﻿New species of the Spiny Mouse genus Neacomys (Cricetidae, Sigmodontinae) from northwestern Ecuador

﻿Abstract Neacomys is a genus of small spiny or bristly sigmodontine rodents that are common components of mammalian faunas in multiple biomes on Central and South America. Recent studies on this group have demonstrated that there is cryptic diversity yet to be discovered within currently recognized species that have not received comprehensive revisions, as well as in areas that have been overlooked. Here we ratify this assertion by describing a new species previously misidentified as the Narrow-footed Spiny Mouse (Neacomystenuipes) from the Chocó biogeographic region in northwestern Ecuador, Neacomysmarci Brito & Tinoco, sp. nov. Distinctiveness of this entity is supported by the combination of the following morphological characters: small size (head-body length 65–85 mm); long tail (69–126% longer than head-body length); pale buff-colored but gray-based belly fur; white throat; hypothenar pad usually absent; long nasals; and a condylar process higher than the coronoid process. Likewise genetic distance analyses and phylogenetic reconstructions based on cytochrome-b (Cytb) sequence data indicate a clear divergence from typical populations of N.tenuipes, and a sister relationship between them. The results presented here increase the diversity of Neacomys to 24 species, placing it among the most diverse genera within the sigmodontine rodents.

New species of the Spiny Mouse genus Neacomys (Cricetidae, Sigmodontinae) from northwestern Ecuador

Introduction
Neacomys is a widely distributed genus of small spiny or bristly rodents that collectively occupy representative regions and habitats in easternmost Panama and the northern half of South America (Patton et al. 2015;Pardiñas et al. 2017 Caccavo and Weksler 2021;Semedo et al. 2021). Currently, 23 species are recognized within this group, occurring its highest concentration in the rainforests of the Amazon region (Hurtado and Pacheco 2017;Semedo et al. 2020Semedo et al. , 2021Brito et al. 2021a;Caccavo and Weksler 2021).
From the years 2017 through 2021, taxonomy of Neacomys has been remarkably dynamic and has resulted in the description of 11 species (Hurtado and Pacheco 2017;Sanchez-Vendizú et al. 2018;Semedo et al. 2020Semedo et al. , 2021Brito et al. 2021a;Caccavo and Weksler 2021;Colmenares-Pinzon 2021). The progress in the understanding of its diversity has been mainly achieved thanks to the exhaustive revision of material deposited in museum collections (Semedo et al. 2020(Semedo et al. , 2021Caccavo and Weksler 2021), as well as increased collection efforts and implementation of molecular analyses (Brito et al. 2021a;Colmenares-Pinzon 2021). However, as there are still many unexplored areas in the heterogenous geography of South America and adjacent Central America (Panama), some of the currently recognized species have not undergone comprehensive taxonomic evaluations, and it is possible that the real diversity of the genus is underestimated.
The Chocó Biogeographic region is considered one of the most diverse hotspots in South America (Myers et al. 2000), yet one of the least studied regions for Neacomys despite its large extension (along the Pacific coasts of Panama, Colombia, and Ecuador). To date, only two species are known to occur in the Chocó, the Painted Bristly Mouse N. pictus, and the Narrow-footed Spiny Mouse N. tenuipes (Patton et al. 2015;Pardiñas et al. 2017). The former has been recorded from one locality in Panama (Goldman 1912), and is scarcely represented in museums, by fewer than 12 specimens collected more than 30 years ago (VertNet Database): the presence of N. tenuipes is supported by only three specimens from two localities in Colombia (Colmenares-Pinzón 2021), and by an unclear number of specimens from at least four localities in Ecuador (Jarrín 2001;Brito et al. 2021a). Poor knowledge about the distribution of Neacomys throughout the Chocó region is accompanied by a lack of genetic characterization. This has prevented the inclusion of N. pictus in phylogenetic analyses of the genus, and thus there are no clues about its relationship with respect to other species. In the case of the populations from the department of Cauca, Colombia, and those from Ecuador, this has precluded the possibility of addressing their degree of differentiation from typical N. tenuipes, or even determine if they represent different species as some authors have hypothesized [e.g., the first one has been treated as N. pusillus Allen, 1912 (Caccavo andWeksler 2021) whereas the second was treated as N. pictus Goldman, 1912 (Jarrín 2001)]. The uncertainty about the affinity of some populations to N. tenuipes illustrates a possible cause of an underestimated diversity within the genus, where some of the currently recognized species have not been reviewed in detail.
With the recent collections of several specimens resembling Neacomys tenuipes in previously unexplored areas of northwestern Ecuador, their genetic and morphological characterization, and their comparison with material from different museums, this work describing a new species constitutes a forward step towards a better understanding of the variation within what has been considered a widely distributed and homogeneous species, as well as of the real diversity of the genus in the Chocó biogeographic region.
A detailed structural scrutiny of the skull of one specimen (MECN 6232; Estación Fisher, Ecuador) was done using a high-resolution micro-computed tomography (micro-CT) desktop scanner device (Bruker SkyScan 1173, Kontich, Belgium) at the Leibniz Institute for the Analysis of Biodiversity Change/ Museum Koenig (LIB, Bonn, Germany). To avoid movements during the scanning process, the material was placed in a small plastic container embedded in cotton wool. Acquisition parameters comprised: an X-ray beam (source voltage 43 kV and current 116 μA) without the use of a filter; 960 projections of 500 ms exposure time each with a frame averaging of 4 recorded over 180° using rotation steps of 0.25 degrees, resulting in a scan duration of 55 min 28 s; a magnification setup generating data with an isotropic voxel size of 12.07 μm. The CT-dataset was reconstructed with N-Recon software (Bruker MicroCT, Kontich, Belgium) and rendered in three dimensions using CTVox for Windows 64 bits v. 2.6 (Bruker MicroCT, Kontich, Belgium).
All specimens were classified into five age classes defined by Semedo et al. (2020) and Caccavo and Weksler (2021) based on the level of eruption of the third molar and the wear of the occlusal surface of the molars. Only specimens between ages 3 and 6 were used in the quantitative morphological comparisons.
We recorded external measurements from tags, and for the craniodental measurements we used digital calipers to the nearest 0.01 mm in all presumed specimens of N. tenuipes recently collected in northwestern Ecuador. We also measured older specimens of the species and other members of the genus housed in museums in Colombia and Ecuador (see above).
The craniodental measurements from 108 specimens tentatively identified as N. tenuipes, N. cf. pictus, and N. rosalindae Sánchez-Vendizú, Pacheco & Vivas-Ruiz, 2018 were compiled in a matrix with 2,376 values. This dataset was analyzed in R v. 4.2.1 (R Core Team 2022) and inferred for missing values using the missMDA package (Josse and Husson 2016). The iterative PCA algorithm was preferred for this purpose with a maximum of 1,000 iterations and a 1e-6 threshold to assess convergence. The estimated number of components needed to predict the missing values were obtained by running 100 simulations with the leave-one-out cross-validation method. Morphological characters were checked for high degrees of correlation using Spearman's coefficient, yet none were discarded since correlation values were ≤ 0.95. Non-parametric methods were preferred in all analyses (Šlenker et al. 2022).
Multivariate analyses performed in this study included Principal Component (PCA), and the K nearest neighbor classificatory Discriminant Analyses (KNN) with the MorphoTools2 package (Šlenker et al. 2022). For the latter, samples were grouped a priori as follows: 1) recently collected samples from northwestern Ecuador presumably belonging to N. tenuipes; 2) older museum specimens from northeastern Ecuador presumably belonging to N. cf. pictus; 3) typical N. tenuipes from Colombia; 4) N. rosalindae. To ensure that only invariant and non-linear characters were used in the KNN analysis, a stepwise discriminant analysis was conducted first and selected the following subset of characters: OL, LR, AW, BPB, CIL, HBL, LM, E, TL, LD, and ZB. Neacomys cf. pictus specimens were excluded because the total number of individuals (n = 2) was smaller than the total number of analyzed characters (Šlenker et al. 2022). The KNN results were plotted by centering and scaling the two variables that contributed the most to the discrimination of groups as predicted by the R 2 and F-values of the stepwise analysis.
Individuals' classification prediction was done using nine neighbors (k = 9) by estimating Euclidean distances through a cross-validation method. The precision of the classification was finally obtained as a percentage by comparing the model's prediction to the a priori classification herein assigned.
Statistical tests for non-uniformly distributed data were calculated and plotted using the ggstatsplot package (Patil 2021), to verify for significant statistical differences in the variables inferred to exert a greater effect on taxon differences. A Kruskal-Wallis test was applied to determine if the groups' medians were significantly different, followed by a Dunn test for a pairwise comparison of groups adjusting the p-value with the Holm method to control for the family-wise error rate (Holm 1979).

DNA extraction, amplification, and sequencing
DNA was extracted from muscle samples of the presumed specimens of N. tenuipes recently collected in northwestern Ecuador. The guanidine thiocinate protocol was used for DNA extraction (Bilton and Jaarola 1996). We amplified between 1000 and 1100 bases pair of mitochondrial gene Cytochrome b (Cytb); we used the forward primer MVZ05, and the reverse primers MVZ16H, MVZ14 (Smith and Patton 1993). The thermal profile for the amplification of Cytb included: an initial denaturation at 94 °C for 180 s, 35 cycles of denaturation at 94 °C for 45 s, primer annealing at 45 °C for 2 min, and the final elongation at 72 °C for 60s (Smith and Patton 1993;Bonvicino and Moreira 2001). The amplicons were sequenced at Macrogen Inc. in South Korea. The Cytb sequences were edited and assembled in the Geneious R11 program (https://www.geneious.com) and then verified to represent endogenous DNA of Neacomys by performing independent searches with the Basic Local Alignments Search Tool (BLAST) (Altschul et al. 1990).

Phylogenetic analyses
We tried to include representatives of the 23 known Neacomys species (Appendix 2), including some sequences from other genera of sigmodontine rodents that were used as outgroups (Appendix 2). The algorithm CLUSTAL-W was used for this purpose as implemented in Geneious R11. The ML tree was inferred using IQ-TREE (Nguyen et al. 2015). The BI analysis was conducted with MrBayes 3.2 (Ronquist et al. 2012), on the CIPRES Science Gateway platform (Miller et al. 2010), the analysis was carried out with two runs and four chains, were run for 10,000,000 generations, with a sampling every 1,000 generations and a burn-in of 0.25. Convergence was evaluated by the effective sample size (EES) and the potential scale reduction factor (PSRF). For most of the parameters the EES should be ≥ 200 and for the PSRF most of the values of the parameters should be between 1.0 and 1.2.

Genetic distances
We calculated an analysis of genetic divergence using an alignment restricted to the genus Neacomys obtained as described above. Uncorrected p-distances (intra and interspecific) were calculated with the MEGA X program (Kumar et al. 2018) and transformed to percentage values. The uncorrected p-distances were calculated in other works (Brito et al. 2021a;Colmenares-Pinzon;Semedo et al. 2020Semedo et al. , 2021.

Morphological qualitative and quantitative comparisons
Morphological qualitative revision and comparisons revealed that recently collected specimens and some older museum specimens from northwestern Ecuador are different from typical Neacomys tenuipes from Colombia in multiple discrete characters.
The two principal components of the PCA analysis explained 56.83% of the variation in the craniodental measurements, with CIL and RW-2 contributing to a greater extent to each one of them, respectively (Table 1). There was a clear overlap in the morphospace between recently collected samples from north- Older museum samples of N. cf. pictus, also from northwestern Ecuador, and typical samples of N. tenuipes from Colombia were recovered as two discrete groups (Fig. 1A). Likewise, typical N. tenuipes was completely discriminated in the KNN while recently collected samples (northwestern Ecuador) and N. rosalindae attained some degree of separation (Fig. 1B); the algorithm achieved accurate classification for samples from northwestern Ecuador, and N. rosalindae, with success rates of 91.3% and 95.8% respectively (Fig. 1B, Table 2). The characters chosen by the stepwise analysis as the greatest contributors to morphologic discrimination were OL (R 2 = 0.96; F = 1287.68; p < 1e-15) and LR (R 2 = 0.52; F = 55.21; p < 1e-15). The Kruskal-Wallis test revealed that the medians significantly differed across all groups (p < 1e-5), and the Dunn pairwise test proved that both characters were significantly different between all species with p-adjusted < 0.001 (Fig. 1C, D). These results constitute additional evidence supporting differentiation of the recently collected Ecuadorian specimens from typical specimens of N. tenuipes.

Genetic comparisons
Neacomys was recovered as a monophyletic group (BS: 100/ PP: 1.00; Fig. 2), with five nested subclades mostly congruent with the species groups mentioned by other authors (Hurtado and Pacheco 2017;Semedo et al. 2020;Brito Figure 1. Morphometric and statistical analyses A scatterplots of the Principal Components B the K neighbor discriminant analyses. Each taxon is enclosed by a convex hull, and color codes are detailed in the legend C, D the distribution of the data is shown in a violin boxplot; the median of each taxon character is indicated with a black dot. Only statistically significant differences among taxa are shown with the p-adjusted Holm method (* p < 1e-3, ** p < 1e-4, *** p < 1e-9). Table 2. Confusion matrix displaying the performance of the K nearest-neighbor classification for three Neacomys species. n represents the number of individuals used as input in the model. Values in "as species" columns represent the number of individuals assigned to each taxon, and the accuracy of the prediction is given as a percentage in the last column.  auriventer to our phylogenetic analyses demonstrated that these morphologically and ecologically similar species are closely related, thus forming the novel "serranensis" group ( Fig. 2). The ML analysis (Fig. 2) obtained the following relationship for the groups: "paracou" + ["spinosus" + {"serranensis" + ("dubosti" + "tenuipes")}]. Relationships between species groups and between species in these groups were mostly consistent with previous phylogenetic hypotheses (Colmenares-Pinzón 2021; Brito et al. 2021a). The samples identified as Neacomys tenuipes from Ecuador and Colombia were grouped into two sister clades (Fig. 2), each clade presents high support Ecuador (100/1.00) and Colombia (86/0.86). Calculated divergence between these two lineages was 4.35%±1.18% (Table 3), a value that is comparable with the divergences between well discrete species such as N. marajoara and N. xingu (4.0%), N. macedoruizi and N. aletheia (4.8%), N. vossi and N. xingu (5.4%), N. marajoara and N. vossi (5.5%), and N. macedoruizi and N. minutus (5.6%).
These results, along with those from the morphological qualitative and quantitative comparisons constitute strong evidence of cryptic diversity within N. tenuipes and that therefore, recently collected specimens from northwestern Ecuador (Chocó Biogeographic region) represent a species clearly distinct from Colombia. Accordingly, this new species is described as follows.  Table 4. Table 3. Uncorrected genetic distances of species of the genus Neacomys formally described (21 species). We calculated the genetic distance based on the Cytochrome b gene. The values to the right of the diagonal are the standard deviation.   Etymology. Named in honor of Marc Hoogeslag of Amsterdam, the Netherlands. He was co-founder and leader of the innovative Land Acquisition Fund of the International Union for the Conservation of Nature -Netherlands, which helps local groups throughout the world to establish new ecological reserves and conserve endangered species. Fundacion EcoMinga's Reserva Manduriacu, the habitat of this new species, is one of the many reserves which have benefited from Marc's program. The species epithet is formed from the surname "Marc" taken as a noun in the genitive case, adding the Latin suffix "i" (ICZN 31.1.2).

Taxonomy
Diagnosis. A species of Neacomys with the following combination of characters: small size (head-body length 65-85 mm), long tail (69-126% longer than head and body length), belly fur pale buff but with gray based hairs, white throat, long nasals (which extend well beyond the plane of the lacrimal), condylar process higher than coronoid process, M1 anterocone divided, M1 with broad protoflexus; m1-m3 with wide hypoflexids.
Morphological description. The following description was based on all specimens available. Neacomys marci sp. nov. is a spiny mouse of small size (head and body length 65-85 mm). The dorsal pelage is dark brown (Fig. 3); soft hairs are mixed with spines; on average dorsal hairs are 9-10 mm in length. The soft hair is tricolor, with a light brown band at the base, an orange band in the middle and a black apical band. The posterior mystacial vibrissae are thick and long (34 mm), surpassing the auricular pinnae when ad pressed back; two superciliary vibrissae, the longest measuring 39 mm, extending to the middle of the dorsum. One mediumsized genal vibrissae (32 mm) are also present, which are more slender than the mystacial vibrissae. The ears are large (12-16 mm) and oval in outline. Although the ears seem to be naked, they are covered with short black fringe of hair. The base of the internal ears is yellowish cream and the edges are dark, the hairs are yellowish and medium in size. A small pale orange postauricular patch is present.
The pelage on the throat is white (Fig. 4A) and extends up to the corners of the mouth. The ventral pelage is pale buff but with gray base, and the hairs are on average 3.0-3.5 mm in length at the middle of the belly. The tail is uniformly dark, slender, and long (69-126% longer than head and body length). It is covered with rectangular scales (13 or 14 rows/cm near the base), with three dark brown hispid hairs emerging from the base of each scale, not longer than 1.5-2 scale rows. The hairs of the terminal portion of the tail form a small tuft (< 3 mm). Females have eight mammae arranged in pectoral, thoracic, abdominal, and inguinal pairs.
The manus is slender and short. The first digit is reduced with a long and wide claw. The other claws are short and curved. Ungual tufts are white and extend beyond the claw ends. The dorsal surface with evident brown scales; each scale has three dark brown hairs and sometimes the central hair is the longest. Long carpal vibrissae can reach the claw of digit V. The digits are relatively large; digit I is substantially shorter than digit II; digit II is shorter than digit III; digit III is slightly larger than digit IV; digit IV is larger than digit V.
Hind feet are long and slender (18-22 mm); the ungual tufts are white, abundant and extend well beyond the edge of the claws (Fig. 5A, D). Their dorsal surface has a small metatarsal patch, with brown scales (Fig. 5D); each scale has three dark brown hairs. Large number of granules covers most of the plantar surface, including the spaces between the pads and reaching the anterior border of the thenar pad. The four interdigital pads are elevated and similar in size; pads II and III are separated by a small interspace, while pads II and IV are separated by an interspace of similar size than pad I (Fig. 5A). The hypothenar pad is very small or absent, while the thenar pad is well developed, large and elevated anteriorly. Digits are relatively short; digit I reaches the base of digit II; digit II is slightly shorter than digit III; digit III is slightly larger than digit IV; digit  IV is larger than digit V; digit V reaches halfway of the first phalanx of digit IV (Fig. 5A, D); claws are short, recurved and basally opened.
The cranium is moderately large for the genus (average CIL = 18.2 mm) with the braincase showing a convex profile (Fig. 6). The dorsal profile of the cranial roof is flat from the nasals to the middle of the frontals, then rises at the back of the frontals and slopes gently down the parietals toward the occiput; the rostrum is long and slender; premaxillae are slightly shorter than nasals, not extending anteriorly beyond incisors, without forming a rostral tube; gnathic process is very small; the suture between the nasal bones and the premaxillary reaches the root of the zygomatic bone; the nasal bone is wide at the base and gradually widens forward (Fig. 7); the interorbital region is narrow; the su- praorbital edges are small and sharp; the zygomatic notches are shallow and wide while seen from above; in the olfactory sagittal plane are two frontoturbinals, one interturbinal and three ethmoturbinals present (Fig. 8F); the lachrymal is small, with contact in equal proportions with the frontal and maxillary; the post-nasal depression is shallow; the fronto-parietal suture is V-shaped; the parietal is restricted to the dorsal portion of the skull; the braincase is rounded and inflated. A gnatic process is not developed; the zygomatic plate is wide and excavated (> M1 length) and slightly inclined backward; the zygomatic arch slender and without a jugal; a squamosal-alisphenoid groove is visible through the translucent braincase (Fig. 8B, E), with a perforation where it crosses the depression for the masticatory nerve; the stapedial foramen is present and small, the carotid canal is small, and the petrotympanic fissure is expressed (Figs. 8C); the cephalic arterial supply is primitive (pattern 1 of Voss 1988); the alisphenoid strut is absent; an anterior opening of the alisphenoid canal is absent; the postglenoid foramen is large; the subsquamosal fenestra is small and the hamular process of the squamosal is long; a small tegmen tympani is present (Fig. 8A); there is no contact between the anterodorsal edge of the ectotympanic and the mastoid tubercle, which leads to an opened ectotympanic ring (Fig. 8A); the orbicular apophysis of the malleus is wide and elongate (oval in shape), with its longitudinal axis inclined towards the manubrium; mastoid bears no dorsolateral fenestra; the paraoccipital process is short.  Abbreviations: ab, auditory bulla; bet, bony Eustachian tube; bmt, buccinators-masticatory trough; bo, basioccipital; bs, basisphenoid; cc, carotid canal; etI, ethmoturbinal I; etII, ethmoturbinal II; etIII, ethmoturbinal III; fo, foramen ovale; ft1, frontoturbinal 1; ft2, frontoturbinal 2; it, interturbinal; ls, lamina semicircularis; mas, mastoid capsule; mlf, middle lacerate foramen; palc, posterior opening of the alisphenoid canal; pet, petrosal; pgf, postglenoid foramen; ppp, posterior palatal pits; ps, presphenoid; sag, squamosal alisphenoid groove; sfr, sphenofrontal foramen; stf, stapedial foramen; ssf, subsquamosal fenestra; tt, tegmen tympani; Pictures are three-dimensional reconstructions based on micro-CT data.
The Hill foramen is tiny; the incisive foramina are short, ending well anterior to the M1s anterior faces; the capsular process of the premaxillary is well developed; the palate is wide and long with the anterior border of the mesopterygoid fossa not reaching M3s posterior faces; the palatal foramina are small; the posterolateral pits are long and paired, and located parallel to the anterior part of the mesopterygoid fossa; the mesopterygoid fossa is broad as the parapterygoid plates, with the anterior margin U-shaped (Fig. 8D); the shape of the pterygoid plate is not expanded, and has straight margin; the sphenopalatine vacuities are elongated and narrow, occupying the posterior part of the presphenoid area; the presphenoid is wide (Fig. 8D); the auditory bullae are small and flask-shaped; the Eustachian tube is short, wide and gradually constricted; the petrosals are well-exposed; the anterior bullae process is in contact with the posterior margin of the pterygoid plate (Fig. 8C); the basioccipital depressions are deep, forming an recognizable crest; the anterior border of the foramen magnum is narrow, with a conspicuous notch.
The mandible with masseteric crest in line with procingulum of m1; the coronoid process is small, slender, and bended backwards; the sigmoid notch is oval; the condylar process is large and robust; the capsular process is forming a rounded elevation that lies below the coronoid process; the angular notch is shallow, and the angular process is blunt. The incisors are opistodont, without grooves, and with yellowish enamel; the molars are brachydont and terraced (Fig. 9A); the main cusps of the upper and lower molars are opposed. The M1 is rounded in outline; the procingulum is narrower than the rest of the molar, with a rounded anteromedian fossette present; anterocone divided; the protoflexus is broad; the mesoflexus is small; the metaflexus is large and wide; the posteroloph is small. The M2 with indistinct protoflexus; the anteroloph is small; the mesoflexus is short and wide; the mesoloph is short; the mesofosette is rounded (Fig. 9A); the posteroloph is similar to M1. The M3 has a small paraflexus and indistinct hypoflexus. The upper molars have three roots each. The m1 is rectangular in outline; the procingulum is not divided into labial and lingual conulids; the protoflexid is short and wide; the hypoflexid is wide; the mesoflexid is large and wide; the mesolophid is large; the posteroflexid is short and broad; the mesofosette is large. The m2 is square in outline; the protoflexid is large and narrow; the hypoflexid is wide and inclined with direction towards the posteroflexid; the mesoflexid is short and wide; the mesolophid is short and wide; the mesofosette is very small. The m3 is anteriorly-posteriorly compressed, having a wide hypoflexid and a small anterolabial cingulum. The lower molars have two roots each.
The tuberculum of the first rib articulates with the transverse processes of the seventh cervical and the first thoracic vertebrae; the second thoracic vertebra has a differentially elongated neural spine; 19 thoracicolumbar vertebrae, the 16 th with moderately developed anapophyses; four sacrals; 33 or 34 caudals, with complete hemal arches in the second, third and fourth; 12 ribs.
The gall bladder is absent. The stomach is unilocular and hemiglandular; the cornified epithelium lines the corpus, while the glandular epithelium occupies the antrum and is slightly extended to the left of the esophageal opening; the bordering fold is notorious for being thick and long, surpassing the left level of the incisura angularis; the incisura angularis is moderately deep and the plica angularis is well expressed with a well-developed pars pyloricus (Fig. 10).
Comparisons with similar species. Neacomys marci sp. nov. differs from its sister species N. tenuipes mainly in ventral coloration, N. marci sp. nov. is pale buff with white throat, while N. tenuipes is completely white to pale orange (Fig. 4). Additionally, N. marci sp. nov. has a slight bicolor at the base tail, while N. tenuipes has a clear bicolor at the base. The condylar process in N. marci sp. nov. is higher than the coronoid process, while in N. tenuipes most are lower than the coronoid process, some are equal to the coronoid process. At the molar level, N. marci sp. nov. has a narrow anterocone of M1, while in N. tenuipes it is wide (Fig. 4). In N. marci sp. nov. the hypoflexus of M3 is indistinct or absent, while in N. tenuipes it is present and well evident.
Another species from the Chocó Biogeographic region with which Neacomys marci sp. nov. could be confused is N. pictus. Both species have white throats, however N. marci sp. nov. has pale buff ventral color and N. pictus is faintly plumbeous basally on the belly. Neacomys marci sp. nov. has the interorbital region (in ventral view) with developed ridges, projecting like ledges; whereas N. tenuipes it is hidden under the maxilla. The mastoid is ossified in N. marci sp. nov. while in N. tenuipes it is most perforated. Other comparisons are summarized in Table 5.
Distribution. Neacomys marci sp. nov. is known from six localities in the provinces of Carchi, Pichincha, and Esmeraldas, in northwestern Ecuador (Fig. 11). Natural history. The distributional range of the species is thus far limited to the Chocó Biogeographical region (Myers et al. 2000), where it occupies the lower subtropical and lower montane ecosystems (Ceron et al. 1999), in an altitudinal range from 450 to 1,630 m (Fig. 12). These forests are characterized by having a tree cover of approximately 30 m height. Most of the vegetation belongs to the families Araceae, Melastomataceae, Cyclanthaceae, Bromeliaceae, and to the ferns. Additionally, the following species of rodents and marsupials were recorded as living in sympatry: Melanomys caliginosus, Mindomys hammondi, Oecomys sp., Rhipidomys latimanus, Tanyuromys thomasleei, Pattonimus musseri, Sigmodontomys alfari, and Transandinomys bolivaris, the heteromyid Heteromys australis, the marsupials Chironectes minimus, Mamosops caucae, and Marmosa isthmica, and the squirrel Microsciurus mimulus.

Discussion
With the description of Neacomys marci sp. nov. the diversity of the genus reaches 24 formally recognized species, of which 14 (60%) have been described in the last five years (Hurtado and Pacheco 2017;Sánchez-Vendizú et al. 2018; Seme-  Semedo et al. 2021). Such dynamism has not been seen recently in the taxonomy of any other group of oryzomyine rodents and places Neacomys as the most diverse group within the tribe, and the third most diverse within the subfamily Sigmodontinae, only comparable to the genus Oligoryzomys (Hurtado 2021).
Results presented here confirm that comprehensive revisions of currently recognized species, i.e., by morphological and genetic characterizations, as well as collection of specimens in unexplored regions are fundamental to unveil cryptic diversity in groups of small mammals. Particularly for Neacomys, species once considered as homogeneous throughout a wide distribution, such as N. minutus, N. spinosus, or N. tenuipes have been split into multiple taxa. For N. tenuipes, Caccavo and Weksler (2021) recognized populations from Venezuela as different (N. leilae Caccavo & Weksler, 2021), whereas in this study, we found enough evidence to propose a separation of the populations from northwestern Ecuador into N. marci sp. nov. Likewise, other authors have noticed clear differences in other populations from the Chocó region in Colombia, whose genetic characterization is pending, to validate the name N. pusillus. On the other hand, it is worth mentioning that other specimens from northwestern Ecuador reviewed here and tentatively identified as N. cf. pictus (QCAZ 708 and MECN 3050), seem to differ notably from this species and from any other species of Neacomys. Further collections and the generation of genetic data from this population could result in the recognition of a new species, which ultimately demonstrates that the number of species within the genus will continue to increase. The rainforests of northwestern Ecuador have both high biodiversity and endemism due to the biogeographic influence of the Chocó and Andes Mountains (Myers et al. 2000). For example, a variety of oryzomyines of the genera Pattonimus, Sigmodontomys, Tanyuromys, Transandinomys, and "Handleyomys" Patton et al. 2015;Brito et al. 2020) are endemic to the Chocó forests. Despite this, our knowledge of the sigmodontine biodiversity of this hotspot is still incomplete.
It is important to mention that after more than two centuries of active research in mastozoology (Tirira 2014), intensive fieldwork was conducted in few places in Ecuador. Examples for those sites in the eastern Andes are Papallacta (Voss 2003), Guandera Biological Reserve (Lee et al. 2015), and Sangay Figure 11. Topographic map of northern South America. Sampling localities of three Neacomys species are shown with color codes described in the legend. Neacomys marci sp. nov. localities correspond to the Chocó biogeographic region in northwestern Ecuador (type locality is shown with black circle). National Park (Brito and Ojala-Barbour 2016), and in the western Andes are Cajas National Park (Barnett 1999 (Brito et al. , 2022a, and Lita (Curay et al. 2022). The interest in complementary biodiversity studies has led to the prioritization of intensive field work, using a variety of trapping techniques (e.g., live traps, spring traps, and pitfall traps), and has also triggered revisions of museum specimens. For example, in the last five years, these approaches have led to the description of at least 14 new sigmodontine: five Chilomys (see Brito et al, 2022a), three Thomasomys (see Brito et al. 2019;Brito et al. 2021b;Lee et al. 2022), one Tanyuromys (see Timm et al. 2018), one Ichthyomys (Fernández de Córdova et al. 2020), two Pattonimus , one Neacomys (Brito et al. 2021a), and one Mindomys (Brito et al. 2022b). This burgeoning richness will surely reorganize part of our understanding of Neotropical cricetids. This context highlights the urgency of establishing national and comprehensive inventory and collection programs, including sampling in previously studied areas as well as improving scholarly access to these resources.

Data availability
All of the data that support the findings of this study are available in the main text.