted to the Guaporé–Ji Paraná interfluve which is allopatric to M. emiliae, and is characterized by a distinctive pelage colour pattern when compared to other congenerics. Mico rondoni was described based on pelage colour and a molecular phylogeny21,50 but, actually, the specimens considered as M. emiliae in the phylogeny belong to Mico schneideri sp. n.
Mico argentatus is also clearly diagnosable by pelage colour patterns. Although M. argentatus and M. leucippe were retrieved as paraphyletic in this and in a previous study3, our path sampling analysis provides decisive evidence for their recognition as separate evolutionary lineages. The path sampling analysis collects the marginal likelihoods of phylogenetic trees differing only in the number of species, thus isolating the effect of number of species in the likelihoods; the marginal likelihoods are then compared using the Bayes factor to test differences in the number of species assumed a priory. Mico argentatus is allopatric to other congeneric species on the east bank of Xingu River and, on the west bank of the Xingu River, M. argentatus is separated from M. emiliae by the Irirí River, and by the range of M. leucippe3,25. Nonetheless, M. argentatus is parapatric to M. leucippe on the west bank of Xingu River and without a clear geographical barrier that could prevent gene flow between populations of both taxa. Moreover, there is no barrier also between populations of M. leucippe and M. emiliae around the Cachimbo highlands.
We consider that our morphological and phylogenomic results provide strong evidence to recognize M. argentatus and M. leucippe as distinct species. Given that most closely related Amazonian primate taxa are separated by physical barriers, we are now exploring the biological and evolutionary processes underlying the existence of such distinct but parapatric marmoset species in a region where there are no apparent physical barriers to gene flow. The taxonomy and distribution of M. marcai is also not well-resolved. It is unclear if this species and M. manicorensis are distinct from each other and from M. nigriceps. All three taxa are morphologically similar and occur in geographically adjacent areas within the Ji-Paraná–Aripuanã interfluve25.
The centenary uncertainties in the taxonomy of M. emiliae emerged from a case of pseudo-cryptic diversity in Neotropical primates. Pseudo-cryptic species are morphologically distinctive51,52 but such distinctness is overlooked due to methodological inadequacies53. Previous taxonomic studies on this species relied on pelage colour patterns obtained from few individuals, some of faded pelage colour, as the sole source of information on which taxonomic decisions were based. The low number of specimens and the few known distribution records in previous studies probably caused the taxonomic confusion surrounding M. emiliae, M. argentatus, M. marcai, M. rondoni, and Mico schneideri sp. n., which in reality are unambiguously diagnosable and distinctive in pelage colour as shown here. Mico schneideri sp. n. was hidden at plain sight: the eight specimens from the Juruena–Teles Pires interfluve, stored in the Museu Paraense Emílio Goeldi since 1995, have been consistently misidentified as M. emiliae21,22,50 and are here designated as Mico schneideri sp. n. paratypes.
Using an integrative taxonomic analysis, carried out with data obtained from a greater number of specimens than in any previous study, we were able to accurately diagnose Mico schneideri sp. n., M. argentatus, M. emiliae, M. marcai, and M. rondoni. The new tissue samples and expanded occurrence data were paramount in allowing a first phylogenomic inference of evolutionary relationships and a robust taxonomic assessment of all these species, as well as a refined understanding of the distribution of M. emiliae and Mico schneideri sp. n. Our research highlights the importance of fieldwork, scientific collections, and the use of multiple data sources within an integrative taxonomic framework of delimiting primate species3,4. Fieldwork is paramount to obtain specimens, samples, and field records, to overcome the scarcity of such materials and data currently available in museums and literature, and to allow the accurate distinction between species and population patterns in morphology and molecular data, as well as to define species geographic distributions.
A century of taxonomic uncertainties surrounding M. emiliae has hopefully come to an end with the description of Mico schneideri sp. n. from the Juruena–Teles Pires interfluve and the recognition of M. argentatus, M. emiliae, M. leucippe, M. marcai, and M. rondoni as distinct species. Nonetheless, the existence of M. argentatus and M. leucippe as distinct species in the absence of physical barriers to gene flow on the west bank of Xingu River is intriguing and the biological processes underlying such scenario deserve further clarification. More studies are also needed on the taxonomy and distribution of M. marcai, M. manicorensis, and M. nigriceps to clarify the species diversity in genus Mico. Criteria-driven test of species hypotheses based on data from pelage colour, genomic DNA, and geographic distribution can resolve the taxonomy and evolutionary relationships of Neotropical primates while reducing subjectivity and adding rigour to the decision-making process. Fieldwork is a critical component of such studies in providing information on phenotypic variation, distribution records, samples and specimens not foreseen in earlier collections and is necessary to further advance the knowledge of diversity and distribution of Neotropical primates. Our research indicates that there are 16 Mico species, 19 marmoset species in Amazonia (including the dwarf marmoset Callibella humilis and two pygmy marmoset species Cebuella pygmaea and C. niveiventris), and 25 marmoset species in Neotropics (including six Callithrix species from eastern Brazil). Delimiting taxa and understanding their distributions is fundamental to support the implementation of effective conservation strategies for Amazonian primates. Such efforts are especially urgent in the Amazonian arc of deforestation, where 52 primate species are largely overlooked by the scientific and conservation community but face high rates of habitat loss54. Primates are effective conservation programme flagship species55, play an important role in ecosystem functioning56 and are the most threatened group of vertebrates in the world2. Therefore, characterizing primate species diversity and distribution in the Amazonian arc of deforestation is a necessary first step on which an entire science-based conservation effort depends and which lends support to biodiversity conservation in this region before the entire biome reaches an environmental point of no return57.
Genomic data generated for this study have been deposited in Genbank (PRJNA552061) and alignment and other data generated during and/or analysed during the current study are available at https://github.com/legalLab/publications.
This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the International Code of Zoological Nomenclature, under the ZooBank Life Science Identifier (LSID) urn:lsid:zoobank.org:pub:B43E02D9-F8CD-4ADA-9B86-CD9378C51723 and urn:lsid:zoobank.org:act:9B4FFB49-FC65-45CC-862B-AA97E1C8F5BA, respectively, and can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix ‘http://zoobank.org/’.
IUCN. Primates of the Amazon. International Union for the Conservation of Nature https://www.primate-sg.org/primate_diversity_by_region (2018).
Estrada, A. et al. Impending extinction crisis of the world’s primates: Why primates matter. Sci. Adv. 3, e1600946. https://doi.org/10.1126/sciadv.1600946 (2017).
Costa-Araújo, R. et al. The Munduruku marmoset: a new monkey species from southern Amazonia. PeerJ 7, e7019. https://doi.org/10.7717/peerj.7019 (2019).
Boubli, J. P. et al. On a new species of titi monkey (Primates: Plecturocebus Byrne et al. 2016) from Alta Floresta, southern Amazon, Brazil. Mol. Phylogenet. Evol. 132, 117–137. https://doi.org/10.1016/j.ympev.2018.11.012 (2019).
Marsh, L. K. A taxonomic revision of the saki monkeys, Pithecia Desmarest, 1804. Neotrop. Primates 21, 1–163. https://doi.org/10.1896/044.021.0101 (2014).
Rylands, A. B. & Mittermeier, R. A. Primate taxonomy: Species and conservation. Evol. Anthropol. 23, 8–10. https://doi.org/10.1002/evan.21387 (2014).
Byrne, H. et al. Uncertainty regarding species delimitation, geographic distribution, and evolutionary history of titi monkey species (Plecturocebus, Pitheciidae) from southwestern Amazonia. Int. J. Primatol. in press.
DeFries, R. S. et al. Carbon emissions from tropical deforestation and regrowth based on satellite observations for the 1980's and 1990's. Proc. Natl. Acad. Sci. 99, 14256. https://doi.org/10.1073/pnas.182560099 (2002).
Carrero, G. C., Fearnside, P. M., Valle, D. R. & Alves, C. S. Deforestation trajectories on a development frontier in the Brazilian Amazon: 35 years of settlement colonization, policy and economic shifts, and land accumulation. Environ. Manag. 66, 966–984. https://doi.org/10.1007/s00267-020-01354-w (2020).
IUCN. The IUCN Red List of Threatened Species. Version 2021-1. https://www.iucnredlist.org (2021).
Rylands, A. B., Coimbra-Filho, A. F. & Mittermeier, R. A. The systematics and distributions of the marmosets (Callithrix, Callibella, Cebuella, and Mico) and Callimico (Callimico) (Callitrichidae, Primates). in The Smallest Anthropoids, Developments in Primatology: Progress and Prospects (eds. Ford, S. M., Porter, L. M. & Davis L. C.) 25–61 (Springer, 2009).
Thomas, O. On mammals from the lower Amazons in the Goeldi Museum, Pará. J. Nat. Hist. 6, 266–282 (1920).
Cruz-Lima, E. Mamíferos da Amazônia Vol. 1 (Museu Paraense Emilio Goeldi, 1944).
Cabrera, A. Catálogo de los mamíferos de America del Sur. Rev. Mus. Argent. Cienc. Nat. 4, 1–307 (1957).
Cabrera, A. & Yepes, J. Mamiferos Sud Americanos 2nd Ed. (Ediar, 1960).
Ávila-Pires, F. D. Taxonomia e zoogeografia do gênero “Callithrix” Erxleben, 1777 (Primates, Callitrichidae). Braz. J. Biol. 29, 49–64 (1969).
Ávila-Pires, F. D. On the validity and geographical distribution of Callithrix argentata emiliae Thomas, 1920 (Primates, Callitrichidae). in A Primatologia No Brasil (ed. Mello, M. T.) Vol. 2, 319–322 (Sociedade Brasileira de Primatologia, 1986).
Hershkovitz, P. Living New World Monkeys (Platyrrhini) Vol. 1 (The University of Chicago Press, 1977).
Vivo, M. D. Taxonomia de Callithrix Erxleben, 1777 (Callitrichidae, Primates) (Fundação Biodiversitas, 1991).
Alperin, R. Callithrix argentata (Linnaeus, 1771): considerações taxonômicas e descrição de subespécie nova. Bol. Mus. Para. Emílio Goeldi Sér. Zool. 9, 317–328 (1993).
Ferrari, S. F., Sena, L., Schneider, M. P. C. & Silva-Jr., J. S. Rondon’s marmoset, Mico rondoni sp. n., from southwestern Brazilian Amazônia. Int. J. Primatol. 31, 693–714. https://doi.org/10.1007/s10764-010-9422-6 (2010).
Garbino, G. S. T. How many marmoset (Primates: Cebidae: Callitrichinae) genera are there? A phylogenetic analysis based on multiple morphological systems. Cladistics 31, 652–678. https://doi.org/10.1111/cla.12106 (2015).
Rylands, A. B., Coimbra-Filho, A. F. & Mittermeier, R. A. Systematics, distributions, and some notes on the conservation status of the Callitrichidae. in Marmosets and Tamarins: Systematics, Behaviour and Ecology (ed. Rylands, A. B.) 11–77 (Oxford University Press, 1993).
Rylands, A. B. et al. An assessment of the diversity of new world primates. Neotrop. Primates 8, 61–93 (2000).
Rylands, A. B. & Mittermeier, R. A. Family Callitrichidae. in Handbook of the Mammals of the World: Primates Vol. 3, 262–347 (eds. Mittermeier, R. A., Rylands, A. B. & Wilson, D. E.) (Lynx, 2013).
Wallace, A. R. On the monkeys of the Amazon. J. Nat. Hist. 14, 451–454 (1854).
Costa-Araújo, R., Farias, I. & Hrbek, T. Mico munduruku. The IUCN Red List of Threatened Species 2020: e.T172269376A172269630. https://dx.doi.org/https://doi.org/10.2305/IUCN.UK.20202.RLTS.T172269376A172269630.en (2020).
Schlick-Steiner, B. C. et al. Integrative taxonomy: A multisource approach to exploring biodiversity. Annu. Rev. Entomol. 55, 421–438. https://doi.org/10.1146/annurev-ento-112408-085432 (2010).
Padial, J. M., Miralles, A., De la Riva, I. & Vences, M. The integrative future of taxonomy. Front. Zool. 7, 16. https://doi.org/10.1186/1742-9994-7-16 (2010).
Rosen, D. Vicariant patterns and historical explanation in biogeography. Syst. Biol. 27, 159–188. https://doi.org/10.2307/2412970 (1978).
Vidal, M. D. et al. Protocolo Para Coleta de Dados Sobre Primatas em Unidades de Conservação da Amazônia. (ICMBio, 2012).
Hershkovitz, P. Metachromism or the principle of evolutionary change in mammalian tegumentary colors. Evolution 22, 556–575. https://doi.org/10.1111/j.1558-5646.1968.tb03992.x (1968).
Hershkovitz, P. Titis, new world monkeys of the genus Callicebus (Cebidae, Platyrrhini): A preliminary taxonomic review. Fieldiana Zool. 55, 1–109 (1990).
Doyle, J. J. & Doyle, J. L. Isolation of plant DNA from fresh tissue. Focus 12, 13–15 (1990).
Peterson, B. K., Weber, J. N., Kay, E. H., Fisher, H. S. & Hoekstra, H. E. Double digest RADseq: An inexpensive method for de novo SNP discovery and genotyping in model and non-model species. PloS One 7, e37135. https://doi.org/10.1371/journal.pone.0037135 (2012).
Boubli, J. P. et al. How many pygmy marmoset (Cebuella Gray, 1870) species are there? A taxonomic re-appraisal based on new molecular evidence. Mol. Phylogenet. Evol. 120, 170–182. https://doi.org/10.1016/j.ympev.2017.11.010 (2018).
Eaton, D. A. PyRAD: Assembly of de novo RADseq loci for phylogenetic analyses. Bioinformatics 30, 1844–1849. https://doi.org/10.1093/bioinformatics/btu121 (2014).
Collins, R. A. & Hrbek, T. An in silico comparison of protocols for dated phylogenomics. Syst. Biol. 67, 633–650. https://doi.org/10.1093/sysbio/syx089 (2018).
Lanfear, R., Frandsen, P. B., Wright, A. M., Senfeld, T. & Calcott, B. Partitionfinder 2: New methods for selecting partitioned models of evolution for molecular and morphological phylogenetic analyses. Mol. Biol. Evol. 34, 772–773. https://doi.org/10.1093/molbev/msw260 (2017).
Bouckaert, R. R. et al. BEAST 2.5: An advanced software platform for Bayesian evolutionary analysis. Preprint at https://www.biorxiv.org/content/https://doi.org/10.1101/474296v1; https://doi.org/10.1101/474296 (2018).
Stamatakis, A. RaxML version 8: A tool for phylogenetic analysis and post- analysis of large phylogenies. Bioinformatics 30, 1312–1313. https://doi.org/10.1093/bioinformatics/btu033 (2014).
Hillis, D. M. & Bull, J. J. An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis. Syst. Biol. 42, 182–192. https://doi.org/10.1093/sysbio/42.2.182 (1993).
Gauthier, J. et al. DiscoSnp-RAD: De novo detection of small variants for RAD-Seq population genomics. PeerJ 8, e9291. https://doi.org/10.7717/peerj.9291 (2020).
Kass, R. E. & Raftery, A. E. Bayes factors. J. Am. Stat. Assoc. 90, 773–795. https://doi.org/10.2307/2291091 (1995).
QGIS. QGIS Development Team—Geographic Information System. Open Source Geospatial Foundation. https://www.qgis.osgeo.org (2014).
46. Silva, F. E. et al. The Roosevelt-Rondon expedition marmoset Mico marcai: unveiling the conservation status of a data deficient species. Oryx 54, 539–545. https://doi.org/10.1017/S0030605318000303 (2018).
Snethlage, E. Die Flüsse Iriri und Curuá im Gebiet des Xingú. in Zeitschrift der Gesellschaft für Erdkunde zu Berlin, 328–354 (1925).
Pimenta, F. E., & Silva-Jr., J. S. An update on the distribution of primates of the Tapajós-Xingu interfluvium, central Amazonia. Neotrop. Primates 13, 23–28 (2005).
Adeney, J. M., Christensen, N. L., Vicentini, A. & Cohn-Haft, M. White-sand ecosystems in Amazonia. Biotropica 48, 7–23. https://doi.org/10.1111/btp.12293 (2016).
Sena, L. et al. Mitochondrial COII gene sequences provide new insights into the phylogeny of marmoset species groups (Callitrichidae, Primates). Folia Primatol. 73, 240–251 (2002).
Medina, R., Lara, F., Goffinet, B., Garilleti, R. & Mazimpaka, V. Integrative taxonomy successfully resolves the pseudo-cryptic complex of the disjunct epiphytic moss Orthotrichum consimile s.l. (Orthotrichaceae). Taxon 61, 1180–1198. https://www.jstor.org/stable/24389106 (2012).
Cornils, A. & Held, C. Evidence of cryptic and pseudocryptic speciation in the Paracalanus parvus species complex (Crustacea, Copepoda, Calanoida). Front. Zool. 11, 19. https://doi.org/10.1186/1742-9994-11-19 (2014).
Knowlton, N. Sibling species in the sea. Annu. Rev. Ecol. Evol. Syst. 24, 189–216. https://doi.org/10.1146/annurev.es.24.110193.001201 (1993).
Soares-Filho, et al. Modelling conservation in the Amazon basin. Nature 440, 520–523. https://doi.org/10.1038/nature04389 (2006).
Dietz, J. M., Dietz, L. A. & Nagagata, E. Y. The effective use of flagship species for conservation of biodiversity: the example of lion tamarins in Brazil. in Creative Conservation: Interactive Management of Wild and Captive Animals (eds. Olney, P. J. S, Mace, G. M. & Feistner, A. T. C.), 32–49 (Chapman & Hall, 1994).
Trolliet, F. et al. Ecosystem services provided by a large endangered primate in a forest-savanna mosaic landscape. Biol. Conserv. 203, 55–66. https://doi.org/10.1016/j.biocon.2016.08.025 (2016).
Lovejoy, T. E. & Nobre, C. Amazon tipping point. Sci. Adv. 4, eaat2340. https://doi.org/10.1126/sciadv.aat2340 (2018).
58. Fialho, S. M. Contribuição à distribuição do gênero Mico (Callitrichidae, Primates) no Médio Teles Pires, Jacareacanga, Pará. Neotrop. Primates 17, 31–32 (2010).
Andrade, R. P., Mourthé, I., Saccardi, V. & Hernández-Ruiz, E. J. Eastern extension of the geographic range of Mico emiliae. Acta Amazon. 48, 259–262. https://doi.org/10.1590/1809-4392201704392 (2018).
Sauer, A. C. et al. Efeito da sazonalidade sobre o padrão comportamental de um grupo de saguis-do-rabo-preto (Mico melanurus) em um fragmento florestal urbano in A Primatologia no Brasil Vol. 14 (eds. Silva, V. L., Ferreira, R. G. & Oliveira, M. A. B.), 266–276 (Sociedade Brasileira de Primatologia, 2017).
Authors thank curators and staff of museums visited. RCA thanks Raimundo Silva, Ivan Baptista, Natal Silva, Sr. Roberto and family, Raony Alencar, Luciano Silva, Thaimaçu Lodge, and Renata Sarmento for invaluable support, and several individuals from local communities. FRM, GC and RCA thank Biota Projetos e Consultoria Ambiental, Consórcio Hidrelétrico de Teles Pires, and ONF Brasil for invaluable support in the field.
This study was funded by Conselho Nacional de Desenvolvimento Científico e Tecnológico (Grant Nos. 563348/2010, 140039/2018-1, 302140/2020-4), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Grant Nos. 3261/2013, 001), Fundação de Amparo à Pesquisa do Amazonas (Grant No. 06200889/2019), Conservation Leadership Programme (Grant No. F02304217), Global Wildlife Conservation’s Margot Marsh Primate Action Fund (Grant No. 6002856), Idea Wild, National Science Foundation (Grant No. 1241066), Fundação de Amparo à Pesquisa de São Paulo (Grant No. 12/50260-6), NERC Environmental Bioinformatics Centre (Grant No. NE/T000341/1) and Fundação de Amparo à Pesquisa do Estado de Mato Grosso (Grant No. 477017/2011).
The authors declare no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Costa-Araújo, R., Silva-Jr., J.S., Boubli, J.P. et al. An integrative analysis uncovers a new, pseudo-cryptic species of Amazonian marmoset (Primates: Callitrichidae: Mico) from the arc of deforestation. Sci Rep 11, 15665 (2021). https://doi.org/10.1038/s41598-021-93943-w