**1. Introduction**

Nematodes are hyper-diverse, abundant, and distributed worldwide [1]. Free-living species play critical ecological roles in benthic energy flow and contribute to the ecosystem by facilitating mineralization and nutrient cycling [2–7]. In the presence of high inputs of organic matter, their abundance increases, helping to regulate this resource. They are also a source of high-quality food for other animals [8–11]. Currently, about 20,000 nematode species—of which 6500 are marine benthic (= meiofaunal)—have been formally described [12,13], with estimates ranging between 0.1 and 100 million species [14]. The number of existing species is still uncertain because such estimates have been made at the local level, whereas little is known on a global scale. Gathering evidence of nematode diversity and distribution, increasing the record of marine nematodes species, especially in overlooked regions, is nowadays crucial [15].

Nematode taxonomy is an overlooked field of study in Mexico, with only about 119 genera and 183 species known for the country [16–22]. The majority of studies of marine nematodes in this region address ecological questions [17–21,23,24], whereas only five studies are focused on taxonomy [16,25–27]. For this reason, faunistic lists for nematodes in Mexico are at the family or genus level in most cases. The slow advance in the taxonomic knowledge of marine nematodes is due to technical difficulties. The identification of marine benthic nematodes is mostly based on morphological traits of male genital structures in mature individuals [28,29]. However, the occurrence of adult specimens is often rare. For this reason, nematode diversity is commonly disentangled to the family or genus level, especially in ecological studies. Morphology-based taxonomy is a time-consuming task that requires well-trained specialists who are becoming rare [30,31]. The use of morphological traits, in most cases, descriptive and potentially affected by convergent evolution and phenotypic plasticity, could also prevent an accurate quantification of the true nematode diversity [32–34]. Hence, there is an increasing need for methods that can rapidly and cost-effectively estimate nematode diversity in marine sediments. Molecular tools for taxonomic identification, delimitation of species, and an approach to the phylogeny hold the potential to overcome difficulties where morphological studies are painstakingly difficult and/or where the number of species far outweighs the availability of taxonomists. The identification of free-living marine nematodes is particularly difficult [35–39], and an integrated approach including genetic, morphological, and ideally, also ecological and behavioral data is needed [40].

Only one study conducted in Mexico (in Baja, CA, USA) [41] considered genetic tools to investigate the diversity of nematodes. Pereira and collaborators [41] revealed both a wide genetic diversity and geographic distribution of populations of *Mesacanthion* species. They used two molecular markers: 28S ribosomal rRNA gene and 18S, with 28S showing a better taxonomic resolution than 18S in delineating also cryptic species (similarly to [42]). The two markers did not show differences in the phylogenetic relationships among the investigated taxa, except for species of the genus *Rhabdodemania* [41].

In the Caribbean, two relevant studies have been conducted on the nematode fauna and none from Mexico [43,44]. In both studies, two molecular markers were considered: cytochrome oxidase subunit I (COI) and 18S ribosomal RNA gene (18S rRNA). Armenteros and collaborators [43] disentangled species of Desmodorid from Punta Francés, Cuba. They generated 34 sequences for COI across five genera and 27 sequences for 18S rRNA gene across six genera. Either marker could fully resolve the phylogenetic relationships of some lineages (i.e., within the subfamilies Desmodorinae and Spiriniinae). However, COI showed a better resolution than 18S among closely related and cryptic species. Macheriotou et al. [44] generated 18S and COI sequences from nematodes sampled in the equatorial North Pacific, Cuba, Italy (Panarea Island), Papua New Guinea, the Netherlands, Tunisia, and Vietnam. They generated 290 COI and 438 18S sequences; using reference databases for marine nematodes, they identified 39 OTUs (Operational Taxonomic Units with High Throughput Sequencing; HTS). Although the ribosomal marker outperformed the mitochondrial marker in terms of species and genus-level detections., they concluded that, for HTS technologies, it is urgent to continue creating high-quality taxon-specific reference sequence databases.

Free-living marine nematode species are poorly represented in public sequence databases. Limited availability of nematode reference sequences, especially from overlooked both localities and habitats such as the deep-sea, seagrass beds, and tropical coral reefs, hinders biogeographic patterns and characterization of ecosystems. Moreover, although DNA taxonomy is most successful when applied to fast-evolving genes such as the mitochondrial gene COI [42,45–48], genetic reference databases for nematodes mostly include nuclear markers such as 18S and 28S [36,43,48,49]. COI is poorly represented [44,47,50] because of the difficulty in amplifying this gene in a wide range of taxa within the phylum using 'universal' primers. Several studies regularly report low success in the amplification of COI and the necessity to design new specific primers to obtain a robust database [44,50,51]. The limited COI sequence datasets for marine nematodes prevent the establishment of an adequate understanding of intraspecific divergence.

The main objectives of this study are to (i) improve our knowledge of the geographic distribution of meiofaunal nematodes in the Mexican Caribbean; (ii) contribute with new COI sequences to the public genetic databases; and, (iii) apply different delimitation models to test the taxonomic resolution of COI in marine nematodes. Integrating different species delineation models should prevent biased conclusions and disclose patterns of diversity and distribution [52]. We aim to disentangle nematode diversity by applying Automatic Barcode Gap Discovery (ABGD) [53], Barcode Index Number system (BINs) [54], and Poisson Tree Processes model (PTP) [55] on COI sequences.

#### **2. Materials and Methods**
