How Many Taxa or Groups are within Nasua nasua and Nasuella olivacea (Procyonidae, Carnivora)? The Mitochondrial Reconstruction of the Complex Evolutionary History of the Coatis throughout the Neotropics and Some Insights into the Systematics of the Genus Bassarycion

Manuel Ruiz-García^1*, María Fernanda Jaramillo¹ and Joseph Mark Shostell^2
*Correspondence: Manuel Ruiz-García, Laboratory of Molecular Population Genetics-Evolutionary Biology, Department of Biology, Faculty of, Pontificia Universidad Javeriana, Cra 7A, No. 43-82, Bogotá DC,1102321, Colombia, Tel: 573144137980, Email:

Keywords

Coatis • Genetic heterogeneity • Nasua • Nasuella • Miocene • Mitochondrial genes • Phylogenetic analyses • Pleistocene • Pliocene • Speciation

Introduction

The Carnivora Order encloses 16 families divided into two monophyletic super-families, Caniformia and Feliformia [1-3]. The first super-family encloses the families Canidae, Ursidae, Mephitidae, Ailuridae, Procyonidae, Mustelidae, Otaridae, Odobenidae, and Phocidae, while the second super-family contains Felidae, Nandiniidae, Prionodontidae, Eupleridae, Herpestidae, Hyaenidae, and Viverridae [4].

The biogeographic history of the Procyonidae in the Americas is complex and the evolution of this family has been very important in understanding the history of the Great American Biological Interchange (GABI) [5,6]. The fossil record shows that Procyonidae originated and dispersed two separate times from North America into South America.

The first dispersion event occurred in the Late Miocene (9-5 Million Years Ago, MYA) with the appearance of the fossils of the genera Cyonasua and Chapalmalania from the Huayqueriense–Montehermosense in South America [7]. This was before the closure of the Isthmus of Panama around 3-2.5 MYA [8- 11]. However, the Procyonidae of this first colonization event apparently went extinct by the end of the Middle Pleistocene [6,12]. The second dispersion of Procyonidae from North America into South America is thought to be the one made by the ancestors of the extant genera during the late Pleistocene, 0.126 MYA [11]. Of the current Procyonidae genera, only Nasua and Procyon had fossils dated to the Lujanense period (0.126-0.009 MYA) [6,7,13]. However, Koepfli, et al. (2007) showed, by using nine nuclear and two mitochondrial (mt) genes that the diversification within the extant genera of Procyonidae occurred in the Middle to Late Miocene. This agrees quite well with the first colonization event but not with the second one as supported by the paleontological record.

One of the most known Procyonidae is the coatis. They are diurnal, highly social mesocarnivore (1.3 – 6.0 kg). Coatis forage both arboreally and terrestrially, and their diet includes primarily fruits, invertebrates and occasionally small vertebrates [14,15]. Females and immature males form permanent groups while mature males are solitary, joining the groups only during the mating season [15,16].

Among coatis, three, or five, extant species are currently recognized: the white-nosed coati (Nasua narica), the South American or brown-nosed coati (Nasua nasua), and the Western Mountain Coati (Nasuella olivacea). Recently, there were claims that the Eastern Mountain Coati (Nasuella meridensis) in the Venezuelan Andean Cordillera should be a new full species [17,18]. In the same way, McFadden, et al. (2008) designated the white-nosed coati from the Cozumel Island (Caribbean Mexico) as a distinct species (Nasua nelsoni) [19]. However, we are reluctant to accept these two species, as we show below.

N. narica is the only coati species distributed in North, Central and South America, from Arizona and New Mexico in the United States to northern Colombia [20]. The brown-nosed coati (N. nasua) is distributed in South America, from Colombia and Venezuela to Uruguay and northern Argentina [14,15,21]. It is found primarily in forested habitats, ranging from tropical rainforest and gallery forest to chaco, cerrado and dry scrub environments [21,22]. N. olivacea is found in the Andes of Venezuela, Colombia, and Ecuador in cloud forest and paramo at altitudes of 1,300-4,250 meters above sea level.

Koepfli, et al. (2007) were the first to try to determine the phylogenetic relationships among the genera within Procyonidae although they did not include samples of N. olivacea. This work was important because it was the first to estimate the temporal splits among and within the extant Procyonidae genera. The study also analyzed the possible phenomena which affected the colonization of this family in Central and South America.

Only a scarce number of molecular studies have covered the coatis. McFadden (2004) and McFadden, et al. (2008) analyzed a small fraction of mtDNA and concluded that the N. nelsoni from Cozumel Island is a full species [19,23]. Helgen, et al. only studied four specimens of Nasuella, three from the Western Colombian Andes (one specimen) and from the northern Ecuadorian Andes (two specimens), and one specimen from the Venezuelan Merida Cordillera. They analyzed only a small fraction of the mtCyt-b gene (366 base pairs, bp). They reached the conclusion, with these preliminary data, that the unique sample from Venezuela was a different species (Nasuella meridensis) from the three samples from the Western Colombian Andean Cordillera and northern Ecuador. Tsuchiya-Jerep (2009) analyzed 90 specimens of N. nasua throughout Brazil. MtDNA analyses showed levels of diversity that were up to ten-fold higher for N. nasua relative to Procyon cancrivorus in Brazil. Five reciprocally monophyletic mtDNA phylogroups were recognized for N. nasua in this country, which were also supported as distinct populations by microsatellite analyses [24]. Neves-Chaves (2011) analyzed 215 specimens of N. nasua from seven localities from Minas Gerais state in Brazil. The control regions of mtDNA and microsatellite loci were analyzed. One of these populations (Mangabeiras Park) showed low intrapopulation genetic diversity, high coefficient of relatedness between individuals, reduced effective population size, Hardy-Weinberg departure, evidence of recent bottleneck, and moderate isolation from other sampled populations. [25] Silva, et al. (2017) analyzed 60 specimens in five populations of N. narica distributed throughout Mexico by means of the mtCyt-b gene and 12 microsatellite loci. They found moderate to high levels of genetic diversity for both genetic markers. There were twentytwo different mtCyt-b haplotypes throughout the five sampled areas and each sampled population harbored unique haplotypes. Only three haplotypes were shared between two different populations that were close geographically. All populations had high haplotype diversity (Hd±=0.968+0.008) but lower levels of nucleotide diversity (π=0.007+0.001) [26]. Niguenda-Morales, et al. (2019) found a high degree of genetic structure and divergence among populations that conform to five evolutionarily significant units of N. narica in Central America. They analyzed three mt genes and 11 microsatellite loci in 85 specimens of N. narica. The most southerly distributed population (Panama) split much earlier (3.8 MYA) than the northern populations (1.2 MYA). Estimated gene flow among populations was low and mostly northwards and westwards. They described phylogeographic patterns within N. narica which were associated with geographic barriers and habitat shifts likely caused by Pliocene-Pleistocene climate oscillations. In addition, they suggested that the dispersal of N. narica was south-to-north beginning in the Pliocene, not in the opposite direction during the Pleistocene as suggested by the fossil record [27]. Jaramillo and Ruiz-García (2021) analyzed the genetic structure of 74 specimens of N. narica sampled in Central America and in northwestern South America (Colombia and Ecuador) by means of two data sets (three mt genes, and complete mitogenomes). They detected six well-differentiated groups of N. narica, one of them, in southern Central America, and another in northern Colombia, were introgressed by mtDNA of N. olivacea [28]. The colonization route was south to north, such as Niguenda-Morales, et al. (2019) claimed. Therefore, for N. narica, there are more molecular phylogeographic and systematic studies than for N. nasua and N. olivacea. For this reason, we centered this work especially in N. nasua and N. olivacea, although we included the sequences of N. narica that we obtained because we want to determine the relationships of the diverse N. narica haplogroups with those of N. nasua and N. olivacea.

We selected the study of mt genes as a first step to understand the phylogeography and the possible existence of full species within Nasua and Nasuella because these markers were those originally chosen to carry out the studies of phylogeography [29,30]. The mitochondrial genes are interesting markers for phylogenetic tasks because they include a rapid accumulation of mutations, rapid coalescence time, lack introns, a negligible recombination rate, and haploid inheritance [31]. They also have a high number of copies per cell which makes mitogenomic data easy to obtain and sequence especially in low quality samples, such as hair, teeth or small pieces of skin [32,33]. Despite representing a single linked locus, selection pressures and evolutionary rates are highly heterogeneous across the mtDNA [34,35]. For all of these reasons, mt genes have a higher discrimination power of genetic groups within species, or very related species, than the nuclear genes do [36-39].

However, the mitochondrial gene poses several problems including heterogeneity in base composition at each codon position, and third-codon position saturation [38,39]. Additionally, extreme care should be taken when using mitochondrial genes for resolving taxonomic problems because gene trees do not necessarily correspond well with species trees. If we use the gene tree to estimate divergence time for the species tree, the species will appear to have diverged more recently than they really did [40]. Furthermore, mitochondrial data show only the evolution of the female lineages, and this could miss hybridization events between close species when males are the gene flow vectors (‘mitochondrial capture’) [41]. To fully understand the evolutionary biology and number of taxa in the coatis, there first needs to be supporting evidence from nuclear data (microsatellites or SNPs), not just mtDNA data. Conscious of this two-pronged approach, we offer our analysis of mitochondrial DNA as the first step towards resolving this question. In fact, mtDNA have been extremely useful in detecting new and previously unnoticed taxa [42-47]. The mt results should be very important in helping to determine conservation strategies based in the systematic classification of the differentiated populations below the species level. The uncertainty about these conservation units can lead to confusion in the establishment of management plans and errors in setting priorities [48] (O’Brien, 1994). Therefore, there will be a necessary second step of analyzing nuclear markers to detect possible hybridization or gene flow events among different lineages.

The main aims of the current work were as follows: 1) To determine how many significant different main haplogroups (and sub-haplogroups) were detected in the three traditional coati species with eight mt genes, and their systematic implications especially in N. nasua and N. olivacea; 2) To determine how many species of coati are really there; 3) To identify which were the oldest and original coati haplotypes, at what species belonged, where, and when they appeared; 4) To determine if there are incomplete lineage sorting, introgression or hybridization among the different traditional species of coatis; 5) To determine if N. olivacea is also distributed outside the current known distribution near the Venezuelan, Colombian, and Ecuadorian Andes; 6) To determine if the sequence of the evolution of coatis occurred from Central America to South America (colonization from north to south) or vice versa (colonization from south to north), and 7) To show some new insights into the systematics of the genus Bassarycion, which was used as outgroup of Nasua- Nasuella.

Materials and Methods

Samples

We analyzed eight mt genes in 345 coati (222 N. nasua, 51 N. olivacea, and 72 N. narica) across 16 Latin American countries. Twenty-four specimens were used as outgroups: 5 Bassarycion neblina (Colombia and Ecuador), 5 Bassarycion medius (Colombia and Ecuador), and 14 Bassarycion alleni (Ecuador, Peru, and Bolivia). For a detailed origin of all the specimens studied, (Table 1 and Figure 1). The samples analyzed were basically coming from specimens hunted in Indian communities as food resources as well as from road kill specimens. A minor fraction of the samples (of Colombian origins) were obtained from the mammologist museum of the Instituto Alexander von Humboldt (Villa de Leyva) with the permissions of their responsibles. No review from the ethics committee was required, as our research work applied on samples of museum skins and killed road and previously hunted animals for human food and did not involve any direct manipulation or disturbance of live animals by the researches.

To clasificate the 345 coati specimens analyzed, we took into accont several facts. First, the geographical origins and the previous reported species in the area sampled. For instance, in Central America, there is not any report of N. nasua or N. olivacea, only N. narica. Second, pelage and body characteristics. N. nasua is distinguished from N. narica by pelage on the muzzle that is brown or gray (white in N. narica), and hairs on the nape of the neck that are in a reversed anterior position (different to N. narica). N. narica has a clear (white) area around the eyes (dark in N. nasua) and the paws are very dark or pale in N. nasua, meanwhile they are of a chocolate color. N. nasua has seven to nine dark bands in the tail, while N. narica has less marked bands in the tail. The hair of N. nasua has generally two color band (reddish and black), meanwhile the hair of N. narica has three color bands (pale yellow, chocolate, and pale yellow). N. olivacea has nine dark bands in the tail but sometime less notorious than in Nasua, because these bands are considerably thinner than in Nasua. The body color is pale yellow and olivaceus. The hair of N. olivacea has four bands (cream, dark coffe, pale yellow, and black). The length of the tail is shorter in Nasuella than in Nasua, and the length of the leg in N. olivacea in smaller than 68 mm (the total length of the body is smaller than 600 mm). Third, when it was possible, skull characters. For N. nasua, the palate is flat along the midline, rather than concave as in N. narica. In the first species, the sides of the nasal bones converge posteriorly rather than being parallel as in the second species. The skull of Nasuella is smaller and more slender than that of Nasua, the middle part of the facial portion is greatly constricted laterally, and the palate extends farther posteriorly [14,49].

Molecular methods

We extracted DNA from skin, muscle and blood samples with a modified phenol-chloroform procedure [50]. DNA from hair/follicles and teeth were extracted using 10-20% Chelex 100 resin (Bio-Rad, USA), with several modifications from Walsh, et al. (1991) [51].

We amplified the eight mt genes with primers from the following papers: 1) 1,140 base pairs (bp) of Cytb [52], 2) 657 bp of COI [53], 3) 720 bp of COII [54], 4) 710 bp of ND4 [55], 5) 1,800 bp of ND5 [56], 6) 329 pb of ATP6 [57], 7) 376 pb of 12s rRNA [58], and 8) 306 bp of D-loop [59]. This equaled a total of 6,038 bp.

PCR conditions for these mt genes were brought to a volume of 25 μl with 13.5 μl of Mili-Q H₂O, 3 μl of MgCl2 1 mM, 1 μl of dNTPs 0.2 mM, 1 μl of each primer (0.1 μM), 2.5 μl of buffer 10X, and one unity of Taq Polymerase with 50-100 ng of DNA. The PCR temperatures were 95° for 5 minutes followed by 10 cycles of 1 minute at 95°C, 1 minute at 58-64°C and 1.5 minute at 72°C, 25 cycles of 1 minute at 95°C, 1 minute at 52-60°C and 1.5 minute at 72°C and one final extension of 15 minutes at 72°C. All amplifications, including positive and negative controls, were checked in 2% agarose gels. Those samples that amplified were purified using membrane-binding spin columns (Qiagen). The PCR products were sequenced in both directions using the Big Dye™ kit in an ABI 377A automated DNA sequences. A consensus of the forward and reverse sequences was determined using the Sequencher program. The sequences were concatenated using the criteria of Vaidya, et al. (2011) [60].

The GenBank accession numbers of the coati specimens analyzed are from MT587713 to MT587788, MW410859 to MW410908, and MW419814 to MW419853.

Phylogenetic studies

jModeltest v2.0 [61] and Mega 6.05 software [62] were applied to determine the best evolutionary mutation model for the sequences analyzed for each individual gene, for different partitions and for the concatenated sequences. Akaike information criterion (AIC) [63,64] and the Bayesian information criterion (BIC) [65] were used to determine the best evolutionary nucleotide model.

Phylogenetic trees were constructed by using two procedures: Maximum Likelihood tree (ML), and Bayesian Inference tree (BI). ML tree was applied to the specimens of the coati taxa analyzed, whilst BI tree was applied to the haplotypes found in these coati taxa.

The ML tree was obtained using RA × ML v7.2.6 software [66]. We used the partitioning scheme and best-fit models chosen by the PartitionFinder software [67]. This program was used to objectively determine the optimal model of evolution and partitioning scheme simultaneously (for the partitions, codons 1+2 combined, and codon 3, for each gene, including control region and 12s rRNA). Therefore, a total of 16 blocks were analyzed, although the major part of them showed similar results. For this reason, the phylogenetic ML tree showed was using simultaneously all the information. Best-fit models were selected using Bayesian information criteria under a ‘greedy’ search scheme using a subset of models specific to RA × ML. The GTR+G+I model (General Time Reversible model+gamma distributed rate variation among sites+proportion of invariable sites) [68] was used to search for the ML tree. We estimated support for nodes using the rapid-bootstrapping algorithm (−f a -x option) for 1,000 non-parametric bootstrap replicates [69]. The haplogroups of coatis were considered significant when bootstraps were higher than 70% (lax limit) [70].

The BI tree was performed also using a GTR+G+I model. BI trees were completed with the BEAST v1.8.1 program [71]. Four independent iterations were run using three data partitions (codon 1, codon 2, codon 3) with six Markov Chain Monte Carlo (MCMC) chains sampled every 1,000 generations for 20 million generations after a burn-in period of two million generations. We checked for convergence using Tracer v1.6 [72]. We plotted the likelihood versus generation and estimated the effective sample size (ESS>200) of all parameters across the four independent analyses. The results from different runs were combined using LogCombiner v1.8.0 software [73] and TreeAnnotator v1.8.0 software [74]. A Birth-Death speciation model and a relaxed molecular clock with an uncorrelated log-normal rate of distribution [75] were used. Posterior probability values provide an assessment of the degree of support of each node on the tree. Values higher than 0.90 were considered strong support for monophyletic haplogroups [76,77]. Trees were visualized in the FigTree v1.4 software [78]. The BEAST v1.8.1 program was run to estimate the time to most recent common ancestor (TMRCA) for different nodes of the BI trees. We used a prior of 12.5+1 MYA (97.5% confidence interval: 10.54-14.46 MYA) for the split of the common ancestor of Bassarycion and Nasua+Nasuella, and a prior of 2.7+0.2 MYA (97.5% confidence interval: 2.31-3.09 MYA) for the split of the common ancestor of Bassarycion medius and Bassarycion alleni. These priors were obtained from Koepfli, et al. (2007).

To reconstruct the possible relationships among the haplogroups of the three “a priori” coati species analyzed and to estimate possible divergence

times among these haplogroups, we constructed a Median Joining Network (MJN) using Network v4.6.0.1 software (Fluxus Technology Ltd) [79]. Additionally, the ρ statistic [80] and its standard deviation [81] were estimated and transformed into years. The ρ statistic is unbiased and highly independent of past demographic events. This approach is named “borrowed molecular clocks” and uses direct nucleotide substitution rates inferred from other taxa [82]. We used an evolutionary rate of 1.3-1.7%per one MY [83] (values for Canidae), which represented one mutation each 12,740-9,742 years for the 6,038 bp analyzed. Network analyses could be more useful to reconstruct the evolutionary history within species or among much related species than bifurcating trees as is the case of the coatis.

The Kimura 2P genetic distance [84] was applied to determine the percentage of genetic differences among the different haplogroups detected in the three species of coatis analyzed. The Kimura 2P genetic distance is a standard measurement for barcoding tasks [85,86]. In this case, the Kimura 2P distance practically offered identical values to the p-distance. Kartavtsev (2011) analyzed sequences of mtCOI from 20,731 vertebrate and invertebrate animal species and obtained 0.89% ± 0.16% for populations within species, 3.78% ± 1.18% for subspecies or semi-species, and 11.06% ± 0.53% for species within a genus [87]. At mtCOII, Ascunce, et al. (2003), and Ruiz-García, et al. (2014) reported an average genetic distance of around 8% among species within a genus, and around 2–5% for subspecies [38,88]. Bradley & Baker (2001) and Baker & Bradley (2006) claimed, for mtCytb, that values<2% would equal intra-specific variation, values between 2% and 11% would merit additional study, and values>11-13% would be indicative of specific recognition [89,90]. Therefore, we take as an average for mitochondrial genes, above 4-10% for possible subspecies, and values higher than 12-13% for different species of the same genus. For species of different genera, this value should be above 16-18% [87].

Results

Phylogenetics of the coatis

The best nucleotide substitution models for the concatenated sequences were TN93+G (20,268.155) and GTR+G+I (13,234.043) for BIC and AIC, respectively.

The ML tree is shown in Figure 2. The first to diverge were the sequences of Bassarycion (100% bootstrap percentage). Within Bassarycion, the situation is confusing. There were three strongly supported clades. The most divergent one consisted of three B. neblina from Ecuador (92%), although, curiously, a specimen of B. alleni from Cochabamba Department (Bolivia) was associated with it. The second robust clade contained two specimens “a priori” also classified as B. neblina (but both from Colombia in this case; 99%), whilst the thrid robust clade was contained two specimens “a priori” classified as B. medius (Valle del Cauca, Colombia, and Santo Domingo de Tsáchilas, Ecuador; 93%). Later, we found various intermixed specimens from “a priori” different species (B. medius and B. alleni). One clade only had specimens of B. alleni (Ecuador, Peru, and Bolivia) but had a low bootstrap (58%). Thus, the systematics of Bassarycion is confusing and in need of a thorough molecular analysis.

This ML tree detected a stronger relationship between N. nasua and N. olivacea than either of these had with N. narica. This suggests that the genus Nasuella should be enclosed in the genus Nasua, as indicated by Ruiz-García, et al. (2020) and Ruiz-García, et al. (2021) [91,92].

The ML tree detected 10 significant clades within N. nasua with geographic significance. Haplogroup 1 (H1-NAS; 80%) was comprised by three sub-haplogroups. The first sub-haplogroup (H1-sh1-NAS; 71%) contained specimens from Bolivia (the major part of the Bolivian specimens analyzed were within this group), and southern Peru (mainly from the Madre de Dios River). The second sub-haplogroup (H1-sh2-NAS; 81%) was integrated by some specimens distributed throughout the Peruvian Amazon (north and south) and in the central Brazilian Amazon (Negro River). The third subhaplogroup (H1-sh3-NAS; 61%), with a low bootstrap support contained Bolivian specimens together with some specimens distributed throughout all the Peruvian Amazon (north and south).

The second haplogroup (H2-NAS; 87%) was integrated by specimens distributed across all the Peruvian Amazon (north and south) and Peruvian mountain zones in the Departments of Cuzco, Pasco, and San Martin.

The third haplogroup (H3-NAS; 90%) was mainly expanded across a large fraction of Peru, both in the Amazon (north, central, south) and within the Andean zones (Apurimac, Cuzco, and San Martín Departments). But this haplogroup was also expanded in part of the Colombian (Leticia) and western Brazilian Amazon (Tabatinga) up to the Ecuadorian Pacific coast. Within this haplogroup, we detected some small clusters of specimens mixed of different geographical areas where this haplogroup was established (65%, 84%, and 71%, respectively). Thus, some sub-haplogroups should be defined within it.

The fourth haplogroup (H4-NAS; 89%) was established at the southern area of South America (southern Brazil, Paraguay, and Uruguay). It is possible that this haplogroup can arrive to more northern Brazilian areas because one specimen from Araguaia (Goias State) was included at this haplogroup.

The fifth haplogroup (H5-NAS; 91%) was a small one integrated by specimens of the northern Peruvian Amazon and it reached the trans-Andean area of Ecuador (Santo Domingo de Tsáchilas Province).

The sixth haplogroup (H6-NAS; 80%) had specimens distributed within the Andean Cordillera areas of Colombia and Ecuador. In Colombia, these specimens were found in the Norte de Santander, Santander, Antioquia, Caldas, Risaralda, Quindío, Tolima, Huila, Cauca, and Nariño Departments, covering practically the entire Colombian Andes. In Ecuador, these specimens were found in the Imbabura, Pichincha, and Esmeraldas Provinces. As we will comment below, a group of Nasuella was associated with this N. nasua’s haplogroup.

The seventh haplogroup (H7-NAS; 76%) was basically distributed within the trans-Andean Pacific area of Ecuador, but included interesting specimens. One of them was a specimen of “a priori” classified as N. narica from the Robinson Crusoe Island (Chile), where the coatis were introduced by a German couple in 1950 to combat rodent plagues. Additionally, we found one specimen sampled in an Andean locality of the Peruvian Apurimac Department in this haplogroup. It was classified by its morphology as N. olivacea (although this species is not reported in Peru) and one specimen from another Andean locality at the Jatun River in the Bolivian Cochabamba Department and also with a morphotype similar to N. olivacea (a species which has not been reported in Bolivia).

The eighth haplogroup (H8-NAS; 81%) was a small one distributed across the Colombian Amazon and the Colombian Eastern Llanos, although we detected within this group a specimen from the trans-Andean Pacific Ecuador.

The ninth haplogroup (H9-NAS; 85%) was composed of two subhaplogroups. The first (H9-sh1-NAS; 75%) was integrated by specimens of trans-Andean and Andean areas of Colombia (Risaralda, Chocó, and Valle de Cauca Departments) and Ecuador (Esmeraldas Province). The second (H9-sh2-NAS; 88%) was composed of specimens of the Colombian Eastern Llanos. Associated to this haplogroup, we found a Nasuella haplogroup as we will comment below.

The tenth haplogroup (H10-NAS; 92%) was composed of specimens mainly distributed by the Colombian and Ecuadorian Amazon (it should be a sub-haplogroup, although its bootstrap percentage was low), but with specimens which were also distributed near the Andean Colombian Departments of Antioquia, Cundinamarca, Valle del Cauca as well as near the Colombian Eastern Llanos, and by the trans-Andean and Pacific areas of Ecuador (Pichincha, Santo Domingo de Tsáchilas, and Guayas Province) (it should also be another sub-haplogroup with a low bootstrap percentage).

In the case of N. olivacea, we detected four different haplogroups, and maybe a fifth haplogroup, as we mentioned earlier. There were some Andean specimens of N. nasua, which were a transition, or “bridge” toward the main group of N. olivacea. These Andean specimens were from the Antioquia Department (Colombia), Imbabura Province (Ecuador), and from the Junín Department (Peru). This last specimen should be indirect proof of the possible existence in the present, or in the past, of N. olivacea in Peru.

The first haplogroup (H1-OLI; 82%) was clustered with H6-NAS, which agrees quite well with that reported by Ruiz-García, et al. (2020) that specimens with the typical morphotypes of N. olivacea shared mitochondrial DNA more related to that of N. nasua than to that of the major fraction of N. olivacea. However, they were also significantly differentiated from the more related mitochondrial DNAs of N. nasua, which invalidates that these N. olivacea are simply specimens of Andean coatis recently introgressed by mitochondrial DNA from N. nasua. The specimens of this haplogroup were extended to the Colombian (Norte de Santander, Antioquia, Caldas, Chocó, Cauca, and Nariño Departments) and Ecuadorian (Carchi Province) Andean Cordilleras.

The second haplogroup (H2-OLI; 88%) was associated, as in the previous

case, to another haplogroup of N. nasua (H9-NAS). This could be explained as in the previous case. The specimens of this haplogroup were also distributed in the Colombian (Caldas Department) and Ecuadorian (Pichincha Department) Andean Cordilleras.

The third haplogroup (H3-OLI; 95%) contained specimens at the Western and Central Colombian (Caldas, Chocó, Tolima, and Cauca Departments) and Ecuadorian (Pichincha, and Cotopaxi Provinces) Andean Cordilleras. Within this haplogroup, there was a very interesting specimen. This specimen was from Alba (Urubamba River; Cuzco Department, Peru). This is the first clear report of N. olivacea in Peru. Unfortunately, this specimen was hunted by a peasant and we sampled a little sample of skin. A picture of this dry skin was taken and it is available from the first author.

A haplogroup of N. narica was present and associated with this haplogroup of N. olivacea, as we will comment below.

The fourth haplogroup (H4-OLI; 90%) was mainly composed by specimens from the Eastern Colombian (Norte de Santander, Boyacá, and Cundinamarca Departments) Andean Cordillera. However, two specimens outside this geographical area were included in this haplogroup. One was from the Los Nevados NP (Caldas Department) in the Central Colombian Andes Cordillera, and another was from the Sangay NP in the Ecuadorian Province of Morona- Santiago.

Potentially, a fifth haplogroup of N. olivacea (H5-OLI; 62%) should be associated with H7-NAS as we aforementioned. If so, the geographical distribution of this haplogroup, in the case that they were Andean coatis, extends from the Apurimac Department in southern Andean Peru to the Cochabamba area in Bolivia. However, we have not complete evidence if this haplogroup is composed by two specimens of real N. olivacea or they are specimens of N. nasua adapted to live in Andean conditions.

In the case of N. narica, we detected four haplogroups, one of them integrated by three sub-haplogroups.

The first haplogroup (H1-NARI; 86%), probably the most ancestral for this species, contained three specimens from Ecuador. Two of them were from the Guayas Province at the trans-Andean and Pacific area of Ecuador. The third specimen was sampled in Fatima at the cis-Andean Amazon area of Ecuador. This haplogroup is extremely interesting because it seems to be the origin of the mainly Central-American N. narica.

The second haplogroup (H2-NARI; 82%) had a couple of specimens from the Yucatan Peninsula in Mexico.

The third haplogroup (H3-NARI; 87%) had a wide geographical distribution in Central America with three well differentiated sub-haplogroups. The first sub-haplogroup (H3-sh1-NARI; 79%) consisted of specimens from Guatemala, Honduras, El Salvador, Nicaragua, and northcentral Costa Rica. The second sub-haplogroup (H3-sh2-NARI; 88%) included specimens from Guatemala and Belize, while the three sub-haplogroup (H3-sh3-NARI; 98%) was distributed near northern Guatemala, and southern Mexico. This subhaplogroup contained one specimen from the Cozumel Island.

The fourth haplogroup (H4-NARI; 94%) was associated, as we aforementioned, to the H3-OLI. This haplogroup was recorded at the southern area of Central America (Puntaarenas Province, southern Costa Rica; Nombre de Dios, Colon Province, Panama), but also in Colombian lands (Chocó and Antioquia Departments) close to the border with Panama. These specimens had undoubtedly morphotypes of N. narica (see Material and Methods) but with mitochondrial DNA of N. olivacea.

Henceforth, with 345 coati specimens and with eight mt genes, we detected 19 main haplogroups (10 N. nasua, five N. olivacea, and four N. narica) and 8-10 sub-haplogroups (five-seven N. nasua, and three N. narica).

The BI tree (Figure 3) was constructed by using haplotypes (not specimens as in the previous tree). It showed a similar perspective in reference to the different haplogroups found within the three species in the ML tree. The major fraction of the haplogroups detected in that analysis was found with the BI tree: H1-NAS (p=0.82), H3-NAS (p=0.88), H4-NAS (p=0.9), H6-NAS (p=0.91) and H1-OLI (p=0.76) both associated, H7-NAS (p=1), H9-sh1-NAS (p=0.98) and H2-OLI (p=0.78) both associated, H10-NAS (p=0.80), H3-OLI (p=0.70), H4-OLI (p=0.88), H2-NARI (p=1), H3-NARI (p=0.93), including H3-sh1-NARI (p=0.87), H3-sh2-NARI (p=0.83), H3-sh3-NARI (p=0.96), and H4-NARI (p=1). Nonetheless, there were some differences with regard to the ML tree. For instance, H2-NAS and H7-NAS disappeared as a significant clades and H5- NAS (p=0.98) was, in this case, integrated within H10-NAS. In the ML tree, the introgressed N. narica‘s haplogroup (H4-NARI) was associated with H3- OLI. In this case, it was associated with H4-OLI. The H1-NARI was the most divergent haplogroup of N. narica in the ML tree. However, in the BI tree, H1- NARI (p=0.73) was included in one sub-haplogroup of the Central American N. narica (H3-sh1-NARI). Last, in the BI tree, two haplotypes (two specimens) of N. olivacea appeared as the most divergent of all the coatis analyzed and potentially their ancestors as the original one (Colombian and Ecuadorian Andes: Risaralda Department, Colombia, and Carchi Province, Ecuador). However, the major difference between both trees was the relationship of Nasuella regards to Nasua. In the ML tree, Nasuella was more related to N. nasua, meanwhile, in the BI tree, Nasuella was more related to N. narica than to N. nasua. In whatever case, Nasuella seems to be a taxon which should be enclosed into Nasua.

If we took the origin of the evolution of the coati’s haplotypes as those less differentiated in reference to the common ancestor with the outgroup (Bassarycion), the first haplotypes which appeared in the MJN for the eight mt genes were H90, H91, and H27 (Figure 4). These haplotypes corresponded to diverse specimens of N. nasua from the trans-Andean Pacific Ecuador, and Western and Central Colombian Andean Cordilleras (Esmeraldas Province, Ecuador, and Risaralda, Chocó, and Valle del Cauca Departments, Colombia). Based on a morphological point of view, the first haplotypes of N. olivacea (H140 and H146; Caldas Department, Colombia, and Pichincha Province, Ecuador, also in the northern Andes) appeared from these. Two routes appeared. The first originated a major fraction of the Andean N. nasua throughout Colombia and Ecuador (H88 and related haplotypes). This in turn, gave origin to some haplotypes of N. olivacea (H181, H159, and associated haplotypes). The specimens with these haplotypes had, undoubtedly, phenotypes of N. olivacea, although these haplotypes were more related to those of the Colombian and Ecuadorian Andean N. nasua than to the next derived group of N. olivacea. These “transition” haplotypes gave origin to a group of N. olivacea mainly (but not exclusively) distributed near the Eastern Colombian Andean Cordillera (H132 and related haplotypes). In the past, haplotypes of this group intruded into a fraction of N. narica distributed in the southern area of Central America (southern Costa Rica, and Panama) and northern Colombian frontier with Panama. Eventually, this group of N. olivacea from the Eastern Colombian Andean Cordillera gave place to another group of N. olivacea in the Western and Central Colombian and Ecuadorian Andes Cordilleras. Thus, this last group of N. olivacea was the most recent one for this species. The second way was initiated with haplotypes from the Colombian Eastern Llanos (H49 and H50). Two additional pathways originated from these. The first one gave rise to some transition haplotypes of N. nasua (H130; Buenaventura, Valle del Cauca Department, Colombia), which finally gave origin to the first haplotypes of N. narica in the Ecuadorian Pacific area, although later some of these N. narica’s haplotypes crossed the Andean Cordillera until they at least reached the Ecuadorian Amazon (H86, H98, and H38). The evolution of the haplogroups of N. narica was discussed in Jaramillo and Ruiz-García (2021) and it is not here discussed [28]. The second one gave origin to all the detected groups of N. nasua. One was composed of haplotypes mainly distributed in the Colombian and Ecuadorian Amazon as well as near the Pacific Ecuador and some Andean areas of Colombia (H3, H31, and related haplotypes). Another group was mainly composed of haplotypes of N. nasua placed from trans-Andean and Pacific Ecuador, but some haplotypes were from more distant locations. One was found in a coati from Robison Crussoe Island (Chile), but two others were found in Peru and Bolivia and these specimens had phenotypes very similar to N. olivacea (H22, H94, H150). There were single and dispersed haplotypes (H51, H62) within these groups of N. nasua, distributed within the Eastern Colombian Llanos (Meta Department) and the Huila Department, both in Colombia. They seem to play a very important role in the origin and generation of three other large groups of N. nasua. One group was situated in southern South America (southern Brazil, but also distributed to more northern areas of Brazil, Paraguay, and Uruguay; H32, H37, and related haplotypes). Another group was mainly in the Peruvian Amazon and Peruvian Andes but also the Colombian and western Brazilian Amazon and even the Ecuadorian Pacific coast (H64, H43, and related haplotypes). Another group (H1, H39, and other related haplotypes) had some sub-groups. The first sub-group was distributed near Andean Peru and some areas of the Peruvian Amazon, which, in turn, originated the majority of the haplotypes of the Madre de Dios River basin (southern Peru) and Bolivia. In turn, these haplotypes generated haplotypes of N. nasua in the central Brazilian Amazon (Negro River).

Temporal divergence between and within different haplogroups of coatis

We mainly showed here the temporal splits estimated throughout the MJN procedure as well as for the BI estimates for the eight mt genes. Generally, the BI estimates, were higher than the MJN estimate especially in the oldest time estimates and lower in the most recent temporal estimates. However, the geological epochs of the main temporal splits were more or less the same. The most ancestral haplotype (H90) found for whatever coati studied belonged to H9-sh1-NAS (N. nasua) (see Figure 4). The temporal

differentiation between this haplotype and the most derived haplotype in H3-OLI (N. olivacea) (13.09 ±0.15 MYA for a substitution rate of 1.3%/MYA; 9.99 ±0.11 MYA for a substitution rate of 1.7%/MYA, respectively), in H3-sh3- NAR (N. narica) (12.97 ± 0.34 MYA; 9.91 ± 0.26 MYA), and in H1-NAS (N. nasua) (8.42 ± 0.29 MYA; 6.44 ± 0.22 MYA). The BI estimation for the initial mitochondrial diversification in the coatis was 11.92 MYA (6.54-13.45 MYA). Therefore, the initial mitochondrial diversification for the coatis was during the Miocene (10-13 MYA) “in situ” in South America. The first haplotypes of H9- sh1-NAS (Colombian and Ecuadorian Andes), that we detected, generated the first haplotypes of N. olivacea (H2-OLI; Colombian and Ecuadorian Andes) around 0.19 ± 0.14/0.15 ± 0.10 MY later. H6-NAS (Colombian and Ecuadorian Andes) appeared next around 2.47 ± 0.32/1.89 ± 0.25 MY after the emergence of H9-sh1-NAS. H9-sh2-NAS (Central Andean Colombia, and Eastern Colombian Llanos) appeared next around 2.61 ± 0.36/1.99 ± 0.27 MY later. The intermediate haplogroup of N. olivacea (H1-OLI; Colombian and Ecuadorian Andes) appeared 4.50 ± 0.32/3.44 ± 0.25 MY later. Approximately, 7.26 ± 0.41/5.54 ± 0.31 MY later, the first haplotypes of N. narica (H1-NARI; trans-Andean Pacific Ecuador) appeared. The last temporal splits support that the evolution of coatis in South America occurred before the complete closure of the Isthmus of Panama around 3 MYA. We previously observed that the H9- sh2-NAS (Central Andean Colombia, and Eastern Colombian Llanos) played an important role in the origin of all the other haplogroups of N. nasua in South America. After the emergence of H9-sh2-NAS, H10-NAS appeared (1.31 ± 0.21/1-00 ± 0.16 MY later), then H7-NAS (3.05 ± 0.32/2.33 ± 0.25 MY later), H4-NAS and H3-NAS (3.48 ± 0.29/2.66 ± 0.22 MY later), and H1-NAS (7.71 ± 0.50/5.90 ± 0.38 MY later).

For N. nasua, within H10-NAS, the mitochondrial diversification began around 0.95 ± 0.31/0.72 ± 0.24 MYA (the BI estimation: 2.04 MYA, 0.92-3.11 MYA), within H4-NAS around 0.61 ± 0.31/0.47 ± 0.23 MYA (the BI estimation: 2.95 MYA, 1.73-5.88 MYA), within H3-NAS around 2.88 ± 0.80/2.20 ± 0.62 MYA (the BI estimation: 2.18 MYA, 1.19-4.57 MYA), within H1-NAS around 0.62 ± 0.31/0.47 ± 0.24 MYA (the BI estimation: 3.09 MYA, 1.70-4.85 MYA), and within H6-NAS around 1.16 ± 0.27/0.89 ±0.20 MYA (the BI estimation: 2.92 MYA, 0.6-5.56 MYA).

For N. olivacea, the time split between H1-OLI and H4-OLI (mainly Eastern Colombian Andean Cordillera) was estimated to have occurred around 4.64 ± 0.70/3.55 ± 0.53 MYA, between H4-OLI and the introgressed haplogroup of N. narica (N4-NARI) around 0.87 ± 0.26/0.67 ± 0.19 MYA, and between H4-OLI and H3-OLI around 1.45 ± 0.14/1.11 ± 0.11 MYA. Within H1-OLI, the mitochondrial diversification began around 2.49 ± 0.47/1.91 ± 0.36 MYA (the BI estimation: 1.72 MYA, 0.70-3.79 MYA), within H4-OLI around 3.05 ± 0.82/2.33 ± 0.63 MYA (the BI estimation: 4.41 MYA, 1.34-5.08 MYA), within H4-NARI around 2.09 ± 0.68/1.60 ± 0.52 MYA, and within H3-OLI around 1.67± 0.42/1.28 ± 0.32 MYA (the BI estimation: 5.34 MYA, 1.53-6.12 MYA).

For N. narica, the time split between H1-NARI (Pacific Ecuador) and H3- NARI (central and northern Central America) was estimated to have occurred around 4.18 ± 0.28/3.56 ± 0.26 MYA, between H3-sh1-NARI and H3-sh3- NARI around 1.16 ± 0.27/0.89 ± 0.21 MYA, within H3-sh1-NARI around 0.46 ± 0.29/0.36 ± 0.22 MYA (the BI estimation: 1.52 MYA, 0.92-3.95 MYA), and within H3-sh3-NARI around 0.74 ± 0.17/0.56 ± 0.13 MYA (the BI estimation: 2.46 MYA, 0.98-3.99 MYA).

Discussion

Some comments on the taxonomy of the out-group: The case of Bassarycion

To date the study of Helgen, et al. (2013) [93] of the Bassarycion systematics is the most complete one. Analyzing the genes CHRNA1 and mtCyt-b, they showed that the most divergent branch was that composed by B. neblina neblina (from western slopes of the western Ecuadorian Andes) and B. neblina osborni (from the eastern slopes of the western Colombian Andes and western slopes of the central Colombian Andes). Our results agree well with the fact that the most divergent Bassarycion branch in our tree was composed by three exemplars of B. n. neblina from Ecuador. However, a specimen of B. alleni from Cochabamba Department (Bolivia) was included in this clade. Other difference in our analysis is that the two specimens classified “a priori” as B. neblina from Colombia did not clustered with the Ecuadorian specimens. Helgen, et al. (2013) defined three possible B. neblina subspecies in Colombia. One B. n. osborni was aforementioned. The other two were B. neblina hershkovitzi (from eastern slopes of central Colombian Andes) and B. neblina ruber (from Urrao, Antioquia Department; western slope of northern western Colombia Andes). However, the two Colombian specimens we analyzed were not of the same areas where these subspecies were described. One is from the Nariño Department in southern Colombia and the other was seized in Bogotá and the real origin is unknown but it is possible that it is from the eastern Colombian Andes. Therefore, it is possible that they are part of undescribed taxa very similar phenotypically to B. neblina. The following branch to diverge was that of B. gabbi from Costa Rica [93]. We did not enclose any specimen from this taxon in our analysis. Finally, the most recent split gave origin to the current B. medius and B.alleni. We also found that the major part of the specimens of B. medius and B. alleni were more related among them than with reference to the other Bassarycion taxa. However, we did not found strictly monophyletic clades for these two species. In fact, we found different polyphyletic clades of B. medius. Even we found some haplotypes of B. alleni (from Ecuador) intermixed between the two B. neblina clades. This could be explained using two hypotheses. (a) Really there are more Bassarycion taxa that those recognized until today and a very exhaustive sampling is urgently needed and/or (b) there are cases of mtDNA introgression between different Bassarycion taxa. Therefore, it is urgently a proof molecular analysis of this genus.

How many species of coatis are within the genus Nasua?

We consider genetic distances above 4-10% for possible subspecies, values around 12-13% for different species of the same genus, and values above 16-18% for species of different genera [87-90]. One problem in distinguishing species of coatis within the genus Nasua is that genetic distance values among multiple haplogroups can conform a continuous range of values. Here, we acknowledge one example of N. nasua.

We begin with genetic distances between H1-sh1-NAS (N. nasua from Bolivia and southern Peru) and other haplogroups. The genetic distance with H1-sh3-NAS was 0.8% (population differentiation), with H4-NAS, it was 5.9% (subspecies differentiation), with H10-sh1-NAS, it was 9.8% (subspecies differentiation near to the limit of full species), with H7-NAS, it was 12% (limit of full species), with H1-NARI (first N. narica from Ecuador), it was 12.8% (limit of full species), with H1-OLI and H2-OLI (the two first haplogroups of N. olivacea), the values were 10.2% and 11.5%, respectively (limit between subspecies and species). If we consider these N. nasua taxa as subspecies (H1-sh1-NAS vs. H10-sh1-NAS, or H7-NAS), then the original Ecuadorian N. narica haplogroup and two original N. olivacea haplogroups should be considered as subspecies of N. nasua too. In contrast, if we consider these N. narica and N. olivacea as different species from the N. nasua taxon, then different haplogroups of N. nasua should be considered different species. This shows the difficulty in distinguishing different taxa of N. nasua as different species and the oldest groups of N. narica and N. olivacea. This is further evidence favouring the existence of one unique genus for the coatis (Nasua). However, the genetic differentiations between H1-sh1-NAS and the most derived N. narica taxa (H2- NARI, H3-sh1-NARI, H3-sh3-NARI), and the most derived N. olivacea (H3- OLI, H4-OLI) were in the range of well-differentiated species (17.3%, 15.8%, 15.9%, 15.7%, and 17%, respectively).

These results put forward that the most derived haplogroups of N. nasua and the most derived haplogroups of N. narica and N. olivacea are within the genetic distance range of well differentiated species. However, a key question remains about the relationships of the most original haplogroups of N. nasua with the original groups of N. narica and N. olivacea, which were typically of subspecies.

The question with N. olivacea is complex. The genetic distances among the “intermediate” (haplotypes between the ancestral haplotypes of N. nasua and the most derived haplotypes of N. olivacea, with morphological characters and geographical distributions typical of N. olivacea) H1-OLI, with the remainder haplogroups of N. olivacea (H2-OLI, H3-OLI, H4-OLI) were 5.7%, 10.9%, and 9.4%, respectively, and between H3-OLI and H4-OLI was 5.4%. These values were in the range of subspecies. Nonetheless, the genetic distances among H2-OLI versus H3-OLI and H4-OLI were 15.4% and 16.4%, respectively, which are clearly in the range of full species. The genetic distances agree quite well with that observed in the phylogenetic trees and in the MJN procedure: the polyphylia of N. olivacea or the existence of groups of Andean N. nasua, which have independently evolved towards a morphotype very similar to that of the real N. olivacea (H3-OLI and H4-OLI).

Therefore, how many species we can discriminate within the genus Nasua? We propose three hypotheses:

1) Each one of the significant and monophyletic haplogroups should be a differentiated species using extreme versions of Phylogenetic Species Concept 1 (PSC) [94] and Phylogenetic Species Concept 2 [95]. If we adopted this strategy, then, we should define, at least, 19 different species of coatis versus the three traditional species that is if we only take into account the main haplogroups detected. However, we do not want to make the same mistake as previous authors and indiscriminately apply the PSC without first understanding the reproductive, ethological, or immunological incompatibilities among the groups of coatis or karyological studies. The application of PSC without the knowledge of these previous items can easily spawn many unrealistic new species that are clearly not sustained by a critical analysis. This was named by Isaac, et al. (2004) as ‘‘taxonomic inflation’’ [96]. Zachos, et al. (2013) suggests that some of the proposed new mammal species are completely unjustifiable [97]. We agree with Zachos (2016) who stated that species are such fundamental units that they should not be introduced carelessly [98]. Descriptions of species and their splitting should not be based on simple morphometric differences (even significant ones) or phylogenetic relationships derived from very limited molecular datasets. These datasets may serve to support conclusions derived from larger and more complete datasets, but are not enough on their own.

2) Another possibly more conservative option is that the three traditional species should be conserved in one unique genus (Nasua nasua, Nasua olivacea, and Nasua narica). Ruiz-García, et al. (2021) discussed the reasons why Nasuella [99] should be enclosed within Nasua [100]. For instance, for the BI tree, N. narica was more related to N. olivacea than to N. nasua. However, for this to be accepted, data must indicate that several groups of N. olivacea are really specimens of N. nasua. They would have convergently evolved to have phenotypes undistinguishable from N. olivacea. Someone could argue that these N. olivacea were more related with the N. nasua than with the most derived N. olivacea. Beyond this last questionable assertion (which means that there was no reproductive isolation between two “a priori” species of two different genera), they are not the product of recent hybridization because the specimens have no intermediate phenotype (they are clearly by size and color specimens of N. olivacea). Furthermore, the haplotypes of these specimens formed well defined groups and spread across a wide geographical area (not in specific points which could be expected in occasional recent hybridization events). Additionally, if these specimens were the product of recent hybridization events, their haplotypes should be un-differentiable from the maternal lineage and the temporal split between these haplotypes should be zero or very sm all. Contrarily, the temporal splits found should be considerable. Thus, these specimens could not be considered product of recent and punctual events. Taking this into consideration, the only “true” N. olivacea should be H3-OLI and H4-OLI. All of the other specimens with a phenotype of N. olivacea outside of these two molecular groups should be classified as N. nasua. Obviously, it would be a challenge to correctly classify specimens of N. olivacea in the field without having a molecular analysis. Likely, some coatis that have intermediate haplotypes between N. nasua and the Central America N. narica should be classified as N. nasua. This should be the case of the original Ecuadorian N. nasua, as well as of one specimen that we classified as N. nasua from the Tamá NP (Norte de Santander Department, Colombia). This last one did show a strong relationship to the origin of N. narica (the same occurred with one specimen from Buenaventura, Valle del Cauca, and Colombia). Then, the unique “true” N. narica should be the Central American one. However, even in this case, there was another problem: the group of N. narica from southern Central America and northern Colombia were introgressed by N. olivacea. These particular specimens had phenotypes of N. narica although their mitochondrial DNA more closely resembled that of N. olivacea than the mitochondrial DNA of the remaining N. narica of Central America. They were clearly not the product of recent and occasional hybridization events, because the temporal split between H4-OLI and the introgressed haplogroup of N. narica (N4-NARI; southern Central America) was around 0.87 ± 0.26/0.67 ± 0.19 MYA.

3) A third hypothesis is that all the coatis belonged to a unique species or super-species. This hypothesis accommodates the polyphylia of N. olivacea, the existence of the “intermediate” haplogroups or haplotypes between N. nasua and N. olivacea, or between N. nasua and N. narica, or the introgression of N. olivacea within a fraction of N. narica. Importantly, this hypothesis supports a broad genetic diversity within species and a continuous evolutionary process especially between much related taxa as it is the case of the coatis. The hypothesis addresses a problem stemming from the fact that the Linnaeus classificatory scheme is Aristotelian, scholastic and the scheme was created to classify static, and discrete, organisms that people thought would never evolve because they were created by God. Thus, in many cases (and the coati should be one of them, as proposed with Ateles; [101]), there would be conflict between this typological philosophy and the evolutionary process which is continuous, especially between very related taxa as it is the case of the coatis.

Molecular intra-generic systematics of Nasua nasua and Nasuella olivacea

We agree more with the second or third hypotheses than with the first one. Indeed, although we believe that future molecular, karyological, ethological, and reproductive data may offer results in favor of a unique species, we adopt the most traditional strategy of three species.

We begin the taxonomic review with N. nasua. It shows the northern and western slopes of the Andes to the Pacific coast of Colombia and Ecuador as the original areas for the initial mitochondrial diversification of the coatis. The first were some Andean N. nasua (Colombian and Ecuador) which originated (polyphyletically) the first ancestral haplogroups of N. olivacea, but also the most derived haplotypes of N. olivacea (which we name the “true” N. olivacea). These N. nasua were also the origin of N. narica. Additionally, the Andean area of central Colombia (Huila Department, for instance) and surrounding areas (Eastern Colombian Llanos) were significant to the origin of all the other haplogroups of N. nasua across South America.

We traditionally consider ten subspecies for N. nasua [21]. Only a small fraction of the haplogroups we found for N. nasua, support the traditional morphological subspecies. Just two cases seem to be absolutely in agreement. These were the cases of H4-NAS with N. n. spadicea (type locality: Paraguay; [102]) and H1-sh1-NAS with N. n boliviensis (type locality: Chaparé, Cochabamba, Bolivia; [103]). One additional case should be the specimen of Goias (Brazil) that we analyzed. This specimen was the most divergent within H4-NAS. If we could have analyzed additional specimens from this Brazilian area (or relatively neighbor zones), we very well might have had a new haplogroup or sub-haplogroup which could agree with N. n. nasua (type locality: Pernanbuco, Brazil; [104]). In fact, Tsuchiya-Jerep (2009) detected five haplogroups (four in Brazil), which could be related to the morphology of the N. nasua subspecies in Brazil. One of these haplogroups was detected in northeastern Brazil (Caatinga and northern Brazilian Atlantic forest). It corresponded to N. n. nasua (our sample probably belongs to this subspecies). These authors detected another one in the central Brazilian Atlantic forest, corresponded to N. n. solitaria (type locality: Bahia, Brazil; [105]). Therefore, we detected two clear cases of correspondence between haplogroups and morphological subspecies, and two other probable cases of correspondence between haplogroups and morfphological subspecies following our result of the specimen of Goias and the data of Tsuchiya-Jerep (2009).

Any possible remaining similarities among haplogroups of N. nasua and traditional subspecies are more complex and, even, dubious. N. n. candace (type locality: Medellín, Antioquia, Colombia; [106]) and N. n. judex (type locality: Bogotá, Cundinamarca, Colombia; [107]) have been defined in Colombia. However, the latter one is considered a synonymous to the former one [21]. Our results showed that, at least, two different haplogroups of N. nasua were present in the Antioquia Department (H6-NAS, and H10-NAS). Therefore, we do not know which of the two (either of them) correlated with N. n. candace. On the other hand, we detected one specimen from Cundinamarca in the H10- NAS. This specimen should be the same as N. n. judex, but then N. n. candace and N. n. judex could not be synonymous. In fact, only in Colombia, did we detect five different haplogroups (and four sub-haplogroups within of two of these haplogroups) for N. nasua although only two subspecies were defined (and one was later synonymized with the senior name) in this country. This demonstrates that the molecular results showed a more complex reality than that described by the putative morphological subspecies. From a conservation point of view, this also introduces a technical complication because many of these haplogroups overlapped geographically.

The same occurred in Peru with N. n. montana (type locality: Umanpuqio, Ceja Region, Peru; [108]). Two different haplogroups of N. nasua were detected throughout the Peruvian Andes (H2-NAS and H3-NAS). Thus, we ignore which of them coincides with that morphological subspecies. In fact, another subspecies has been described in Peru, N. n. dorsalis (type locality: Peru and Ecuador; [109]). However, we detected, at least, seven haplogroups and sub-haplogroups of N. nasua in Peru and many of them are geographically overlapped. This introduces considerable complications for a traditional systematic nomenclature approach.

Ecuador is an even more complex case. Three morphological subspecies have been defined in this country: N. n. manium (type locality: Balzar, Guayaquil, Ecuador; [106]), N. n. quichua (type locality: Jima, Azuay Province, Ecuador; [110]), and the quoted N. n. dorsalis. Only for the Andean, and trans-Andean- Pacific Ecuador, we detected seven different haplogroups (H3-NAS, H5-NAS, H6-NAS, H7-NAS, H8-NAS, H9-NAS, and H10-NAS) and one additional haplogroup in the Ecuadorian Amazon (H10-NAS). This did not consider the different groups found in N. olivacea and N. narica which are highly related to N. nasua and which are also located in the Ecuadorian territory.

Therefore, the taxonomic and conservation strategies could be more complex for N. nasua in Ecuador, Colombia, and Peru. This puts forward that the northern Andes were transcendental in the origin and diversification of the coatis.

The situation in Bolivia was easier to understand. The majority of the specimens of N. nasua in this country were part of H1-NAS. Only three Bolivian specimens were outside of this group. One was from Ibiato (Beni Department), which did not have clear clusters of haplogroups. Another specimen sampled in the Santa Cruz Department, was clustered in the H10-NAS. The third was from Jatun River (Cochabamba Department). The last two specimens are interesting. The first is part of a haplogroup that originated in northern Amazonia (current Colombian and Ecuador), and dispersed to the Bolivian Amazon. The second specimen was even more interesting. It was sampled in an Andean area from the Cochabamba Department and it was closely related with another specimen from the Andean area of the Apurimac Department (southern Peru) and both were included within H7-NAS. Both specimens showed a morphology similar to that of N. olivacea, or, at least, intermediate between N. nasua and N. olivacea, although they are related to an N. nasua haplogroup originating in the trans-Andean and Pacific Ecuador. Three hypotheses arose from this finding: 1) It is one additional polyphyletic group of N. olivacea distributed in Peru and Bolivia, where this species has not yet been reported; 2) These specimens represent cases of convergent adaptative morphology evolution of N. nasua to the phenotype of N. olivacea which is adapted to live in high altitudes; or 3) Because these haplotypes were very similar to those of the N. nasua haplogroup where they were enclosed, it could be a case of relatively recent hybridization or introgression between N. nasua and N. olivacea. However, as N. olivacea has not been described in these areas of Peru and Bolivia, this could be a proof in favor of the existence of N. olivacea in some Andean areas of Peru and Bolivia.

There only seems to be one group of N. nasua in Paraguay and Uruguay as well as for southern Brazil. It is these southern areas of South America where these groups of N. nasua arrived more recently and therefore, the genetic situation is simpler.

Our analyses indicated that the coati introduced to the Juan Fernández Archipielago (Robinson Crusoe Island) in Chile was N. nasua (H7-NAS). It was previously thought that the trans-Andean Pacific Ecuadorian coati was N. narica, and the coatis in the Juan Fernández Archipielago probably were taken from the Pacific Ecuador as our molecular results demonstrated, but also our findings demonstrated that the majority (but not all) of coatis of this Ecuadorian area are N. nasua.

As we did not analyze specimens from Venezuela, Guianas, and northern Brazil, we cannot determine if any haplogroups is related to the morphological subspecies N. n. vittata (type locality: Mt. Roraime, Venezuela; [108]).

One really unique way to determine the similarity between the haplogroups and the morphological subspecies would be to sequence the holotypes used to define the different subspecies of N. nasua.

We analyzed the systematics of N. olivacea in Ruiz-García, et al. (2020). Here, we will very briefly discuss the issue. Helgen, et al. (2009) determined that the population of the mountain coati in Merida Cordillera in Venezuela was an endemic species that they named Nasuella meridensis. However, Ruiz-García, et al., (2020, 2021) showed that the Venezuelan N. meridensis was undifferentiated from the H4-OLI that we mainly detected in the Eastern Colombian Andean Cordillera. Therefore, N. olivacea olivacea [111] and N. meridiensis [110] are synonyms. This means that H4-OLI agrees quite well with N. o. olivacea [111]. H3-OLI should correlate with N. o. quitensis [112] because it is the most common haplogroup of the mountain coati in the Andes region of Ecuador. However, H1-OLI and H2-OLI were also in the northern Ecuadorian Andes. Henceforth, only the molecular analysis of the holotype of N. o. quitensis will completely resolve this issue.

Nevertheless, this work brings forth new information on N. olivacea: 1) we augmented the number of mountain coatis and we detected one specimen from Tamá NP (Norte de Santander Department, Colombia) belonging to H1-OLI. All of the specimens from Tamá NP were previously analyzed by Ruiz-García, et al. (2020) and belonged to H4-OLI. Thus, we did detect one specimen with other haplogroups of N. olivacea on the Eastern Colombian Andean Cordillera. Thus, we detect more cases of geographic sympatry between different groups as we increase the sampling size. 2) The H1-OLI was a haplogroup of phenotypic specimens similar to N. olivacea but with the mitochondrial DNA more related to the Colombian and Ecuadorian Andean N. nasua than to the N. olivacea H3-OLI and H4-OLI. We named this haplogroup as “intermediate” between N. nasua (H6-NAS) and the most derived haplogroups of N. olivacea (H3-OLI and H4-OLI). This shows the strong relationships between different groups of Andean N. nasua and different groups of N. olivacea. Again this supports that the mountain coatis are really polyphyletic. 3) Additionally, in this work, we detected an opposing case. Specimen’s phenotypically N. nasua presented mitochondrial DNA more related to that of the most derived N. olivacea (H3-OLI and H4-OLI) than to any other haplogroups of N. nasua. Any specimens that were phenotypically N. nasua but had similar (but differentiated) mitochondrial DNA to the derived haplogroups of N. olivacea (rather than to the other N. nasua’ DNAs) were Andean specimens from Antioquia (Colombia), Imbabura and Pichincha Provinces in Ecuador, and Junin Department in central Peruvian Andes. Thus, we detected not only “intermediate” haplotypes of phenotypic specimens of N. olivacea from N. nasua towards the most derived N. olivacea, but we also detected “intermediate” haplotypes of specimens phenotypically N. nasua towards the most derived N. olivacea haplogroups, which is highly related with reticulate evolution among very related taxa without genetic isolation when these taxa are living in sympatry. For this reason, it would be better to consider all of the coatis as a unique species. 4) But, possibly, the most relevant finding was the clear detection of a specimen of N. olivacea from H3-OLI in southern Peru at the Urubamba River (Cuzco Department). This specimen had a morphotype similar to the N. olivacea from Colombia and Ecuador and, also, an mtDNA of an undisputable N. olivacea from H3- OLI. Therefore, this is the first indubitable proof in favour of the presence of N. olivacea in Peru. Until now, there was no clear evidence of mountain coati in Peru. So, we provide the first irrefutable proof of the presence of a “true” N.

olivacea in Peru (photos of the dry skin of this specimen could be provided by the first author)

Temporal splits in the evolution of the coatis

Traditionally, from a paleontological point of view, the earliest fossils of Nasua were found in North America and dated to the late Hemphillian (6.7- 4.7 MYA) to early Irvingtonian (1.6-1.0 MYA) [113-115]. Based on these paleontological records and that the South American fossil record of Nasua was dated to the Pleistocene, the current coatis derived from North American lineages that migrated into South America during the Pleistocene following the emergence of the Panamanian isthmus [113,116,117]. In fact, the entry in South America (Argentina) of the first Holartic mammals during the Huayquerian (Late Miocene; 9-6 MYA) were procyonids of the endemic genera Cyonasua and Chapalmalania [118,119] dated, at least, around 7.3 MYA. Paleontologist considered, however, that these procyonids disappeared and there were not linked with the current Nasua. They considered that the current Nasua in South America is the descendent of a second wave of procyonids that entered into South America during the GABI 4 (Great American Biotic Interchange, phase 4; [11]) developed during the Lujanian and Platan ages (beginning at 0.125 MYA; Late Pleistocene).

Nevertheless, our molecular data completely disagree with this view. The molecular evolution of the coati seems to appear around 13-10 MYA “in situ” in South America and therefore they are correlated with the first procyonid wave, represented by Cyonasua, arrived to South America during the Miocene. With the MJN procdure, we estimated around 13 MY (for the nucleotide substitution rate of 1.3%/MY), or 9.9 MY (for the rate of 1.7%/MY) passed between the first and oldest N. nasua haplotype and the most derived haplotype of N. narica The Bayesian analyses indicated that the original mitochondrial diversification process in the coatis began around 11.9 MYA. In agreement with our own results, Nigenda-Morales, et al. (2019) determined that the initial evolution of the coatis occurred from South America to Central America. They estimated around the beginning of the evolution of coati in South America to have begun 7-6 MYA. They only studied five specimens of N. nasua and therefore they did not detect the oldest haplotypes of the Andean N. nasua. Therefore, they did not detect the oldest mitochondrial diversification events that we detected in the current study. Be it as it may be, the initial lineage diversification in the coatis began in South America (and not in North or Central America) in the Late Miocene.

Koepfli, et al. (2007) showed that the ancestors of Bassaricyon and Nasua and the ancestors of Bassariscus and Procyon diverged contemporaneously near the end of the Middle Miocene around 12–13 MYA. Additionally, the ancestors of the extant species Bassariscus astutus and B. sumichrasti diverged around 9.1–10.6 MYA in the Late Miocene. Recently, it was determined the initial diversification process of Potos flavus occurred around 9.6 MYA [120].

These correlated patterns of divergence between independent procyonid lineages agreed with the influence of major regional environmental processes on radiation of the Procyonidae [121]. This period coincides with a process of major global cooling, along with formation of the Antarctic ice sheet and a short-term but important drop in global sea level [122-124]. In addition, the Neotropical forests reduced and transformed into savanna-like environments that expanded [125]. These climatic events played an outstanding role in differentiation of other mammalian species [121,126].

The time elapsed from the first and original N. nasua haplotype to the most derived haplotypes of N. nasua (H1-NAS; Bolivia and southern Peru) was around 8.4-6.4 MY. We detected that the initial Ecuadorian N. narica haplotypes originated around 4.2-3.6 MY, whilst Nigenda-Morales, et al. (2019) detected the initial evolution of the southern Central American N. narica around 3.8 MYA. All these diversification processes were during the Pliocene.

Two processes were important during the Pliocene, which could help to understand the important diversification of coati during this period. The first was climatic. The cold and dry climate during the Pliocene coincided with the onset of high-latitude glacial cycles, causing an explosive expansion of low-biomass vegetation, in many places of the world. These changes were correlated with the diversification of prey species that exploited new habitats, which in turn could provide new niches for many carnivores as the coatis. Additionally, it caused fragmentation of some Neotropical rain forests. The second was geologic. Strong geological changes occurred in the Northern Andes, which had a noteworthy repercussion in the Andean Cordilleras and in the rivers related with these Cordilleras. It was very important the formation of the Fitzcarrald Arch (4 MYA). It is demonstrated that the Nazca Ridge subduction imprint had a significant influence on the eastern side of the Andes by means of the Fitzcarrald Arch [127,128]. This uplift is responsible for the atypical three-dimensional shape of the Amazonian foreland basin. Related to Nazca Ridge subduction, arc volcanism in the Peruvian Andes ceased around 4 MYA [129]. These geological changes in the Northern Andes were fundamental because many of the oldest N. nasua and N. olivacea groups appeared in that area of the Northern Andes and, even, the oldest Ecuadorian N. narica haplotypes were also generated at this time.

Finally, the diversification among the youngest haplotypes of N. narica or within the youngest haplogroups of N. nasua and N. olivacea were basically during the Pleistocene (2.5-0.01 MYA). We dated the diversification of N. narica from middle Central America to northern Guatemala and southern Mexico to around 1.1 MYA. This estimate is very similar to the 1.2 MYA estimate made by Nigenda-Morales, et al. (2019). The Pleistocene Refugia Hypothesis (PRH) [130-132] could be extremely important in helping to explain why in some areas, especially in Colombia, Ecuador, and Peru, many haplogroups of N. nasua and N. olivacea were found in sympatry. The PRH could be also being extended to the Pliocene (until 2.8 MYA) [131]. This hypothesis claims that alternative dry-wet cycles caused by the Milankovitch cycles [133] affected the forest of the Neotropics. It was provided evidence about the climatic dryness that affected the Amazon and humid tropical environments of the Neotropics [134]. Large inactive dune fields, typical of dry environments, extended several thousand km2 between the Negro and Branco rivers (Pantanal do Norte, northern Brazilian Amazon) [135]. Today, this area is covered by the caatinga (open vegetation), which suggests that this area was much dryer in the past compared to today. Correlated with this, numerous smaller Aeolian sand fields exist further north around Boa Vista [136]. Some rainforests, 250 kilometers north of Manaus, are underlain by layers of coarse and extremely poorly sorted sediments which must have been deposited under dry climatic conditions of the recent geological past when dense rainforests were absent at this geographical area [137]. Henceforth, extreme sympatry of many groups of N. nasua and N. olivacea, rather than N. narica originated by contraction (isolation and apparition of the haplogroups found) and posterior expansion of these haplogroups which meet in some identical geographical areas (both in high and low altitudes) today.

Conclusion

Our experience shows us that in order to determine how many groups or taxa are within a species, or taxon, we should initially conduct an analysis of mitochondrial DNA. In the second step, we analyze nuclear markers. This allows us to determine the real number of groups or ESUs and if genetic interchange (s) occurred amongst these groups. The second step also helps to determine when the genetic interchanges occurred, and if gene flow is sexually biased. Therefore, this mitochondrial analysis of the coatis is a first step in completely reconstructing the evolutionary history of these amazing species.

References

1. Eisenberg, J. F. “Mammals of the Neotropics: The Northern Neotropics: Panama, Colombia, Venezuela, Guyana, Suriname, French Guyana.” J Mammal 71 (21 May 1990): 267â??268.

Google Scholar, Crossref

2. Wyss, A. R. and J.J. Flynn. “A phylogenetic analysis and definition of the Carnivora.” Mammal Phylogeny: Placentals (1993): 32-52.

Google Scholar

3. Wozencraft, W. C. “Order Carnivora.” Mammal Species of the World: A Taxonomic and Geographic Reference (2005): 532â??628.

Google Scholar

4. Wilson, Don E. and Russell A. Mittermeier. “Handbook of the Mammals of the World. Vol. 1.Carnivores.” (2009).

Google Scholar

5. Koepfli, Klaus-Peter, Matthew E. Gompper, Eduardo Eizirik and Cheuk-Chung Ho, et al. “Phylogeny of the Procyonidae (Mammalia: Carnivora): molecules, morphology and the Great American Interchange.” Mol Phylogenet Evol 43 (2007): 1076-1095.

Google Scholar, Crossref

6. Soibelzon, Leopoldo and Francisco Juan Prevosti. “Fossils of South American Land Carnivores (Carnivora, Mammalia).” Nova Science Publisher, New York, NY, USA. (2013): 509â??527.

Google Scholar

7. Ruiz-Ramoni, Damián, Ascanio Rincón and Marisol Montellano-Ballesteros. “Evidencias del origen de Nasua y Procyon (Procyonidae: Carnivora) en América del Sur.” Rev Brasil Paleontol 21 (2018): 87-94.

Google Scholar, Crossref

8. Marshall, L. G., R. F. Butler, R. E. Drake and G. H. Curtis, et al. “Calibration of the great American interchange.” Science 204 (1979): 272-279.

Google Scholar, Crossref, Indexed at

9. Webb, S. David. “Late Cenozoic of mammal dispersals between the Americas.” 4 (1985): 357â??386.

Google Scholar, Crossref

10. Webb, S. David. “The Great American Biotic Interchange: patterns and processes.” Annals Missouri Botanical Garden 93 (2006): 245â??257.

Google Scholar, Crossref

11. Woodburne, Michael O. “The Great American Biotic Interchange: Dispersals, tectonics, climate, sea level and holding pens.” J Mammal Evol 17 (2010): 245-264.

Google Scholar, Crossref, Indexed at

12. Marshall, Larry G. “Geochronology and land-mammal biochronology of the transamerican faunal interchange.” The Great American Biotic Interchange. Plenum Press, New York, NY, USA. (1985): 357-386.

Google Scholar, Crossref

13. Ramírez, Mariano A., Francisco J. Prevosti, Alejandro Acosta and Natacha B. U. C., et al. “On the presence of Nasua Storr (Carnivora, Procyonidae) in the Buenos Aires province in the late Holocene.” Rev Mus Argentino Cienc Nat 17 (2015): 51â??58.

Google Scholar

14. Nowak, Ronald M. “Walker’s Mammals of the World.” (1999).

Google Scholar

15. Redford, Kent Hubbard and John Frederick Eisenberg. “Mammals of the Neotropics: The Southern Cone: Chile, Argentina, Uruguay, Paraguay.” University of Chicago Press, Chicago and London. (1992).

Google Scholar

16. Russell, James K. “Exclusion of adult male coatis from social-groups - protection from predation.” J Mammal 62 (1981): 206-208.

Google Scholar, Crossref

17. Helgen, Kristofer M., Roland W. Kays, Lauren E. Helgen and Mirian Tieko Nunes Tsuchiya-Jerep, et al. “Taxonomic boundaries and geographic distributions revealed by an integrative systematic overview of the mountain coatis, Nasuella (Carnivora: Procyonidae).” Small Carnivore Conservation 41 (2009): 65-74.

Google Scholar

18. Kays, R. “Family Procyonidae (raccoons).” (2009): 504â??530.

Google Scholar

19. McFadden, Kate, Matthew E. Gompper, David Valenzuela-Galván and Juan Carlos Morales. “Evolutionary history of the critically endangered Cozumel dwarf carnivores inferred from mitochondrial DNA analyses.” J Zool 276 (2008): 176â??186.

Google Scholar, Crossref

20. Gompper, Matthew E. “Nasua narica.” Mammalian Species 487 (1995): 1-10.

Google Scholar, Crossref

21. Gompper, Matthew E. and Denise M. Decker. “Nasua nasua.” Mammalian Species 580 (1998): 1-9.

Google Scholar

22. Emmons, L. H. “Carnivores (Procyonidae).” (1990): 136-138.

Google Scholar

23. McFadden, Katherine Walton. “The Ecology, Evolution and Natural History of the Endangered Carnivores of Cozumel Island, Mexico.” (2004).

Google Scholar

24. Tsuchiya-Jerep, Mirian Tieko Nunes. “Filogeografia, história demográfica e diversidade molecular de duas espécies neotropicais da família Procyonidae (Mammalia, Carnivora): Nasua nasuae Procyon cancrivorus.” (2009).

Google Scholar

25. Neves-Chaves, B. R. “Diversidade genética e dinâmica populacional de quatis (Nasua nasua) em Minas Gerais.” (2011).

Google Scholar

26. Silva Caballero, Adrián, Gloria León-Ávila, David Valenzuela-Galván and Jesús E. Maldonado, et al. “Patterns of Genetic Diversity of the White- Nosed Coati Reveals Phylogeographically Structured Subpopulations in Mexico.” Natural Resources 8 (2017): 31-53.

Google Scholar, Crossref

27. Nigenda-Morales, Sergio F., Matthew E. Gompper, David Valenzuela-Galván and Anna R. Lay, et al. “Phylogeographic and diversification patterns of the white-nosed coati (Nasua narica): Evidence for south-to-north colonization of North América.” Mol Phylogenet Evol 131 (2019): 149â??163.

Google Scholar, Crossref

28. Jaramillo, María F. and Manuel Ruiz-García. “Evidencia de estructura genética y espacial muy robusta en el coatí de nariz blanca (Nasua narica; Procyonidae, Carnivora) en Centroamérica y norte de Sudamérica mediante análisis mitogenómicos.” Therya (2022).

Google Scholar

29. Avise, John C. “Molecular markers, natural history, and evolution.” New York, Chapman and Hall. (1994).

Google Scholar

30. Avise, John C. “Phylogeography: The History and Formation of Species.” Harvard University Press, Cambridge. (2000).

Google Scholar

31. Avise, John C., Jonathan Arnold, R. Martin Ball and Eldredge Bermingham, et al. “Intraspecific phylogeographic: the mitochondrial DNA bridge between population genetics and systematics.” Ann Review Ecol Syst 18 (1987): 489-522.

Google Scholar

32. Guschanski, Katerina, Johannes Krause, Susanna Sawyer and Luis M. Valente, et al. “Next- generation museomics disentangles one of the largest primate radiations.” Syst Biol 62 (2013): 539â??554.

Google Scholar, Crossref

33. Mason, Victor C., Gang Li, Kristofer M. Helgen and William J. Murphy. “Efficient cross-species capture hybridization and next-generation sequencing of mitochondrial genomes from noninvasively sampled museum specimens.” Genome Research 21 (2011): 1695â?? 1704.

Google Scholar, Crossref

34. Galtier, Nicolas, David Enard, Yoan Radondy and Eric Bazin, et al. “Mutation hotspots in mammalian mitochondrial DNA.” Genome Research 16 (2006): 215â??222.

Google Scholar, Crossref, Indexed at

35. Nabholz, Benoit, Hans Ellegren and Jochen B. W. Wolf. “High levels of gene expression explain the strong evolutionary constraint of mitochondrial protein-coding genes.” Mol Biol Evol 30 (2012): 272â??284.

Google Scholar, Crossref

36. Collins A. C. and J. M. Dubach. “Phylogenetic relationships of spider monkeys (Ateles) based on mitochondrial DNA variation.” Int J Primatol 21 (2000): 381â??420.

Google Scholar, Crossref

37. Cortés-Ortiz, Liliana, Eldredge Bermingham, Ciro Rico and E. Rodríguez-Luna. “Molecular systematics and biogeography of the Neotropical monkey genus Alouatta.” Mol Phylogenet Evol 26 (2003): 64â??81.

Google Scholar, Crossref

38. Ruiz-García, Manuel, Myreya Pinedo-Castro and Joseph Mark Shostell. “How many genera and species of woolly monkeys (Atelidae, Platyrrhine, Primates) are there? First molecular analysis of Lagothrix flavicauda, an endemic Peruvian primate species.” Mol Phylogenet Evol 79 (2014): 179-198.

Google Scholar, Crossref

39. Ruiz-García, Manuel, Sebastián Sánchez-Castillo, María Ignacia Castillo and Kelly Luengas. “How many species, taxa, or lineages of Cebus albifrons (Platyrrhini, Primates) inhabit Ecuador? Insight from mitogenomics.” Intl J Primatol 39 (2018): 1068-1104.

Google Scholar, Crossref

40. Freeman, Scott and Jon C. Herron. “Evolutionary Analysis.”Upper Saddle River, Prentice Hall. (1998).

Google Scholar

41. Burrell, Andrew S., Clifford J. Jolly, Anthony J. Tosi and Todd R. Disotell. “Mitochondrial evidence for the hybrid origin of the kipunji, Rungwecebus kipunji(Primates: Papionini).” Mol Phylogenet Evol 51 (2009): 340â??348.

Google Scholar, Crossref

42. Derenko, Miroslava, Boris Malyarchuk, Galina Denisova and Maria Perkova, et al. “Complete Mitochondrial DNA Analysis of Eastern Eurasian Haplogroups Rarely Found in Populations of Northern Asia and Eastern Europe.” PLoS ONE 7 (2012): e32179.

Google Scholar, Crossref, Indexed at

43. Krause, Johannes, Qiaomei Fu, Jeffrey M. Good and Bence Viola, et al. “The complete mitochondrial DNA genome of an unknown hominin from southern Siberia.” Nature 464 (2010): 894â??897.

Google Scholar, Crossref

44. Pääbo, Svante. “Neanderthal Man: In search of lost genomes.” New York: Basic Books. (2014).

Google Scholar

45. Posth, Cosimo, Christoph WiÃ?ing, Keiko Kitagawa and Luca Pagani, et al. “Deeply divergent archaic mitochondrial genome provides lower time boundary for African gene flow into Neanderthals.” Nature Communications 8 (2017): 16046.

Google Scholar, Crossref

46. Sawyer, Susanna, Gabriel Renaud, Bence Viola and Jean-Jacques Hublin, et al. “Nuclear and mitochondrial DNA sequences from two Denisovan individuals.” Proc Nat Acad Science USA 112 (2015): 15696â??15700.

Google Scholar, Crossref

47. Slon, Viviane, Charlotte Hopfe, Clemens L. Weiss and Fabrizio Mafessoni, et al. “Neandertal and Denisovan DNA from Pleistocene sediments.” Science 356 (2017): 605-608.

Google Scholar, Crossref

48. O’Brien, S. J. “A role for molecular genetics in biological conservation.” Proc Nat Acad Science USA 91 (1994): 5748-5755.

Google Scholar, Crossref

49. Guzmán-Lenis, Angélica R. “Revisión preliminar de la familia Procyonidae en Colombia.” Acta Biológica Colombiana 9 (2004): 60-76.

Google Scholar

50. Sambrock, J., E. Fritsch and T. Maniatis. “Molecular Cloning: A Laboratory manual.” 2nd edition. V1. New York: Cold Spring Harbor Laboratory Press. (1989).

Google Scholar

51. Walsh, P. Sean, David A. Metzger and Russell Higuchi. “Chelex 100 as a medium for simple extraction of DNA for PCR-based typing from forensic material.” BioTechniques 10 (1991): 506-513.

Google Scholar, Crossref

52.  Irwin, David M., Thomas D. Kocher and Allan C. Wilson. “Evolution of the cytochrome b gene of mammals.” J Molec Evol 32 (1991): 128â??144.

Google Scholar, Crossref

53. Folmer, Ole, Michael B. Black, Hoeh Wr and R. Lutz, et al. “DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates.” Molec Marine Biol Biotechnol 3 (1994): 294-299.

Google Scholar, Indexed at

54. Ruiz-García, Manuel, Maria Ignacia Castillo, Andrea Ledezma and Norberto Leguizamon, et al. “Molecular Systematics and Phylogeography of Cebus capucinus (Cebidae, Primates) in Colombia and Costa Rica by means of mitochondrial COII gene.” Amer J Primatol 74 (2012): 366-380.

Google Scholar, Crossref

55. Björnerfeldt, Susanne, Matthew T. Webster and Carles Vilà. “Relaxation of selective constraint on dog mitochondrial DNA following domestication.” Genome Research 16 (2006): 990-994.

Google Scholar, Crossref

56. Trigo, T. C., T. R. O. Freitas, G. Kunzler and L. Cardoso, et al. “Inter-species hybridization among Neotropical cats of the genus Leopardus, and evidence for an introgressive hybrid zone between L. geoffroyiand L. tigrinusin southern Brazil.” Mol Ecol 17 (2008): 4317â??4333.

Google Scholar, Crossref

57. Trinca, Cristine S., Benoit de Thoisy, Fernando C. W. Rosas and Helen F. Waldemarin, et al. “Phylogeography and demographic history of the neotropical otter (Lontra longicaudis).” J Heredity 103 (2012): 479â??492.

Google Scholar, Crossref

58. Zhang, Ya-ping and Ryder Oliver A. “Phylogenetic relationships of bears (the ursidae) inferred from mitochondrial DNA sequences.” Mol Phylogenet Evol 3 (1994): 351â??359.

Google Scholar, Crossref

59. Hoelzel, A. Rus, Jose V. Lopez, Gabriel A. Dover and Stephen J. O'Brien. “Rapid evolution of a heteroplasmic repetitive sequence in the Mitochondrial DNA control region of carnivores.” J Mol Evol 39 (1994): 191-199.

Google Scholar, Crossref

60. Vaidya, Gaurav, David J. Lohman and Rudolf Meier. “SequenceMatrix: concatenation software for the fast assembly of multi-gene datasets with character set and codon information.” Cladistics 27 (2011): 171â??180.

Google Scholar, Crossref

61. Darriba, Diego, Guillermo L. Taboada, Ramón Doallo and David Posada. “jModelTest2: more models, new heuristics and parallel computing.” Nature Methods 9 (2012): 772.

Google Scholar, Crossref

62. Tamura, Koichiro, Glen Stecher, Daniel Peterson and Alan Filipski, et al. “MEGA6: Molecular Evolutionary Genetics Analysis version 6.0.” Mol Biol Evol 30 (2013): 2725-2729.

Google Scholar, Crossref, Indexed at

63. Akaike, H. “A new look at the statistical model identification.” IEEE Transactions on Automatic Control 19 (1974): 716â??723.

Google Scholar, Crossref

64. Posada, David and Thomas R. Buckley. “Model selection and model averaging in phylogenetics: advantages of akaike information criterion and Bayesian approaches over likelihood ratio tests.” Syst Biol 53 (2004): 793â??808.

Google Scholar, Crossref

65. Schwarz, Gideon. “Estimating the dimension of a model.” Annals of Statistics 6 (1978): 461-464.

Google Scholar

66. Stamatakis, Alexandros. “RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models.” Bioinformatics 22 (2006): 2688â??1243.

Google Scholar, Crossref

67. Lanfear, Robert, Brett Calcott, Simon Y. W. Ho and Stephane Guindon. “PartitionFinder: combined selection of partitioning schemes and substitution models for phylogenetic analyses.” Mol Biol Evol 29 (2012): 1695â??1701.

Google Scholar, Crossref

68. Lanave, Cecilia, Giuliano Preparata, Cecilia Sacone and Gabriella Serio. “A new method for calculating evolutionary substitution rates.” J Mol Evol 20 (1984): 86-93.

Google Scholar, Crossref

69. Stamatakis, Alexandros, Paul Hoover and Jacques Rougemont. “A rapid bootstrap algorithm for the RAxML Web servers.” Syst Biol 57 (2008): 758â??771.

Google Scholar, Crossref

70. Hillis, David M. and James J. Bull. “An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis.” Syst Biol 42 (1993): 182â??192.

Google Scholar, Crossref

71. Drummond, Alexei J., Marc A. Suchard, Dong Xie and Andrew Rambaut. “Bayesian phylogenetics with BEAUti and the BEAST 1.7.” Mol Biol Evol 29 (2012): 1969â??1973.

Google Scholar, Crossref, Indexed at

72. Rambaut, A., M. A. Suchard, W. Xie and A. J. Drummond. “Tracer v1.6.” (2013).

Google Scholar

73. Rambaut, A. and A. J. Drummond. “Log Combiner v1.8.0.” (2013a).

Google Scholar

74. Rambaut, A. and A. J. Drummond. “TreeAnnotator v1.8.0.” (2013b).

Google Scholar

75. Drummond, Alexei J., Simon Y. W. Ho, Matthew J. Phillips and Andrew Rambaut. “Relaxed phylogenetics and dating with confidence.” PLOS Biol 4 (2006): e88.

Google Scholar, Crossref

76. Erixon, Per, Bodil Svennblad, Tom Britton and Bengt Oxelman. “Reliability of Bayesian posterior probabilities and bootstrap frequencies in phylogenetics.” Syst Biol 52 (2003): 665â??673.

Google Scholar, Crossref

77. Huelsenbeck, John P. and Bruce Rannala. “Frequentist properties of Bayesian posterior probabilities of phylogenetic trees under simple and complex substitution models.” Syst Biol 53 (2004): 904â??913.

Google Scholar, Crossref

78. Rambaut, A. “FigTree v1.4.” (2012).

Google Scholar

79. Bandelt, H. J., P. Forster and A. Rohl. “Median-joining networks for inferring intraspecific phylogenies.” Mol Biol Evol 16 (1999): 37-48.

Google Scholar, Crossref

80. Morral, N., J. Bertrantpetit, X. Estivill and V. Nunes, et al. “The origin of the major cystic fibrosis mutation (delta F508) in European populations.” Nature Genetics 7 (1994): 169-175.

Google Scholar, Crossref

81. Saillard, J., P. Forster, N. Lynnerup and H. J. Bandelt, et al. “mtDNA variation among Greenland Eskimos: the edge of the Beringian expansion.” Amer J Human Genet 67 (2000): 718-726.

Google Scholar, Crossref, Indexed at

82. Pennington, R. Toby and Christopher W. Dick. “Diversification of the Amazonian flora and its relation to key geological and environmental events: a molecular perspective.” (2010): 373-385.

Google Scholar

83. Wayne, Robert K., Eli Geffen, Derek J. Girman and Klaus P. Koepfli, et al. “Molecular Systematics of the Canidae.” Syst Biol 46 (1997): 622-653.

Google Scholar, Crossref

84. Kimura, Motoo. “A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences.” J Mol Evol 16 (1980): 111-121.

Google Scholar, Crossref

85. Hebert, Paul D. N., Alina Cywinska, Shelley L. Ball and Jeremy R. deWaard. “Biological identifications through DNA barcodes.” Proc Royal Soc London B 270 (2003): 313-321.

Google Scholar, Crossref

86. Hebert, Paul D. N., Mark Y. Stoeckle, Tyler S. Zemlak and Charles M. Francis. “Identification of birds through DNA barcodes.” PLoS Biol 2 (2004): 1657-1663.

Google Scholar, Crossref, Indexed at

87. Kartavtsev, Yuri Ph. “Divergence at Cyt-b and Co-1 mtDNA genes on different taxonomic levels and genetics of speciation in animals.” Mitochondrial DNA 22 (2011): 55-65.

Google Scholar, Crossref

88. Ascunce, Marina S., Esteban Hasson and Marta D. Mudry. “COII: a useful tool for inferring phylogenetic relationships among New World monkeys (Primates, Platyrrhini).” Zool Scripta 32 (2003): 397-406.

Google Scholar, Crossref

89. Bradley, Robert D. and Robert J. Baker. “A test of the genetic species concept: cytochrome-b sequences and mammals.” J Mammal 82 (2001): 960â??973.

Google Scholar, Crossref

90. Baker, Robert J. and Robert D. Bradley. “Speciation in mammals and the genetic species concept.” J Mammal 87 (2006): 643â??662.

Google Scholar, Crossref, Indexed at

91. Ruiz-García, Manuel, Maria Fernanda Jaramillo, Carlos Herney Cáceres-Martínez and Joseph Mark Shostell. “The Phylogeographic Structure of the mountain coati (Nasuella olivacea; Procyonidae, Carnivora), and phylogenetic relationships with the other coati species (Nasua nasuaand Nasua narica) as inferred by mitochondrial DNA.” Mammal Biol 100 (2020): 521-548.

Google Scholar, Crossref

92. Ruiz-García, Manuel, María F. Jaramillo, Juan B. López and Yudrum Rivillas, et al. “Mitochondrial and karyotypic evidence reveals a lack of support for the genus Nasuella (Procyonidae, Carnivora).” J Vertebrate Biol 71 (2021): 21040.

Google Scholar, Crossref

93. Helgen, Kristofer M., C. Miguel Pinto, Roland Kays and Lauren E. Helgen, et al. “Taxonomic revision of the olingos (Bassarycion), with description of a new species, the Olinguito.” ZooKeys 324 (2013): 1-83.

Google Scholar, Crossref, Indexed at

94. Cracraft, Joel. “Speciation and its ontology: the empirical consequences of alternative species concepts for understanding patterns and processes of differentiation.” (1989): 28â??59.

Google Scholar

95. De Queiroz, Kevin and Michael J. Donoghue. “Phylogenetic systematics and the species problem.” Cladistics 4 (1988): 317-338.

Google Scholar, Crossref

96. Isaac, Nick J. B., James Mallet and Georgina M. Mace. “Taxonomic inflation: its influence on macroecology and conservation.” Trends Ecol Evol 19 (2004): 464â??469.

Google Scholar, Crossref

97. Zachos, Frank E., Marco Apollonio, Eva V. Bärmann and Marco Festa-Bianchet, et al. “Species inflation and taxonomic artefactsâ??a critical comment on recent trends in mammalian classification.” Mammal Biol 78 (2013): 1â??6.

Google Scholar, Crossref

98. Zachos, Frank E. “Species Concepts in Biology. Historical Development, Theoretical Foundations and Practical Relevance.” Springer International Publishing, Switzerland. (2016).

Google Scholar

99. Hollister, Ned. “The genera and subgenera of raccoons and their allies.”  Proceedings of the United States National Museum 49 (1915): 143â??150.

Google Scholar, Crossref

100. Storr, G. C. C. “Prodromus methodi mammalium.” Litteris Reissimis, Tubingae (1780): 43.

Google Scholar

101. Ruiz-García, Manuel, Nicolás Lichilín, Pablo Escobar-Armel and Geven Rodríguez, et al. “Historical Genetic demography and some insights into the systematics of Ateles (Atelidae, Primates) by means of diverse mitochondrial genes.” (2016): 435-476.

Google Scholar

102. Olfers, Von I. “Bemerkungen zu Illiger’s ueberblick der Säugthire nach ihrer vertheilung über die Welttheile, rück-sichtlich der Südamericanishen Arten (Species).” 15 (1818): 192-237.

Google Scholar

103. Cabrera, A. “Una nueva forma del género Nasua.” Neotropica 2 (1956): 1-4.

Google Scholar

104. Linnaeus, C.“Systema naturae per regna tria nature, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. Regnum animale.” Twelfth ed. Laurentii Salvii. Holmiae 1 (1766): 1-824.

Google Scholar

105. Schinz, Heinrich Rudolf. “Das Thierreich eingetheilt nach dem Bau der Thiere als grundlage ihre Naturgeschichte und der vergleichenden anatomia von dem Herrn Ritter von Cuvier, Staatsrath von Frankeich und beständiger Secretär der Academie der Wissenschaften U.S.M.” Aus dem Französischen frei übersetzt und mit vielen Zusätzen versehen von H. R. Schinz. J. S. Gotta, Stuuugart und Tubiengen 1 (1821): 1-894.

Google Scholar, Crossref

106. Thomas, Oldfield. “Two new species of Nasua.” Annals and Magazine of Natural History Series 8 10 (1912): 228-230.

Google Scholar

107. Thomas, Oldfield. “New Nasua, Lutra, and Proechimys from South America.”  Annals and Magazine of Natural History Series 8 14 (1914): 57-61.

Google Scholar, Crossref

108. Tschudi, Johann Jakob von and Jean Louis Cabanis. “Untersuchungen über die Fauna Peruana.” St Gallen 1 (1845): 1-262.

Google Scholar

109. Gray, J. E. “Notice of a new species of Nasua.” Proc Zool Society of London (1866): 169-170.

Google Scholar

110. Thomas, Oldfield. “New Neotropical mammals, with a note on the species of Reithrodon.” Annals and Magazine of Natural History (series 7) 8 (1901): 246â?? 255.

Google Scholar, Crossref

111. Gray, John Edward. “A revision of the genera and species of ursine animals (Ursidae) founded on the collection in the British Museum.” Proc Zool Society of London (1865): 677â??709.

Google Scholar

112. Lönnberg, E. “Mammals from Ecuador and related forms.” Arkiv för Zoologi 8 (1913): 1â??36.

Google Scholar

113. Baskin, Jon Alan. “Tertiary Procyoninae (Mammalia: Carnivora) of North America.” J Vertebrate Paleontology 2 (1982): 71-93.

Google Scholar, Crossref

114. Cassiliano, Michael L. “Biostratigraphy of Blancan and Irvingtonian mammals in the Fish Creek-Vallecito section, southern California, and a review of the Blancanâ??Irvingtonian boundary.” J Vertebrate Paleontology 19 (1999): 169-186.

Google Scholar, Crossref

115. Emmert, Laura G. and Rachel A. Short. “Three new procyonids (Mammalia, Carnivora) from the Blancan of Florida.” Bull Florida Museum Nat Hist 55 (2018): 157-173.

Google Scholar

116. Baskin, Jon Alan. “Comments on New World Tertiary Procyonidae (Mammalia; Carnivora).” J Vertebrate Paleontology 9 (1989): 110â??117.

Google Scholar, Crossref

117. Baskin, Jon Alan. “New procyonines from the Hemingfordian and Barstovian of the Gulf Coast and Nevada, including the first fossil record of the Potosini.”Bull American Museum Nat Hist 279 (2003): 125â??146.

Google Scholar, Crossref

118. Prevosti, Francisco J. and Leopoldo H. Soibelzon. “Evolution of the South American carnivores (Mammalia, Carnivora): a paleontological perspective.” (2012): 102â??122.

Google Scholar, Crossref

119. Prevosti, Francisco J., Analía Forasiepi and Natalia Zimicz. “The evolution of the Cenozoic terrestrial mammalian predator guild in South America: competition or replacement?” J Mammal Evol 20 (2013): 3-21.

Google Scholar, Crossref

120. Ruiz-García, Manuel, Maria Fernanda Jaramillo and Joseph Mark Shostell. “Mitochondrial phylogeography of kinkajous (Procyonidae, Carnivora): maybe not a single ESU.” J Mammal 100 (2019): 16311652.

Google Scholar, Crossref

121. Delsuc, Frédéric, Sergio F. Vizcaíno and Emmanuel J. P. Douzery. “Influence of Tertiary paleoenvironmental changes on the diversification of South Américan mammals: a relaxed molecular clock study with xenarthrans.” BMC Evol Biol 4 (2004): 11-24.

Google Scholar, Crossref

122. Haq, Bilal U., Jan Hardenbol and Peter R. Vail. “Chronology of fluctuating sea levels since the Triassic.” Science 235 (1987): 1156-1167.

Google Scholar, Crossref

123. Miller, Kenneth G., Michelle A. Kominz, James V. Browning and James D. Wright, et al. “The Phanerozoic record of global sea-level change.” Science 310 (2005): 1293â??1298.

Google Scholar, Crossref

124. Zachos James, Mark Pagani, Lisa Sloan and Ellen Thomas, et al. “Trends, rhythms, and aberrations in global climate 65 Ma to present.” Science 292 (2001): 686-693.

Google Scholar, Crossref

125. Potts, R. and A. K. Behrensmeyer. “Late Cenozoic terrestrial ecosystems.” (1992): 419â??541.

Google Scholar

126. Johnson, Warren E., Eduardo Eizirik, Jill Pecon-Slattery and William J. Murphy, et al. “The Late Miocene radiation of the modern Felidae: a genetic assessment.” Science311 (2006): 73â??77.

Google Scholar

127. Espurt, Nicolas, Patrice Baby, Stéphane Brusset and Martin Roddaz, et al. “The Nazca ridge and uplift of the Fitzcarrald arch: implications for regional geology in northern South America.” (2010): 89â??102.

Google Scholar

128. Espurt, Nicolas, Patrice Baby, Stéphane Brusset and Martin Roddaz, et al. “How does the Nazca ridge subduction influence the modern Amazonian foreland basin?” Geology 35 (2007): 515â??518.

Google Scholar, Crossref

129. Rosenbaum, Gideon, David Giles, Mark Saxon and Peter G. Betts, et al. “Subduction of the Nazca ridge and the Inca plateau: insights into the formation of ore deposits in Peru.” Earth Planet Science Letters 239 (2005):18â??32.

Google Scholar, Crossref

130. Haffer, Jürgen. “Speciation in Amazonian forest birds.” Science 165 (1969): 131-137.

Google Scholar, Crossref

131. Haffer, Jürgen. “Alternative models of vertebrate speciation in Amazonia: an overview.” Biol Conser 6 (1997): 451-476.

Google Scholar, Crossref

132. Haffer, Jürgen. “Hypotheses to explain the origin of species in Amazonia.” Brazilian J Biol 68 (2008): 917-947.

Google Scholar, Crossref

133. Bennett, ?Keith D. “Evolution and ecology. - The Pace of Life.” (1997): 241.

Google Scholar

134. Burnham, Robyn J. and Alan Graham. “The History of Neotropical Vegetation: New Developments and Status.” Annals Missouri Botanical Garden 86 (1999): 546-589.

Google Scholar, Crossref

135. Santos, L. O. S, B. W. Nelson and C. A. Giovannini. â??Corpos de areia sob leitos abandonados de grandes ríos.â? Ciência Hoje (São Paulo) 16 (1993): 22-25.

Google Scholar

136. Carneiro, Filho A. and J. A. Zinck. “Mapping paleo-aeolian sand cover formations in the northern Amazon basin from TM images.” ITC J 3 (1994): 270-282.

Google Scholar

137. Veiga, A. T. C., M. A. Dardenne and E. P. Salomão. â??Geologia dos aluviões auriferos e estaniferos da Amazónia.â? Anais XXXV Congresso de Geologia (Belém, Pará) 1 (1988): 164-177.

Google Scholar